Chiral α-substituted carbonyls and alcohols from the S(N)2' displacement of cuprates on chiral carbonates: An alternative to the alkylation of chiral enolates

Article abstract of DOI:10.1021/jo000810t

A highly stereoselective sequence of reactions, based on the anti-selective S(N)2' addition of cuprates to allylic carbonates, transforms alkynes or alkenyl halides into carbonyls having α-chiral centers. The method, which uses menthone as a chiral auxiliary, is a useful alternative to the alkylation of chiral enolates with the added advantage of allowing for the 'alkylation' of sec- and tert-alkyl and aryl groups.

Full text of DOI:10.1021/jo000810t

J . Org. Chem. 2000, 65, 7091-7097
7091
Ch ir a l r-Su bstitu ted Ca r bon yls a n d Alcoh ols fr om th e SN2′
Disp la cem en t of Cu p r a tes on Ch ir a l Ca r bon a tes: An Alter n a tive
to th e Alk yla tion of Ch ir a l En ola tes
Claude Spino,* Christian Beaulieu, and J ulie Lafrenie`re
Universite´ de Sherbrooke, De´partement de Chimie, 2500 Boulevard Universite´,
Sherbrooke, Qc, J 1K 2R1, Canada
cspino@courrier.usherb.ca
Received May 26, 2000
A highly stereoselective sequence of reactions, based on the anti-selective SN2′ addition of cuprates
to allylic carbonates, transforms alkynes or alkenyl halides into carbonyls having R-chiral centers.
The method, which uses menthone as a chiral auxiliary, is a useful alternative to the alkylation of
chiral enolates with the added advantage of allowing for the “alkylation” of sec- and tert-alkyl and
aryl groups.
Sch em e 1
In tr od u ction
Among the most widely used methodologies to prepare
compounds with chiral carbon centers sits the alkylation
of chiral enolates and derivatives.1-4 These include
enolates and equivalents derived from chiral amides,
esters, imines, enamines, and hydrazones and enolates
possessing chirality at the metal.1,2,4-6 The ubiquitous
nature of carbonyl compounds possessing R chiral centers
and the huge synthetic versatility of the carbonyl func-
tional group are responsible for the impressive arsenal
of such methodologies now available to the synthetic
chemist. Each methodology has its own advantages, but
they all share a common limitation: reactivity. In
general, only methyl or ethyl iodide, benzyl halides,
R-alkoxy halides, and similarly reactive electrophiles are
adequate. Some methodologies allow for the alkylation
of â-branched iodides,7 but secondary, tertiary, vinyl, and
aryl halides often are either unreactive or lead to second-
ary reactions such as elimination. The Lewis acid induced
alkylation of silyl enol ethers allows reaction with SN1-
prone electrophiles but is of limited use with ester- or
amide-derived O-silyl ketene acetals, and sec-alkyl elec-
trophiles are poor electrophiles with this method.8,9
We recently communicated our preliminary results on
an alternative to the alkylation of chiral enolates involv-
ing the anti-selective addition of cuprate reagents onto
chiral carbonates derived from menthone (Scheme 1).10
We now report the full account of this study, disclosing
the wide generality and applicability of the method to
construct chiral carbon centers R to a carbonyl or hy-
droxymethylene unit. There are several stereochemical
issues to be dealt with in this highly stereoselective
sequence; the stereoselective addition of alkynyl or vinyl
anions to menthone (1 f 4 or 1 f 5); restriction of the
rotational freedom of the vinyl group in 6 during the SN2′
displacement step; the stereoselectivity of the cuprate
addition process itself (6 f 7), and finally the configu-
rational stability of the final carbonyl compound during
the ozonolysis step (7 f 8). We have divided this paper
in sections dealing with each issue.
* To whom correspondence should be addressed. Tel: (819)-821-
7087, fax: (819)-821-8017.
(1) Evans, D. A. Aldrichimica Acta 1982, 15, 23.
(2) Evans, D. A. In Asymmetric Synthesis. Stereodifferentiating
Reactions, Part B; Morrison, J . D., Ed.; Academic Press: New York,
1984; Vol. 3, p 1.
(3) Caine, D. In Comprehensive Organic Synthesis; Trost, B. M.,
Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 3, p 1.
(4) Enders, D. In Asymmetric Synthesis. Stereodifferentiating Reac-
tions, Part B; Morrison, J . D., Ed.; Academic Press: New York, 1984;
Vol. 3, p 275.
(5) Bergbreiter, D. E.; Newcomb, M. In Asymmetric Synthesis.
Stereodifferentiating Reactions, Part A; Morrison, J . D., Ed.; Academic
Press: New York, 1983; Vol. 2, p 243.
Ad d ition of Alk en yl a n d Alk yn ylm eta ls to Men -
th on e. Several factors motivated our choice of menthone
as the chiral auxiliary. It is inexpensive and available
in both enantiomeric forms, each in high enantiomeric
purity, but more importantly, we believed that the 1,2-
(6) Corey, E. J .; Xu, F.; Noe, M. C. J . Am. Chem. Soc. 1997, 119,
12414.
(7) Myers, A. G.; Yang, B. H.; Chen, H.; McKinstry, L.; Kopecky, D.
J .; Gleason, J . L. J . Am. Chem. Soc. 1997, 119, 6496.
(8) Reetz, M. T. Angew. Chem., Int. Ed. Engl. 1982, 21, 96.
(9) Examples of asymmetric alkylation with SN1 prone electrophiles,
other than sec- or tert-alkyl electrophiles, can be found in Evans, D.
A.; Urpi, F.; Somers, T. C.; Clark, S. J .; Bilodeau, M. T. J . Am. Chem.
Soc. 1990, 112, 8215.
(10) Spino, C.; Beaulieu, C. J . Am. Chem. Soc. 1998, 120, 11832.
10.1021/jo000810t CCC: $19.00 © 2000 American Chemical Society
Published on Web 09/13/2000

7092 J . Org. Chem., Vol. 65, No. 21, 2000
Spino et al.
Ta ble 1. Ad d ition s of Cu p r a tes on Ch ir a l Ca r bon a te 6b
axial alcohols 5. The choice between the two methods
should be made based on the overall yield of pure axial
alcohol 5.
(R ) Me).
entry
cuprate
producta yieldb (%) % de
Ad d ition of Cu p r a te Rea gen ts on Ch ir a l Ca r bon -
a tes. The carbonate was chosen as the leaving group
because it was one of the few into which the tertiary
alcohol could be derivatized. The axial tertiary alcohol
in 5 is sterically congested and rather unreactive. Ace-
tates, which are more commonly used for cuprate dis-
placement reactions, were obtained in very low yields
with much decomposition. The same was true for mesy-
lates and carbamates. Carbonates could only be produced
using dimethyl carbonate or methyl chloroformate and
n-butyllithium in THF at -78 °C. Not surprisingly, the
carbonates were unstable and had to be used immedi-
ately in the next step without further purification.
Sometimes, some of the starting alcohol was recovered
after the subsequent cuprate addition. We have deter-
mined that an incomplete carbonate formation or carbon-
ate hydrolysis prior to the cuprate step was responsible
for this observation. Careful and prompt handling of the
carbonate compounds alleviated the problem.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Et2CuLi
EtCuCNLi
n-Bu2CuLi
n-BuCuCNLi
(R)-7a
(R)-7a
(R)-7b
(R)-7b
(S)-7c
(R)-7d
(R)-7e
(R)-7f
(S)-7g
(S)-7g
(S)-7h
(S)-7i
(R)-7je
(S)-7k
(S)-7l
(S)-7m
75
67
70
55
78
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
t-BuCuCNLi
H2CdCH(CH2)2CuMgBr
(Cyhex)2CuMgBr
(H2CdCH)2CuLi
Ph2CuLi
Ph(thiophene)CuCNLi
(p-MeO-Ph)2CuLi
(m-MeO-Ph)2CuLi
(4-i-BuPh)2CuLi
(furanyl)2CuLi
(3-pyridinyl)2CuLi
(n-Bu3Sn)2CuLi
73
91
20c
75
40d
44
64
80
65
17f
>100g
a
The annotation R or S designates the newly formed chiral
center starting from (-)-menthone unless otherwise indicated.b
Isolated yield for two steps from the corresponding alcohol 3.
The SN2 addition, the reduction, and the elimination products were
also isolated. 80% based on recovered starting material. Made
c
d
e
f
from (+)-menthone. Alcohol 5b was also isolated in 73% yield.
Contaminated with n-Bu4Sn.
g
The pivotal step in this sequence relied on the highly
anti-selective addition of cuprate reagents on allylic
carbonates and related systems. The first example of a
cuprate reagent displacing an acetate with allylic re-
arrangement was published in 1969 by Crabbe´ and co-
workers.11 It was followed immediately by two reports
from a team of Zoecon Corp. on the stereospecificity of
the cuprate addition on allylic acetates.12,13 Cuprates have
since been recognized to add preferentially anti to acyclic
or cyclic allylic, propargylic, and allenic halides, acetates,
carbonates, and epoxides.14-26 To produce a stereogenic
center of high optical purity, the addition of the cuprate
reagents had to occur anti to the carbonate on only one
of the two reactive conformers A or C, respectively
(Figure 1). Previous work by Crabbe´27 and ourselves28,29
indicated that a group as small as methyl or a bromine
atom adjacent to the carbonate on cyclic systems, such
as 6, was sufficient to drive the conformational equilib-
rium completely toward conformer A. Our own MMX
addition of organometallics to menthone would be highly
selective. The experimental confirmation came as no
surprise. The isopropyl group of menthone effectively
directs nucleophiles to the opposite face of the carbonyl.
For example, the addition of vinylmagnesium bromide
prepared from 2b (R ) H) or trans-1-propenyllithium
made from trans-1-bromopropene 2a (R ) Me) to men-
thone proceeded with complete stereoselectivity to give
the axial alcohols 5a and 5b, respectively (Scheme 1).
The addition of alkynyllithium or magnesium reagents
was somewhat less stereoselective but gave good yield
of the axial alcohols 4 accompanied with 5-25% of the
corresponding equatorial alcohols. However, prior addi-
tion of dry cerium trichloride to the alkynyllithium
reagent mixture increased significantly the selectivities
which are reported in Table 2. We have also noticed that
slow addition of a solution of menthone to a solution of
the alkynylcerate leads to the highest selectivities.
Importantly, in every instance the diastereomeric alco-
hols were easily separated by flash chromatography, and
pure axial alcohols 4 could be isolated in good to excellent
yields. The tertiary propargyl alcohols 4 could be reduced
to the corresponding trans-allylic alcohols with Red-Al
or lithium aluminum hydride in THF at room tempera-
ture or refluxing THF. No trace of the corresponding cis-
alcohols were detected. A good yield of the allene was
isolated when alkynyl silyl ether 4g was submitted to
the reducing conditions overnight at reflux. In this case,
the reduction is best carried out at -10 °C for a short
period of time in which case only 12% of the correspond-
ing allene was formed vs 73% of the desired allylic
alcohol. It was also possible to hydrogenate the triple
bonds using Lindlar’s catalyst under one atmosphere of
hydrogen to obtain the cis-propargylic alcohols.
(11) Rona, P.; To¨kes, L.; Tremble, J .; Crabbe´, P. J . Chem. Soc., Chem.
Commun. 1969, 43.
(12) Anderson, R. J .; Henrick, C. A.; Siddall, J . B. J . Am. Chem.
Soc. 1970, 92, 735.
(13) Anderson, R. H.; Henrick, C. A.; Siddall, H. B.; Zurflu¨h, R. J .
Am. Chem. Soc. 1972, 94, 5379.
(14) Ibuka, T.; Yamamoto, Y. Organocopper Reagents: A Practical
Approach; Taylor, R. J . K., Ed.; Oxford University Press: Oxford, 1994;
Part 7.
(15) Krause, N.; Gerold, A. Angew. Chem., Int. Ed. Engl. 1997, 36,
186.
(16) Goering, H. L.; Kantner, S. S.; Seitz, E. P. J . Org. Chem. 1985,
50, 5495.
(17) Underiner, T. L.; Goering, H. L. J . Org. Chem. 1990, 55, 2757.
(18) Marshall, J . A.; Blough, B. E. J . Org. Chem. 1991, 56, 2225.
(19) Marshall, J . A.; Crute III, T. D.; Hsi, J . D. J . Org. Chem. 1992,
57, 115.
(20) Marshall, J . A.; Pinney, K. G. J . Org. Chem. 1993, 58, 7180.
(21) Corey, E. J .; Boaz, N. W. Tetrahedron Lett. 1984, 25, 3063.
(22) Fleming, I.; Thomas, A. P. J . Chem. Soc., Chem. Commun. 1985,
411.
(23) Fleming, I.; Thomas, A. P. J . Chem. Soc., Chem. Commun. 1986,
1456.
(24) For a review, see: Marshall, J . A. Chem. Rev. 1989, 89, 1503.
(25) Marshall, J . A.; Trometer, J . D.; Cleary, D. G. Tetrahedron 1989,
45, 391.
(26) Goering, H. L.; Kantner, S. S. J . Org. Chem. 1984, 49, 422.
(27) Crabbe´, P.; Dollat, J .-M.; Gallina, J .; Luche, J .-L.; E., V.;
Maddox, M. L.; To¨ke`s, L. J . Chem. Soc., Perkin Trans. 1 1978, 730.
(28) Spino, C.; Liu, G. J . Org. Chem. 1993, 58, 817.
(29) Spino, C.; Liu, G.; Tu, N.; Girard, S. J . Org. Chem. 1994, 59,
5596.
Alternatively, the alkynes can be hydroaluminated and
quenched with iodine or NBS to give the corresponding
vinyl halides. Each vinyl halide can undergo a metal-
halogen exchange with tert-butyllithium to give the
corresponding vinyllithium which add with complete
stereoselectivity to menthone to give high yields of pure

Chiral R-Substituted Carbonyls and Alcohols
J . Org. Chem., Vol. 65, No. 21, 2000 7093
Ta ble 2. Ad d ition s of Alk en yl a n d Alk yn yl Meta ls on Men th on e a n d SN2′ Ad d ition of Cu p r a tes on Ch ir a l Ca r bon a tes 6.
SN2′Cent addition of
addition on menthone
cuprates on carbonates 6
yielda of
4-(âOH) (%)
ratiob of
4-(âOH)/4-(ROH)
yieldc of
7 (%)
entry
3
R
R′ in R′2CuLi
Prd
% ded
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
3b
3c
3d
3d
3e
3e
3f
3g
3h
3i
3i
3i
3j
3j
3j
3k
3l
3m
3n
3o
3o
3o
Me
85
90
87
(5:1)e
20:1
18:1
see Table 1
Me
Ph
p-Cl-Ph
Me
Ph
Me
PhMe2Si
Me
Me
i-Pr
t-Bu
Me
i-Pr
t-Bu
Me
i-Pr
Me
Me
Me
i-Pr
- -
70
87
91
61
72
75
59
82
61
67
40
58
59
49
84
- -
>99
>99
>99
>99
>99
>99
>99
>99
>99
82
33
90
33
14
>99
82
>99
>99
>99
60
n-Bu
i-Pr
i-Pr
t-Bu
t-Bu
BnO-CH2
TBDMSO-CH2
PMBO-(CH2)2
Ph
(S)-7b
(R)-7n f
(R)-7of
(R)-7c
(R)-7p
(R)-7q
(R)-7r
(R)-7sf
(R)-7g
(R)-7t
(S)-7p
(R)-7h
(R)-7u
(S)-7v
(R)-7i
(R)-7w
(R)-7x
(R)-7k
(R)-7l
(R)-7y
(S)-7z
70
20:1
96
26:1
83 (56)e
90
17:1 (2:1)e
50:1
80
20:1
Ph
Ph
p-MeO-Ph
p-MeO-Ph
p-MeO-Ph
m-MeO-Ph
p-Cl-Ph
6-MeO-1-Npht
2-furanyl
3-pyridinyl
3-pyridinyl
3-pyridinyl
86
12:1
89
74
72
59
50
27:1
20:1
16:1
4:1
63
42g
43
16:1
38 (21)h
20 (48)h
16 (28)h
t-Bu
60
a
b
Isolated yield of pure 4 (â-OH) after separation of diastereomers starting with (-)-menthone unless otherwise stated. Determined
by GC or NMR on crude mixtures. c Isolated yield for two steps from corresponding alcohol 5. de ) diastereomeric excess. e No CeCl3
d
added. f Made from (+)-menthone. g Contaminated with SN2 displacement product (see text). Yield of recovered alcohol 5o in parentheses.
h
Sch em e 2
F igu r e 1. Energies of three conformations of 6.
conformer A (Figure 1). The ORTEP diagram of 9
(Supporting Information) shows the six-membered ring
calculations placed conformer A (R ) H) at more than 4
in a chair conformation with both the methyl and
kcal/mol lower in energy than conformer C.
isopropyl groups in the axial position. A1,3 strain is
The SN2′ displacement reactions proved highly selec-
presumably responsible for this observation.30 We could
tive indeed. The addition of various cuprate reagents to
not confirm that this conformation was also predominant
chiral carbonates 6b (R ) Me) gave essentially only one
in solution.
diastereomer of each adduct 7 (Table 1). We measured
Changing the type of the cuprate reagent (Gilman’s,
the ratio of diastereomers by gas chromatography (GC)
cyanocuprates, higher-order cuprates, etc.) affected the
against an authentic mixture of both diastereomers in
yield of the reaction but never the stereoselectivity (Table
most cases. Thus, by interchanging the substituents on
1). Monoalkylcyano cuprates were an appealing choice
the cuprate and alkene, we were able to prepare the
in view of the fact that none of the transferable alkyl
diastereomers of 7b,c,g-i,k ,l. We then prepared equi-
portion was wasted. It gave good yields in many cases,
molar mixtures of each pair of diastereomers and found
but we have found that dialkyl cuprates were generally
that they were easily separated by GC. Injection of the
superior. Higher order cuprates (not included in Table
pure diastereomers from the crude reaction mixture in
1) also gave satisfactory yields but offered no advantage
these cases showed a single peak. This result was
over Gilman’s type cuprates. The choice of cuprate
correlated with the proton and 13C NMR spectra. The
reagent should thus be made based on an optimum yield
NMR spectra and GC traces of all the other adducts were
of product, taking into account the cost of the alkyl being
found to be clean, and their diastereomeric purity was
transferred.
confirmed when we verified the enantiomeric purities of
the corresponding alcohol compounds after ozonolysis
(vide infra). The absolute stereochemistry of the adducts
could be obtained from the sign of the optical rotation of
some known aldehydes or acids after ozonolysis. By
replacing the p-methoxybenzyl (PMB) group of compound
(R)-7s by a TBDPS protecting group ((R)-7s f 9, Scheme
Table 1 lists several different alkyl and arylcuprates
that reacted with 6b (R ) Me) with high stereoselectivity.
(+)-Ibuprofen31 was obtained in high enantiomeric purity
from (+)-menthone (cuprate adduct (R)-7j, Table 1, entry
13) and our method seems well suited to make secondary
and tertiary alkyl analogues of this important anti-
2), we were able to grow crystals of compound 9 of
sufficient quality to perform a single-crystal X-ray analy-
sis which confirmed that the adducts possessed the
(30) Eliel, E. L.; Wilen, S. H. Stereochemistry of Organic Compounds;
Wiley-Interscience: New York, 1994.
(31) Lombardino, G. J . Non-Steroidal Antiinflammatory Drugs;
stereochemistry corresponding to an anti-addition on
Wiley-Interscience: New York, 1985.

7094 J . Org. Chem., Vol. 65, No. 21, 2000
Spino et al.
Sch em e 3
Sch em e 4
inflammatory drug, something not easily achieved by
existing methods.32-35 Heterosubstituted aromatic (Table
1, entries 11 and 12) and heteroaromatic cuprates
(entries 14 and 15) also reacted with carbonate 6, giving
the corresponding adducts 7 with complete stereoselec-
tivity. Finally, stannyl and silyl cuprates added highly
stereoselectively as well (Table 1, entry 16, and Table 2,
entry 8).
any other substrate we have tried. Compound 6n started
reacting with the cuprate at ca. 10 °C. Adding the cuprate
at -30 °C and letting the mixture slowly warm to room
temperature led to the formation of 18% of the SN2
product. In contrast, adding the cuprate at -30 °C and
warming quickly to room-temperature kept the amount
of SN2 adduct to a minimum. Since both SN2 and SN2′
adducts were diasteromerically pure, radicals or ionic
intermediates are unlikely in this case. The fact that 6j
did not give any noticeable quantity of SN2 adduct (Table
2, entry 13) is rather difficult to interpret, and we cannot
explain this discrepancy at the moment.
Fortunately, the desired aryl-substituted adducts 7 can
be prepared with high stereoselectivity if the aromatic
nucleus is part of the cuprate reagent. In those cases,
high stereoselectivity was observed even when the allylic
carbonate contained a bulky substituent such as isopro-
pyl or tert-butyl (Table 2, entries 3, 4, and 6). Presumably,
radical or ionic intermediates are not stable enough in
these cases, and reductive coupling occurs without com-
petition from homolytic or heterolytic cleavage. While
most Ar2CuLi reagents gave satisfactory results, addition
of lithium divinylcuprate to allylic carbonate 6b gave only
20% of the desired adduct (R)-7f and, in addition to some
SN2 adduct, much reduction and elimination products
were isolated. Although vinyl cuprates have been exten-
sively used in Michael-type additions, examples of a
successful SN2′ or SN2 addition of a vinyl cuprate on
allylic systems are scarce. Ibuka reported the addition
of vinyl copper species, prepared by transmetalation from
divinyl zinc, onto allylic mesylate. He also reported
reduction products from the addition of vinyl cuprates
on allylic mesylates, perhaps via the in situ formation of
copper hydride species.14,39,40
Groups of different sizes and nature were tolerated on
the vinylic moiety of the carbonate molecules leading to
hindered adducts with high diastereomeric purity (Table
2). For example, (R)-tert-butylphenylacetic acid was
prepared in >99% ee after ozonolysis of adduct (R)-7p
(Table 2, entry 6), and, as far as we are aware, this
constitutes the only highly stereoselective synthesis of
this compound. Metzger and co-workers prepared its
ester in 32% ee using a chiral tin hydride reagent on the
corresponding radical.36,37 We also prepared (R)-2-(4-
chlorophenyl)-3-methylbutanoic acid, a fragment of the
important class of pyrethroid insecticides, from the
ozonolysis of the corresponding cuprate adduct (R)-7o
(Table 2, entry 4).38 However, addition of some cuprate
reagents on styryl-substituted carbonates gave substan-
tially lower stereoselectivities (entries 11-15, 17, 21, 22).
Curiously, the stereoselectivity seemed to depend on the
size of the cuprate reagent. Entries 10, 13, 16, and 18-
20 of Table 2 show that lithium dimethylcuprate added
to styryl-substituted carbonates 6i-k ,m -o with high
selectivity while the corresponding diisopropyl and tert-
butyl cuprate gave products with lower % de. Larger
cuprate reagents take higher temperatures and longer
times to react and may produce a biradical intermediate
III (or cationic intermediate) from a copper complex I in
a sequence where the rate-determining step is presum-
ably the reductive coupling (I f II) (Scheme 3). Interest-
ingly, reactions between the small lithium dimethyl-
cuprate and allylic carbonates 6m and 6n gave, along
with the main SN2′ adduct, 5-25% of the SN2 displace-
ment products (entries 18 and 19). The SN2 displacement
product was not observed in the addition of cuprates on
We have investigated the addition of lithium dimeth-
ylcuprate on the carbonate prepared from the minor
equatorial alcohol 5f (R-OH) (Scheme 4) obtained from
the addition of 3-benzyloxypropynyllithium on 3f on
menthone followed by Red-Al reduction. The major
cuprate adduct (S)-7q corresponded to an anti addition
of the cuprate on conformation R-6f-A as expected, but
the 5:1 selectivity observed was much lower than the
near-complete selectivity obtained from the carbonate of
5f (â-OH) under the same reaction conditions. We at-
tribute this variance to a lower energy difference between
(32) For a review on arylpropanoic acids see: Sonawane, H.; Bellur,
N. S.; Ahuja, J . R.; Kulkarni, D. G. Tetrahedron: Asymmetry 1992, 3,
163.
(33) Fuji, K.; Node, M.; Tanaka, F.; Hosoi, S. Tetrahedron Lett. 1989,
30, 2825.
(34) Noyori, R.; Nagai, K.; Kitamura, M. J . Org. Chem. 1987, 52,
3176.
(35) Stille, J . K.; Parrinello, G. J . Am. Chem. Soc. 1987, 109, 7122.
(36) Blumenstein, M.; Schwarzkopf, K.; Metzger, J . O. Angew.
Chem., Int. Ed. Engl. 1997, 36, 235.
(37) Metzger, J . O.; Blumenstein, M.; Schwarzkopf, K. Eur. J . Org.
Chem. 1998, 177.
(38) Davies, J . H. In The Pyrethroid Insecticides; Leahey, J . P., Ed.;
Taylor & Francis: London, 1985; p 1.
(39) Ibuka, T.; Akimoto, N.; Tanaka, M.; Nishii, S.; Yamamoto, Y.
J . Org. Chem. 1989, 54, 4055.
(40) Ibuka, T.; Habashita, H.; Kawasaki, T.; Takameto, Y. J . Org.
Chem. 1998, 63, 2392.

Chiral R-Substituted Carbonyls and Alcohols
J . Org. Chem., Vol. 65, No. 21, 2000 7095
Sch em e 5
Sch em e 7
are susceptible to racemization upon chromatography.
When the crude aldehydes were reduced to the corre-
sponding alcohols prior to purification, the Mosher esters
of the latter indicated no racemization had taken place.
Earlier, we reported modest yields of carboxylic acids 11
using H2O2 as workup conditions for the ozonolysis.10 We
have solved this problem by utilizing J ones reagent
instead of hydrogen peroxide. Acids can now be obtained
in reproducible 60-80% yields.
Sch em e 6
The trisubstituted double bond exocyclic to the men-
thone nucleus is in fact quite unreactive. Treatment of
(R)-7b with osmium tetraoxide for days led to recovery
of that double bond intact. We could selectively cleave
the monosubstituted alkene in (R)-7d with the Lemieux
conditions (OsO4/NaIO4) with no concomitant cleavage
of the auxiliary (Scheme 7). Unlike the amide or ester
auxiliaries used in alkylation reactions, our auxiliary is
virtually insensitive to strongly basic and acidic condi-
tions. For example, compound (S)-7g was recovered intact
from treatment with n-BuLi in ether at room tempera-
ture or 1 N sulfuric acid in methanol, each for a 12 h
period. The chiral auxiliary can therefore be left on as a
protecting group for the carbonyl until later in a synthe-
sis. Of course, the final target should not contain a
functionality somewhere else in the molecule that can
react with ozone because it is probable that it will be
cleaved along with the menthone auxiliary. For example,
the adducts with a pyridine or a furan ring or the
tributyltin adduct gave several products upon ozonolysis.
This constitutes a limitation to the present methodology
which we are addressing in the next generation chiral
auxiliary. One possibility involves a chiral auxiliary
bearing a functional group adequately positioned for the
selective cleavage of the desired double bond.
In conclusion, we have demonstrated the utility of our
sequence to produce R-chiral carbonyls or hydroxymeth-
ylenes of high enantiomeric purity. Our method is
complementary to the classical approach of alkylating a
chiral enolate, both from an electronic point of view (an
alkylating agent is electrophilic whereas the cuprate
reagent involved here is nucleophilic) and from the point
of view of the nature of the “alkylating” agent (our
methods allow tertiary and secondary alkyl as well as
aryl groups to be introduced in a highly stereoselective
manner). Finally, the very high enantiomeric purity of
the final compounds obtained via our strategy compares
favorably to those obtained by the alkylation of chiral
enolates. Further developments are under way in our
laboratory.
the two reactive conformations R-6f-A and R-6f-B (Scheme
4). Calculations gave 2.2 kcal/mol energy difference
between these two conformations while the energy dif-
ference between the two reactive conformations having
the equatorial vinyl group (cf. A and C in Figure 1) was
over 4 kcal/mol. We cannot exclude a syn addition of the
cuprate on conformation R-6f-A as a source of the minor
diastereomer. Interestingly, the cuprate addition on the
carbonate of propargyl alcohols 4i (â-OH) or 4i (R-OH)
proceeded with high stereoselectivity in both cases to give
the corresponding chiral tetrasubstituted allenes 10a and
10b in 86 and 80% yield, respectively (Scheme 5). We
did not get confirmation yet that both additions proceeded
anti to the carbonate, but a highly selective syn addition
is unlikely. We are currently investigating the chemistry
of such chiral allenes.
Oxid a tive Clea va ge of th e Au xilia r y. Oxidative
cleavage of the trisubstituted double bond was best
achieved with ozone in an appropriate solvent. Initially,
we found that the lipophilic nature of some of our adducts
led to a sluggish ozonolysis due to incomplete solubility
in normal solvents such as dichloromethane and metha-
nol at -78 °C. Addition of hexanes to the solution
sometimes solved the problem, but eventually acetone
was found to be a better solvent. Most ozonolyses were
slow at -78 °C and required monitoring of the reaction
by thin-layer chromatography since the usual blue color
of excess ozone was not a reliable indication that cleavage
was complete. Nonetheless, ozonolysis of all adducts gave
high yields of aldehydes, alcohols, or acids depending on
the workup conditions. Ozonolysis and treatment of the
ozonide with sodium borohydride in methanol furnished
enantiomerically pure alcohols 12 which were derivatized
to their corresponding Mosher’s esters 13 (Scheme 6).
Each ester was checked for enantiomeric purity by proton
and fluorine NMR or by GC to confirm the earlier GC
analysis of the adducts 7. Some of the R-aryl aldehydes
Exp er im en ta l Section
Gen er a l. Reactions were performed under a nitrogen or
argon atmosphere with oven-dried glassware unless otherwise
stated. Diethyl ether, hexanes, and tetrahydrofuran were
distilled over sodium-benzophenone. Dichloromethane and
triethylamine were distilled over calcium hydride. Flash
chromatography was done using Merck Kieselgel silica gel 60
(230-400 Mesh A.S.T.M). Concentration of organic solutions
implies evaporation on a rotary evaporator followed by pump-
ing under reduce pressure (0.5 mmHg).

7096 J . Org. Chem., Vol. 65, No. 21, 2000
Spino et al.
Melting points were determined on a Reichert 7905 melting
point apparatus integrated into an Omega Engineering Model
199 Chromel-alumel thermocouple. Infrared spectra were
recorded on a Perkin-Elmer Paragon 1000 spectrometer, and
only the major bands are reported. Nuclear magnetic reso-
nance (NMR) spectra were recorded on a Bruker AC-300
instrument. The following abbreviations were used: s, singlet;
d, doublet; dd, doublet of doublet; q, quadruplet; m, multiplet.
Chemical shifts are reported in δ relative to chloroform or
acetone. Proton decoupled carbon NMR spectra used chloro-
form (77.0 ppm) and methanol (49.0 ppm) as internal stand-
ards. Mass spectra (MS) were obtained on a VG Micromass
ZAB-2F instrument.
Gen er a l P r oced u r e for th e Ad d ition of Alk yn yl or
Alk en ylm eta ls to Men th on e. The appropriate alkyne or
vinyl bromide (1.2 equiv) was dissolved in dry cold (-78 °C)
THF (0.7M) and n-BuLi (1.2 equiv) or t-BuLi (2.4 equiv)
respectively was slowly added. Dry cerium trichloride (1.0
equiv) was added and after stirring for 60 min, (-)-menthone
(1.0 equiv) was added over 5 min. The reaction mixture was
stirred at -78 °C for 1 h after which time saturated sodium
chloride was poured into the cold solution. The aqueous phase
was separated and extracted with ether, and the combined
organic phases were washed with brine, dried over magnesium
sulfate, filtered, concentrated in vacuo, and purified by flash
chromatography on silica gel (hexanes-EtOAc 9:1) to give
exclusively the axial alcohol compound.
3594, 3494, 3014, 2943, 2863, 1496, 1446, 1171; LRMS (m/z
(relative intensity)): 258 (M+, 20), 223 (5), 202 (20), 173 (100);
Exact mass calcd for C18H26O: 258.1984. Found: 258.1989.
[R]20D: -51.5 (c 3.06, CHCl3)
Gen er a l P r oced u r e for th e Con ver sion of th e Alcoh ols
to Ca r bon a tes. The appropriate alcohol was dissolved in dry
cold (-78 °C) THF (to a concentration of ca. 0.1 mol/L), n-BuLi
(2 equiv) was added dropwise, and the yellow solution was
stirred at -78 °C for 30 min and then at 0 °C for another 30
min. Then, dimethyl carbonate (2 equiv) was added, and the
reaction mixture was stirred at 0 °C for 30 min and then at
room temperature for 1.5 h, after which time sat. ammonium
chloride was poured into the cold solution. The aqueous phase
was separated and extracted with ether, and the combined
organic phases were washed with brine, dried over magnesium
sulfate, filtered, and concentrated in vacuo to give the carbon-
ates which were used crude in the next reaction.
Ca r bon a te 6b. 1H NMR (CDCl3, 300 MHz): δ 5.72 (dm,
1H, J ) 16.2 Hz), 5.45 (dq, 1H, J ) 16.2, 6.1 Hz), 3.71 (s, 3H),
2.71 (dm, 1H, J ) 15.2 Hz), 2.23-2.13 (m, 1H), 1.83-1.72 (m,
4H), 1.70-1.05 (m, 6H), 0.97-0.83 (m, 9H).
Ca r bon a te 6i. 1H NMR (CDCl3, 300 MHz): δ 7.40-7.28
(m, 5H), 6.48 (d, 1H, J ) 16.2 Hz), 6.35 (d, 1H, J ) 16.2 Hz),
3.75 (d, 3H), 2.83 (dm, 1H, J ) 15.3 Hz), 2.25-2.16 (m, 1H),
1.90-1.75 (m, 2H), 1.69-1.15 (m, 5H), 0.98-0.83 (m, 9H).
Gen er a l P r oced u r e for th e Ad d ition of Cu p r a tes to
Allyl Ca r bon a tes. P r oced u r e A: Dia lk yl Cu p r a tes. The
cuprates were prepared by adding 2.4 equiv of the organo-
lithium to a suspension of purified copper iodide (1.2 equiv)
and lithium iodide (1.2 equiv) in dry THF (0.1 M) at -30 °C.
After the second equivalent of organolithium was added, the
solution was stirred 30 min at -30 °C, and then the crude
carbonate (1 equiv) dissolved in THF was added dropwise. The
reaction mixture was stirred at -30 °C for 30 min and then
for 1 h at 0 °C (adduct 11 h required rt for completion) after
which time a solution of sat. ammonium chloride and am-
monium hydroxide (9:1) was poured into the cooled solution
(0 °C). The aqueous phase was separated and extracted with
pentane, and the combined organic phases were washed with
brine, dried over magnesium sulfate, filtered, concentrated in
vacuo, and purified by flash chromatography on silica gel
(100% hexanes) to give exclusively the anti SN2′ diastereomer.
P r oced u r e B: Alk yl Cya n ocu p r a tes. The cuprates were
prepared by adding 1.5 equiv of the organolithium to a
suspension of copper cyanide (1.5 equiv) in dry THF (0.1 M)
at -78 °C. The suspension was stirred for 30 min at -78 °C
after which time the crude carbonate (1 equiv) dissolved in
THF was added. The reaction mixture was heated to room
temperature for 4 h after which time a solution of sat.
ammonium chloride and ammonium hydroxide (9:1) was
poured into the reaction mixture. The aqueous phase was
separated and extracted with ether, and the combined organic
phases were washed with brine, dried over magnesium sulfate,
filtered, concentrated in vacuo, and purified by flash chroma-
tography on silica gel (100% hexanes) to give exclusively the
anti SN2′ diastereomer.
(1S,2S,5R)-2-Isop r op yl-5-m eth yl-1-{2-p h en yl-1-eth yn -
1-yl}cycloh exa n -1-ol (4i). Yield: 80%; 1H NMR (CDCl3, 300
MHz): δ 7.44-7.39 (m, 2H), 7.31-7.28 (m, 3H), 2,53-2.44 (m,
1H), 2.06 (dt, 1H, J ) 13.7, 2.7 Hz), 1.85-1.73 (m, 2H), 1.71-
1.65 (m, 1H), 1.58-1.41 (m, 5H), 1.01 (d, 3H, J ) 9.9 Hz), 0.98
(d, 3H, J ) 9.9 Hz), 0.90 (d, 3H, J ) 5.9 Hz); 13C NMR (CDCl3,
75 MHz): δ 2 × 131.6 (s), 3 × 128.2 (s), 123.0 (s), 94.0 (s),
83.5 (s), 72.1 (s), 50.6 (d), 50.1 (t), 34.8 (t), 28.5 (d), 27.3 (d),
23.9 (q), 21.9 (q), 20.7 (t), 18.8 (q); IR (neat, cm-1): 3474, 3015,
2943, 2863, 1491, 1451; LRMS (m/z (relative intensity)): 256
(M+, 10), 241 (15), 213 (20), 171 (100); Exact mass calcd for
C
18H24O: 256.1827. Found: 256.1825. [R]20D: +11.5 (c 0.88,
CHCl3).
(1S,2S,5R)-2-Isopr opyl-5-m eth yl-1-{1-pr open -1-yl}cyclo-
1
h exa n -1-ol (5b). Yield: 99%; H NMR (CDCl3, 300 MHz): δ
5.65 (dq, 1H, J ) 15.4, 6.7 Hz), 5.45 (d, 1H, J ) 15.4 Hz), 2.05-
1.90 (m, 1H), 1.80-1,65 (m, 2H), 1.71 (dd, 1.71, 3H, J ) 6.7,
1.7 Hz), 1.53-1.38 (m, 4H), 1.20 (s, 1H), 1.13-1.00 (m, 2H),
0.90-0.80 (m, 9H); 13C NMR (CDCl3, 75 MHz): δ 139.4 (d),
121.6 (d), 76.0 (s), 49.6 (t), 49.5 (d), 35.1 (t), 27.9 (d), 26.7 (d),
23.8 (q), 22.3 (q), 20.9 (t), 18.5 (q), 17.7 (q); IR (neat, cm-1):
3600, 3507, 2943, 2871, 1451, 1169; LRMS (m/z (relative
intensity)): 196 (M+, 10), 181 (5), 163 (5), 111 (100), 69 (25);
Exact mass calcd for C13H24O: 196.1827. Found: 196.1832.
[R]20D: -31.0 (c 1.30, CHCl3).
Gen er a l P r oced u r e for th e Red u ction of th e P r op a r -
gyl Alcoh ols to th e Cor r esp on d in g Allyl Alcoh ols. The
alkyne was dissolved in THF (to a concentration of ca. 0.3 mol/
L), and Red-Al (1.5 equiv), dissolved in THF (at a concentration
of ca. 0.6 mol/L), was added dropwise at room temperature.
The reaction mixture was stirred at reflux for 14 h (except for
3,3-dimethyl-1-butyne where the reaction was done at room
temperature to avoid allene formation) after which time H2SO4
(10%) was poured slowly into the solution precooled at 0 °C.
The aqueous phase was separated and extracted with ether,
and the combined organic phases were washed with sat.
sodium carbonate solution, dried over magnesium sulfate,
filtered, concentrated in vacuo, and purified by flash chroma-
tography on silica gel (hexanes-EtOAc 15:1) to give allyl
alcohol.
Ad d u ct (S)-7g. Yield: 75%; 1H NMR (CDCl3, 300 MHz): δ
7.32-7.17 (m, 5H), 5.24 (d, 1H, J ) 9.2 Hz), 3.77-3.72 (m,
1H), 2.41 (dm, 1H, J ) 8.9 Hz), 1.95 (octet, 1H, J ) 6.7 Hz),
1.85-1.65 (m, 4H), 1.63-1.54 (m, 1H), 1.45-1.25 (m, 1H), 1.30
(d, 3H, J ) 6.6 Hz), 1.18-1.07 (m, 1H), 0.90-0.85 (m, 9H);
13C NMR (CDCl3, 75 MHz): δ 148.2 (s), 138.7 (s), 128.3 (d), 2
× 127.5 (d), 126.9 (d), 2 × 125.6 (d), 51.1 (d), 37.2 (d), 35.2 (t),
32.4 (d), 31.7 (t), 26.6 (t), 26.5 (d), 22.9 (q), 22.0 (q), 20.6 (q),
19.8 (q); IR (neat, cm-1): 3015, 2974, 2872, 1595, 1446; LRMS
(m/z (relative intensity)): 256 (M+, 30), 213 (100), 157 (50),
131 (65); Exact mass calcd for C19H28
: 256.2191. Found:
2-Isop r op yl-5-m et h yl-1-{2-p h en yl-1-et h en -1-yl}cyclo-
256.2195. [R]20D: +47.3 (c 1.90, CHCl3)
1
h exa n -1-ol (5i). Yield: 86%; H NMR (CDCl3, 300 MHz): δ
7.41-7.19 (m, 5H), 6.64 (d, 1H, J ) 16.2 Hz), 6.22 (d, 1H, J )
16.2 Hz), 2.05-1.94 (m, 1H), 1,87-1.72 (m, 2H), 1.62-1.37 (m,
4H), 1.31-1.07 (m, 3H), 0.98-0.87 (m, 9H); 13C NMR (CDCl3,
75 MHz): δ 138.2 (d), 137.2 (s), 2 × 128.4 (d), 127.0 (d), 126.5
(d), 2 × 126.2 (d), 76.6 (s), 49.7 (d), 49.2 (t), 35.0 (t), 27.8 (d),
27.1 (d), 23.8 (q), 22.0 (q), 20.9 (t), 18.6 (q); IR (neat, cm-1):
Ad d u ct (R)-7g. Yield: 61%; 1H NMR (CDCl3, 300 MHz): δ
7.28-7.12 (m, 5H), 5.24 (d, 1H, J ) 9.2 Hz), 3.77-3.68 (m,
1H), 2.36 (dd, 1H, J ) 13.3, 4.2 Hz), 1.94 (octet, 1H, J ) 7.8
Hz), 1.82-1.54 (m, 5H), 1.45-1.35 (m, 1H), 1.32 (d, 3H, J )
7.3 Hz), 1.18-1.07 (m, 1H), 0.95-0.80 (m, 9H). [R]20D: -102.9
(c 0.99, CHCl3)

Chiral R-Substituted Carbonyls and Alcohols
J . Org. Chem., Vol. 65, No. 21, 2000 7097
Gen er a l P r oced u r e for th e Ozon olysis of th e Cu p r a te
Ad d u cts to Alcoh ols. The alkenes were dissolved in dichlo-
romethane (0.1 M), the solution was cooled to -78 °C, and
ozone was bubbled through. When the solution remained light
blue, indicating excess ozone, the ozone flow was stopped and
N2 was bubbled to remove excess of ozone. Then, dichloro-
methane was evaporated, the ozonide was dissolved in MeOH
(0.1M), and sodium borohydride was added (5 equiv) at 0 °C.
The resulting slurry was stirred at room temperature for 10
h. HCl (1 N) was then slowly added to the mixture cooled at 0
°C to destroy the excess reagent. Water and ether were added,
the aqueous phase was separated and extracted with ether,
and the combined organic phases were washed with brine,
dried over magnesium sulfate, filtered, concentrated in vacuo.
and purified by flash chromatography on silica gel (hexanes-
EtOAc 4:1) to give the desired alcohol.
However, the [R]20 of the alcohol 12g obtained from the
D
reduction of the crude aldehyde was identical with that of the
alcohol 12g obtained by direct NaBH4 reduction of the ozonide.
Gen er a l P r oced u r e for th e Ozon olysis of th e Cu p r a te
Ad d u cts to Acid s. Same procedure as above. After bubbing
N2, the solvent was evaporated and the ozonide was dissolved
in a mixture of EtOAc, H2O2 (30%) and water (2.5:1:1 respec-
tively) (0.1 M) and heated at 60 °C for 16 h. Then the solution
was cooled to room temperature, and NaOH (2 N) was added
followed by addition of ether. The organic and aqueous phase
were separated, and the aqueous was washed with ether and
acidified with HCl (12 N) to pH ) 1. The aqueous phase was
extracted with ether, and the combined organic phases were
dried over magnesium sulfate, filtered, and concentrated in
vacuo to give the desired acids.
(R)-2-P h en ylp r op a n oic Acid (11g). Yield: 70%. [R]20
-71.8 (c 1.57, CHCl3), lit. [R]20D: -76.5 (neat).42
:
D
(R)-2-P h en ylp r op a n ol (12g). Yield: 75%. [R]20D: +14.3 (c
1.65, CHCl3), lit. [R]20D: +15.8 (neat).41
Ack n ow led gm en t. We are grateful to the Natural
Sciences and Engineering Research Council of Canada
and the Universite´ de Sherbrooke for financial support.
Gen er a l P r oced u r e for th e Ozon olysis of th e Cu p r a te
Ad d u cts to Ald eh yd es. Same procedure as above. After
bubbing N2, triphenylphosphine was added (1 equiv) at -78
°C, and the resulting solution was stirred at room temperature
for 2 h. Dichloromethane was evaporated, and the residue was
purified by flash chromatography on silica gel (hexanes-
EtOAc 15:1) to give the desired aldehyde
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures, characterization data, and NMR spectra for new
compounds and X-ray data and ORTEP for 9. This material
is available free of charge via the Internet at http://pubs.acs.org.
(R)-2-P h en ylp r op a n a ld eh yd e (8g). Yield: 86%. [R]20
:
D
product racemizes upon flash chromatography on silica gel.
J O000810T
(41) J anssen, A. J . M.; Klunder, A. J . H.; Zwanenburg, B. Tetrahe-
dron 1991, 47, 7645.
(42) Mosher, H. S.; Aaron, C.; Dull, D.; Schmiegel, J . L.; J aeger, D.;
Ohashi, Y. J . Org. Chem. 1967, 32, 2797.