A Novel Enantioselective Access to Enantiomerically Pure Phospholanes Using the Chiral Lithium Amide Base Approach

Full text of DOI:10.1021/jo972016m

912
J . Org. Chem. 1998, 63, 912-913
Ta ble 1. En a n tioselective Dep r oton a tion of P h osp h in e
Oxid e 3
A Novel En a n tioselective Access to
En a n tiom er ica lly P u r e P h osp h ola n es Usin g
th e Ch ir a l Lith iu m Am id e Ba se Ap p r oa ch
b
electrophile
MeI
product
yield (%)
eea (%)
[R]D
5a
5b
5c
5d
5e
5f
87
89
62
72
85
82d
74
60
85
90
c
36
55
EtI
Stephen C. Hume and Nigel S. Simpkins*
BnBr
allylBr
MeCOMe
PhCHO
cyclohex-CHO
PhSSO2Ph
117
92
100
97
Department of Chemistry, The University of Nottingham,
University Park, Nottingham NG7 2RD, U.K.
87
87
92e
82f
82
Received November 4, 1997
5g
5h
100
In tr od u ction
a
Measured by HPLC using a Chiralcel OD column, with 30%
iPrOH in hexane as eluant (1 mL/min) and UV detection at 256
Enantiomerically pure phosphines are without doubt the
most powerful and versatile ligands for use in asymmetric
catalysis involving transition metals.1 The C2-symmetric
bis-phosphines, such as BINAP and DuPHOS, have been
most widely applied,2 although unsymmetric chiral phos-
phines, usually incorporating additional metal-coordinating
groups, have also been shown to have important applica-
tions, Hayashi’s ferrocene derivatives being prime ex-
amples.3 Although many chiral phosphines and bis-phos-
phines have been prepared, there remains a need for novel
and easily accessible examples, with the potential for
structural fine-tuning, which may give improved results in
processes that at present give only modest selectivities.4
We became interested in the development of a novel access
to chiral phosphines, incorporating an alicyclic backbone (cf.
the phospholane unit of DuPHOS), involving a symmetry-
breaking chiral base reaction, exemplified in its simplest
form by the conversion of 1 into 2.5,6
b
nm. All positive in sign, measured in CHCl3 (c ) ca. 0.8-1.1).
c The ee could not be determined by HPLC. Yield of 7:1 mixture
d
of diastereomers. e The ee shown is for the major diastereomer.
f Determined using the inseparable 1:1 mixture of diastereomers
([R]D values not measured for this mixture).
oxide product having at least two asymmetric centers.
Bearing in mind structural features that might be desirable
in the final chiral phosphine, we realized this plan using
the readily available 1,2,5-triphenylphospholane oxide (3),8
by reaction with the chiral base 4 in THF at -100 °C, Table
1.
As can be seen, useful levels of asymmetric induction of
around 90% ee were achieved, it being possible to further
enrich several of the products to at least 97% ee by simple
recrystallization.9
Limited literature precedent suggested that the relative
configuration of the alkylated phosphine oxides should be
as shown for 5.10,11 Using enriched samples, X-ray crystal-
lography allowed both the relative and absolute configura-
tions to be assigned for 5e and for the major adduct 5f from
reaction with PhCHO (as shown);12 the other products are
assumed to belong to the same diastereo- and enantiomeric
series. Evidence that the anion alkylations had indeed
occurred in the same diastereomeric sense as the carbonyl
addition reactions (i.e., syn to the PdO bond) was obtained
from NOE studies of 5a , 5b, and 5d (see structure).
As in the analogous reactions of cyclic sulfoxides,7 kineti-
cally controlled discrimination between the two acidic sites
(R and R′) should result in a chiral metalated intermediate,
which on electrophilic quenching would give a phosphine
* To whom correspondence should be addressed. Tel.: (0115) 9513533.
Fax: (0115) 9513564. E-mail: nigel.simpkins@nottingham.ac.uk.
(1) For recent reviews, see: (a) Noyori, R. Asymmetric Catalysis in
Organic Synthesis; J ohn Wiley and Sons: New York, 1994. (b) Advanced
Asymmetric Synthesis; Stephenson, R., Ed.; Blackie Academic and
Professional: London, 1996. (c) Catalytic Asymmetric Synthesis; Ojima, I.,
Ed.; VCH Publishers Inc.: New York, 1993. For more complete listings,
see the bibliographies in ref 4a-d.
(2) BINAP: Noyori, R.; Takaya, H. Acc. Chem. Res. 1990, 23, 345.
DuPHOS: Burk, M. J . J . Am. Chem. Soc. 1991, 113, 8518.
(3) See, for example: Hayashi, T.; Ohno, A.; Lu, S.; Matsumoto, Y.;
Fukuyo, E.; Yanagi, K. J . Am. Chem. Soc. 1994, 116, 4221 and previous
papers in the series.
(8) Fiaud, J -C.; Legros, J -Y. Tetrahedron Lett. 1991, 32, 5089.
(9) Recrystallization: 5a from MeOH-H2O (×3), g99% ee, 30% recovery
(or from EtOH-H2O 96% ee, 69% recovery); 5d from EtOAc, g99% ee, 59%
recovery; 5e from EtOH-H2O, 97% ee, 69% recovery; 5h from EtOH-H2O,
g99% ee, 40% recovery.
(4) For recent examples of C2-symmetric bis-phosphine syntheses, see:
(a) Longmire, J . M.; Zhang, X. Tetrahedron Lett. 1997, 38, 1725. (b) Zhu,
G.; Cao, P.; J iang, Q.; Zhang, X. J . Am. Chem. Soc. 1997, 119, 1799. (c)
Zhu, G.; Chen, Z.; J iang, Q.; Xiao, D.; Cao, P.; Zhang, X. J . Am. Chem. Soc.
1997, 119, 3836. See also: (d) Langer, F.; Pu¨ntener, K.; Stu¨rmer, R.;
Knochel, P. Tetrahedron: Asymmetry 1997, 8, 715.
(5) (a) The enantioselective deprotonation of phosphine-boranes, using
sBuLi-sparteine, has been described; see: Muci, A. R.; Campos, K. R.;
Evans, D. A. J . Am. Chem. Soc. 1995, 117, 9075. (b) Chiral lithium amides
have been shown to be ineffective with acyclic phosphine oxides; see:
O’Brien, P.; Warren, S. Synlett 1996, 579.
(6) Complementary studies of enantioselective protonation of metalated
phosphine oxides have appeared; see: (a) Vedejs, E.; Garcia-Rivas, J . A. J .
Org. Chem. 1994, 59, 6517. (b) Guillen, F.; Moinet, C.; Fiaud, J -C. Bull.
Soc. Chim. Fr. 1997, 134, 371 (this report concerns the phospholane system
employed in our work). See also ref 5b.
(7) (a) Armer, R.; Begley, M. J .; Cox, P. J .; Persad, A.; Simpkins, N. S. J .
Chem. Soc., Perkin Trans. 1 1993, 3099. (b) Blake, A. J .; Westaway, S. M.;
Simpkins, N. S. Synlett 1997, 919.
(10) (a) Marinetti, A.; Kruger, V.; Le Menn, C.; Ricard, L. J . Organomet.
Chem. 1996, 522, 223 and references therein. For related results involving
phospholanic acid derivatives, see: (b) Polniaszek, R. P. J . Org. Chem. 1992,
57, 5189. In some cases the stereochemical outcome seems not to have been
assigned; see: Mathey, F.; Muller, G.; Bonnard, H. Bull. Soc. Chim. Fr.
1972, 4021.
(11) The intermediate lithiated phosphine oxide would be expected to
be configurationally unstable, on the basis of previous studies; see O’Brien,
P.; Warren, S. Tetrahedron Lett. 1995, 36, 8473. Our results can be
interpreted as
a highly diastereoselective electrophilic quench of the
intermediate, leading to substitution with overall retention. Only in the
case of anion protonation have we seen substitution with inversion, in accord
with the observations of Fiaud and co-workers (see ref 5b).
(12) The absolute configurations for 5e and 5f were established by the
collection of low-temperature data, including Friedel equivalents, and by
refinement of a Flack parameter (values 0.05(8) and 0.0(3), respectively);
see: Flack, H. D. Acta Crystallogr., Sect. A 1983, 39, 876. We thank Drs.
W-S. Li and A. J . Blake of this department for these determinations; full
details will be published elsewhere.
S0022-3263(97)02016-1 CCC: $15.00 © 1998 American Chemical Society
Published on Web 02/04/1998

Communications
J . Org. Chem., Vol. 63, No. 4, 1998 913
feature,16 cf. 8, in which access to the reactive center is
possible via two equivalent open quadrants.
In order to allow access to additional types of product,
especially those having a functional group “handle”, we
briefly examined further metalation-electrophilic quench
sequences using enantiomerically pure 5a as the starting
Although these systems have been very successful, the
requirement for C2-symmetry poses quite a serious limita-
tion on the structures available. We considered that a
design in which one of the quadrants is closed to give a
situation such as 9 where only one carefully tailored “open
site” is available may offer some advantages, particularly
with respect to steric fine-tuning. The product phosphines
available from our new chiral base chemistry, described
above, appear to conform to this design, as indicated by 10,
and give substantial flexibility in the choice of R. The ability
to prepare simple phosphines and bifunctional systems, of
either absolute configuration, starting with a readily avail-
able phospholane is clearly attractive. Further studies of
the asymmetric deprotonation, along with application of the
product ligands in asymmetric catalysis, are ongoing.
n
material. For example, reaction with BuLi in THF, followed
by addition of either PhCHO or Ac2O, gave 6a and 6b in
87% and 53% yields, respectively.13
Finally, we have shown that reduction of such phosphine
oxides can be carried out employing a mixture of Cl3SiH and
pyridine,14 for example, to give 7a -c in the yields indi-
cated.15
Ack n ow led gm en t. We thank the EPSRC for a student-
ship in support of S.C.H. and for the provision of a four-
circle diffractometer.
Many recent syntheses of phosphines and bis-phosphines
have focused on the C2-symmetry element as a key design
Su p p or tin g In for m a tion Ava ila ble: Selected data for 5a ,c-
f,h and 6b, including HPLC determination of ee and a typical
experimental procedure (23 pages).
(13) Compound 6a was formed as a 15:1 mixture of diastereomers at
the new carbinol center.
J O972016M
(14) Naumann, L.; Zon, G.; Mislow, K. J . Am. Chem. Soc. 1969, 91, 7012.
(15) At present we have not proved that the phosphine oxide reduction
has occurred with retention of configuration.
(16) For
a review of C2-symmetry and asymmetric induction, see:
Whitesell, J . K. Chem. Rev. 1989, 89, 1581.