On the Reactivity of o-Lithioaryl Ethers: Tandem Anion Translocation and Wittig Rearrangement

Article abstract of DOI:10.1021/ol0258029

(matrix presented) Allyl and benzyl 2-lithioaryl ethers, generated by bromine-lithium exchange in THF, undergo a new tandem anion translocation-[1,2]-Wittig rearrangement allowing the isolation of the corresponding benzylic alcohols.

Full text of DOI:10.1021/ol0258029

ORGANIC
LETTERS
2002
Vol. 4, No. 9
1587-1590
On the Reactivity of o-Lithioaryl Ethers:
Tandem Anion Translocation and Wittig
Rearrangement
,†
Jose´ Barluenga,* Francisco J. Fan˜ana´s,† Roberto Sanz,‡ Ce´sar Marcos,‡ and
Marta Trabada‡
Instituto UniVersitario de Qu´ımica Organometa´lica “Enrique Moles”, Unidad
Asociada al C.S.I.C. Julia´n ClaVer´ıa, 8, UniVersidad de OViedo, 33071 OViedo, Spain,
and Departamento de Qu´ımica, AÄ rea de Qu´ımica Orga´nica, Facultad de Ciencias,
Pza. Missael Ban˜uelos s/n, UniVersidad de Burgos, 09001 Burgos, Spain
barluenga@sauron.quimica.unioVi.es
Received March 1, 2002
ABSTRACT
Allyl and benzyl 2-lithioaryl ethers, generated by bromine−lithium exchange in THF, undergo a new tandem anion translocation-[1,2]-Wittig
rearrangement allowing the isolation of the corresponding benzylic alcohols.
The “anion translocation” reaction, in which an organolithium
compound formed under kinetic control undergoes an inter-
or intramolecular proton transfer, offers an easy entry to new
organolithium compounds that are difficult to obtain fol-
lowing the traditional lithiation methods.1 R-Oxygen car-
banions are unstable2 and usually undergo protophilic
cleavage (elimination) or [1,2]-Wittig rearrangement,3 which
consists of the isomerization of these carbanions to metal
alkoxides.4 In most of the examples of [1,2]-Wittig rear-
rangement the migrating groups are alkyl and benzyl, and
the migration of vinyl5 and aryl groups6 is not so common.
Another feature of this kind of rearrangement is the presence
in the starting ether, in most of the cases, of a group that
can stabilize the negative charge in a position R to the
oxygen.7 In general, the yields of this process are moderate,
and harsh conditions are often employed. The mechanism
traditionally accepted for the [1,2]-Wittig rearrangement
involves the homolysis of an R-anion intermediate followed
by recombination of the radical and radical anion fragments.8
Despite many mechanistic studies, synthetic applications9 of
the [1,2]-Wittig rearrangement have remained limited, mainly
because of the rather low yields and the restricted range of
(6) (a) Eisch, J. J.; Kovacs, C. A.; Rhee, S.-G. J. Organomet. Chem.
1974, 65, 289-301. (b) Kiyooka, S.; Tsutsui, T.; Kira, T. Tetrahedron Lett.
1996, 37, 8903-8904.
(7) Scho¨llkopf, U. Angew. Chem., Int. Ed. Engl. 1970, 9, 763-773.
(8) (a) Scha¨fer, H.; Scho¨llkopf, U.; Walter, D. Tetrahedron Lett. 1968,
2809-2814. (b) Garst, J. F.; Smith, C. D. J. Am. Chem. Soc. 1976, 98,
1526-1537.
(9) See, for example: (a) Hoffmann, R. W.; Bru¨ckner, R. Chem. Ber.
1992, 125, 1957-1963. (b) Hoffmann, R. W.; Ru¨ckert, T.; Bru¨ckner, R.
Tetrahedron Lett. 1993, 34, 297-300. (c) Tomooka, K.; Igarashi, T.; Nakai,
T. Tetrahedron Lett. 1993, 34, 8139-8142. (d) Tomooka, K.; Igarashi, T.;
Nakai, T. Tetrahedron 1994, 50, 5927-5932. (e) Tomooka, K.; Yamamoto,
H.; Nakai, T. J. Am. Chem. Soc. 1996, 118, 3317-3318. (f) Maleczka, R.
E., Jr.; Geng, F. J. Am. Chem. Soc. 1998, 120, 8551-8552. (g) Tomooka,
K.; Yamamoto, H.; Nakai, T. Angew. Chem., Int. Ed. 2000, 39, 4500-
4502. (h) Tomooka, K.; Kikuchi, M.; Igawa, K.; Suzuki, M.; Keong, P.;
Nakai, T. Angew. Chem., Int. Ed. 2000, 39, 4502-4505. (i) Kitagawa, O.;
Momose, S.; Yamada, Y.; Shiro, M.; Taguchi, T. Tetrahedron Lett. 2001,
42, 4865-4868.
† Universidad de Oviedo.
‡ Universidad de Burgos.
(1) (a) Ahmed, A.; Clayden, J.; Rowley, M. Chem. Commun. 1998, 297-
298. (b) Ahmed, A.; Clayden, J.; Rowley, M. Tetrahedron Lett. 1998, 39,
6103-6106. (c) Fleming, I.; Mack, S. R.; Clark, B. P. Chem. Commun.
1998, 715-716. (d) Ahmed, A.; Clayden, J.; Yasin, S. A. Chem. Commun.
1999, 231-232.
(2) The R-carbanion of an alkyl benzyl ether is able to add to a carbonyl
moiety in the alkyl chain without Wittig rearrangement or protophilic
decomposition: Matsumoto, M.; Watanabe, N.; Ishikawa, A.; Murakami,
H. Chem. Commun. 1997, 2395-2396.
(3) Wittig, G.; Do¨ser, H. Justus Liebigs Ann. Chem. 1942, 550, 260-
268.
(4) Marshall, J. A. In ComprehensiVe Organic Synthesis; Trost, B. M.,
Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 3, pp 975-1014.
(5) Tomooka, K.; Inoue, T.; Nakai, T. Chem. Lett. 2000, 418-419.
10.1021/ol0258029 CCC: $22.00 © 2002 American Chemical Society
Published on Web 04/11/2002

substrates.10 On the other hand, we have recently described
that cuprates derived from 2-lithiophenyl allyl ethers undergo
a Claisen-type rearrangement with retention of the regio- and
stereochemistry.11 In the context of the research of new
reactivities of this kind of substrates, we now report the
transformation of 2-lithioaryl allyl and benzyl ethers into allyl
and benzyl alcohol derivatives through a tandem anion
translocation-[1,2]-Wittig rearrangement.
The treatment of different allyl and benzyl 1-bromo-2-
naphthyl ethers 1 with 2 equiv of tert-butyllithium in THF
at -78 °C afforded, after warming to different temperatures
and subsequent hydrolysis, the benzylic alcohols 3 in
moderate to good yields along with variable amounts of other
compounds, depending on the starting material (Scheme 1
Table 1. Rearranged Alcohols 3 from Naphthyl Ethers 1
starting
ether
T (°C)/
yield
(%)a
R1
R2
R3
R4
time (h) product
1a
1b
1cb
1d
1e
1f
H
Me
H
H
H
-(CH)4-
-(CH)4-
H
H
Me
Me
H
H
H
H
H
Me
-30/3
-30/3
0/12
0/12
0/12
-30/3
-30/12
3a
3b
3cc
3d
3e
3f
81
82
72d
61e
23f
83
H
H
H
-(CH2)3-
H
H
H
Me
1g
3g
78
a Isolated yield based on the starting ether 1. b A 4:1 mixture of E/Z-
isomers. c E-isomer (>95% by 1H NMR). d Ether 8c (15%) was also
obtained. e Ether 8d (22%) and â-naphthol (14%) were also obtained. f Ether
8e (51%) and ketone 7 (20%) were also isolated.
Scheme 1a
stabilize the negative charge in 5. So with allyl, methallyl,
or benzyl ethers 1a,b,f the anion translocation takes place
at -30 °C, whereas with R- and/or γ-substituted ethers 1c-e
it occurs at 0 °C. Probably as a result of this increase in the
reaction temperature, a competitive protonation of the
intermediate anions 4 or 5 by the solvent, giving the
corresponding naphthyl ether 8c-e, could explain the lower
yield of 3c-e. Furthermore, we have observed that starting
from R-substituted benzyl ether 1g the reaction also proceeds
at -30 °C but to be completed it needs longer times (12 h)
compared to the analogous benzyl ether derivative 1f. These
results show that substitution at the R- or γ-position of the
starting allyl or benzyl ether delays the anion translocation.
Moreover, the presence of an allyl or benzyl substituent in
the starting ether is necessary to promote the anion translo-
cation process, as was demonstrated by the fact that 1-bromo-
2-naphthyl methyl ether does not undergo the tandem
reactions described, probably because of the lack of a group
that stabilizes the R anion to the oxygen.
With respect to the nature of the allyl fragment some other
considerations could also be made. The reaction works
efficiently for all kinds of allyl and benzyl moieties tested
except for the case of cyclohexenyl ether 1e, where a mixture
of R,â-unsaturated ketone 7, debrominated ether 8e, and
alcohol 3e was obtained. The formation of the unexpected
spiro-derivative 7 could be explained by taking into account
that intermediate organolithium 5 could undergo an R-elim-
ination process leading to carbenoid reactions. In this case a
cyclopropanation of a double bond of the aromatic ring is
assumed, affording intermediate 6, which rearranges to the
more stable benzo-fused cycloheptenone derivative 7 after
hydrolysis. This fact and the formation of â-naphthol in the
reaction of ether 1d supports the idea that the R-elimination
competes with the Wittig rearrangement in those cases where
less stable tertiary carbanions, such as 5d,e, are formed.
Although 5g is also a tertiary carbanion, formation of
â-naphthol is not observed with ether 1g, probably because
of its benzylic character that stabilizes it. Interestingly, with
crotyl ether 1c the tandem sequence affords alcohol 3c as
almost isomerically pure E-alkene (20:1 by 1H NMR) despite
the fact that the starting ether is a 4:1 mixture of E and
Z-diastereoisomers. This fact could be explained by assuming
a Reagents and conditions: (a) t-BuLi (2 equiv), THF, -78 to T
°C; (b) H2O.
and Table 1).12 The formation of the resulting products could
be explained by assuming that the initial organolithium
compounds 4, generated by bromine-lithium exchange,13
undergo an anion translocation process generating new anions
5, which suffer a [1,2]-Wittig rearrangement to afford the
alkoxides 2, which upon hydrolysis allow the isolation of
alcohols 3. The temperature at which the translocation takes
place was determined for each lithiated ether 4 by the
deuteriolysis of aliquots and analysis by GC-MS.
From the results summarized in Table 1 it can be
concluded that the temperature at which the reactions occur
is related to the ability of the allyl or benzyl fragment to
(10) Tomooka, K.; Yamamoto, H.; Nakai, T. Liebigs Ann./Recueil 1997,
1275-1281.
(11) Barluenga, J.; Sanz, R.; Fan˜ana´s, F. J. Tetrahedron Lett. 1997, 38,
6103-6106.
(12) Alcohols 3, derived from a Wittig rearrangement, could also be
obtained from the corresponding â-naphthyl ethers without the bromine
atom by treatment with t-BuLi (1 equiv).
(13) Neumann, H.; Seebach, D. Chem. Ber. 1978, 111, 2785-2812.
1588
Org. Lett., Vol. 4, No. 9, 2002

that intermediate organolithiums 5 are allyloxy carbanions
and the E,Z-mixture of 5c can therefore be transformed into
the more stable E-isomer through an allylic rearrangement.
To support the proposed mechanism of an anion translo-
cation, inter- or intramolecular, we prepared the deuterium-
labeled ether 1h, by reduction of methyl benzoate with
lithium aluminum deuteride, chlorination14 of the resulting
alcohol, and further alkylation with 1-bromo-2-naphthol.
When this substrate was used as starting material the
dideuterated alcohol 3h was obtained, showing that the
deuterium-lithium exchange takes place in the intermediate
4h, leading to the benzyllithium derivative 5h. Further [1,2]-
Wittig rearrangement of 5h affords the alcohol 3h after
hydrolysis (Scheme 2). We have also observed that although
under the same temperature conditions. In this way, different
ratios of the rearranged alcohol 3b and the hydrolyzed ether
8b were obtained, depending on the concentration (Table
2). These results seem to indicate that the anion translocation
Table 2. Effect of Concentration in the Reaction of 1b
concentration
(mmol of 1b/mL THF)
ratio 3b:8bb
1/30
1/10
1.5/10
1/3
1:6.5
3:1
3.3:1
7.5:1
a The reaction was carried out from -78 to 20 °C for 2 h. b The overall
yield was about 95% in all the examples.
Scheme 2a
is intermolecular as shown by the decrease in the ratio of
the rearranged product 3b with an increase of the dilution.
Also, crossover experiments were carried out in order to
know about the intermolecularity of this process. So, a 1:1
mixture of lithium derivatives 4a and 4h was allowed to
warm from -78 to 20 °C to obtain alcohols 3a and 3h. The
lack of incorporation of deuterium in 3a could be due to the
different temperatures at which the translocation takes place
in organolithiums 4a and 4h (see above). However, a more
successful experiment was accomplished when a 1:1 mixture
of allyl â-naphthyl ether and organolithium compound 4h
was allowed to react at -30 °C for 3 h, affording a mixture
of alcohol 3a and unreacted allyl â-naphthyl ether besides
the hydrolyzed ether derived from 4h. This result also
supports the intermolecular nature of the translocaction.
To extend the scope of this process we investigated the
reaction of 2-bromophenyl allyl and benzyl ethers 9. So,
treatment of 9 with 2 equiv of tert-butyllithium at -78 °C
in THF afforded the organolithium compounds 10, which
after warming to 0 °C and further stirring for 3 h led to a
mixture of compounds 12, 13, and 14 in different ratio,
depending on the R1 and R2 substituents in the allyl moiety
of 9 (Scheme 3).
It is important to note that organolithium compounds 10
are stable at -30 °C, in contrast to the related intermediates
4 derived from naphthyl ethers 1, and it is necessary to warm
the reaction mixture to 0 °C and stir it for 3 h to obtain the
final products. This increase in the reaction temperature could
be the reason these phenyl ethers undergo the tandem anion
translocation-Wittig rearrangement less efficiently than the
naphthyl derivatives. The formation of these compounds
could be again explained through the intermediate R-oxygen-
ated organolithium 11 generated from 10 by an anion
translocation process. When R1 ) R2 ) H (pathway a), a
competitive intramolecular carbolithiation followed by a
γ-elimination, as described by Bailey,15 took place, and the
corresponding o-cyclopropylphenol 13a was obtained along
with the expected alcohol 12a in a 2.5:1 ratio (63% overall
a Reagents and conditions: (a) t-BuLi (2 equiv), THF, -78 °C;
(b) 0 °C, 3 h; (c) H2O; (d) rt, 3 h; (e) D2O, -78 °C.
the intermediate 4f, derived from benzyl ether 1f, undergoes
the anion translocation at -30 °C in 3 h (see Table 1), the
lithiated intermediate 4h is completely stable in these reaction
conditions and its evolution to the final product 3h is
observed at 0 °C in 3 h. In addition, with diphenylmethyl
ether 1i the hydrogen-lithium exchange takes place at -78
°C, probably as a result of the additional stabilization
introduced by the second phenyl group in 4i (Scheme 2). In
contrast with the rest of the ethers 1, the corresponding
translocated anion 5i is stable and could be deuterated to
afford ether 8i(D) in 84% yield. If the reaction mixture is
allowed to reach room temperature for 3 h the corresponding
rearranged alcohol 3i is obtained in 71% yield after hydroly-
sis.
To establish the inter- or intramolecular nature of the anion
translocation process, several experiments were performed
with varying concentrations of the starting ether 1b in THF,
(14) The chlorination was carried out with thionyl chloride and 1H-
benzotriazole following the procedure described by Chaudhari, S. S.;
Akamanchi, K. G. Synlett 1999, 1763-1765.
(15) An isomerization of 2-(2-propenoxy)phenyllithium to afford 2-(cy-
clopropyl)phenol in 40% yield has been reported: Bailey, W. F.; Punzalan,
E. R. Tetrahedron Lett. 1996, 37, 5435-5436.
Org. Lett., Vol. 4, No. 9, 2002
1589

hydrolysis of intermediate 11’b, generated by an allylic
rearrangement in 11b.16 The formation of 14b only in this
case could be due to the additional stabilization of the
resulting olefin by the methyl group at the R1 position. The
best result in the rearranged alcohol was obtained in pathway
c, starting from benzyl 2-bromophenyl ether, which affords
the expected diphenylmethanol 12c in 74%. In this case the
formation of the corresponding products 13 and 14 is not
possible.
Scheme 3
In summary, we have described a new tandem anion
translocation-Wittig rearrangement on o-lithioaryl ethers that
have a moiety able to stabilize a carbanion at the position
adjacent to the oxygen. The final products are the corre-
sponding allylic or benzylic alcohols, and this sequence
points to the potential development of anion translocation
as a new synthetic strategy in organic chemistry.
Acknowledgment. Financial support from the Direccio´n
General de Investigacio´n Cient´ıfica y Te´cnica (DGICYT,
PB97-1271) and Junta de Castilla y Leo´n (BU-09/99) is
gratefully acknowledged. C. Marcos thanks the Universidad
de Burgos and Junta de Castilla y Leo´n for a fellowship.
Supporting Information Available: Experimental pro-
cedures, characterization data for all compounds, and 2D-
NMR spectra for 7. This material is available free of charge
via the Internet at http://pubs.acs.org.
yield). However, when R1 ) Me and R2 ) H (pathway b)
the intermediate 10b afforded an equimolecular mixture of
alcohol 12b and enol ether 14b (60% overall yield), derived
from the translocated organolithium 11b, together with a
small amount (12%) of the corresponding cyclopropyl
derivative 13b. The unexpected formation of 14b also
supports that R-oxygenated anions 11 are intermediates in
these reactions because its formation could be explained by
OL0258029
(16) (a) Still, W. C.; Macdonald T. L. J. Am. Chem. Soc. 1974, 96, 5561-
5563. (b) Still, W. C.; Macdonald T. L. J. Org. Chem. 1976, 41, 3620-
3622.
1590
Org. Lett., Vol. 4, No. 9, 2002