Synthesis of both enantiomers of a P-chirogenic 1,2-bisphospholanoethane ligand via convergent routes and application to rhodium-catalyzed asymmetric hydrogenation of CI-1008 (pregabalin)

Article abstract of DOI:10.1021/ja034715o

Both enantiomers of a P-chirogenic 1,2-bisphospholanoethane ligand are synthesized via two convergent methods. The first method relies on the chiral alkylation of 1-((-)-menthoxy)phospholaneborane using a s-BuLi/(-)-sparteine derived chiral base. Only one enantiomer of the catalyst could be synthesized via this method because only one antipode of sparteine is available in nature. The second route relies on the combination of methylphosphine borane and a chiral 1,4-diol. Either enantiomer of the ligand can be synthesized via the second route from the appropriate enantiomer of the 1,4-diol. Asymmetric hydrogenation using catalyst precursor 36 on acetamidoacrylic acid derivatives provided modest to good enantioselectivity (77-95% ee) under low H₂ pressure (30 psi). Asymmetric hydrogenation of CI-1008 (pregabalin) precursors, 39 and 40, provided good enantioselectivities (92%) at high catalyst loading (1 mol%) and low pressure (30 psi). Enantiomeric excesses dropped sharply with catalyst loading at this pressure. Increasing the pressure of H₂ caused a significant increase in enantiomeric excess for low catalyst loading reactions. Several studies were undertaken to further investigate the enantioselectivity dependence on both pressure and catalyst loading.

Full text of DOI:10.1021/ja034715o

Published on Web 08/01/2003
Synthesis of Both Enantiomers of a P-Chirogenic
1,2-Bisphospholanoethane Ligand via Convergent Routes and
Application to Rhodium-Catalyzed Asymmetric Hydrogenation
of CI-1008 (Pregabalin)
Garrett Hoge
Contribution from Pfizer Global Research and DeVelopment, 2800 Plymouth Road,
Ann Arbor, Michigan 48105
Received February 17, 2003; E-mail: garrett.hoge@pfizer.com
Abstract: Both enantiomers of a P-chirogenic 1,2-bisphospholanoethane ligand are synthesized via two
convergent methods. The first method relies on the chiral alkylation of 1-((-)-menthoxy)phospholaneborane
using a s-BuLi/(-)-sparteine derived chiral base. Only one enantiomer of the catalyst could be synthesized
via this method because only one antipode of sparteine is available in nature. The second route relies on
the combination of methylphosphine borane and a chiral 1,4-diol. Either enantiomer of the ligand can be
synthesized via the second route from the appropriate enantiomer of the 1,4-diol. Asymmetric hydrogenation
using catalyst precursor 36 on acetamidoacrylic acid derivatives provided modest to good enantioselectivity
(77-95% ee) under low H2 pressure (30 psi). Asymmetric hydrogenation of CI-1008 (pregabalin) precursors,
39 and 40, provided good enantioselectivities (92%) at high catalyst loading (1 mol %) and low pressure
(30 psi). Enantiomeric excesses dropped sharply with catalyst loading at this pressure. Increasing the
pressure of H2 caused a significant increase in enantiomeric excess for low catalyst loading reactions.
Several studies were undertaken to further investigate the enantioselectivity dependence on both pressure
and catalyst loading.
Introduction
ligand produced low enantiomeric excesses (<20%) in the
rhodium-catalyzed asymmetric hydrogenation of N-acetamido-
The development of 1,2-bisphospholane ligands and their
application to the highly enantioselective asymmetric hydroge-
nation of olefin substrates has profoundly influenced the
direction of chiral ligand research during the past decade. Some
of the highest enantiomeric excesses in asymmetric hydrogena-
tion reported to date have been gleaned from 1,2-bisphospholane
catalysts on classes of substrates including N-acetamidoacrylic
acid derivatives, ene-amines, enol acylates, imines, and
itaconates.1-3 These ligands have also been useful for high
asymmetric induction in catalytic [4+2] cycloisomerization
reactions,4 copolymerization between alkenes and carbon mon-
oxide,5 [4+1] cycloaddition reactions,6 allylic substitution,7 and
hydroacylation reactions.8
cinnamic acid. The much-hailed BPE and Duphos ligands, 2
and 3, respectively, were reported in 1991.10,11 The overall
success and popularity of these ligands is anchored not only in
the fact that they gave unprecedented high enantiomeric excesses
(for 1991) for a variety of substrate classes in hydrogenations,
but also in their commercialization, marketing, and accessibility
through chemical supply houses.12 The success of these ligands
has spurred research into an array of novel bisphospholanes and
their four-membered analogues, bisphosphetanes, during the past
decade.
Groups have focused on the development of C2-symmetric
analogues with a variety of chiral arrays around the five-
membered phospholane rings, as well as one report of a
bisphospholane with chirality located on the ethane bridge
separating the two phosphorus atoms.13 Some of these ligands,
4-6, demonstrate unique chiral substituents at the 2 and 5
positions of the phospholanes.14,15 In the case of ligand 4, the
substituents render the ligand’s corresponding rhodium catalyst
soluble in water. Other groups have prepared phospholanes that
In 1987, the first chiral bisphospholanoethane, 3,4-dimethoxy-
1,2-bisphospholanoethane, 1, was reported (Figure 1).9 This
(1) Ohkuma, T.; Kitamura, M.; Noyori, R. In Catalytic Asymmetric Synthesis,
2nd ed.; Ojima, I., Ed.; Wiley-VCH: New York, 2000; pp 1-110.
(2) Brown, J. M. In ComprehensiVe Asymmetric Catalysis; Jacobsen, E. N.,
Pfalz, A., Yamamoto, H., Eds.; Springer-Verlag: Berlin, 1999; Vol. I, pp
121-182.
(3) Burk, M. J. Acc. Chem. Res. 2000, 33, 363.
(4) Gilbertson, S. R.; Hoge, G. S.; Genov, D. G. J. Org. Chem. 1998, 63,
10077.
(5) Jiang, Z.; Sen, A. J. Am. Chem. Soc. 1995, 117, 4455.
(6) Murakami, M.; Itami, K.; Ito, Y. J. Am. Chem. Soc. 1999, 121, 4130.
(7) Dierkes, P.; Ramdeehul, S.; Barloy, L.; De Cian, A.; Fischer, J.; Kamer,
P. C. J.; van Leeuwen, P. W. N. M.; Osborn, J. A. Angew. Chem., Int. Ed.
1998, 37, 3116.
(8) Barnhart, R. W.; McMorran, D. A.; Bosnich, B. Chem. Commun. 1997,
589.
(9) Brunner, H.; Net, G. Z. Naturforsch., B: Chem. Sci. 1996, 51, 1210.
(10) Burk, M. J. J. Am. Chem. Soc. 1991, 113, 8518.
(11) Burk, M. J.; Feaster, J. E.; Harlow, R. L. Tetrahedron: Asymmetry 1991,
2, 569.
(12) STREM Chemicals currently offers research quantities of the DuPhos and
BPE ligands and catalysts.
(13) Burk, M. J.; Pizzano, A.; Martin, J. A. Organometallics 2000, 19, 250.
(14) Holz, J.; Sturmer, R.; Schmidt, U.; Drexler, H.; Heller, D.; Krimmer, H.;
Borner, A. Eur. J. Org. Chem. 2001, 24, 4615.
(15) RajanBabu, T. V.; Yan, Y.; Shin, S. J. Am. Chem. Soc. 2001, 123, 10207.
9
10.1021/ja034715o CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 10219-10227
10219

A R T I C L E S
Hoge
Figure 2. Comparison of P-chirogenic versus non-P-chirogenic bisphos-
pholanes.
larly, phosphetanes are synthesized from chiral 1,3-diols.
Seemingly, endless arrays of novel C2-symmetric bisphos-
pholanes can be synthesized from unique 1,4-chiral diols.22
Despite the fact that the steric environments around a bound
transition metal of C2-symmetric P-chirogenic bisphospholane
ligands would mimic those of their non-P-chirogenic analogues
(given the cis orientation of substituents on the phospholane
ring with respect to the lone pair on the phosphorus), P-
chirogenic 1,2-bisphospholanes have not been reported. The
diagrams in Figure 2 depict the similarities between the steric
environments around the metal center of the BPE ligand (used
here as a generic non-P-chirogenic 1,2-bisphospholane ligand)
and a generic P-chirogenic 1,2-bisphospholane. The upper right
and lower left quadrants of each complex are similarly hindered.
The upper left and lower right quadrants of each complex are
unhindered. The corresponding enantiomers, of course, would
display the opposite array.
Figure 1. Bisphospholanes and bisphosphetanes.
The hindered quadrants of both ligand types exhibit similar
steric environments for identical R groups. However, the
P-chirogenic ligand’s unhindered quadrants are less sterically
hindered than the unhindered quadrants of BPE. Although the
R substituents of BPE in the upper left and lower right quadrants
are trans with respect to the metal, the carbon atom attached to
the alkyl substituent is tertiary. In the P-chirogenic complex,
secondary carbons occupy the nonsterically hindered quadrants.
This might be a structural advantage for higher enantioselectivity
in asymmetric catalysis over traditional bisphospholanes. Ad-
ditionally, P-chirogenic ligands also offer the potential advantage
of having chirality attached directly to the metal center instead
of chirality that is one atom removed as displayed by all of the
ligands shown in Figure 1.23
That P-chirogenic bisphospholanes have not been reported
is perhaps partially the result of the ease of synthesis of ligands
in Figure 1 (Scheme 1) in contrast to the potential synthetic
difficulty in preparing a P-chirogenic ligand. Although one of
the first chiral bisphosphine ligands was, in fact, P-chirogenic
(DiPamp),24 synthetic methodology to synthesize ligands con-
taining phosphorus chirality has not been extensively developed.
Noteworthy exceptions are Imamoto’s Bis-P ligand25 and
Zhang’s Tangphos (another class of P-chiral bisphospholane!),26
Scheme 1. Retrosynthetic Analysis of Bisphospholanes 1-15a
a Backbone ) 1,2-ethane, 1,2-benzene, 1,1′-ferrocene, or 2,3-ben-
zothiophene.
contain chirality at every carbon atom on the phospholane ring
such as the D-mannitol derived ligands, 7-10.16 Still others have
focused on changing the “backbone” of the bisphospholanes
from 1,2-ethane or 1,2-benzene to ferrocene, such as 11,17 or
even heterocycles, such as 12.18 Yet another pathway that groups
have taken is the synthesis of four-membered-ring phosphetanes
such as the ferrocenyl backbone ligand, 13,19 and ligands 14-
15.20,21
The ligands in Figure 1 have two commonalities: (1) They
were all synthesized from a bis-primary phosphine and C2-
symmetric chiral 1,4-diol precursors, and (2) they are all C2-
symmetric (except for 12). Scheme 1 depicts a retrosynthetic
analysis of the bisphospholanes shown in Figure 1. Typically,
a chiral 1,4-cyclic sulfate or bismesylate (synthesized from a
chiral 1,4-diol) is reacted with the phosphine anions of a 1,2-
bis-primary phosphine to form the phospholane ligands. Simi-
(22) This concept has been described in the patent literature. See: Berens, H.
International publication number: WO 99/24444, filed 05/11/1998.
(23) It must also be pointed out that the phosphorous atom is rendered chiral
because of the existence of stereogenic carbons in the phospholane ring.
Therefore, while the described ligands are P-chirogenic, the stereogenic
carbon atoms play the major role in defining the steric environment of the
ligands and their corresponding rhodium catalysts.
(24) Knowles, W. S. Angew. Chem., Int. Ed. 2002, 41, 1998.
(25) Imamoto, T.; Watanabe, J.; Wada, Y.; Masuda, H.; Yamada, H.; Tsuruta,
H.; Matsukawa, S.; Yamaguchi, K. J. Am. Chem. Soc. 1998, 120, 1635.
(16) Holz, J.; Quirmbach, M.; Schmidt, U.; Heller, D.; Sturmer, R.; Borner, A.
J. Org. Chem. 1998, 63, 8031.
(17) Chiral Quest ligand.
(18) Solvias ligand.
(19) Chirotech Ligand.
(20) Marinetti, A.; Kruger, V.; Buzin, F. Tetrahedron Lett. 1997, 38, 2947.
(21) Marinetti, A.; Jus, S.; Genet, J. Tetrahedron Lett. 1999, 8365.
9
10220 J. AM. CHEM. SOC. VOL. 125, NO. 34, 2003

P-Chirogenic 1,2-Bisphospholanoethane Ligands
A R T I C L E S
Scheme 3. Formation and Chiral Alkylation of
Menthoxyphospholane Boranea
Figure 3. P-Chirogenic 1,2-bisphospholanoethane, 19.
Scheme 2. Retrosynthetic Analysis of 19: Chiral Alkylation Route
both of which were synthesized via similar chiral deprotonation/
oxidative coupling strategies. Therefore, not only are P-
chirogenic 1,2-bisphospholanes interesting from a practical
standpoint (i.e., their use in catalysis), but they also represent a
worthy synthetic challenge.
The synthesis of both enantiomers of a P-chirogenic 1,2-
bisphospholanoethane ligand, 19 (Figure 3), is reported.27 Two
convergent synthetic strategies were developed to synthesize
the 4-chiral center containing ligand. In addition, the rhodium
complex of the ligand was prepared, and the results of catalytic
asymmetric hydrogenation on both acetamidoacrylic acid de-
rivatives as well as a substrate precursor to an antipsychotic
pharmaceutical, CI-1008, are reported. The enantioselectivity
dependence on both H2 pressure and catalyst loading during
the asymmetric hydrogenation of a pharmaceutical precursor
of CI-1008 is also reported.
a Reagents and conditions: (a) (1) s-BuLi, THF, -78 °C, (2) benzyl-
bromide; (b) (1) s-BuLi, (-)-sparteine, THF, -78 °C, (2) benzylbromide.
(Scheme 3). Successful selective alkylation at the R-carbons
on the five-membered ring of 1-alkoxyphospholane oxides has
been documented.36 Bias was recorded for cis-alkylation with
respect to the oxygen atom. Use of s-BuLi promoted deproto-
nation of 21 and, after alkylation with benzylbromide, produced
a 1:1 mixture of inseparable diastereomers, 23 and 24, in 94%
yield. The only observed alkylation occurred cis on the
phospholane ring with respect to the borane. In efforts to
increase the diastereomeric ratio of 23:24, we attempted
deprotonation of 21 with a chiral base generated from s-BuLi/
(-)-sparteine and then alkylation of the resulting anion with
benzylbromide. The ratio of 23:24 increased to 9:1, and the
reaction yielded 77%.37
Results and Discussion
P-Chirogenic Phospholane Synthesis: Chiral Alkylation
Route. Our synthesis revolved around assembly of a chiral
1-methylphospholaneborane, 20, and then coupling this phos-
pholane with itself to give 19 after borane deprotection (Scheme
2). There is some precedent for coupling methylphosphine
oxides, sulfides, and boranes,28-30 so we felt confident that 20
could be coupled and then deprotected to provide 19. The real
synthetic challenge for this synthesis is the construction of 20,
not 19 itself. However, very few chiral methylphospholanes have
been reported, and none of the reports were applicable to our
target.31-34
Nucleophilic displacement reactions are capable of inverting
or retaining phosphorus stereochemistry.38,39 The course of
reaction often depends on the structure of the molecule, and, in
fact, there is some precedent for retention to be observed in
nucleophilic displacements of phosphorus atoms in a five-
membered ring.40 Retention of stereochemistry at phosphorus
is important for our synthesis because inversion would place
the phospholane ring substituents on the opposite side of the
ring with respect to the lone pair, an undesirable stereochemical
conformation for our ligand synthesis. We treated the 9:1
mixture of 23 and 24 with MeLi to produce 20 in 53% yield
with complete retention of stereochemistry (Scheme 4).41
Synthesis began with treatment of (-)-menthoxyphosphorus
dichloride,35 22, with a bis-Grignard reagent generated from 1,4-
dibromobutane followed by complexation of the free phosphine
with borane-methyl sulfide complex to form compound 21
The methylphospholane borane, 20, was then treated with
s-BuLi to form the methyl anion and then oxidatively coupled
with CuCl2 to produce 25. Although the ee of 20 was only 77%,
(26) Tang, W.; Zhang, X. Angew. Chem., Int. Ed. 2002, 41, 1612.
(27) A similar P-chirogenic phospholane ligand was reported during the review
process of this paper. See: Shimizu, H.; Saito, S.; Kumobayashi, H. AdV.
Synth. Catal. 2003, 345, No. 1+2, 185.
(28) Imamoto, T.; Tsuruta, H.; Wada, Y.; Masuda, H.; Yamaguchi, K.
Tetrahedron Lett. 1995, 36, 8271.
(29) Corey, E. J.; Chen, Z.; Tanoury, G. J. J. Am. Chem. Soc. 1993, 115, 11000.
(30) Juge, S.; Genet, J. P. Tetrahedron Lett. 1989, 30, 2783.
(31) Hammond, P. J.; Lloyd, J. R.; Hall, C. D. Phosphorus Sulfur 1981, 10, 67.
(32) Beer, P. D.; Hammond, P. J.; Hall, C. D. Phosphorus Sulfur 1981, 10,
185.
(33) Borleske, S. G.; Quin, L. D. Phosphorus 1975, 5, 173.
(34) Quin, L. D.; Roser, C. E. J. Org. Chem. 1974, 39, 3423.
(35) Imamoto, T.; Yoshizawa, T.; Hirose, K.; Wada, Y.; Masuda, H.; Yamaguchi,
K.; Seki, H. Heteroat. Chem. 1995, 6, 99.
(36) Polniaszek, R. P.; Foster, A. L. J. Org. Chem. 1991, 56, 3137.
(37) Muci, A. R.; Campos, K. R.; Evans, D. A. J. Am. Chem. Soc. 1995, 117,
9075.
(38) Hall, C. R.; Inch, T. D. Tetrahedron 1980, 36, 2059.
(39) Wadsworth, W. S., Jr.; Larsen, S.; Horten, H. L. J. Org. Chem. 1973, 38,
256.
(40) Corriu, R.; Lanneau, G.; Leclercq, D. Tetrahedron 1986, 42, 5591.
(41) It is noteworthy that the 9:1 ratio of diastereomers 28 and 29 translates to
80% de. That compound 20 is produced in 77% ee is consistent with
complete retention of configuration during the displacement of the menthoxy
group with MeLi.
9
J. AM. CHEM. SOC. VOL. 125, NO. 34, 2003 10221

A R T I C L E S
Hoge
Scheme 4. Synthesis of P-Chirogenic Bisphospholane, 19, and Its
Rhodium Complex, 26a
Scheme 5. Synthesis of Catalyst 36 from a Chiral Epoxidea
a Reagents and conditions: (a) 5 equiv of MeLi, THF, 25 °C; (b) (1)
s-BuLi, THF, -78 °C, (2) CuCl2, (3) recrystallization from 2-propanol; (c)
(1) HBF4, CH2Cl2,
0
f
25 °C, (2) K2CO3; (d) [Rh-
a Reagents and conditions: (a) (1) diethylmalonate, NaOEt, EtOH/THF,
25 °C, (2) 130 °C, DMSO; (b) LAH, THF, 25 °C; (c) MsCl, Et3N, 0 °C;
(d) (1) 2 equiv of n-BuLi, THF, -78 °C, (2) 30, THF, -78 °C f 25 °C;
(e) (1) s-BuLi, THF, -78 °C, (2) CuCl2, (3) recrystallization from
2-propanol; (f) (1) HBF4, CH2Cl2, 0 f 25 °C, (2) K2CO3; (g) [Rh-
(NBD)2]+BF4-, THF/MeOH, -15 °C.
(NBD)2]+BF4-, THF/MeOH, -15 °C.
the coupling of 20 with itself resulted in an enantiomeric
amplification from 77% ee to 96% ee (meso ligand was the
side product). After one recrystallization from 2-propanol, 25
was isolated in 41% yield with enantiomeric excess >99%. The
borane groups were removed via treatment of 25 with HBF4
followed by hydrolysis with K2CO3 to form ligand 19 in 89%
yield. Ligand 19 was immediately complexed to rhodium (due
to O2 sensitivity) to produce catalyst precursor 26 in 80% yield.
equiv of n-BuLi which produced the phosphorus dianion
followed by the addition of 30 to the anion produced a 1:1
diastereomeric mixture of 32 and 33. These diastereomers were
separated via meticulous column chromatography over silica
gel to produce the desired phospholane, 32, in 38% yield (of
the expected yield). After separation, 32 was easily converted
to 36 via previously described chemistry. Furthermore, either
enantiomer of the catalyst can be synthesized via this route
depending upon which enantiomer of chiral epoxide is used as
starting material.44
P-Chirogenic Phospholane Synthesis: Chiral 1,4-Diol
Route. The unfortunate reality of all syntheses utilizing (-)-
sparteine derived chiral bases is that only one antipode of
sparteine is available in nature. Therefore, the (S,S) enantiomer
of catalyst 26 cannot be synthesized via the original route. An
alternative synthesis was needed to synthesize the (S,S) enan-
tiomer. We chose a convergent route to the opposite enantiomer
of 20, the key intermediate in the chiral alkylation route.
Epoxide 27, generated from the hydrolytic kinetic resolution42
of its corresponding racemate, was treated with diethylmalonate
and K2CO3 followed by decarboxylation to form 28 (Scheme
5). The resulting lactone was reduced with LAH to produce 29
which was then converted to bis-mesylate 30. Methylphosphine
is facile to generate, but the malodorous gas is difficult to handle.
An alternative is the use of methylphosphine borane, a
compound recently reported to be stable, isolatable, and capable
of forming phosphorus anions that can be alkylated with alkyl
halides.43 Treatment of methylphosphine borane, 31, with 2
X-ray Crystal Structure of Rhodium Complex 36. Figure
4 shows the ORTEP diagram generated from the X-ray crystal
structure of 36. Both norbornadiene and BF4- counteranion were
removed for clarity. The diagram is positioned so that the
rhodium atom is in front of the ethane bridge. This view clearly
displays the C2-symmetric binding environment for the substrate
around the rhodium atom. The positions of the benzyl groups
in the crystalline form give the molecule a propeller-like
appearance. Hydrogen atoms are displayed on the stereogenic
carbons to demonstrate that the stereochemical configuration
(43) Bourumeau, K.; Gaumont, A.; Denis, J. J. Organomet. Chem. 1997, 529,
205.
1
(44) We have also succeeded in synthesizing the racemic chiral diol on
/
2
kilogram scale via this route and then separating the enantiomers via SMB
chromatography.
(42) Jacobsen, E. N. Acc. Chem. Res. 2000, 33, 421.
9
10222 J. AM. CHEM. SOC. VOL. 125, NO. 34, 2003

P-Chirogenic 1,2-Bisphospholanoethane Ligands
A R T I C L E S
Table 2. Pressure Effects versus Enantioselectivity on CI-1008
Precursors, 39 and 40, Using Catalyst Precursor 36a
+
c
entry
X
mol % catalystb
H pressure (psi)
2
ee (%)
+
+
1
2
3
4
5
6
7
8
9
t-BuNH3
1
1
30
30
30
92 (S)
92 (S)
47 (S)
69 (S)
90 (S)
97 (S)
97 (S)
97 (S)
71 (S)
91 (S)
96 (S)
K+
t-BuNH3+
t-BuNH3+
t-BuNH3+
t-BuNH3+
t-BuNH3+
t-BuNH3
K+
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
100
250
400
500
750
100
250
400
10
11
K+
K+
-
Figure 4. ORTEP diagram of 36 (BF4 anion and norbornadiene are
removed for clarity).
a All reactions were performed at room temperature with a substrate
concentration of 0.3 M. All reactions afforded conversion to a single product.
b All reactions performed with 1 mol % catalyst used 1 mmol of substrate
and were complete within 1 h. All reactions performed with 0.1 mol %
catalyst used 3 mmol of substrate and were complete within 4 h.
c Enantiomeric excesses were determined by chiral GC as described in the
Supporting Information.
Table 1. Hydrogenation of Acetamidoacrylic Acid Derivatives
Using Catalyst Precursor 36a
loading decreased the enantioselectivities drastically (entries
8-10). This pressure effect is in accord with observations and
predictions based on complex equilibria in asymmetric hydro-
genations using bisphosphine ligands.46,47
1
2
c
entry
R
R
mol % catalystb
H pressure (psi)
ee (%)
2
1
2
3
4
5
6
7
8
9
Me
H
Me
H
Me
Me
H
Me
H
H
H
1
1
1
1
0.1
0.1
0.1
0.1
0.1
0.1
30
30
30
30
30
30
30
400
400
400
95 (R)
86 (R)
84 (R)
77 (R)
93 (R)
80 (R)
77 (R)
68 (R)
26 (R)
37 (R)
Rh-Catalyzed Asymmetric Hydrogenation of a Substrate
Precursor to CI-1008. Catalyst precursor, 36, was then used
in the asymmetric hydrogenation of two substrate precursors
to CI-1008 (pregabalin), 39 and 40. CI-1008 is a pharmaceutical
used to treat psychotic disorders, seizure disorders, and pain.48
Under mild conditions (30 psi H2, 25 °C), substrate 39 was
converted to product 41 in 92% enantiomeric excess with
complete conversion (Table 2, entry 1). Results using substrate
40 under the same conditions provided 42 with similar selectivity
(Table 2, entry 2).
Enantioselectivity Dependence on Catalyst Loading. In-
terestingly, when the catalyst loading was lowered to 0.1 mol
% for the hydrogenation of the CI-1008 substrate, 39, and the
pressure of the reaction was kept constant at 30 psi H2, the
enantiomeric excess of the product dropped significantly (Table
2, entry 3). This is in direct contrast to the acetamidoacrylic
acid substrate results that were detailed in Table 1 where it was
demonstrated that the enantiomeric excesses remained essentially
the same at lower catalyst loadings. This inconsistency suggests
that during the progression of hydrogenating 39 in entry 3
catalytically active species are being formed that produce low
enantiomeric excesses of 41. One possible explanation for this
phenomenon is that the nitrile functionality of the substrate or
product might complex with the active catalyst as the hydro-
genation progresses.49 This could alter the highly ordered C2-
symmetric steric environment of the active catalyst, thus
preventing the ordered binding of substrate molecules for
hydrogenation that is necessary for high enantiomeric excesses.
Ph
Ph
H
Ph
Ph
H
H
Ph
10
H
a All reactions were performed at room temperature with a substrate
concentration of 0.3 M. All reactions afforded conversion to a single product.
b All reactions performed with 1 mol % catalyst used 1 mmol of substrate
and were complete within 1 h. All reactions performed with 0.1 mol %
catalyst used 3 mmol of substrate and were complete within 4 h.
c Enantiomeric excesses were determined by chiral HPLC or GC as
described in the Supporting Information.
of 36 was as expected. The stereochemistry of the phosphorus
atoms is also confirmed by observing that the benzyl groups
on the phospholane were positioned cis with respect to the
coordinated metal.
Rh-Catalyzed Asymmetric Hydrogenation of Acetami-
doacrylic Acid Derivatives. Catalyst precursor, 36, was then
screened for the catalytic asymmetric hydrogenation of classic
acetamidoacrylic acid substrates, 37 (Table 1). We found
selectivity to be modest (77-95% ee) at 1 mol % catalyst
loading and 30 psi H2 (entries 1-4). However, these enantio-
selectivities were similar to those gleaned from BPE ligands,
2, on the same substrates.45 Lowering the catalyst loading to
0.1 mol % at 30 psi H2 showed a slight decrease in enantio-
selectivity for these substrates (entries 5-7). Finally, we found
that increasing the pressure to 400 psi at 0.1 mol % catalyst
(46) Brown, J. M.; Chaloner, P. A. J. Chem. Soc., Chem. Commun. 1980, 344.
(47) Halpern, J. Science 1982, 217, 401.
(48) Burk, M. J.; Goel, O. P.; Hoekstra, M. S.; Mich, T. F.; Mulhern, T. A.;
Ramsden, J. A. PCT Int. Appl., 2001.
(45) Murk, M. J.; Feaster, J. E.; Nugent, W. A.; Harlow, R. L. J. Am. Chem.
Soc. 1993, 115, 10125.
9
J. AM. CHEM. SOC. VOL. 125, NO. 34, 2003 10223

A R T I C L E S
Hoge
Table 3. Enantioselectivity versus Time during Asymmetric
Enantioselectivity Dependence on H2 Pressure. We next
investigated the dependence of enantioselectivity on H2 pressure
during the hydrogenation of 39 and 40 with catalyst precursor
36 (Table 2, entries 4-11). At 0.1 mol % catalyst loading, we
observed that the enantiomeric excess of product 41 dramatically
increases as the H2 pressure is increased. When the reaction
was performed at 100 psi H2, the enantiomeric excess of the
product at 100% conversion increased to 69% (entry 4). In entry
5, the reaction was performed at 250 psi H2, and the enantio-
meric excess increased to 90%. Further increase in pressure to
400 psi (entry 6) produced 97% enantiomeric excess. Reactions
performed at increased pressures of 500 and 750 psi H2, entries
7 and 8, respectively, showed no further increase in enantiomeric
excess of product 41. The potassium salt substrate, 40, followed
the same trend as that of the tert-butylammonium salt substrate,
39 (entries 9-11).50
Evidence of enantiomeric excess dependence on pressure has
been reported.51 Our results corroborate the findings that an
increase in hydrogen concentration in solution (caused by an
increase in hydrogen pressure) can increase the enantiomeric
excess of some asymmetric hydrogenation reactions. There are
also reports of enantiomeric excess dependence on catalyst
loading.52,53 These reports demonstrate that lower catalyst
loadings lead to lower enantiomeric excesses of products for
their catalyst systems. Our results also corroborate these reports.
However, we have demonstrated that asymmetric hydrogenation
enantiomeric excesses can change as a function of both catalyst
loading and H2 pressure.
Hydrogenation of CI-1008, 39, Using Catalyst Precursor 36 at Low
Catalyst Loadinga
entry
time (h)
conversion (%)
eeb (%)
1
2
3
4
5
6
0.5
1
2
4
8
3
10
22
43
68
88 (S)
47 (S)
42 (S)
35 (S)
27 (S)
18 (S)
16
100
a The reaction was performed at room temperature using 3 mmol of
substrate with a concentration of 0.3 M in MeOH. The reaction afforded
conversion to a single product. b Enantiomeric excesses were determined
by chiral GC as described in the Supporting Information.
Scheme 6. Pressure Increase during Hydrogenation of 39 with 36
Increases Selectivity for the Remainder of the Reactiona
Enantiomeric Excess Monitoring as a Function of Reac-
tion Completion. We monitored the enantiomeric excess of
product 41 as a function of reaction completion during a low
catalyst loading hydrogenation of 39 (Table 3). The enantiomeric
excess was 88% after 30 min reaction time and then quickly
declined to less than 50% ee after 1 h (entry 2). After complete
conversion of 39 to 41, the enantiomeric excess of product 41
was less than 20% (entry 6). Given the data from Tables 2 and
3, it is apparent that increases in pressure (400 psi or greater)
are capable of preventing the catalytically active species leading
to low enantiomeric excesses from forming, or, if they have
formed during the course of the reaction, the increases in
pressure prevent them from catalytically converting substrate
to product.
a The reaction was performed at room temperature using 1 mmol of
substrate with a concentration of 0.3 M in MeOH. The reaction afforded
conversion to a single product with predominantly an S configuration.
Enantiomeric excesses were determined by chiral GC as described in the
Supporting Information.
Enantioselectivity Dependence on Increasing Pressure
during Hydrogenation. We investigated the effect of a
significant increase in H2 pressure on enantioselectivity during
a low pressure and low catalyst loading hydrogenation of 39
with catalyst 36 (Scheme 6). In step 1 of the experiment, the
reaction was begun at 30 psi H2 pressure and 0.1 mol % catalyst
loading. A sample at the 2 h mark demonstrated, not unexpect-
edly, that the ee was 67% after 22% conversion to product.
The reaction mixture was then pressurized to 400 psi H2 in step
2 of the experiment and stirred for an additional 4 h. The
enantiomeric excess at 100% conversion was 88% (step 1 +
step 2). By calculation, the enantiomeric excess of the hydro-
genation of 78% of the remaining substrate (step 2 only) was
94%, a result consistent with the high enantiomeric excesses at
increased H2 pressures recorded in Table 2.
(49) It has been speculated that substrate 39 can bind with a rhodium complex
of Me-DuPhos. See: Burk, M. J.; de Koning, P. D.; Grote, T. M.; Hoekstra,
M. S.; Hoge, G.; Jennings, R. A.; Kissel, W. S.; Le, T. V.; Lennon, I. C.;
Mulhern, T. A.; Ramsden, J. A.; Wade, R. A. J. Org. Chem. 2003, 68,
5731.
(50) It should be noted that substrates 39 and 40 are 1:5.7 E/Z mixtures and
that the reaction rates and enantiomeric excesses observed at different
reaction pressures might be different for each diastereomer. Pure E substrate
could not be isolated for these comparisons, but we have observed that
pure Z substrate follows the same enantioselectivity trends versus pressure
as the 1:5.7 E/Z mixture.
Summary and Conclusions
Both enantiomers of a new addition to the chiral phospholane
family, a P-chirogenic 1,2-bispholanoethane, have been suc-
cessfully synthesized. Two convergent methods highlight solu-
tions to the synthesis of this P-chirogenic ligand.
While the enantiomeric excesses observed in the hydrogena-
tion of acetamidoacrylic acid derivatives with catalyst 36 were
(51) Yongkui, S.; Landau, R. N.; Wang, J.; LeBlond, C.; Blackmond, D. G. J.
Am. Chem. Soc. 1996, 118, 1348.
(52) Bell, D.; Davies, M. R.; Finney, F.; Geen, G. R.; Kincey, P. M.; Mann, I.
S. Tetrahedron Lett. 1996, 37, 3895.
(53) Gao, Y.; Hanson, R. M.; Klunder, J. M.; Ko, S. Y.; Masamune, H.;
Sharpless, K. B. J. Am. Chem. Soc. 1987, 109, 5765.
9
10224 J. AM. CHEM. SOC. VOL. 125, NO. 34, 2003

P-Chirogenic 1,2-Bisphospholanoethane Ligands
A R T I C L E S
1.36-1.48 (m, 1H), 1.57-1.64 (m, 2H), 1.89-1.96 (m, 9H), 2.05-
2.08 (m, 1H), 3.90-3.99 (m, 1H); 13C NMR (100 MHz, CDCl3) δ
16.0, 20.9, 22.1, 23.0, 25.4, 25.80, 25.84, 29.1 (d, JC-P ) 42.0 Hz),
30.5 (d, JC-P ) 37.4 Hz), 31.3, 34.2, 43.4, 48.6, 79.0 (d, JC-P ) 3.8
Hz); 31P NMR (162 MHz, CDCl3) δ 142 (m). Anal. Calcd for C14H30-
BOP: C, 65.64; H, 11.80. Found: C, 66.24; H, 10.55.
modest, excellent enantioselectivity was observed for the
asymmetric hydrogenation of CI-1008 precursors, 39 and 40.
An interesting combination of pressure and catalyst loading
effects on enantioselectivity was also observed during hydro-
genation studies on substrates 39 and 40.
The methodology described in this Article is ideal for the
synthesis of a series of P-chirogenic bisphospholanes with
varying alkyl substituents. We are currently investigating the
synthesis of derivatives of 19 and attempting to synthesize
P-chirogenic phospholanes with a variety of “backbones”.
Furthermore, we are examining the hydrogenation of 39 and
40 with catalyst 36 in more detail to elucidate the mechanism
of observed pressure and catalyst loading effects on enantio-
selectivity.
Preparation of (1S,2R)-1-((-)-Menthoxy)-2-benzylphospholane-
borane (23). Into a 500 mL round-bottom flask was dissolved 21 (8.83
g, 0.0345 mol) in 150 mL of Et2O. The solution was placed under N2
and then cooled to -78 °C. In a separate 250 mL flask was dissolved
(-)-sparteine (9.9 mL, 0.043 mol) in 100 mL of Et2O. The sparteine
solution was placed under N2 and then cooled to -78 °C. To the
sparteine solution was added s-BuLi (33.2 mL of 1.3 M in cyclohexane
solution reagent, 0.043 mol) via syringe. The s-BuLi/sparteine solution
was stirred for 30 min at -78 °C and then delivered to the flask
containing 21 via cannula over 30 min. After the addition, the reaction
was stirred for 2 h at -78 °C. Into a separate 100 mL round-bottom
flask was dissolved benzyl bromide (5.3 mL, 0.0449 mol) in 50 mL of
Et2O under N2. The benzyl bromide solution was delivered quickly to
the anion solution via cannula. The cold bath was removed, and the
reaction was allowed to warm slowly to room temperature. The reaction
was quenched with 500 mL of 1 N HCl, the organic layer was separated,
and then the aqueous layer was extracted with 2 × 75 mL of Et2O.
The combined organic layers were dried over MgSO4, and then the
solvent was removed on a rotary evaporator. The crude product was
flash chromatographed over silica gel (1.5% ethyl acetate in hexanes)
providing the product (9.14 g, 77%) as a 9:1 mixture of 23 and 24,
respectively.
23: 1H NMR (400 MHz, CDCl3) δ 0.2-1.1 (m, 14H), 1.2-1.4 (m,
3H), 1.3-1.5 (m, 1H), 1.7-2.0 (m, 8H), 2.18-2.30 (m, 1H), 2.20-
2.25 (m, 1H), 2.55-2.61 (m, 1H), 3.15-3.21 (m, 1H), 3.96-4.02 (m,
1H), 7.15 (m, 3H), 7.24-7.29 (m, 2H); 13C NMR (100 MHz, CDCl3)
δ 16.2, 20.9, 22.1, 23.1, 23.7, 26.0, 30.3, 31.2, 31.3, 34.2, 34.9, 43.4,
44.3, 48.7, 79.0 (d, JC-P ) 4.6 Hz), 126.2, 128.5, 128.7, 140.7 (d, 14.5
Hz); 31P NMR (162 MHz, CDCl3) δ 143.5 (m). HRMS (EI): (M +
Na)+ 369.2251 ((M + Na)+, exact mass calcd for C21H36BNaOP:
369.2494).
Experimental Section
General. THF was either distilled from sodium prior to use or
obtained from Aldrich Sure-Seal bottles supplied by Aldrich Chemical
Co. as 99.9% anhydrous. Dichloromethane (anhydrous, 99.8%) and
ether (anhydrous, 99.8%) were used as needed from Aldrich Sure-Seal
bottles supplied by Aldrich Chemical Co. (1R,2S,5R)-(-)-Menthol,
borane methyl sulfide complex (approximately 10-10.2 M), phosphorus
trichloride (98%), s-BuLi (1.3 M in cyclohexane), n-BuLi (2.5 M in
hexanes), MeLi (1.0 M in THF/cumene), (-)-sparteine, benzyl bromide
(98%), tetrafluoroboric acid-dimethyl ether complex, trimethylsilyl-
diazomethane, sodium metal (stick, dry, 99%), diethylmalonate (99%),
lithium aluminum hydride (powder, 95%), methanesulfonyl chloride
(99.5+%), triethylamine (99.5%), methyl 2-acetamidoacrylate, 2-ac-
etamidoacrylic acid, and R-acetamidocinnamic acid were obtained from
Aldrich Chemical Co. AgBF4 (99%) and chloronorbornadiene rhodium-
(I) dimer (99%) were supplied by Strem Chemicals, Inc. Hydrogen
gas (99.995%) was used from a lecture bottle supplied by Specialty
Gas. (-)-Menthoxyphosphorus dichloride was synthesized according
to literature procedures.35 (S)-(2,3-Epoxypropyl)benzene (99.9% chemi-
cal purity, 98.2% enantiomeric excess) was purchased from Rhodia-
Chirex on a custom synthesis contract. Methylphosphine borane was
synthesized according to literature procedures.41 Hydrogenations were
performed in a Griffin-Worden pressure vessel supplied by Kimble/
Kontes.
Preparation of 1-((-)-Menthoxy)phospholaneborane (21). Into
a three-neck 250 mL round-bottom flask equipped with a reflux
condenser under N2 was placed 150 mL of freshly distilled THF and
magnesium turnings (2.36 g, 0.0973 mol). An iodine crystal was added,
and then 1,4-dibromobutane (4.65 mL, 0.0389 mol) was added dropwise
over 30 min with a syringe pump while the reaction was stirred with
a magnetic stir bar. The reaction became hot during the addition and
refluxed near the end of the addition (an ice bath was kept on hand to
cool the reaction to prevent it from becoming uncontrollable). After
the addition, the reaction was refluxed for 1 h. The Grignard solution
(dark gray after reflux) was then cooled to room temperature. Into a
separate 500 mL flask were placed 22 (10.0 g, 0.0389 mol) and 250
mL of freshly distilled THF under N2. The solution of 22 was cooled
to 0 °C, and then the Grignard solution was delivered to the flask
quickly via cannula. The reaction mixture was warmed to room
temperature and then stirred 3 h whereupon BH3‚Me2S (3.9 mL of a
10.0 M solution, 0.0389 mol) was delivered to the reaction via syringe
followed by stirring overnight. The reaction was quenched cautiously
with 250 mL of H2O, the organic layer was separated, and the aqueous
layer was extracted with 3 × 100 mL of Et2O. The combined organics
were dried over MgSO4, and then the solvent was removed on a rotary
evaporator. Flash column chromatography over silica gel (1% ethyl
acetate in hexanes) provided the product (5.13 g, 52%). bp 146.9 °C;
[R]24D ) -55.6° (c 1.0, MeOH); 1H NMR (400 MHz, CDCl3) δ 0.1-
1.0 (br m, 3H), 0.78 (d, J ) 7.1 Hz, 3H), 0.87 (d, J ) 5.4 Hz, 3H),
0.88 (d, J ) 4.6 Hz, 3H), 0.91-1.04 (m, 2H), 1.17-1.24 (m, 2H),
Preparation of (1S,2R)-1-Methyl-2-benzylphospholaneborane
(20). A 9:1 mixture of 23 and 24 (9.14 g, 44.37 mmol), respectively,
was placed in a 250 mL round-bottom flask and dissolved in 120 mL
of THF under N2. The solution was warmed to 50 °C. To the solution
was added MeLi (1.0 M, 92.4 mL, 92.4 mmol) via syringe. The solution
became yellow after addition, and after being stirred for 1 h the solution
was light red. The reaction was quenched carefully with 300 mL of 1
N HCl, and then the organic layer was separated. The aqueous layer
was extracted with 2 × 100 mL of Et2O. The combined organic layers
were dried over MgSO4, and then the solvent was removed on a rotary
evaporator. Flash column chromatography of the crude product over
silica gel (2.5% ethyl acetate in hexanes) provided the product (2.83
g, 53%). The product was determined to be 77% ee by HPLC analysis
[Chiracel OD-H, 20% 2-propanol in hexanes, 1 mL/min, 214 nm UV
detection, enantiomers eluted at 5.29 min (1R,2S, minor) and 5.59 min
(1S,2R, major)]. The assignment of the relative stereochemistry was
accomplished by NOE difference 1H NMR spectra (see Supporting
Information). bp 167.6 °C; 1H NMR (400 MHz, CDCl3) δ 0.05-0.97
(br m, 3H), 1.25 (d, J ) 10.7 Hz), 1.42-1.70 (m, 2H), 1.70-2.20 (m,
5H), 2.64-2.73 (m, 1H), 3.06-3.13 (m, 1H), 7.17-7.29 (m, 5H); 13
C
NMR (100 MHz, CDCl3) δ 11.8 (d, JC-P ) 31.3 Hz), 24.4, 25.8 (d,
JC-P ) 37.4 Hz), 25.8, 33.2, 35.4, 41.0 (d, J ) 33.6 Hz), 41.0, 126.3,
128.5, 128.8, 140.5 (d, JC-P ) 11.5 Hz); 31P (162 MHz, CDCl3) δ
29.4 (m). HRMS (EI): (MH - BH3)+ 193.0980 ((MH - BH3)+, exact
mass calcd for C21H36BNaOP: 193.1146).
Preparation of 1,2-Bis((1R,2R)-2-(benzylphospholaneborane))-
ethane (25). To a 250 mL round-bottom flask was dissolved 20 (2.83
g, 13.74 mmol) in 80 mL of THF. The solution was placed under N2
and cooled to -78 °C. To this solution was added s-BuLi (1.3 M, 11.6
9
J. AM. CHEM. SOC. VOL. 125, NO. 34, 2003 10225

A R T I C L E S
Hoge
mL, 15.1 mmol) via syringe, and the solution turned red. After the
mixture was stirred for 2 h at -78 °C, CuCl2 powder was added to the
reaction in one portion with vigorous stirring. The reaction was allowed
to warm to room temperature, and then it was stirred overnight. The
mixture was quenched with 100 mL of concentrated NH4OH, and the
organic layer was separated. The aqueous layer was then extracted with
3 × 50 mL of ethyl acetate. The combined organic layers were washed
successively with 5% NH4OH, 1 N HCl, and brine. The organic layer
was dried over MgSO4, and the solvent was removed on a rotary
evaporator. The crude product was from hot 2-propanol to provide the
product (1.15 g, 41%). The product was determined to be >99% ee
and >98% purity by HPLC analysis [Chiracel OD-H, 20% 2-propanol
in hexanes, 0.5 mL/min, 214 nm UV detection, enantiomers eluted at
12.81 min ((1S,2S), minor) and 16.84 min ((1R,2R), major), meso
complex eluted at 14.51 min]. mp ) 128.8 °C; (R)25D -13.7° (c 0.95,
then washed successively with 1 N HCl, saturated NaHCO3, brine, and
deionized water. The organic layer was dried over MgSO4. The volatiles
were removed in vacuo yielding 13.6 g of a yellow oil. The diol was
then distilled at 178 °C (8 mmHg) to provide the product (10.1 g, 61%).
The product was determined to be 95% ee via chiral HPLC analysis
[Chiralcel OD-H, 20% 2-propanol in hexane, 1.0 mL/min, 214 nm UV
detection, enantiomers eluting at 4.78 min (S, minor) and 5.20 min (R,
1
major)]. bp 178 °C (8 mmHg). H NMR (400 MHz, CDCl3) δ 1.51-
1.58 (m, 1H), 1.69-1.77 (m, 3H), 2.12 (br s, 2H), 2.70 (dd, J ) 13.4
Hz, J ) 8.5 Hz, 1H), 2.82 (dd, J ) 13.7 Hz, J ) 4.4 Hz, 1H), 3.63-
3.72 (m, 2H), 3.83-3.89 (m, 1H), 7.20-7.33 (m, 5H); 13C NMR (100
MHz, CDCl3) δ 29.7, 34.0, 44.2, 63.2, 72.8, 126.8, 129.0, 129.7, 138.5.
HRMS (EI): (M + Na)+ 203.0855 ((M + Na)+, exact mass calcd for
C11H16NaO2: 203.1048).
Preparation of (R)-1-Phenyl-2,5-pentane-bis-mesylate (30). The
chiral diol, 29 (10.1 g, 0.061 mol), was dissolved in 300 mL of CH2-
Cl2 and placed in a 1 L round-bottom flask equipped with a pressure
equalizing dropping funnel. The flask was purged with nitrogen, and
then the solution was cooled to 0 °C using an ice bath. To the solution
was added triethylamine (21.3 mL, 0.15 mol) via syringe. Into the
dropping funnel was placed methanesulfonyl chloride (10.4 mL, 0.135
mol) dissolved in 50 mL of CH2Cl2. The methanesulfonyl chloride
solution was then delivered to the reaction over a period of 30 min.
After the addition, the reaction was stirred 30 min at 0 °C and then
warmed to room temperature and stirred for 4 h. The reaction was
cooled to 0 °C and then quenched cautiously with 1 N HCl. To this
quenched solution was added 100 mL of 1 N HCl, and then the reaction
mixture was transferred to a separatory funnel. The CH2Cl2 layer was
separated. The aqueous layer was extracted with 300 mL of CH2Cl2,
and then the combined CH2Cl2 layers were washed successively with
1 N HCl, saturated NaHCO3, brine, and deionized water. The organic
layer was dried over MgSO4, and then the solvent was removed in
vacuo, providing the product (18.5 g, 90%). The compound was used
1
CHCl3); H NMR (400 MHz, CDCl3) δ 0.0-1.0 (m, 6H), 1.15-1.30
(m, 2H), 1.40-1.60 (m, 6H), 1.70-1.80 (m, 4H), 1.90-2.10 (m, 6H),
2.68-2.78 (m, 2H), 2.92-2.99 (m, 2H), 7.16-7.24 (m, 6H), 7.27-
7.31 (m, 4H); 13C NMR (100 MHz, CDCl3) δ 19.9 (d, JC-P ) 26.7
Hz), 24.5, 24.6 (d, JC-P ) 36.6 Hz), 24.8, 33.9, 35.5, 40.4 (d, J ) 32.0
Hz), 126.5, 128.5, 129.1, 140.3; 31P NMR (162 MHz, CDCl3) δ 40
(m). HRMS (EI): (M + Na)+ 433.2293 ((M + Na)+, exact mass calcd
for C24H38B2NaP2: 433.2533).
Preparation of (5R)-Benzyldihydrofuran-2-one (28). Sodium metal
(6.43 g, 0.28 mol) was dissolved in 200 mL of EtOH. To the solution
was added 100 mL of anhydrous THF. Diethylmalonate (51 mL, 0.33
mol) was poured into the reaction, and the reaction was stirred for 5
min and then cooled to 0 °C in an ice bath. To the solution was added
(S)-(2,3-epoxypropyl)benzene, 27 (15.0 g, 0.11 mol), via syringe, and
then the ice bath was removed. The reaction was stirred overnight at
room temperature. During the course of the reaction, the reaction
mixture turned from a clear solution to a white gel. This gel could be
stirred magnetically. After the mixture was stirred overnight, 60 mL
of 5 N HCl was added to the reaction making the pH of the reaction
5. The volatiles were then removed under reduced pressure on a rotary
evaporator leaving a yellow oil suspended in water. To the suspension
was added 65 mL of DMSO, and then the flask was heated to 150 °C
in an oil bath. Gas evolved from the solution. After 16 h, no further
gas evolution was observed. The reaction was cooled to 0 °C, and then
300 mL of deionized water was added. The resulting solution was
extracted with 3 × 150 mL of diethyl ether. The combined organic
layers were then washed with 400 mL of deionized water, separated,
and then dried over MgSO4. The volatiles were removed on a rotary
evaporator, producing the product (17.3 g, 90%). The product was
determined to be 97% ee via chiral HPLC analysis [Chiralcel OD-H,
20% 2-propanol in hexane, 1.0 mL/min, 214 nm UV detection,
enantiomers eluting at 8.02 min (S, minor) and 8.89 min (R, major)].
The compound was used in the next step without further purification.
bp 199.4 °C; 1H NMR (400 MHz, CDCl3) δ 1.90-2.00 (m, 1H), 2.21-
2.29 (m, 1H), 2.33-2.51 (m, 2H), 3.07 (dd, J ) 14.16 Hz, J ) 6.10
Hz, 1H), 3.48 (dd, J ) 14.16 Hz, J ) 7.08 Hz, 1H), 4.12-4.77 (m,
1H), 7.19-7.34 (m, 5H); 13C NMR (100 MHz, CDCl3) δ 27.5, 28.3,
41.9, 81.8, 127.5, 128.2, 129.2, 136.1, 177.3. HRMS (EI): (M + Na)+
199.0679 ((M + Na)+, exact mass calcd for C11H12O2: 199.0735).
Preparation of (R)-1-Phenyl-2,5-pentanediol (29). To a 1 L round-
bottom flask equipped with a 500 mL pressure equalizing dropping
funnel was dissolved LAH (4.6 g, 0.12 mol) in 300 mL of THF under
N2. Into the dropping funnel was placed 28 (17.7 g, 0.1 mol) dissolved
in 300 mL of THF. The reaction was cooled to 0 °C in an ice bath,
and then the solution of 28 was added dropwise over a period of 30
min. After addition, the reaction was warmed to room temperature and
then stirred overnight. The reaction was cooled to 0 °C and then
quenched cautiously with 1 N HCl. Deionized water (200 mL) was
then added to the reaction mixture, and the contents of the flask were
transferred to a separatory funnel. The aqueous solution was extracted
with 3 × 200 mL of ethyl acetate. The combined organic layers were
1
in the next step without further purification. mp 151.3 °C; H NMR
(400 MHz, CDCl3) δ 1.82-2.03 (m, 6H), 2.43 (s, 3H), 3.00 (s, 3H),
4.27 (m, 2H), 4.89 (m, 1H), 7.23-7.32 (m, 3H), 7.33-7.35 (m, 2H);
13C NMR (100 MHz, CDCl3) δ 25.1, 31.2, 37.6, 38.1, 41.4, 69.4, 84.1,
127.5, 129.0, 129.9, 136.7. HRMS (EI): (M + Na)+ 358.9923 ((M +
Na)+, exact mass calcd for C13H20NaO6S2: 359.0599).
Preparation of (1R,2S)-1-Methyl-2-benzylphospholaneborane
(32). Methylphosphineborane was distilled trap-to-trap (2 mmHg) at
room temperature prior to the reaction. The receiving trap was kept at
-78 °C during distillation. The methylphosphineborane (4.1 g, 0.065
mol) was weighed quickly in air and then dissolved in 600 mL of THF
in a 2 L round-bottom flask. The flask was purged with N2, and then
the solution was cooled to -78 °C. To the solution was added n-BuLi
(52 mL, 2.5 M, 0.13 mol) via syringe over a period of 2-3 min. The
reaction was stirred for 1 h at -78 °C. Into a separate 500 mL round-
bottom flask was dissolved 30 (18.2 g, 0.054 mol) in 300 mL of THF
under N2. The solution of 30 was then added to the methylphos-
phineborane anion over a period of 2-3 min via cannula. The reaction
was allowed to warm to room temperature over 2 h and then was stirred
overnight. The reaction was quenched with 1 N HCl, and then 600 mL
of diethyl ether was added to the reaction mixture. The reaction mixture
was transferred to a separatory funnel. The organic layer was separated,
and then it was washed successively with 1 N HCl, brine, and deionized
water. After the layer was dried over MgSO4, the volatiles were
removed in vacuo, producing 13 g of a yellow oil. Flash column
chromatography over silica gel (1.5% ethyl acetate in hexane) produced
product 32 (2.1 g, 38% of expected). The product was determined to
be 91% ee via chiral HPLC analysis [Chiralcel OJ-R, 35% water in
acetonitrile, 1.0 mL/min, 214 nm UV detection, enantiomers eluting
at 3.90 min ((1S,2R), minor) and 4.29 min ((1R,2S), major)]. Analytical
samples of diastereomer 33 were also isolated (eluted after 32 during
flash column chromatography).
9
10226 J. AM. CHEM. SOC. VOL. 125, NO. 34, 2003

P-Chirogenic 1,2-Bisphospholanoethane Ligands
A R T I C L E S
1
32: bp 167.6 °C; [R]24 ) 0.5 °C (c 0.8, MeOH); H NMR (400
Preparation of 1,2-Bis((1R,2S)-2-benzylphospholano)ethane-
rhodium(I) Tetrafluoroborate (36). In a 25 mL vial was placed
[Rh(norbornadiene)2]+BF4- (232 mg, 0.571 mmol) under nitrogen. To
the metal complex was added 2 mL of MeOH via syringe, and the
resulting solution was cooled to -15 °C. In a separate vial, 35 (240
mg, 0.6283 mmol) was dissolved in 4 mL of THF under N2, and then
the resulting solution was taken up in a syringe. The solution of 35
was added dropwise to the metal complex solution over a period of 5
min while the temperature of the reaction was maintained at -15 °C.
The resulting solution was then allowed to warm to room temperature
and was stirred for 2 h. The solvent was removed in vacuo, and then
the red powder was recrystallized from hot methanol in a glovebox
producing a red crystalline product (213 mg, 55%). X-ray crystal-
lography confirmed the structure and stereochemistry of 36 (see
D
MHz, CDCl3) δ 0.05-0.97 (br m, 3H), 1.25 (d, J ) 10.7 Hz), 1.42-
1.70 (m, 2H), 1.70-2.20 (m, 5H), 2.64-2.73 (m, 1H), 3.06-3.13 (m,
1H), 7.17-7.29 (m, 5H); 13C NMR (100 MHz, CDCl3) δ 11.8 (d, JC-P
) 31.3 Hz), 24.4, 25.8 (d, JC-P ) 37.4 Hz), 25.8, 33.2, 35.4, 41.0 (d,
JC-P ) 33.6 Hz), 41.0, 126.3, 128.5, 128.8, 140.5 (d, JC-P ) 11.5 Hz);
31P (162 MHz, CDCl3) δ 29.4 (m). HRMS (EI): (MH - BH3)+
193.0980 ((MH - BH3)+, exact mass calcd for C12H18P: 193.1146).
1
33: bp 173.3 °C; [R]24 ) 18.3 °C (c 1.2, MeOH); H NMR (400
D
MHz, CDCl3) δ 0.19-1.00 (br m, 3H), 1.30 (d, J ) 10.3 Hz, 3H),
1.38-1.44 (m, 1H), 1.61-1.74 (m, 2H), 1.96-2.10 (m, 3H), 2.21-
2.26 (m, 1H), 2.43-2.51 (m, 1H), 3.03-3.09 (m, 1H), 7.19-7.33 (m,
5H); 13C NMR (100 MHz, CDCl3) δ 7.9 (d, JC-P ) 29.8 Hz), 24.7,
25.9 (d, JC-P ) 35.9 Hz), 32.7, 34.8, 39.8 (d, JC-P ) 34.3 Hz), 126.8,
128.8, 139.9, 140.0; 31P (162 MHz, CDCl3) δ 27.8 (m). HRMS (EI):
(MH - BH3)+ 193.0981 ((MH - BH3)+, exact mass calcd for
C12H18P: 193.1146).
Preparation of 1,2-Bis((1S,2S)-2-benzylphospholano)ethane (35).54
The phosphine borane, 34 (100 mg, 0.2439 mmol), was dissolved in 5
mL of degassed CH2Cl2 in a Schlenk tube under N2. The solution was
cooled to 0 °C, and then HBF4‚Me2O (0.45 mL, 3.66 mmol) was added
dropwise via syringe. The reaction was then warmed to room temper-
ature and stirred overnight. The reaction was quenched with a degassed
mixture of 6 mL of Et2O and 6 mL of saturated K2CO3. The aqueous
layer was removed via pipet while N2 was blown across the solution.
The organic layer was washed with 5 mL of degassed brine. The
aqueous layer was removed via pipet, and the organic layer was dried
over MgSO4 and then filtered through basic alumina. The solvent was
evaporated in vacuo to produce product 35 (85 mg, 91%). The free
phosphine ligand was immediately bound to rhodium to prevent
Supporting Information). mp 192.4 °C (decomp.); [R]24 ) 0.5 °C (c
D
0.8, MeOH); 31P NMR (162 MHz, CDCl3) δ 66.5 (d, JRh-P ) 144.6
Hz). HRMS (EI, direct insert): (M - BF4)+ 577.0685 ((M - BF4)+,
exact mass calcd for C31H40P2Rh: 577.1660).
Acknowledgment. The author thanks Dainius Macikenas and
Ron Rubin for expert assistance in X-ray crystallography and
Cassandra Farren for administrative assistance. Om Goel, James
Davidson, and Daniel Belmont are acknowledged for stimulating
discussion and project support.
Supporting Information Available: Instrumentation descrip-
tions, a general procedure for asymmetric hydrogenation,
methods for enantiomeric excess determinations for asymmetric
hydrogenation products, proof of relative and absolute stereo-
chemistry for 20 and 23, X-ray crystal data for 36, and selected
1H and 13C NMR spectra (PDF and CIF). This material is
available free of charge via the Internet at http://pubs.acs.org.
1
oxidation. H NMR (400 MHz, CDCl3) δ 0.78-0.99 (m, 4H), 1.18-
1.32 (m, 6H), 1.53-1.60 (m, 4H), 1.79-1.82 (m, 6H), 2.62-2.69 (m,
4H), 7.11-7.13 (m, 6H), 7.18-7.22 (m, 4H); 31P NMR (162 MHz,
CDCl3) δ -7.9.
(54) McKinstry, L.; Livinghouse, T. Tetrahedron Lett. 1994, 35, 9319.
JA034715O
9
J. AM. CHEM. SOC. VOL. 125, NO. 34, 2003 10227