N-BuLi/LiCH2CN-mediated one-carbon homologation of aryl epoxides into conjugated allyl alcohols

Article abstract of DOI:10.1021/ol402466y

A series of styrene oxides in the presence of a 1:1 mixture of n-butyllithium (n-BuLi) and lithioacetonitrile (LiCH₂CN) in THF are converted into one-carbon homologated allyl alcohols in an unusual regioselective manner.

Full text of DOI:10.1021/ol402466y

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
n‑BuLi/LiCH2CN-Mediated One-Carbon
Homologation of Aryl Epoxides into
Conjugated Allyl Alcohols
Takashi Tomioka,* Rambabu Sankranti, Tsuyoshi Yamada, and Courtney Clark
Department of Chemistry and Biochemistry, University of Mississippi, University,
Mississippi 38677, United States
tomioka@olemiss.edu
Received August 27, 2013
ABSTRACT
A series of styrene oxides in the presence of a 1:1 mixture of n-butyllithium (n-BuLi) and lithioacetonitrile (LiCH2CN) in THF are converted into one-
carbon homologated allyl alcohols in an unusual regioselective manner.
A lithiated acetonitrile (LiCH2CN), first introduced by
Kaiser1 and Seebach2 independently in 1968, is a readily
available chemical species that can be prepared from
acetonitrile (CH3CN) and n-butyllithium (n-BuLi) in tet-
rahydrofuran. Due to the simplicity of use as well as
functional versatility, the species has now been widely
employed as a useful multipurpose reagent in the synthetic
community.3,4
4-hydroxy-4-phenylbutanenitrile 1 as an authentic sample
(Scheme 1). Based on a literature procedure (Scheme 1,
eq 1),5 we performed the reaction under essentially the
same conditions, but at a lower temperature (À78 °C).6
The reaction mixture was then gradually warmed up to
room temperature (Scheme 1, eq 2). Strangely, the reaction
proceeded poorly and a significant amount of starting
material was recovered. Although desired alcohol 1 was
obtained in modest yield (less than 10%), a non-negligible
amount of allyl alcohol 2a (13%) was also unexpectedly
isolated. Another allyl alcohol 3a, presumably an elimina-
tion product of 1, was detected as well, but only in a trace
amount. To confirm this curious observation, we then
tested the exact literature condition (i.e., 0 °C). As antici-
pated, only product 1 was obtained nearly quantitatively
and allyl alcohol 2a was not observed at all. Therefore,
our incidentally applied lower reaction temperature (i.e.,
À78 °C) seemed to be important to lead to 2a.
Recently, our group conducted a simple ring-opening
addition of LiCH2CN to styrene oxide due to the need of
(1) Kaiser, E. M.; Hauser, C. R. J. Org. Chem. 1968, 33, 3402.
(2) Crouse, D. N.; Seebach, D. Chem. Ber. 1968, 101, 3113.
(3) For recent selected examples (LiCH2CN in organic synthesis),
see: (a) Taber, D. F.; Green, J. H.; Zhang, W.; Song, R. J. Org. Chem.
2000, 65, 5436. (b) Taber, D. F.; Bui, G.; Chen, B. J. Org. Chem. 2001, 66,
3423. (c) Fleming, F. F.; Shook, B. C. J. Org. Chem. 2002, 67, 3668. (d)
Pound, M. K.; Davies, D. L.; Pikington, M.; de Pina Vaz Sousa, M. M.
Tetrahedron Lett. 2002, 43, 1915. (e) Bolshan, Y.; Chen, C.-Y.; Chilenski,
J. R.; Gosselin, F.; Mathre, D. J.; O’Shea, P. D.; Roy, A.; Tillyer, R. D.
Org. Lett. 2004, 6, 111. (f) Zenouz, A. M. Tetrahedron Lett. 2004, 45,
2967. (g) Cainelli, G.; Galletti, P.; Giacomini, D.; Gualandi, A.; Quin-
tavalla, A. Tetrahedron 2005, 61, 69. (h) Kawashima, T.; Kashima, H.;
Wakasugi, D.; Satoh, T. Tetrahedron Lett. 2005, 46, 3767. (i) Singh, K.;
Arora, D.; Singh, S. Tetrahedron Lett. 2007, 48, 1349. (j) Robertson, J.;
Tyrell, A. J.; Chovatia, P. T.; Skerratt, S. Tetrahedron Lett. 2009, 50,
7141. (k) Michon, C.; Sharma, A.; Bernardinelli, G.; Lacour, J. Chem.
Commun. 2010, 46, 2206. (l) Saitoh, H.; Satoh, T. Tetrahedron Lett. 2010,
51, 3380. (m) Schwartz, B. D.; Banwell, M. G.; Cade, I. A. Tetrahedron
Lett. 2011, 52, 4526. (n) Davis, M. C.; Groshens, T. J. Synth. Commun.
2012, 42, 2664.
Although it is known that the allylic alcohols 2a and
3a can be directly prepared from styrene oxide by using
dimethylsulfonium methylide, Me2SdCH2 (Scheme 2),7
the regioselectivity (2a vs 3a) is opposite to ours. Also,
(5) Eagon, S.; Ball-Jones, N.; Haddenham, D.; Saavedra, J.; DeLieto,
C.; Buckman, M.; Singaram, B. Tetrahedron Lett. 2010, 51, 6418.
(6) Since our group was accustomed to prepare/use LiCH2CN at
À78 °C (see refs 4aÀ4c), we simply applied the temperature as usual,
instead of 0 °C.
(4) (a) Tomioka, T.; Takahashi, Y.; Vaughan, T. G.; Yanase, T. Org.
Lett. 2010, 12, 2171. (b) Tomioka, T.; Sankranti, R.; Vaughan, T. G.;
Maejima, T.; Yanase, T. J. Org. Chem. 2011, 76, 8053. (c) Tomioka, T.;
Takahashi, Y.; Maejima, T. Org. Biomol. Chem. 2012, 10, 5113.
(7) Alcaraz, L.; Harnett, J. J.; Mioskowski, C.; Martel, I. P.; La Gall,
T.; Shin, D.-S.; Falck, J. R. Tetrahedron Lett. 1994, 35, 5449.
r
10.1021/ol402466y
XXXX American Chemical Society

Scheme 1. Ring-Opening Addition of LiCH2CN to Styrene
Oxide
Scheme 3. Plausible Mechanism I
(below À78 °C) due to the thermal instability.11 Thus, this
alternative path seemed to be more reasonable to explain
why alcohol 2a was not produced at 0 °C, but À78 °C (see
Scheme 1).
bis(trimethylstannyl)methane can alternatively lead to
allylic alcohol 2a as a major product (65%) from styrene
oxide,8,9 but no synthetic generality of this class of trans-
formation has been studied/established yet. Herein, we report
an unusual, regioselective transformation of styrene oxide
into a one-carbon homologated conjugated allyl alcohol.
Scheme 4. Plausible Mechanism II
Scheme 2. Me2SdCH2 Mediated Transformation of Epoxide
into One-Carbon Homologated Allyl Alcohol
Based on these assumptions, we hypothesized that, if the
carbenoid path (mechanism II) was mainly involved, the
use of a stronger base than LiCH2CN (e.g., n-BuLi) could
more efficiently deprotonate the R-hydrogen and generate
the key intermediates 5 and 6 for subsequent transforma-
tion. Hence, we treated styrene oxide with a 1:1 mixture of
n-BuLi as a base and LiCH2CN as a nucleophile, which
was simply prepared by exposing 1 equiv of CH3CN to
2 equiv of n-BuLi in THF at À78 °C for 15À20 min.12 To
our delight, this attempt (Scheme 5) dramatically im-
proved the yield of 2a (13%f80%).13 The protocol was
also reproducible.
As to the reaction mechanism, a simple anionic path
(mechanism I) as illustrated in Scheme 3 was initially
proposed. That is, a nucleophilic attack of LiCH2CN to
the benzylic carbon of styrene oxide, followed by either
intramolecular or intermolecular β-elimination from oxy-
anion intermediate 4, seemed to be reasonable. Based on
the mechanism, various reaction conditions were subse-
quently examined (i.e., reaction time, temperature, con-
centration, amount of LiCH2CN reagent, solvent, and
additive); however, the yield of 2a was hardly improved,
and even worse, the reaction was poorly reproducible.
Prior to further optimization, we then proposed a
carbenoid mediated mechanism (Scheme 4). The oxiranyl
anion 5, generated upon deprotonation of the acidic
R-hydrogen of styrene oxide, often exhibits carbene-like
reactivities via a presumed intermediate 6 (or a hybrid of 5
and 6).10 Subsequent nucleophilic addition of LiCH2CN to
carbene 6, followed by elimination of a cyanide ion from
dianion 7, should afford alcohol 2a. It should be noted that
anion 5 is typically formed at a very low temperature
Scheme 5. Synthesis of Allyl Alcohol 2a from Styrene Oxide
Accordingly, substrate generality of the protocol to
various aryl epoxides was examined (Table 1). Styrene ox-
idesthatare ortho-, meta-, and para-alkylated(entries 1À3)
as well as an arylated one (entry 4) were converted into the
corresponding alcohols in fair to good yields (51À81%).
(8) Maruyama, E.; Kikuchi, T.; Nishio, H.; Uematsu, M.; Sasaki, K.;
Saotome, N.; Sato, T. Nippon Kagaku Kaishi 1985, 350.
(9) Only two examples, styrene oxide (65%) and 1-phenylpropylene
oxide (55%), are demonstrated.
(10) (a) Doris, E.; Dechoux, L.; Mioskowski, C. J. Am. Chem. Soc.
1995, 117, 12700. (b) Morgan, K. M.; O’Connor, M. J.; Humphrey, J. L.;
Buschman, K. E. J. Org. Chem. 2001, 66, 1600. (c) Magnus, A.;
Bertilsson, S. K.; Andersson, P. G. Chem. Soc. Rev. 2002, 31, 223.
(11) Nagaki, A.; Takizawa, E.; Yoshida, J. Chem.;Eur. J. 2010, 16,
14149.
(12) To confirm the existence of n-BuLi and LiCH2CN, the 1:1
mixture was treated with benzaldehyde and both addition products
were observed by NMR (∼80% conversion).
(13) Attempts to examine the effect of TMEDA led to inferior yields
(∼45%).
B
Org. Lett., Vol. XX, No. XX, XXXX

Table 1. Synthesis of Allyl Alcohol 2 from Aryl Epoxidea
Scheme 6. Modified Addition Sequence
Scheme 7. Trapping Carbene Species
Naphthalene-based aromatic epoxides (entries 5 and 6)
also smoothly underwent the transformation (71% and
77%, respectively). More substituted epoxides such as di-
and trisubstituted epoxides (entries 7 and 8) provided
desired products with similar efficiency (65%). Other than
the simple styrene oxides, functionalized epoxides were
further investigated. Chlorinated styrene oxide (entry 9)
gave a slightly inferior result (44%) presumably due to the
susceptibility of the halogen group to n-BuLi. A methoxy
group (entry 10) and an acetal protecting group (entry 11)
were compatible under the conditions (60% and 65%,
respectively).
Foroperationalsimplicity, amodifiedconditionwithout
prior preparation of LiCH2CN was subsequently tested;
viz., n-BuLi (2 equiv) was directly added into a solution of
aryl epoxide (1 equiv) and CH3CN (1 equiv) in THF at
À78 °C (Scheme 6). This concise protocol yielded essentially
identical results as our original stepwise procedure.
Lastly, to provide support for the presence of a proposed
carbene intermediate in the reaction, an epoxide having an
alkene moiety 8 was examined (Scheme 7). Along with allyl
alcohol 2m (40%), an intramolecular cyclopropanation
adduct 9 (27%) was also isolated as a competitive side
product. This observation implies that the carbene species
is at least generated under the reaction conditions and
possibly supports the carbenoid mechanism as well.
In summary, we reported our serendipitous discovery
which involves a novel carbene-mediated transformation
of styrene oxide into 2-phenyl-2-propen-1-ol. A 1:1 mix-
ture of n-butyllithium and lithiated acetonitrile in tetrahy-
drofuran converted aryl epoxides into one-carbon ho-
mologated allyl alcohols in a highly regioselective manner.
Further synthetic applications of this protocol are cur-
rently under study in our group.
a CH3CN (2.1 mmol), n-BuLi (4.1 mmol), aryl epoxide (2.0 mmol),
dry THF (3.0 mL).
Acknowledgment. We thank Dr. Amal Dass and Mr.
Nuwan Kothalawala (University of Mississippi) for their
analytical assistance. We gratefully acknowledge the Na-
tional Science Foundation (1153105) for financial support.
Supporting Information Available. Experimental pro-
cedures and characterization data for compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
The authors declare no competing financial interest.
Org. Lett., Vol. XX, No. XX, XXXX
C