Asymmetric Allylic Alkylation
COMMUNICATION
activity. However, the corresponding benzyl ethers 2 f and
2d gave excellent SN2’ selectivity (>95:5) and e.r. (>95:5)
(Table 2, entries 15 and 16). The presence of a methyl sub-
stituent in the allylic ether 2 f, allows for the preparation of
a range of synthons with a methyl group at the stereocenter;
an important structural moiety that is present in many natu-
ral compounds.[18] The compatibility of two ether moieties in
the same molecule was demonstrated by the use of allyl sub-
strate 1g, which bears both an allylic-OMe and -OBn ether,
and 2g, which contains two allylic -OBn moieties (Table 2,
entries 17 and 18). Although the conversion was lower in
both cases in comparison with the previous results, the SN2’
selectivity observed was complete, with excellent enantio-
meric ratios. Surprisingly, in the case of 1g, with internal
competition between -OMe and -OBn moieties as leaving
group, we observed that substitution occurred towards the
-OMe group instead of the -OBn group (Scheme 2). Despite
Scheme 3. Synthesis of Arundic acid. Conditions: a) BnBr, NaH (60%),
THF, 08C to RT, 90%; b) i) BF3·OEt2 (2.0 equiv)/TMSOTf (6.0 equiv),
L2 (11 mol%), CuTC (5 mol%), n-HexLi (sol. in hexane), CH2Cl2,
À808C, ii) O3, MeOH, À808C. iii) NaBH4, MeOH, À808C to RT, 72%
for two steps; c) TEMPO (cat.), IBD, CH3CN/H2O (1:1), 94%,>99:1 e.r.
(61% overall yield).
(-OBn). The potential for application in organic synthesis is
demonstrated by a three-step synthesis of pharmaceutically
important (S)-Arundic acid. Mechanistic studies on the
exact nature of this highly selective transformation based on
the combination of Lewis acids and organolithium reagents
are ongoing.
Experimental Section
Typical procedure: A Schlenk tube equipped with septum and stirring
bar was charged with CuTC (0.01 mmol, 1.90 mg, 5 mol%) and the
ligand L2 (0.022 mmol, 11 mol%). Dry dichloromethane (0.5 mL) was
added and the solution was stirred under nitrogen at room temperature
for 15 min. In another Schlenk tube BF3·OEt2 (0.4 mmol, 50 mL, 2.0
equiv) was added to a solution of TMSOTf (1.2 mmol, 215 mL, 6.0 equiv)
in dichloromethane (0.5 mL) at À808C, and the resulting “BF2OTf” solu-
tion was stirred for 15 min. Then, the corresponding allyl ether 1–2
(0.2 mmol) in CH2Cl2 (1 mL) was added at À808C to the copper/ligand
solution prepared earlier and subsequently a solution of “BF2OTf” in di-
chloromethane (0.4 mmol, 600 mL, 2.0 equiv) was added. In a separate
Schlenk tube, the corresponding organolithium reagent (0.30 mmol,
1.5 equiv) was diluted with hexane (combined volume of 1 mL) under ni-
trogen and added dropwise to the reaction mixture over 2 h using a sy-
ringe pump. The reaction was quenched with a saturated aqueous NH4Cl
solution (2 mL) and the mixture was warmed up to room temperature,
diluted with dichloromethane and the layers were separated. The aque-
ous layer was extracted with dichloromethane (3ꢄ5 mL) and the com-
bined organic layers were dried with anhydrous Na2SO4, filtered and the
solvent was evaporated in vacuo. The crude product was purified by flash
chromatography on silica gel using different mixtures of n-pentane/Et2O
as the eluent.
Scheme 2. Compatibility and selectivity with two distinct ether function-
alities.
the fact that allylic -OBn ethers proved to be more reactive
substrates in previous cases (see for example, Table 2, en-
tries 13–16), we envision that a possible interaction between
the aromatic ring of the benzyl group and the copper com-
plex could control the regioselectivity of the reaction.[19]
The application of this new concept is illustrated by the
shortest enantioselective synthesis of (S)-Arundic acid 7.[20]
Arundic acid 7 is a small but important molecule with out-
standing pharmacological properties that is currently under-
going phase II trials for the treatment of acute ischemic
stroke as well as clinical development for Alzheimerꢃs and
Parkinsonꢃs disease.[21]
The synthesis starts with the treatment of commercially
available alcohol 5 with BnBr and NaH in THF. Alcohol 6
was obtained by asymmetric allylic alkylation of 2d followed
by direct reductive ozonolysis of the alkene 3g in MeOH/
CH2Cl2. The resulting alcohol 6 was oxidized to afford Ar-
undic acid 7 in three steps from alcohol 5 with>99:1 e.r.
and in 61% overall yield (Scheme 3).
Acknowledgements
We thank the Netherlands Organization for Scientific Research (NWO-
CW) and the National Research School Catalysis (NRSC-C) for financial
support. M.P. thanks the Xunta de Galicia for an Angeles AlvariÇo con-
tract. V.H. thanks the Spanish Ministry of Science and Innovation
(MICINN) for a postdoctoral fellowship.
In summary, the first example of the highly enantioselec-
tive asymmetric allylic alkylation of inert allylic ethers with
organolithium reagents is presented. The Lewis acid com-
patibility with organolithium reagents, under copper/phos-
phoramidite-catalyzed conditions, is essential to allow this
novel highly enantioselective transformation to occur. Ex-
cellent results reported for allylic methyl ethers (-OMe),
with similar selectivities than those recently reported for al-
lylic halides, were even improved for allylic benzyl ethers
Keywords: allylic alkylation · allylic ethers · copper · Lewis
acids · organolithium reagents
[1] a) Comprehensive Asymmetric Catalysis: Suppl. 2. (Eds.: E. N. Ja-
cobsen, A. Pfaltz, H. Yamamoto), Springer, 2004; b) Z. Lu, S. Ma,
c) G. Helmchem, U. Kazmaier, S. Fçrster in Catalytic Asymmetric
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
3
&
ÞÞ
These are not the final page numbers!