In situ recycling of chiral ligand and surplus nucleophile for a noncatalytic reaction: Amplification of process throughput in the asymmetric addition step of efavirenz (DMP 266)

Article abstract of DOI:10.1021/op025595s

The synthesis of efavirenz (DMP 266) involves a highly enantioselective asymmetric reaction of ketone 2a with lithium cyclopropylacetylide 3a in the presence of (1R,2S)-pyrrolidinylnorephedrine (PNE) 4b as the chiral mediator to produce 6b, a key advance

Full text of DOI:10.1021/op025595s

Organic Process Research & Development 2003, 7, 324−328
Technology Reports
In Situ Recycling of Chiral Ligand and Surplus Nucleophile for a Noncatalytic
Reaction: Amplification of Process Throughput in the Asymmetric Addition
Step of Efavirenz (DMP 266)
,
†
Anusuya Choudhury,* James R. Moore, Michael E. Pierce, Joseph M. Fortunak, Ioannis Valvis, and Pat N. Confalone
Chemical Process R&D Department, Process Research Facility, The DuPont Pharmaceuticals Company,‡
Deepwater, New Jersey 08023-0999, U.S.A.
Abstract:
tetrameric aggregate 8 by structural reorganization of the
product-incorporated inactive aggregate 9.
The synthesis of efavirenz (DMP 266) involves a highly enan-
tioselective asymmetric reaction of ketone 2a with lithium
cyclopropylacetylide 3a in the presence of (1R,2S)-pyrrolidi-
nylnorephedrine (PNE) 4b as the chiral mediator to produce
Aggregation in lithium-based anion chemistry is an
important phenomenon in organic synthesis, which has been
exploited to achieve high facial selectivities observed in
asymmetric reactions. However, such aggregative properties
often become a liability due to high stoichiometric require-
ments when the product is incorporated as a ligand, resulting
in the disruption of reactive aggregates. An example is the
synthesis of efavirenz (1), a potent HIV-1 reverse tran-
6b, a key advance intermediate bearing a tetrasubstituted chiral
carbon. The major drawback of this reaction is that it requires
at least 2 mol of cyclopropylacetylene (CPA) 3b, 2 mol of the
chiral mediator 4b, and 4 mol of an n-alkyllithium base to
generate a 1 mol of addition product 6b. In a program to
improve the cost-effectiveness of the process, we have studied
this asymmetric addition reaction to reduce the stoichiometry
of the chiral moderator and CPA nucleophile. The initial
experimental designs to improve the stoichiometry were based
on the assumption that the second equivalent of lithium
cyclopropylacetylide simply acted as a base to deprotonate the
N-H of ketones 2a or 2b, leaving 1 equiv of unlithiated and
presumably unreactive CPA. It was reasoned that if we lithiate
this CPA in the postreaction mixture, it would complex with
the already lithiated ligand in the reaction which could then
be used to convert additional 2a or 2b into product thus
improving the stoichiometry. After some trial experiments, a
dramatic decrease in stoichiometry was achieved and it was
found that by simply adding one more equivalent of an
n-alkyllithium to the system after the first reaction cycle, it was
possible to obtain at least a 14% throughput increase with just
a 5% dilution of the reaction volume in this simple and more
cost-efficient process. The chiral addition reactions could be
run with multiple cycles in the same pot with the sequential
addition of n-alkyllithium 5, CPA 3b, and ketone 2a. However,
after four cycles, there was some decrease in the enantioselec-
tivities (90.8%) that ultimately places a practical limit on the
number of recycles possible. The process throughput increase
can be explained in the light of a recent mechanistic investiga-
tion by Collum and co-workers. We postulate that the introduc-
tion of an n-alkyllithium to the reaction mixture after the
completion of the first cycle regenerates the reactive cubic
1
scriptase inhibitor, which is designed around the key
asymmetric reaction of ketone 2a with lithium cyclopropy-
lacetylide 3a in the presence of (1R,2S)-pyrrolidinylnorephe-
2
drine (PNE) 4b as the chiral mediator (Scheme 1). The
asymmetric addition step, which creates a chiral tetrasub-
stituted carbon, proceeds with exceptionally high enantiose-
lectivity. The major drawback of this reaction is that it
requires at least 2 mol of cyclopropylacetylene (CPA) 3b, 2
mol of the chiral mediator 4b, and 4 mol of an n-alkyllithium
base to generate a 1 mol of addition product 6b. These
superstoichiometric requirements for asymmetric addition,
which has been postulated to be aggregation related, greatly
increase the cost of the drug, particularly for the use of a
3
custom designed chiral auxiliary 4b and the CPA as the
†
Present address: Johnson&Johnson Process Research & Development/Drug
Evaluation, Raritan, NJ 08869. Telephone: 908-704-4975. Fax: 908-231-7429.
E-mail: achoudhu@prdus.jnj.com.
‡
Current name: Bristol-Myers Squibb Pharma Company.
(1) (a) Romero, D. L. Ann. Rep. Med. Chem. 1994, 29, 123. (b) Young, S. D.;
Britcher, S. F.; Tran, L. O.; Payne, L. S.; Lumma, W. C.; Lyle, T. A.;
Huff, J. R.; Anderson, P. S.; Oleson, D. B.; Carroll, S. S.; Pettibone, D. J.;
O’Brien, J. A.; Ball, R. G.; Balani, S. K.; Lin, J. H.; Chen, L.-W.; Schleif,
W. A.; Sardana, V. V.; Long, W. J.; Byrnes, V. W.; Emini, E. A.
Antimicrob. Agents Chemother. 1995, 39, 2602.
(2) (a) Thompson, A.; Corley, E. G.; Huntington, M. F.; Grabowski, E. J. J.
Tetrahedron Lett. 1995, 36, 8937. (b) Pierce, M. E.; Parsons, R. L., Jr.;
Radesca, L. A.; Lo, Y. S.; Silverman, S.; Moore, J. R.; Islam, Q.;
Choudhury, A.; Fortunak, J. M. D.; Nguyen, D.; Luo, C.; Morgans, S. J.;
Davis, W. P.; Confalone, P. N.; Chen, C.-Y.; Tillyer, R. D.; Frey, L.; Tan,
L.; Xu, F.; Zhao, D.; Thompson, A. S.; Grabowski, E. J. J.; Reider, P. J.
J. Org. Chem. 1998, 63, 8536.
(
3) (a) Soai, K.; Yokoyama, S.; Hayasaka, T. J. Org. Chem. 1991, 56, 4264.
b) Zhao, D.; Chen, C.-Y.; Tillyer, R. D.; Pierce, M. E.; Moore, J. R. Org.
Synth. 1999, 77, 12.
(
3
24
•
Vol. 7, No. 3, 2003 / Organic Process Research & Development
10.1021/op025595s CCC: $25.00 © 2003 American Chemical Society
Published on Web 03/14/2003

Scheme 1
Table 1. Asymmetric Addition of Lithium Cyclopropylacetylide 3a to Ketones 2a/2b in the Presence of Chiral Ligand 4b
ketone
conversiona
%, first cycle
(overall)
% (S)-isomerb
first cycle
(overall)
(
equiv in
CPA (3b)
(equiv)
n-hexyllithium
(5) (equiv)
ligand 4b
(equiv)
entry
each cycle)
1
2
3
4
5
6
7
8
9
2a (1)
2b (1)
2a (1)
2.15
2.15
1. 07
2.2
2.3
2.36
4.4
4.4
2.15
4.8 + 1
4.4 + 1
4.44 + 1.07
4.4 + 1.05
4.4 + 1.5
4.33 + 2
4.8 + 2 + 2
4.8 + 2 + 2 + 2
2.3
2.3
1.15
2.6
2.1
2.3
2.3
2.3
2.3
>98
>98
>98
94
50
>98
99.4 (97.4)
(97.8)
99 (99.03)
99 (96.9)
97.9 (96)
98.1
94
90.8
2a (1 + 0.5)
2a (1 + 0.4)
2a (1 + 0.3)
2a (1 + 1)
2a (1 + 1)
2a (1 + 1)
2a (1 + 1 + 1)
2a (1 + 1 + 1 + 1)
92.7 (87)
97 (91)
>99 (100)
>99 (73)
98 (91)
>99 (99.0)
>96
2.1 + 1.05
2.1 + 1.05
c
d
3.33
e
1
1
0
1
2.2 + 1 + 1
2.2 + 1 + 1 + 1
2.65
2.65
>95
a
Conversion was monitored by HPLC. b Enantiomeric ratios were determined by chiral HPLC or chiral SFC. c The total amount of 3b intended for a number of
cycles can be introduced in the reaction flask from the beginning of the first cycle, but generation of 3a in each cycle is controlled by sequential addition of hexyllithium,
as shown in the column 4. er of the isolated product; also see ref 7 for conversion. e An aliquot of the reaction was used to determine the er and % of conversion.
The reaction was continued for the next cycle (see entry 11), which was subsequently quenched.
d
starting material. Although one can isolate and reuse most
of the excess reagents, this additional processing and the high
n-alkyllithium stoichiometry requirement still negatively
impact the manufacturing cost. We have developed an
alternate strategy where the chiral ligand 4a and the
nucleophile 3a can be recycled in the same reaction pot, by
in situ regeneration of the reactive aggregate, to amplify
product generation with excellent enantiomeric purity and
chemical yield for this super stoichiometric process.
In a program to improve the cost-effectiveness of the
process, we have been studying various aspects of this
asymmetric addition reaction particularly in the direction of
reducing the stoichiometry of the chiral moderator and CPA
nucleophile. Our standard asymmetric addition procedure was
to combine the components 3b, 4b, and an n-alkyllithium
reagent 5 (2.15, 2.3, and 4.4 equiv, respectively) between
the enantiomerically enriched (>99% er) product 6b. When
the reaction was carried out under similar conditions but with
stoichiometric ratios of 3a and 4a, only about 50% conver-
sion was obtained (Table 1, entry 3).
Recently, Collum et al. have performed a spectroscopic
investigation to elucidate the origin of the enantiofacial bias
4
of this chiral addition reaction. On the basis of these studies,
they proposed possible structures of the solution aggregates
of the nucleophile and the chiral ligand, which explains the
superstoichiometric requirements for this reaction. The salient
features of their findings are as follows: (1) The N-H of
the starting ketone 2a or the product 6a is unexpectedly not
lithiated in the presence of 2 equiv of lithium cyclopropy-
lacetylide 3a. (2) A cubic tetrameric structure 8 (Scheme 2)
comprised of 3a and 4a in a ratio of 2:2 is proposed as the
reactive aggregate. (3) The origin of the 50% conversion
(when 3a and 4a are used in stoichiometric ratios with respect
to 2a) has been postulated partly due to the tetrameric nature
of the reactive aggregate and then the participation of product
alkoxide 6a in an aggregate of 3a:4a:6a in a ratio of 1:2:1.
The resulting complex 9 (Scheme 2) inhibits the reaction to
-
20 °C and 0 °C to allow complexation and then to cool
the mixture below -40 °C followed by addition of either
ketone 2a or 2b (1 equiv) while maintaining the temperature
below -40 °C in order to avoid degradation of the lithiated
intermediates 6a or 7a. Typically after 1 h, HPLC analysis
showed 95-99% conversion and a 94-99% enantiomeric
ratio (er) (see Table 1, entries 1 and 2). Workup resulted in
(4) Thompson, A.; Corley, E. G.; Huntington, M. F.; Grabowski, E. J. J.;
Remenar, J. F.; Collum, D. B. J. Am. Chem. Soc. 1998, 120, 2028.
Vol. 7, No. 3, 2003 / Organic Process Research & Development
•
325

Scheme 2
of PNE, since the enantiomeric ratio of the formed desired
enantiomer is typically >99%.
Three additional experiments were conducted to confirm
that the postreaction mixture is giving extra enantio-pure
product after the addition of 1 equiv of n-alkyllithium. Each
of these experiments used 2.3 equiv of chiral ligand, 2.1
equiv of CPA, and 4.3 equiv of n-hexyllithium to generate
the reactive tetrameric complex under our standard condi-
tions, and the chiral additions were run at -43 °C. The
reactors, 1st, 2nd, and 3rd, each containing the tetrameric
aggregate, were charged, respectively, with 1.3, 1, and 1
equiv of ketone 2a. After 1 h at -43 °C, the 1st reactor was
left as such, 2nd reactor was charged with additional 0.3
equiv of 2a and the 3rd reactor was charged with an
additional 1 equiv of n-hexyllithium followed by 0.3 equiv
of ketone 2a. The reactions were quenched after another hour
of aging. The enantiomeric ratios of all three reaction
mixtures were experimentally the same, 98%. The significant
finding was that the first two reactions gave a total of 1.1
equiv of product 6b, but the third reaction gave a complete
conversion (1.3 equiv) of product. Thus, by simply adding
one more equiValent of an n-alkyllithium to the system after
the first reaction cycle, it was possible to obtain at least a
proceed further as it places the ketone 2a and the nucleophile
5
3a distal to each other in the transition structure. If the
product-incorporated complex would have been the entire
reason for inhibition of the reaction, then, with 1 equiv each
of 3a and 4a used, the yield could only be 33%.
1
4% throughput increase with just a 5% dilution of the
Our initial experimental designs to improve the stoichi-
ometry were carried out prior to mechanistic findings by
Collum et al. and were based on the assumption that the
second equivalent of lithium cyclopropylacetylide simply
acted as a base to deprotonate the N-H of ketones 2a or 2b
leaving 1 equiv of unlithiated and presumably unreactive
CPA. We reasoned that if we lithiated this CPA in the
postreaction mixture, it would complex with the already
lithiated ligand in the reaction. Unless the product interferes,
this assembly could then be used to convert additional 2a or
reaction Volume in this simple and more cost-efficient
process.
We next focused on how to obtain 2 equiv of 6b from a
single pot (Table 1, entries 7, 8, and 9). Here, we added an
extra 1 equiv of CPA, either before or after the first reaction
cycle, followed by the addition of 1.05, 1.5, and 2 equiv of
n-hexyllithium after the first reaction cycle, and then finally
charged an additional 1 equiv of 2a. These reactions gave
73, 91, and 99% additional conversions to product 6b,
7
2b to product thus improving the stoichiometry.
respectively. A significant decrease in reaction rate was
This hypothesis was tested as follows: After completion
of the reaction between 1.0 equiv of the PMB-protected
ketone 2a with the preformed chiral complex (at 3b:4b:5 )
observed in the second reaction cycle, and the reaction
showed a slight decrease in enantioselectivity compared to
that of the first reaction cycle.
2
.2:2.6:4.8 equiv, respectively), an additional 0.5 equiv of
The chiral addition reactions could be run with multiple
cycles in the same pot (Table 1, entries 10 and 11). The
amounts of n-hexyllithium 5, CPA 3b, and ketone 2a were
adjusted at the end of each reaction cycle. After four cycles
(entry 11), the overall stoichiometry of 2a:3b:4b:5 is 1:1.25:
0.5:2.5 compared to 1:2.1:2.3:4 in the nonrecycled process.
However, there was some further decrease in the enantiose-
lectivities (90.8%) that ultimately places a practical limit on
the number of recycles possible.
PMB-protected ketone 2a was introduced into the same
reaction mixture. As expected, there was essentially no
further reaction with continued aging at -43 °C. On the other
hand, when 0.5 equiv of 2a was introduced after the addition
of 1 equiv of n-hexyllithium 5 to the reaction mixture, an
8
1
7% additional product conversion with a 97.4% er (Table
, entry 4) was obtained. Decreasing the amount of addition-
ally charged 2a to 0.4 equiv and 0.3 equiv (Table 1, entries
and 6, respectively) resulted in 91% and 100% conversions
5
The process throughput increase can be explained in the
light of recent mechanistic investigations by Collum and co-
workers. We postulate that the introduction of an alkyllithium
to the reaction mixture after the completion of the first cycle
regenerates a reactive aggregate. As shown in Scheme 2,
with 97.8% and 99.03% er’s, respectively. This overall
process is shown in Scheme 2.
6
The possibility of autostereoamplification by having the
lithiated product (6a) acting as a chiral ligand in lieu of PNE
was next examined. Analysis of the product of this reaction
showed to have 70% ee favoring the wrong enantiomer. This
8
the aggregate 9 (3a:4a:6a in a ratio of 1:2:1), which causes
autoinhibition, undergoes structural reorganization in the
“
background” reaction must be insignificant in the presence
(
7) This reaction was run in 0.579 mol scale in a 5 L reactor. For the initial
chiral addition: at T130 conversion ) 98.8%, at T160 conversion ) 99.1%,
er ) 99.2%. For the second cycle reaction in the same pot: at T230
conversion ) 93.8%, at T260 conversion ) 97.6%, T290 conversion )
98.98%. T1 and T2 represent conversion times in minutes for original
addition and the next iterative addition, respectively.
(
5) Xu, F.; Reamer, R. A.; Tillyer, R.; Cummins, J. M.; Grabowski, E. J. J.;
Reider, P. J.; Collum, D. B.; Huffman, J. C. J. Am. Chem. Soc. 2000, 122,
1
1212.
(
6) Shibata, T.; Morioka, H.; Hayase, T.; Choji, K.; Soai, K. J. Am. Chem.
Soc. 1996, 118, 471-472.
326
•
Vol. 7, No. 3, 2003 / Organic Process Research & Development

9
presence of an additional 1 equiv of alkyllithium. It might
-1 to -23 °C to give the tetrameric complex 8 and was
cooled to -43 °C. The THF solution of the ketone (200 g,
99.5% pure, 0.5789 mol/250 mL of THF) was added to the
complex, allowing the internal temperature to rise to -40
°C during the addition. The resulting red solution was aged
at -43 ( 3 °C for 1 h. A half hour sample aliquot was
quenched into 1 N HCl, and HPLC analysis showed >99%
conversion of starting ketone to product. The reaction mass
was aged for an additional 0.5 h. n-Hexyllithium (179.08 g,
32.2% solution in hexanes, 0.625 mol) was charged at -43
( 3 °C, and the reaction mass was aged for 15 min. A THF
solution of the ketone (60 g/90 mL of THF) was charged
dropwise at -43 ( 3 °C. HPLC analysis of an aliquot after
be possible that the product separates from the aggregate as
a bis lithiated species 6c.10 The remainder CPA-Li 3a and
ligand 4a can reaggregate to generate 8 (or mixture of
aggregates), which reacts with added ketone 2a to generate
additional product.
We postulate that the drop in enantiomeric ratios for the
iterative addition cycles (Table 1, entries 10 and 11) could
be due to the presence of an increased percentage of lithiated
product 6b (or 6c) which was previously shown to lead to
the opposite enantioselectivity.
In conclusion, we have demonstrated that the high
stoichiometric requirement for the chiral ligand and the
lithium cyclopropylacetylide in the enantioselective addition
step of efavirenz can be significantly reduced and the product
can be isolated with excellent enantiomeric purity. We have
also demonstrated that multiple asymmetric reactions can be
carried out in an iterative fashion in the same reaction pot,
and with process throughput increased, controlling the
amounts of subsequent charges of n-hexyllithium and the
ketone 2a at the end of each cycle.
0.5 h quenched into 1 N HCl showed a ratio of 99:1 product:
starting ketone. The cooling bath was removed, and the
reaction mixture was quenched into 1 N HCl (3.4 L). The
organic layer was separated and was washed with 2 × 617
mL of 1 N HCl. The aqueous layers were combined and
back extracted with 700 and 300 mL of toluene. The
combined organic solution was washed with 750 mL of water
and was concentrated under reduced pressure to 650 g.
Toluene (208 mL) was charged, and the solution was heated
to +65 °C and was aged for 0.5 h. n-Heptane (930 mL) was
charged slowly at +65 °C to maintain a solvent ratio of 60:
Experimental Procedure:
Generation of 1.3 Equiv of Product (Table 1, Entry
). A 5-L, four-neck round-bottom flask was equipped with
6
40 heptane:toluene and was aged for 0.5 h. The mixture was
a mechanical stirrer, Dean-Stark trap with condenser,
nitrogen inlet, and thermocouple. The system was purged
with nitrogen for 15 min, and a nitrogen atmosphere was
maintained throughout the reaction. A stock solution of the
ligand 4b (693.83 g, 39.4 wt % in toluene, 779.6 mL, 2.3
slowly cooled to -5 °C and aged for 0.5 h at that temperature
to complete the crystallization. The solids were collected by
filtration and washed with 700 mL of 60:40 heptane:toluene
chilled to -1 °C. The product was dried at +50 °C with
vacuum and nitrogen. The isolated yield of the product was
3
equiv, 1.33 mol) was charged to the reactor followed by Ph -
25
2
+
84 g: chemical yield 91.0%, 99.3 wt %, ep ) 98.06, [R]
8.07 (c, 1.006, methanol) (lit. ) 8.15).
D
CH (1.44 g, 0.0058 mol, 0.01 equiv, used as indicator). The
reaction mixture was heated to boiling, and 82 mL of solvent
was distilled off. The mixture was cooled to about room
temperature, and anhydrous THF (827.3 mL) was charged.
The reaction mixture was cooled to -20 °C, and n-
hexyllithium (379.9 g, 32.2 wt % solution in hexanes, 1.327
mol, 2.29 equiv) was charged at such a rate as to maintain
a temperature below 0 °C until the reaction mixture turned
to an orange-red color. Additional n-hexylithium (357.3 g,
2
b
Generation of 2 Equiv of Product (Table 1, Entry 9).
A 5 L, 4-neck round-bottom flask was equipped with a
mechanical stirrer, Dean-Stark trap with condensor, nitrogen
inlet, and thermocouple. The system was purged with
nitrogen for 15 min, and a nitrogen atmosphere was
maintained throughout the reaction. A stock solution of the
ligand 4b (1872.4 g, 14.6 wt % solution in toluene, 2.1 L,
2
.3 equiv) was charged to the reactor. Triphenylmethane
3
2.2 wt % solution in hexanes, 1.248 mol, 2.157 equiv) was
(1.44 g, 0.0058 mol, 0.01 equiv, used as indicator) was
charged beyond this red point at such a rate as to maintain
the temperature below 0 °C. Cyclopropylacetylene 3b (90.36
g, 1.367 mol, 2.36 equiv) was added dropwise, keeping the
temperature below 0 °C. The mixture was aged for 1 h at
charged to the reactor. The reaction mixture was heated to
boiling, and 1.2 L of solvent was distilled off. The mixture
was cooled to about room temperature, and anhydrous THF
(
884.6 mL) was charged. The reaction mixture was cooled
to -20 °C and n-hexyllithium (552 mL, 385.96 g, 32.2 wt
solution in hexanes, 2.33 equiv) was charged at such a
(
8) The monolithiated addition product 6a decomposes at higher temperature.
1
H NMR analysis showed that the crude mixture appears to consist of a
%
mixture of indolines as the major constituents.
rate to maintain a temperature below 0 °C until the reaction
mixture turns to a red color. Actual charge was 375.7 g.
Additional n-hexyllithium (353.31 g, 32.2 wt % solution in
hexanes) was charged beyond this red point at such a rate
as to maintain the temperature below 0 °C. Cyclopropy-
lacetylene (128.56 g, 1.947 mol, 3.36 equiv) was charged at
such a rate as to maintain the temperature below 0 °C until
the color changed from orange to yellow. The mixture was
stirred at 0-10 °C and chilled to -40 °C. The ketone
solution in THF (200 g/300 mL of THF, 99.5% pure, 0.5789
mol) was charged below -40 °C. The reaction mixture was
(
9) For solution structures derived from n-butyllithium and vicinal amino
alkoxides, see: Sun, X.; Winemiller, M. D.; Xiang, B.; Collum, D. B. J.
Am. Chem. Soc. 2001, 123, 8039. In this case, the addition of n-butyllithim
to benzaldehyde seems to show no conversion dependence because the
product separates out of the mixed aggregate, thus regenerating the C2-
symmetric reactive aggregate.
(
10) It might be possible that the addition of n-hexyllithium causes the metalation
of N-H possibly with a consequent formation of its own dianionic
homoaggregate and regeneration of the reactive form (we thank reviewer
1
for this comment). However, if this would have been the reason, the
original ee of the process (after multiple iterations) would have been
restored. But some drop in enantiomeric ratio was observed (Table 1, entries
1
(
0 and 11) which implicated the involvement of another mechanism
possibly product acting as a competing ligand causing low enantioselec-
tivity).
Vol. 7, No. 3, 2003 / Organic Process Research & Development
•
327

stirred at -43 ( 3 °C. A 0.5 h sample aliquot was withdrawn
and quenched into 2 N HCl. The conversion was 98.8:1.2
have a toluene content of 954 mL. It was heated to 65 °C
and was aged for 0.5 h. n-Heptane (1430 mL) was charged
slowly at +65 °C to maintain a solvent ratio of 60:40
heptane:toluene and was aged for 0.5 h. The mixture was
cooled to room temperature and aged for 1 h. The solids
were collected by filtration and washed with 1000 mL of
60:40 heptanes:toluene. They were washed again with 200
mL of 60:40 heptanes:toluene. The wet cake was dried
overnight at 50 °C with a nitrogen purge to generate 400.3
(product:starting material). Another aliquot after 1 h of
reaction showed a conversion of 99.1:0.9 (product:starting
material) with a chiral purity of 99.2%. n-Hexyllithium was
charged (347.81 g, 32.2% solution in hexane, 1.2156 mol,
2
.099 equiv) to the reaction mixture at the chiral addition
temperature. The mixture was stirred for 15 min at -43 (
°C. A solution of the ketone (200 g/300 mL of THF) was
charged slowly at -43 ( 3 °C. An aliquot withdrawn after
.5 h of the reaction showed 93.8% product, whereas the 1
3
2
5
g (83.9%), 99.5 wt %, ep 96.9%, [R]
D
+7.95 (c, 1.006,
2
b
0
methanol) (lit. ) 8.15).
h sample showed 97.6% and 1.5 h sample showed 98.98%
product being formed. The cooling bath was removed, and
the reaction mixture was quenched into 1 N HCl (4.16 L).
It was stirred for 20 min, and the layers were separated. The
organic layer was extracted out and washed with 2 × 378
mL of 2 N HCl. The aqueous layers were combined and
back extracted with 700 and 300 mL of toluene. The organics
Acknowledgment
We are thankful to Professor D. F. Taber for valuable
suggestions and to Dr. Roger Stringham, Barbara Lord, and
Bill Cummings for their analytical support.
Received for review September 17, 2002.
OP025595S
were combined and washed with 750 mL of H
2
O. The
organic layer was concentrated under reduced pressure to
328
•
Vol. 7, No. 3, 2003 / Organic Process Research & Development