.
Angewandte
Communications
DOI: 10.1002/anie.201303527
Synthetic Methods
Enantioselective Organocatalytic Fluorination-Induced Wagner–
Meerwein Rearrangement**
Fedor Romanov-Michailidis, Laure Guꢀnꢀe, and Alexandre Alexakis*
Despite the fact that only 21 fluorinated molecules are known
to be biosynthesized,[1] it is notable that about 20–25% of all
modern pharmaceuticals and agrochemicals incorporate at
least one fluorine atom.[2] Nevertheless, the practice of
introducing fluorine atoms into bioactive compounds is
rather recent, as the first synthetic fluorinated drug, namely
5-fluorouracil, was synthesized as late as in 1957.[3] Ever since,
the research aimed at incorporating fluorine atoms into small
organic molecules has attracted and intrigued synthetic
organic chemists.[4] Yet, despite the importance of fluorine,
carbon–fluorine bond formation remains a challenge.[5] Per-
forming this task in concert with bringing chirality into the
target molecule, and doing so with only a catalytic quantity of
the enantioinducing reagent is of even greater practical
importance.[6]
Halocyclization of olefins is an important class of organic
transformations. Among these reactions, halolactonizations
have been studied extensively and applied in the synthesis of
many bioactive molecules.[7] Very recently, catalytic enantio-
selective versions of the aforementioned transformation were
also reported.[8] Far less studied is the related halogenation/
semipinacol rearrangement cascade.[9] In this last reaction, an
allylic alcohol (1) undergoes a Wagner–Meerwein alkyl
migration, which is initiated by the formation of the halonium
ion intermediate 2 (Scheme 1). Whereas the chlorination- and
bromination-induced Wagner–Meerwein rearrangements of
electron-rich cyclic enol ethers were recently shown to be
amenable to asymmetric catalysis,[10,11] the development of
a truly enantioselective catalytic fluorination-induced variant
remains underexplored.[12] This is partly related to the
inherently high reactivity of most electrophilic fluorinating
reagents, and leaves little room for the introduction of a chiral
catalyst.[13]
In the course of the last ten years, it has been extensively
demonstrated that ionic catalysts incorporating at least one
chiral ion are able to render enantioselective those trans-
formations which proceed through reaction intermediates
bearing an opposite electrostatic charge.[14,15] Specifically in
the field of asymmetric-counteranion-directed catalysis,
binol-derived phosphoric acids have been disclosed as
privileged precursors of chiral anions.[16,17] Reasoning that
our postulated halonium ion intermediate 2 bears a net
positive charge, we were interested to see if a chiral anion
could induce asymmetry into the Wagner–Meerwein rear-
rangement. Among the four possible halogen atoms, we were
particularly attracted by fluorine as initiator for the Wagner–
Meerwein transposition. Furthermore, to the best of our
knowledge, chiral-binol-derived phosphoric acids have only
been used to promote semipinacol rearrangements of elec-
tron-rich cyclic enol ethers.[18] The transposition of simple
allylic alcohols remains a great challenge because it cannot be
initiated by a proton alone. Nevertheless, such a reaction
could be of great synthetic interest, as it would lead to the
formation of valuable all-carbon quaternary stereogenic
centres. Herein, we report some of our recent results on this
subject. Our catalytic system was inspired from the one
recently reported for related fluorocyclization reactions.[19]
Optimization studies were carried out with the strained
allylic alcohol A1, Selectfluor as the fluorinating reagent, and
a set of synthetic axially chiral phosphoric acids (L), derived
from (Ra)-binol (Table 1).
In the course of the preliminary catalyst screening, the
employment of highly sterically congested phosphoric acids
(L4–L8), which are related to the known (Ra)-TRIP scaf-
fold,[20] turned out to be crucial for accessing practical
enantioselectivities of the product b-fluoro spiroketone B1
(Table 1, entries 4–6). Interestingly, acids bearing isopropyl
(L4) and cyclopentyl (L6) substituents at positions X and Y
outperformed the acid L5 which bears cyclohexyl groups at
these same positions. Among the numerous solvents tested,
highly hydrophobic, yet strongly solubilizing solvents (tolu-
ene, fluorobenzene, and diisopropylether) were better than
hydrophobic solvents of lower solubilizing ability (cyclohex-
ane). Since nonpolar solvents favor ion pairing, the present
reaction is an example of anionic phase-transfer catalysis
(PTC), where a lipophilic chiral anion extracts the insoluble
Scheme 1. The concept of halonium-ion-promoted Wagner–Meerwein
transposition of allylic alcohols, and the idea of inducing chirality by
means of a chiral counterion. Hal=F, Cl, Br, or I.
[*] F. Romanov-Michailidis, Prof. Dr. A. Alexakis
Department of Organic Chemistry, University of Geneva
Quai Ernest Ansermet 30, 1211 Geneva 4 (Switzerland)
E-mail: Alexandre.Alexakis@unige.ch
Dr. L. Guꢀnꢀe
Laboratory of Crystallography, University of Geneva
Quai Ernest Ansermet 24, 1211 Geneva 4 (Switzerland)
[**] The authors thank the Swiss National Research Foundation (Grant
No. 200020-126663) and COST action CM0905 (SER Contract No.
C11.0108) for financial support.
Supporting information for this article is available on the WWW
2
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!