A chiral auxiliary cleavable by ring-closing alkene metathesis - Efficient synthesis of chiral nonracemic cycloalkenes

Article abstract of DOI:10.1139/V06-095

p-Menthane-3-carboxaldehyde is a readily available chiral auxiliary used to prepare cycloalkenes and heterocycles bearing a chiral tertiary or quaternary carbon of high enantiomeric purity. The auxiliary is available in both enantiomeric forms and is inexpensive and recyclable. It is cleaved by a ring-closing alkene metathesis reaction directly yielding the cycloalkene.

Full text of DOI:10.1139/V06-095

1290
A chiral auxiliary cleavable by ring-closing alkene
metathesis — Efficient synthesis of chiral
nonracemic cycloalkenes1
Luc Boisvert, Francis Beaumier, and Claude Spino
Abstract: p-Menthane-3-carboxaldehyde is a readily available chiral auxiliary used to prepare cycloalkenes and
heterocycles bearing a chiral tertiary or quaternary carbon of high enantiomeric purity. The auxiliary is available in
both enantiomeric forms and is inexpensive and recyclable. It is cleaved by a ring-closing alkene metathesis reaction
directly yielding the cycloalkene.
Key words: chiral auxiliary, cleavage reaction, cyclization, ring-closing alkene metathesis, enantioenriched cycloalkenes.
Résumé : Le p-menthane-3-carboxaldéhyde est utilisé comme auxiliaire chiral pour fabriquer des cycloalcènes et des
hétérocycles contenant un centre carboné tertiaire ou quaternaire de pureté énantiomérique élevée. L’auxiliaire est dis-
ponible dans les deux séries énantiomériques et est peu dispendieux et recyclable. Il est clivé par une réaction de fer-
meture de cycle par métathèse d’alcènes menant directement au cycloalcène.
Mots clés : auxiliaire chiral, réaction de clivage, cyclisation, fermeture de cycle par métathèse d’alcènes, cycloalcènes
énantioenrichis. Boisvert et al. 1293
Introduction
knowledge, this constitutes the first example of a chiral aux-
iliary cleavable by a RCM reaction (5, 6). Using our strat-
egy, the cleavage of the auxiliary becomes intrinsic to the
synthetic strategy and does not add extra steps en route to
the final target.
Chiral auxiliaries are exceptionally useful tools in syn-
thetic organic chemistry because many of them achieve high
levels of asymmetric induction on a wide range of substrates
(1). One inherent drawback to their use is the need to install
and cleave the auxiliary, which adds two chemical steps to
the synthesis of the target molecule. This disadvantage can,
however, be minimized if one of these two steps consists of
a chemical transformation that would have been carried out
eventually as part of the synthetic plan. This is not the case
for the majority of chiral auxiliaries found in the literature
that are cleaved either by addition–elimination reactions on
polarized π systems (e.g., hydride reduction of C=O func-
tional groups) or by hydrolysis of acetal-like functionalities
(1). In addition, while other auxiliary-specific cleaving reac-
tions have been developed (e.g., ozonolysis of C=C bonds
(2) or hydrogenolysis of benzylic groups (3)), there is still a
need to widen the arsenal of available methods to cleave
chiral auxiliaries.
Synthesis of RCM precursors3
The sequence of reactions starts with the synthesis of al-
lylic alcohols 7a–7e from p-menthane-3-carboxaldehyde 3
(7). These allylic alcohols were prepared by one of three
methods (Scheme 1 and Table 1).
Method A involves the AlMe3-promoted addition of
vinyllithium reagents 6a to aldehyde 3 (8). In this manner, a
mixture of diastereomeric allylic alcohols, which are easily
separated by flash chromatography, is obtained with
selectivities ranging from ca. 10:1 to more than 100:1,
favouring the Felkin–Ahn adducts (i.e., 7 as illustrated).
Method B involves the addition of vinylalanes 6b, generated
by the zirconium-catalyzed methylalumination of alkynes (9)
to aldehyde 3. In Method C, lithium acetylides 2 are added
to aldehyde 3 in the presence of dry cerium trichloride to af-
ford a mixture of diastereomeric propargylic alcohols 4 (10).
In this report, we disclose the cleavage of a chiral auxil-
iary by ring-closing alkene metathesis (RCM) (4) to yield
highly enantioenriched cycloalkenes. To the best of our
Received 19 October 2005. Accepted 13 December 2005. Published on the NRC Research Press Web site at http://canjchem.nrc.ca
on 28 July 2006.
This article is dedicated to Dr. Alfred Bader for his precious contributions to the science of chemistry.
L. Boisvert, F. Beaumier, and C. Spino.2 Université de Sherbrooke, Département de Chimie, 2500 Boul. Université, Sherbrooke,
QC J1K 2R1, Canada.
1This article is part of a Special Issue dedicated to Dr. Alfred Bader.
2Corresponding author (e-mail: claude.spino@USherbrooke.ca).
3Supplementary data for this article are available on the journal Web site (http://canjchem.nrc.ca) or may be purchased from the
Depository of Unpublished Data, Document Delivery, CISTI, National Research Council Canada, Ottawa, ON K1A 0R6, Canada.
DUD 5067. For more information on obtaining material refer to http://cisti-icist.nrc-cnrc.gc.ca/irm/unpub_e.shtml.
Can. J. Chem. 84: 1290–1293 (2006)
doi:10.1139/V06-095
© 2006 NRC Canada

Boisvert et al.
1291
Scheme 1.
Scheme 2.
PivO
R2
Pr
H
i-
i-Pr
i-Pr
R1 R2
H
H
MgBr
R'O
Pr
i-
n
R2
R2
t-BuLi
Et2O
Me
OHC
R1
H
3
n = 1,2
R1
I
M
R1
R1
6a M = Li
6b M = AlMe2
Method A
AlMe3
n
5
CuCN, Et2O
41%–91%
Me
Me
Me
8a–8d
9a–9g
PivCl, Et3N
CH2Cl2, 86%–100%
7a–7e R' = H
8a–8e R' = Piv
Method B
Cp2ZrCl2
AlMe3, CH2Cl2
Scheme 3.
Red-Al
THF
i-
Pr
R2
R1
HO
Pr
i-
Solvent, heat
H
H
BuLi
n-
THF
9a–9g
+
3
R1
R1
Li
R1
CeCl3
Method C
MesN
NMes
Cl
1
2
n
Me
4
Me
Ru
10a–10g
12
Ph
Cl
PCy3
Table 1. Preparation of allylic alcohols 7a–7e (cf. Scheme 1).
11
Entry R1
R2
Compound Yield of 7a (%), (dr)
1
2
Bn
TBSOCH2
H
H
7a
7b
64,b 170:1
3). The only detectable (1H NMR) products in the crude re-
action mixtures were the cycloalkenes 10 and compound 12.
Cyclization of precursors of six-membered rings bearing a
tertiary carbon center (9b and 9d) necessitated harsher con-
ditions (11 (10 mol%), ClCH2CH2Cl, 83 °C, 3 h) to avoid
the formation of the corresponding dimers 13b and 13d by a
competitive cross-metathesis reaction (Fig. 1). Under these
conditions, the RCM reactions occurred smoothly to provide
cycloalkenes 10b and 10d in good yields (Table 2, entries 2
and 4), with only trace amounts of 13 being detected.
Adducts 9e–9g, bearing a quaternary carbon, presented a
greater challenge. Minimizing the production of dimer 13
was crucial because when 13e (derived from 9e) was treated
under RCM conditions (with or without an atmosphere of
ethylene), no useful amount of 10e could be obtained. Ulti-
mately, it was found that higher dilution and temperatures
were efficient in decreasing the amount of 13. However,
these conditions sometimes caused side-reactions, like the
well-documented alkene isomerization (13), such that
cycloalkenes 10e and 10g were contaminated with small
amounts of inseparable alkene regioisomers 14e and 14g, re-
spectively (Fig. 1).
Many strategies were tried to inhibit the formation of 14
(see Supplementary data),3 but none gave reproducible re-
sults. Although the isomerization could not be completely
suppressed, cycloalkene 10e could be obtained in 79% yield
as an acceptable 31:1 mixture of 10e and 14e by maintaining
strictly anhydrous conditions and limiting the reaction time.
In the same manner, 10g was obtained in 70% yield as a
42:1 mixture of 10g and 14g. The fact that the RCM pro-
ceeds at all on substrates 9e and 9g is very satisfying, given
the presence of an allylic quaternary center and a bulky
menthyl fragment on each side of the E double bond (14).
It is perhaps not surprising, given the above results, that
we have not yet succeeded in forming six-membered rings
bearing a quaternary carbon. Compound 9f afforded mostly
dimer 13f and degradation products under various condi-
tions; RCM adduct 10f was never observed, nor was by-
product 12 (Table 2, entry 6).
43,c 3.5:1d
3
4
Bn
Me 7c
76,e 11:1
68,e 11:1
37,c 3.1:1d
TBSO(CH2)4 Me 7d
5
MOMOCH2 7e
H
aIsolated yield of diastereomerically pure allylic alcohol 7.
bMethod A.
cMethod C (yield of 7 over two steps).
dDR of 4 from the addition of 2 to 3.
eMethod B.
These propargylic alcohols are easily separated and con-
verted to geometrically pure E allylic alcohols 7 by reduc-
tion with Red-Al. The alcohols 7a–7e were then converted
to the corresponding allylic pivalate esters 8a–8e in excel-
lent yields.
Cyanocuprates, derived from Grignard reagents, displaced
the pivalate esters stereospecifically (anti to the leaving
group with complete transfer of chirality) with 100% stereo-
selectivity (reaction on the conformer with minimized A1,3
strain, yielding E alkenes) and with exclusive SN2′ regio-
selectivity (no SN2 adduct detected by GC) (7, 11). Thus,
cuprate adducts 9a–9g (precursors to five- and six-
membered rings for the subsequent RCM reactions) were
prepared from pivalate esters 8a–8d (Scheme 2). It should
be noted that both absolute stereochemistries at the newly
created chiral center in adducts 9 are accessible since both
enantiomers of p-menthane-3-carboxaldehyde 3 are avail-
able, the double-bond geometry in 8 can be controlled, and
the order of introduction of the different substituents on the
new chiral center can be varied (7).
RCM reactions
After screening different ruthenium-based metathesis cata-
lysts (see Supplementary data),3 it was found that the forma-
tion of cycloalkenes 10 by a RCM reaction was best
performed using the Grubbs–Nolan catalyst 11 (12)
(Scheme 3). For precursors of five-membered rings bearing
a tertiary carbon center (9a and 9c) very mild conditions (11
(1 mol%), CH2Cl2, 40 °C, 3 h) resulted in smooth transfor-
mation to cycloalkenes 10a and 10c (Table 2, entries 1 and
The enantiomeric purity (determined by GC or HPLC
analysis against racemic samples) of each RCM adduct was
© 2006 NRC Canada

1292
Can. J. Chem. Vol. 84, 2006
Table 2. Yields and enantiomeric ratios of cycloalkenes 10a–10g.
Entry
9
R1
R2
n
10
Yield of 10 (%)a
er (%)
b
1
2
3
9a
9b
9c
Bn
Bn
TBSOCH2
H
H
H
1
2
1
10a
10b
10c
81
84
87
—
>99:1c
>98:2d
>98:2d
4
9d
TBSOCH2
H
2
10d
73
5
6
7
9
9f
9g
Bn
Bn
Me
Me
Me
1
2
1
10
10f
10g
79b
0
70b
97:3e
—
—
b
TBSO(CH2)4
Note: General conditions: 11 (1–10 mol%), CH2Cl2 or ClCH2CH2Cl (0.01–0.002 mol/L), reflux, 3 h (see Supplementary
data).3
aIsolated yield of 10 after flash chromatography.
bSee text.
cDetermined by HPLC analysis.
dDetermined by GC analysis on the free alcohol.
eDetermined by HPLC analysis on the allylic oxidation derivative.
Scheme 4.
Fig. 1. By-products from RCM reactions.
Pr
i-
Me
H
Ar
1. ArMgBr, CuCN, 94%
2. HCl, MeOH, 97%
H
H
Ar
11 (5 mol%)
R2
i-Pr
R1 R2
R1
8e
H
CH2Cl2, 40oC
81%, >99%
O
3. AllylBr, NaH, 92%
O
n
ee
n
H
n-1
Me
R1 R2
i-Pr
Ar = -MeOPh
m
15
16
14
13
Me
( )n
NHBoc
11
(1 mol%)
Boc
( )n
N
a
R = Bn, n = 1
R = -Bu, n = 0
t
n-Pr, n = 1
R
b
c R =
found to be excellent. For entries 1 and 7 of Table 2, we
were unable to achieve complete resolution of the
cycloalkene enantiomers, but their enantiomeric ratios are
likely to be equally high. By-product 12 can be easily recov-
ered by flash chromatography in yields of 70%–80% and, if
desired, can be recycled back to aldehyde 3 by ozonolysis.
It is possible to envisage the application of our strategy to
the preparation of a wide variety of highly enantioenriched
cycloalkenes. Among the most interesting candidates,
heterocycles are highly desirable (15). As an example, diene
15 (obtained in three steps from pivalate ester 8e) was sub-
mitted to a RCM reaction to furnish dihydropyran 16 in 81%
yield and >99% ee (as determined by HPLC analysis)
(Scheme 4). N-Heterocycles, prepared by a different route
(16), were also obtained by a similar RCM cleavage strategy
as shown in Scheme 4. Dehydropiperidine 18a and 18c,
pyrrolidine 18b, and dihydroquinoline 20 were formed in ex-
cellent yield under much milder conditions (1 mol% catalyst
in refluxing dichloromethane). Dihydropyran 22 required
higher catalyst loading and higher temperature. We are cur-
rently embarked on the total synthesis of several alkaloids
using this methodology.
CH2Cl2, 40oC
, >99%
R
-Pr
i
82%–96%
ee
17a–17c
NHBoc
18a–18c
11
(1 mol%)
BocHN
CH2Cl2, 40oC
99%, >99%
-Pr
19
i
ee
20
-Pr
NHBoc
O
NHBoc
i
11
(5 mol%)
DCE,reflux
81%, >99%
O
ee
21
22
We are currently expanding the methodology to include
more diverse and complex structures.
Acknowledgement
We thank the Natural Sciences and Engineering Research
Council of Canada (NSERC), the ACS Petroleum Research
Fund (Grant No. 36256-AC1), and the Université de
Sherbrooke for financial support.
Conclusion
We have achieved the first cleavage of a chiral auxiliary
by a RCM reaction. This unprecedented method of cleavage
has the merit of increasing the complexity of the substrate
during the cleavage of the auxiliary, while not adding any
extra steps to a synthetic plan aimed at synthesizing
cycloalkenes. The sequence described herein allows the syn-
thesis of enantioenriched cycloalkenes bearing a tertiary or
quaternary chiral carbon in just four steps starting from
readily available starting materials (vinyliodides or alkynes).
References
1. (a) J. Seyden-Penne. Chiral auxiliaries and ligands in asym-
metric synthesis. John Wiley and Sons, New York. 1995;
(b) R.E. Gawley and J. Aubé. Principles of asymmetric synthe-
sis. Elsevier Science, New York. 1996.
2. (a) C. Spino, C. Beaulieu, and J. Lafrenière. J. Org. Chem. 65,
7091 (2000); (b) M. Braun. Angew. Chem. Int. Ed. 37, 430
(1998).
© 2006 NRC Canada

Boisvert et al.
1293
3. Y. Aoyagi, R.P. Jain, and R.M. Williams. J. Am. Chem. Soc.
123, 3472 (2001).
hedron Lett. 37, 6787 (1996); (c) D.A. Conlon, D. Kumke, C.
Moeder, M. Hardiman, G. Hutson, and L. Sailer. Adv. Synth.
Catal. 346, 1307 (2004); (d) P.E. Harrington and M.A. Tius. J.
Am. Chem. Soc. 123, 8509 (2001).
4. Recent reviews: (a) R.H. Grubbs (Editor). Handbook of me-
tathesis. Edited by Wiley-VCH, Weinheim, Germany. 2003;
(b) K.C. Nicolaou, P.G. Bulger, and D. Sarlah. Angew. Chem.
Int. Ed. 44, 4490 (2005); (c) R.H. Grubbs. Tetrahedron, 60,
7117 (2004); (d) A.H. Hoveyda, D.G. Gillingham, J.J. Van
Veldhuizen, O. Kataoka, S.B. Garber, J.S. Kingsbury, and
J.P.A. Harrity. Org. Biomol. Chem. 2, 8 (2004).
5. Although we know of no other RCM-cleavable chiral auxil-
iary, the underlying concept is similar to RCM-cleavable link-
ers developed for solid-phase synthesis. For example, see:
(a) M.S.M. Timmer, J.D.C. Codée, H.S. Overkleeft, J.H. van
Boom, and G.A. van der Marel. Synlett, 2155 (2004); (b) J.-D.
Moriggi, L.J. Brown, J.L. Castro, and R.C.D. Brown. Org.
Biomol. Chem. 2, 835 (2004).
11. For the use of CuCN in the SN2′ displacement of allylic esters
with Grignard reagents, see: C.C. Tseng, S.D. Paisley, and
H.L. Goering. J. Org. Chem. 51, 2884 (1986).
12. (a) J. Huang, E.D. Stevens, S.P. Nolan, and J.L. Peterson. J.
Am. Chem. Soc. 121, 2674 (1999); (b) M. Scholl, T.M. Trnka,
J.P. Morgan, and R.H. Grubbs. Tetrahedron Lett. 40, 2247
(1999).
13. For recent discussions, see: (a) C.D. Edlin, J. Faulkner, D.
Fengas, C.K. Knight, J. Parker, I. Preece, P. Quayle, and S.N.
Richards. Synlett, 572 (2005); (b) S.H. Hong, M.W. Day, and
R.H. Grubbs. J. Am. Chem. Soc. 126, 7414 (2004); (c) B.
Schmidt. Eur. J. Org. Chem. 1865 (2004).
6. While RCM reactions on dienes attached to a chiral auxiliary
are numerous, in these systems the auxiliary is cleaved by
transformations different from the RCM reaction. For exam-
ple, see: M.T. Crimmins and B.W. King. J. Org. Chem. 61,
4192 (1996).
7. (a) C. Spino and C. Beaulieu. Angew. Chem. Int. Ed. 39, 1930
(2000); (b) C. Spino, C. Godbout, C. Beaulieu, M. Harter,
T.M. Mwene-Mbeja, and L. Boisvert. J. Am. Chem. Soc. 126,
13312 (2004).
8. C. Spino, M.-C. Granger, and M.-C. Tremblay. Org. Lett. 4,
4735 (2002).
9. E.-I. Negishi and D.Y. Kondalov. Chem. Soc. Rev. 25, 417
(1996).
14. Successful RCM reactions on such sterically demanding sub-
strates are rare in the literature. For RCM on substrates with
the same general features, see: (a) S. Suga, M. Watanabe, and
J.-I. Yoshida. J. Am. Chem. Soc. 124, 14824 (2002); (b) E.
Roulland, C. Monneret, J.-C. Florent, C. Bennejean, P. Renard,
and S. Léonce. J. Org. Chem. 67, 4399 (2002); (c) Y. Chong,
G. Gumina, and C.K. Chu. Tetrahedron: Asymmetry, 11, 4853
(2000). In all cases, the terminal alkyl group (opposite a qua-
ternary center) on the E alkene is much less hindered than in
our case (i.e., menthyl fragment) and the RCM proceeds much
more easily than here, allowing the use of the less active cata-
lyst, Cl2(Cy3P)2Ru=CHPh.
15. For the synthesis of heterocycles by RCM and other methods,
see the thematic issue on heterocycles: Chem. Rev. 104(5),
(2004).
16. D. Gagnon, S. Lauzon, C. Godbout, and C. Spino. Org. Lett. 7,
4769 (2005).
10. The use of carefully dried CeCl3 was beneficial in increasing
both the yields of the propargylic alcohols 4 and the Felkin–
Ahn selectivity. For discussions on the drying of CeCl3 and its
effects, see: (a) N. Takeda and T. Imamoto. Org. Synth. 76,
228 (1999); (b) V. Dimitrov, K. Kostova, and M. Genov. Tetra-
© 2006 NRC Canada