Low-temperature n-butyllithium-induced rearrangement of allyl 1,1-dichlorovinyl ethers

Article abstract of DOI:10.1021/ol052685j

Upon treatment with n-BuLi at low temperatures, a variety of allyl 1,1-dichlorovinyl ethers 2 undergo rearrangement to furnish γ,δ- unsaturated esters 3 after alcohol addition. Compounds containing quaternary centers (3e: R₁ = H, R₂,

Full text of DOI:10.1021/ol052685j

ORGANIC
LETTERS
2006
Vol. 8, No. 3
451-454
Low-Temperature n-Butyllithium-Induced
Rearrangement of Allyl 1,1-Dichlorovinyl
Ethers
Aaron Christopher, Dahniel Brandes, Stephen Kelly, and Thomas Minehan*
Department of Chemistry, California State UniVersitysNorthridge,
Northridge, California 91330
thomas.minehan@csun.edu
Received November 4, 2005
ABSTRACT
Upon treatment with n-BuLi at low temperatures, a variety of allyl 1,1-dichlorovinyl ethers 2 undergo rearrangement to furnish
esters 3 after alcohol addition. Compounds containing quaternary centers (3e: R1 H, R2, R3 ) −(CH2)5 ; 3f: R1 H, R2
(CH2)2CH(CH3)2) may be formed in high yield and under mild conditions utilizing this protocol. The reaction is stereospecific and may be
applied to the preparation of -disubstituted lactones.
2,3 â-C-glycosides and
γ
,
δ
-unsaturated
)
−
)
)
CH3, R3
)
∆
-
r,â
The [3,3] sigmatropic rearrangement of allyl vinyl ethers (the
Claisen rearrangement) is a powerful strategy for forming
carbon-carbon bonds that has been extensively used in the
synthesis of complex molecules (Figure 1, eq 1).1 Uncata-
significantly lower temperatures under conditions that allow
excellent stereocontrol and that are compatible with complex
substrates containing thermally sensitive functionality.2
Despite the enormous practical utility of these processes, the
[3,3]-sigmatropic rearrangement of the analogous allyl alkyn-
yl ethers has not been explored extensively (Figure 1, eq 2).
Early investigations by Arens3 and Schmid4 revealed that
substituted and unsubstituted benzyl alkynyl ethers undergo
sigmatropic rearrangement followed by intramolecular ketene
trapping to form indanones in high yield. Although Arens
refers to the rapid low-temperature rearrangements of allyl
alkynyl thioethers,5 to our knowledge, there have been no
formal reports of aliphatic [3,3]-sigmatropic rearrangements
of allyl alkynyl ethers. In this paper, we describe our attempts
to study this potentially useful process.
Figure 1. [3,3]-Sigmatropic rearrangements.
(2) (a) Wipf, P. ComprehensiVe Organic Synthesis; Trost, B. M., Fleming,
I., Eds.; Pergamon: Oxford, 1991; Vol. 5, p 827. (b) Wilson, S. R. Organic
Reactions; John Wiley & Sons: New York, 1993; Vol. 43, p 93. (c)
Paquette, L. A. Tetrahedron 1997, 53, 13971.
(3) Olsman, H.; Graveland, A.; Arens, J. F. Rec. TraV. Chim. Pays-Bas
1964, 83, 301.
lyzed Claisen rearrangements typically require heating of
allyl vinyl ether substrates in excess of 100 °C. Variants of
the allyl vinyl ether rearrangement can be performed at
(4) Wunderli, A.; Zsindely, J.; Hansen, H. J.; Schmid, H. Chimia 1972,
26, 643.
(5) Brandesma, L.; Bos, H. J.; Arens, J. F. In The Chemistry of
Acetylenes; Viehe, H. G., Ed.; Marcel Dekker: New York, 1969; pp 751-
860.
(1) For reviews of the Claisen Rearrangment, see: (a) Ziegler, F. E.
Chem. ReV. 1988, 88, 1423. (b) Ito, H.; Taguchi, T. Chem. Soc. ReV. 1999,
28, 43. (c) Hiersemann, M.; Abraham, L. Eur. J. Org. Chem. 2002, 1461.
(d) Nubbemeyer, U. Synthesis 2003, 961.
10.1021/ol052685j CCC: $33.50
© 2006 American Chemical Society
Published on Web 01/05/2006

A current practical method for the preparation of terminal
and internal alkynyl ethers involves treatment of dichloro-
vinyl ethers with excess n-BuLi followed by quenching with
an alcohol or an electrophile.6,7 We thus decided to synthesize
allyl dichlorovinyl ethers as direct precursors of allyl alkynyl
ethers. Greene’s method for forming 1,2-dichlorovinyl ethers,
involving reaction of potassium alkoxides with trichloro-
ethylene,8 proved unsatisfactory when applied to allylic
alcohols.9 Instead, Bruckner’s mild two-step protocol involv-
ing alcohol formylation (acetic formic anhydride10/pyridine)
and subsequent dichlorovinylation (PPh3/CCl4) furnished high
yields of allyl 1,1-dichlorovinyl ethers 211 from a variety of
allylic alcohols 1 (Scheme 1).12 Addition of 2.2 equiv of
Table 1. Scope of the Rearrangement of Allyl
1,1-Dichlorovinyl Ethers
entry
2
3
R1
R2
R3
yield (%)
1
2
3
4
5
6
7
2a 3a
2b 3b
2c 3c
2d 3d
2e 3e
2f 3f
2g 3g
H
H
H
H
H
C5H11
t-BuSiPh2OCH2
H
H
91
90
95
92
85
75
t-BuSiPh2OCH2
-(CH2)3-
H
H
H
-(CH2)5-
CH3
H
(CH3)2CH(CH2)2
Ph
Scheme 1. Rearrangement of Substrate 2a
same reaction conditions gave <10% of the rearranged
homoallylic esters, providing instead a cis/trans mixture of
monodebromination products.14 Dichlorovinyl ether 2g de-
rived from cinnamyl alcohol (Table 1, entry 7) also failed
to furnish useful amounts of rearranged product, likely
because of intramolecular allylic deprotonation by the
initially formed vinyl anion.15
In light of these results, we explored the stereospecificity
of this process by independently subjecting diastereomeri-
cally pure dichlorovinyl ethers to the rearrangement protocol.
Upon treatment with n-BuLi followed by ethanol quench,
cis-carvyl dichlorovinyl ether 5a produced rearranged cis-
ethyl ester 6a exclusively, while trans-carvyl dichlorovinyl
ether 5b furnished rearranged trans-ethyl ester 6b exclusively
(Scheme 2).16 These results indicate that the rearrangement
n-BuLi to a THF solution of allyl 1,1-dichlorovinyl ether
2a at -78 °C, followed by warming to -40 °C and
quenching with excess ethanol, gave, after aqueous workup,
not the expected terminal alkynyl ether 4a but rather
rearranged ester 3a in 91% yield.13 Quenching the reaction
mixture with methanol or menthol instead of ethanol gave
rise to the corresponding rearranged esters 3am and 3an in
85% and 72% yields, respectively.
Subjection of a variety of cyclic and acylic allyl dichlo-
rovinyl ethers to these reaction conditions gave similarly high
yields of rearrangement products (Table 1); notably, com-
pounds containing quaternary centers (Table 1, entries 5 and
6) can be formed efficiently using this protocol. Interestingly,
exposure of the corresponding dibromovinyl ethers to the
Scheme 2. Stereospecific Rearrangements
(6) For a review of methods for the synthesis of ynol ethers, see: Stang,
P. J.; Zhdankin, V. V. The Chemistry of Triple-Bonded Functional Groups;
Patai, S., Ed.; John Wiley & Sons: New York, New York, 1994; Chapter
19.
(7) (a) Smithers, R. H. Synthesis 1985, 556. (b) Himbert, G.; Loffler, A.
Synthesis 1992, 495.
(8) Moyano, A.; Charbonnier, F.; Greene, A. E. J. Org. Chem. 1987,
52, 2, 2919.
is indeed highly stereospecific, with carbon-carbon bond
(9) Only trace amounts of 1,2-dichlorovinyl ethers were formed using
the Greene protocol, with starting allylic alcohols recovered in >90% yield
from the reaction mixture.
(14) A similar result was noted by Bruckner in an attempted synthesis
of alkynylsulfonamides from dibromovinyl sulfonamides; see ref 12.
(15) Treatment of 2g with n-BuLi at -78 °C gave a dark-colored solution
(presumably due to formation of a highly delocalized carbanion), which
upon ethanol quench furnished a cis/trans mixture of monodechlorination
products.
(16) (a) 5a and 5b were obtained from cis- and trans-carveol, respec-
tively, by treatment with acetic formic anhydride/pyridine followed by PPh3/
CCl4/THF (see supporting info for details). cis-Carveol was synthesized
by stereoselective reduction of (R)-(-)-carvone: Mowery, M. E.; DeShong,
P. J. Org. Chem. 1999, 64, 1684. trans-Carveol was prepared by Mitsunobu
inversion/hydrolysis of cis-carveol: Uesaka, N.; Saitoh, F.; Mori, M.;
Shibasaki, M.; Okamura, K.; Date, T. J. Org. Chem. 1994, 59, 5633. (b)
1H NMR data for 6a and 6b matched those reported by Bermejo: Rico,
R.; Zapico, J.; Bermejo, F.; Sanni, S. B.; Garcia-Granda, S. Tetrahedron:
Asymmetry 1998, 9, 293.
(10) Huffman, C. W. J. Org. Chem. 1958, 23, 727.
(11) The allyl 1,1-dichlorovinyl ethers prepared were stable at room
temperature and at 60 °C in THF under the conditions for their formation
from the corresponding formate esters. The [3,3]-sigmatropic rearrangement
of allyl 1,1-dichlorovinyl ethers at temperatures >100 °C has been
described: Morimoto, T.; Sekiya, M. Synthesis 1981, 308.
(12) Bruckner, D. Synlett 2000, 1402. See also: Lakhrissi, M.; Chapleur,
Y. J. Org. Chem. 1994, 59, 5752.
(13) (a) The use of LDA instead of n-BuLi under the same reaction
conditions failed to afford rearrangement products. (b) It was subsequently
found that warming the reaction to -40 °C before quench with ethanol
was unnecessary; high yields of rearranged products 3 are consistently
obtained after stirring the reaction mixture for 15-30 min at -78 °C
followed by ethanol quench and warming to room temperature (vide infra).
452
Org. Lett., Vol. 8, No. 3, 2006

formation occurring on the same face of the molecule as
carbon-oxygen bond cleavage.
Scheme 4. Rearrangement with Electrophile Incorporation
To explore the potential application of this process to the
synthesis of C-glycosides, we subjected glucal derivatives 7
and 9 to the rearrangement protocol (Scheme 3). Exposure
Scheme 3. C-Glycoside Synthesis
occurs suggests that a pathway different from that originally
proposed by Arens and Schmid for benzyl alkynyl ether
sigmatropy (Scheme 5, 15 f 16b)22 may be operative. In
Scheme 5. Possible Mechanism for the n-BuLi-Induced
of 3,4-dibenzyl glucal 7 to n-BuLi at -78 °C followed by
ethanol quench at -40 °C gave rise to ∆2,3-C-glycoside 8
with exclusive â-anomeric stereochemistry in 76% yield.17
In contrast, conformationally restricted glucal 9 gave none
of the desired rearrangement product when subjected to the
same conditions; TLC analysis indicated that a compound
other than 918 present at -40 °C after ethanol quench was
decomposing as the reaction mixture was warmed to ambient
temperatures. Clearly, geometric factors in the substrate are
important for the rearrangement, and this result points toward
the existence of a pathway involving a cyclic transition state.
For glucal derivatives, such a transition-state geometry may
only be achieved by ring flip, which is impossible for
substrate 9.
Rearrangement of Allyl 1,1-Dichlorovinyl Ethers
Finally, to assess if the anionic intermediate generated
upon n-BuLi treatment of allyl dichlorovinyl ethers would
react with electrophiles other than the proton, the rearrange-
ment reaction of 2c was performed with addition of either
methyl triflate (2.0 equiv) or TMSCl (2.0 equiv) at -40 °C
prior to addition of ethanol (Scheme 4). Indeed, R-substituted
ethyl esters 10a,b and 11a,b were isolated in 58% and 73%
yields, respectively; in each case a ∼1:1 mixture of diaster-
eomers was obtained.19 Exposure of 10a,b to 3% HCl in
CH3OH for 7 h at room temperature gave a separable 1.4:1
cis/trans mixture of racemic butyrolactones 12a and 12b.20
Although the mechanism of this rearrangement has not
been established,21 the low temperature at which the reaction
experiments to probe the reaction mechanism, it was found
that treatment of dichlorovinyl ether 2a with 3 equiv of
n-BuLi in THF at -78 °C for 10 min followed by addition
of excess ethanol gave, upon warming to room temperature,
a 10:1 mixture of rearranged product 3a and monochlorovinyl
ether 13b (Scheme 5, R1dC5H11). Treatment of 2a instead
with 1.5 equiv n-BuLi in THF at -78 °C for 10 min followed
by ethanol quench and warming to room temperature gave
a ∼1:1 mixture of 2a/3a; again, approximately 9% of the
reaction mixture consisted of monochlorovinyl ether 13b.
These results, and also the observation that the use of LDA
instead of n-BuLi in the reaction does not furnish rearrange-
ment products,12 suggest that halogen-metal exchange is a
likely first step in the mechanism. Once formed, the
alkenyllithium species 13a may undergo transformation to
another intermediate that consumes an additional 1 equiv of
n-BuLi before the next transmetalation can take place. A
possible reaction pathway, outlined in Scheme 5, involves
(17) In a two-dimensional NMR (NOESY) experiment performed on 8,
an NOE was observed between carbohydrate H1′ and H5′, thus confirming
the assignment of the â stereochemistry shown above. See the Supporting
Information for details. For a discussion of the stereochemical assignment
of 2,3-unsaturated C-glycosides, see: Brakta, M.; Farr, R. N.; Chaguir, B.;
Massiot, G.; Lavaud, C.; Anderson, W. R.; Sinou, D.; Daves, G. D. J. Org.
Chem. 1993, 58, 2992.
(18) Based on our proposed mechanism (Scheme 5, vide infra), we
suggest that this unstable species present at low temperature is an allyl
alkynyl ether (e.g., 15a, Scheme 5).
(19) Trapping the anionic intermediate with benzaldehyde (1.5 equiv, 1
h, -78 °C, followed by 1.5 equiv of TMSCl, -78 to +25 °C) was also
successful, giving rise to a mixture of aldol diastereomers.
(20) The stereochemistry of 12a and 12b was assigned by comparison
of their 1H NMR spectral data with literature values: Tamaru, Y.; Furukawa,
Y.; Mizutani, M.; Kitao, O.; Yoshida, Z. J. Org. Chem. 1983, 48, 3631.
(21) Attempts to trap the anionic intermediate by reaction with acetyl
chloride, tosyl chloride, ethyl chloroformate, TIPS-Cl, or TBDPS-Cl led
only to a complex mixture of unidentifiable products.
(22) See refs 3 and 4 above.
Org. Lett., Vol. 8, No. 3, 2006
453

formation of an allyl alkynyl ether 15a (via carbene 14),
followed by deprotonation (f 15b) and accelerated rear-
rangement,23 producing lithioketene 16a prior to addition to
ethanol; protonation (16 f 16b), ethoxide trap of the ketene,
and ester enolate protonation would furnish the product.
Experiments to elucidate the latter part of the pathway shown,
in particular to distinguish whether species 15a or 15b is
undergoing sigmatropic rearrangement, are currently under-
way.
This rearrangement reaction occurs at low temperatures
and under mild conditions, affording high yields of products
containing up to two new carbon-carbon bonds. A potential
advantage of this method over other variants of the Claisen
rearrangement2 is the ability to add a variety of alcohol
nucleophiles to the proposed ketene intermediate, allowing
the one-pot preparation of diverse esters of γ,δ-unsaturated
carboxylic acids (Scheme 1, 3a, 3am, and 3an). Further
studies on the utility and the scope of this reaction are in
progress.
Acknowledgment. We thank the ACS Petroleum Re-
search Fund (No. PRF 38271-GB1) and Research Corpora-
tion (No. CC6343) for their generous support of our program.
Mass spectra were provided by the UC Riverside mass
spectrometry department. We also thank the manuscript
reviewers for their helpful comments and Dr. Paul Shin
(CSUN) for helping perform two-dimensional NMR experi-
ments.
(23) As of yet we can suggest no obvious kinetic advantage for the
rearrangement to occur via the lithioalkynyl ether. A number of anion-
accelerated [3,3]-sigmatropic rearrangements are known and take place at
temperatures below their neutral counterparts: (a) Paquette, L. A. Tetra-
hedron 1997, 53, 13971. (b) Paquette, L. A. Angew. Chem., Int. Ed. Engl.
1990, 29, 609. (c) Hill, R. K. In ComprehensiVe Organic Synthesis; Trost,
B. M., Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 5, Chapter
7.1. (d) Wilson, S. R. Org. React. 1993, 43, 93.
Supporting Information Available: Experimental details
for all reactions as well as characterization data. This material
is available free of charge via the Internet at http://pubs.acs.org.
OL052685J
454
Org. Lett., Vol. 8, No. 3, 2006