Birch reductive alkylation of biaryls: Scope and limitations

Article abstract of DOI:10.1021/jo901395m

(Chemical Equation Presented) Birch reductive alkylation of biaryls has been carried out by varying the nature of the substituents on the aromatic rings. Our investigations have focused on electron-rich substituents such as OMe, OH, and NR₂ gro

Full text of DOI:10.1021/jo901395m

pubs.acs.org/joc
Birch Reductive Alkylation of Biaryls: Scope and Limitations
€
Raphael Lebeuf, Julie Dunet, Redouane Beniazza, Dawood Ibrahim, Gopal Bose,
ꢀ ꢀ
Muriel Berlande, Frederic Robert, and Yannick Landais*
ꢀ ꢀ ꢀ
Universite de Bordeaux, Institut des Sciences Moleculaires, UMR-CNRS 5255, 351, Cours de la Liberation,
F-33405 Talence Cedex, France
y.landais@ism.u-bordeaux1.fr
Received July 2, 2009
Birch reductive alkylation of biaryls has been carried out by varying the nature of the substituents on
the aromatic rings. Our investigations have focused on electron-rich substituents such as OMe, OH,
and NR2 groups as they are present on the skeleton of targeted alkaloids. The regioselectivity is
strongly affected by the electronic nature of these substituents on both rings. The 3,5-dimeth-
oxyphenyl moiety is selectively reduced and then alkylated, while phenols and anilines do not react
under these conditions. A biaryl possessing both a 3,5-dimethoxyphenyl moiety and a phenol ring
may, however, be reduced and alkylated provided the acidic phenolic proton is removed prior to the
treatment with Li in NH3. Similarly, biaryls possessing a o-sulfonamide group are reduced
regioselectively and alkylated with R-chloroacetonitrile or N-tosylaziridine to provide the corre-
sponding dienes in reasonable to good yields. A survey of the alkylating agents was also performed
showing that various functional groups may be introduced at the benzylic position, including esters,
primary and tertiary amides, nitriles, epoxides, and acetals and also unfunctionalized sterically
hindered t-Bu groups and cyclopropyl substituents. The introduction of the latter indicates that both
a SN2 and a SET mechanism may take place during the alkylating step.
Introduction
stereogenic centers and particularly that of the quaternary
center3 present in most representative members of these
families. Although these alkaloids are biogenetically unre-
lated, it was possible to identify in most of them, a common
substructure, incorporating a substituted aryl group linked
to a functionalized cyclohexane moiety (in blue, Figure 1). In
addition, an ethylamino group on a stereogenic quaternary
center (in red, Figure 1) is present in certain cases (morphine,
strychnine, crinine). Owing to our long-standing interest in
desymmetrization processes,4 it occurred to us that a sym-
metrical arylcyclohexadiene bearing (or not) an ethylamino
group at the benzylic center (e.g., I, Figure 1) would con-
stitute a common building block for all of these alkaloids.
Suitable functionalization would provide a somewhat unified
The alkaloids isolated from Amaryllidaceae, strychnos,
and also morphinans have attracted considerable interest1
and still today constitute a source of inspiration for synthetic
chemists2 and valuable targets for biological evaluation. The
polycyclic nature of these complex natural products induces
many synthetic problems, including the stereocontrol of all
(1) Hesse, M. Alkaloids-Nature’s Curse or Blessing?; Wiley-VHCA:
Zurich, 2002; 413 pp, ISBN 3-906390-24-1.
€
(2) (a) Hoshino, O. The Amaryllidaceae Alkaloids. In The Alkaloids; Cordell,
G. A., Ed.; Academic Press: San Diego, 1998; Vol. 51, pp 323-424. (b) Rinner,
U.; Hudlicky, T. Synlett 2005, 365–387. (c) Zezula, J.; Hudlicky, T. Synlett 2005,
ꢀ
388–405. (d) Bonjoch, J.; Sole, D. Chem. Rev. 2000, 100, 3455–3482. (d)
Creasey, W. A. Monoterpenoid Indole Alkaloids; Saxton, J. E., Ed. In The
Chemistry of Heterocyclic Compounds; Taylor, E. C., Ed.; Wiley: New York,
1994; Supplement to Vol. 25, part 4, pp 715-754.
(3) For reviews on methodologies to install quaternary centers, see: (a)
Trost, B. M.; Jiang, C. Synthesis 2006, 369–396. (b) Denissova, I.; Barriault,
L. Tetrahedron 2003, 59, 10105–10146. (c) Corey, E. J.; Guzman-Perez, A.
Angew. Chem., Int. Ed. 1998, 37, 388–401. (d) Peterson, E. A.; Overman, L. E.
Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 11943–11948.
(4) (a) Angelaud, R.; Babot, O.; Charvat, T.; Landais, Y. J. Org. Chem.
1999, 64, 9613–9624. (b) Abd. Rahman, N.; Landais, Y. Curr. Org. Chem.
2002, 6, 1369–1395. (c) Landais, Y.; Zekri, E. Eur. J. Org. Chem. 2002, 4037–
4053. (d) Lebeuf, R.; Robert, F.; Schenk, K.; Landais, Y. Org. Lett. 2006, 8,
4755–4558. (e) Rousseau, G.; Robert, F.; Schenk, K.; Landais, Y. Org. Lett.
2008, 10, 4441–4444.
DOI: 10.1021/jo901395m
r
Published on Web 08/05/2009
J. Org. Chem. 2009, 74, 6469–6478 6469
2009 American Chemical Society

Lebeuf et al.
JOCFeatured Article
SCHEME 1. Birch Reduction and Birch Reductive Alkylation of
Biaryls
FIGURE 1. Arylcyclohexadienes I as precursors of various classes
of alkaloids.
deau and co-workers,9 who showed that the introduction of
an aryl substituent on a substituted arene modifies to a large
extent the regioselectivity of the reduction as compared to
the reduction of the same arene lacking the aryl substituent.
Birch reduction of biaryls bearing various substituents in-
cluding Me, SiMe3, F, CO2H, CO2R was thus investigated.
Interestingly, few studies have been performed on Birch
reduction of electron-rich biaryls (substituted only by
OMe, NR2, etc.).10 As shown in the examples above, reduc-
tion of a p-methoxybiphenyl 6 leads to various regioisomers
and over-reduced products, indicating that regioselectivity
may not be so easy to predict in this case (Scheme 1).10a We
found no investigation on Birch reductive alkylation of
electron-rich biaryls (substituted with OMe groups for
instance). Harvey first reported on the Birch reductive
alkylation of biaryls.11 Its studies on the Birch reductive
methylation of biphenyl 7 revealed that alkylation mainly
takes place at the benzylic position as well as at the bis-allylic
position when Na is used as a reducing agent (Scheme 1).12
Dialkylation is also observed in this case, which was attrib-
uted to further deprotonation of the bis-allylic position by
the highly basic sodium amide formed upon reaction.
strategy toward the total synthesis of representative mem-
bers 1-5 of these distinct families of alkaloids.
We recently showed that arylcyclohexadienes of type I
could be prepared through a straightforward but multistep
synthesis including the generation of the diene moiety from
the corresponding silyl enol ethers.4e,5 A simpler approach
would involve the regioselective Birch reduction of a suitably
substituted biaryl precursor followed by a regioselective
protonation or alkylation of the resulting anion. This Birch
reductive alkylation (BRA) strategy brings up two main
problems: (i) the regioselectivity issue during the reduction
of aryl moieties and (ii) the competition between alkylation
and protonation of the resulting anion, which may either
lead to the diene (I, X=H)6 or to the alkylated product
(I, X ¼ H). The first issue will mainly depend on the nature of
the substituents on both aromatic groups, while the second
relies on the basicity vs nucleophilicity of the last formed
anion in ammonia, the medium in which the Birch reduction
is generally performed. While the Birch reductive alkylation
on simple arenes has been studied intensively,7 its extension
to biaryls has been so far little explored.8 The influence of
substituents on the aromatic rings of the biaryl system on the
course of the Birch reduction was mainly studied by Rabi-
Owing to the lack of information on the regioselectivity of
such reductions and the scarce results on the Birch reductive
alkylation of biaryls, it was decided to examine the scope and
limitations of the Birch reductive alkylation of diversely
substituted biaryls (with electron-rich substituents). We thus
engaged on a systematic study, varying the nature of the
€
(5) (a) Rousseau, G. Ph.D. Thesis dissertation, November 13, 2008. (b) Brase,
S. Synlett 1999, 1654–1656.
(6) Such arylcyclohexadienes may also be prepared by addition of an aryl
radical onto benzene, see: (a) Crich, D.; Grant, D.; Wink, D. J. J. Org. Chem.
2006, 71, 4521–4524. (b) Crich, D.; Rumthao, S. Tetrahedron 2004, 60, 1513–
1516. (c) Crich, D.; Krishnamurthy, V. Tetrahedron 2006, 62, 6830–6840.
(7) For exhaustive reviews on Birch reductive alkylation, see: (a) Rabideau,
P. W.; Marcinow, Z. Org. React. 1992, 42, 1-334 and references therein.
(b) Schultz, A. G. Chem. Commun. 1999, 1263–1271. (c) Rabideau, P. W.
Tetrahedron 1989, 45, 1579–1603.
(9) (a) Rabideau, P. W.; Huser, D. L.; Nyikos, S. J. Tetrahedron Lett. 1980,
21, 1401-1404. (b) Rabideau, P. W.; Nyikos, S. J.; Huser, D. L.; Burkholder, E. G.
J. Chem. Soc., Chem. Commun. 1980, 210–211. (c) Marcinow, Z.; Clawson, D.
K.; Rabideau, P. W. Tetrahedron 1989, 45, 5441–5448. (d) Harvey, R. G.;
Lindow, F.; Rabideau, P. W. J. Am. Chem. Soc. 1972, 94, 5412–5420. (e) Jessup,
D. W.; Paschal, J. W.; Rabideau, P. W. J. Org. chem. 1977, 42, 2620–2621. (f)
Harvey, R. G. Synthesis 1970, 161–172.
(10) (a) Birch, A. J.; Nadamuni, G. J. Chem. Soc., Perkin Trans. 1 1974,
545–552. (b) Tanaka, H.; Shibata, M.; Ito, K. Chem. Pharm. Bull. 1984, 32,
3271–3272. (c) Tanaka, H.; Takamura, Y.; Shibata, M.; Ito, K. Chem.
Pharm. Bull. 1986, 34, 24–29.
(11) (a) Lindow, D. F.; Cortez, C. N.; Harvey, R. G. J. Am. Chem. Soc.
1972, 94, 5406–5412. (b) Lindow, D. F.; Harvey, R. G. J. Am. Chem. Soc.
1971, 93, 3786–3787.
(12) Rabideau, P. W.; Peters, N. K.; Huser, D. L. J. Org. Chem. 1981, 46,
1593–1597.
€
(8) (a) Muller, P. M.; Pfister, R. Helv. Chim. Acta 1983, 66, 771–779.
(b) M€uller, P. M.; Pfister, R.; Urban, R. European Patent 12801 1980; Chem.
Abstr. 1980, 93, 185839s. (c) Rabideau, P. W.; Peters, N. K.; Huser, D. L. J. Org.
Chem. 1981, 46, 1593–1597. (d) Guo, Z.; Schultz, A. G. Org. Lett. 2001, 3,
1177–1180. (e) Casimiro-Garcia, A.; Schultz, A. G. Tetrahedron Lett. 2006, 47,
2739–2742. (f) Tapas, P.; Malachowski, W. P.; Lee, J. Org. Lett. 2006, 8, 4007–
4010. (g) Tapas, P.; Malachowski, W. P.; Lee, J. J. Org. Chem. 2007, 72, 930–937.
(h) Malachowski, W. P.; Tapas, P.; Phounsavath, S. J. Org. Chem. 2007, 72, 6792–
6796.
6470 J. Org. Chem. Vol. 74, No. 17, 2009

Lebeuf et al.
JOCFeatured Article
biaryl and that of the electrophile. Preliminary results have
shown that the process is applicable to a variety of biarylic
systems and is readily amenable to large-scale synthesis.13
This process which is part of a methodology called
the BRAD strategy (for Birch Reductive Alkylation-De-
symmetrization)4d,e,14 offers a rapid entry toward useful
building blocks for the synthesis of alkaloids such as
those depicted in Figure 1. We provide here a full account
of these investigations, focusing on various aspects of this
study, including (1) the regiocontrol as a function of the
nature of the biaryl system; (2) the variation of the nature of
the electrophile; and (3) some mechanistical aspects of the
BRA.
SCHEME 2. BRA of Biaryls 8 and 10
Results and Discussion
Birch Reductive Alkylation of Biaryls with Electron-Rich
Substituents. The investigations started with various biaryls
bearing electron-rich substituents including methoxy and
methylenedioxy groups frequently found on naturally occur-
ring alkaloids (Figure 1). Our first attempt on biaryl 8 was
encouraging, producing with 2.4 equiv of lithium in ammo-
nia under reflux the alkylated arylcyclohexadiene 9b along
with phenol 9a (Scheme 2). Similarly, biaryl 10 afforded
alkylated product 11b and the demethylated biaryl 11a,
along with some recovered starting material (17%). In both
cases, a para (in 8) or an ortho (in 10) substituent was lost
during the process, a well-precedented observation on simple
arenes.7a,15 This can be explained by the formation of a
radical anion on the most electron-rich arene, which would
then decompose into an alkoxide and a phenyl radical.16 The
latter would then be reduced further into the corresponding
anion which would be protonated by NH3, leading to biaryls
9a and 11a. Consequently, it was deduced that 9b and 11b
were probably generated in a second phase through BRA of
9a and 11a, respectively (vide infra).
SCHEME 3. BRA of Biaryls Bearing m-OMe Groups
The reductive alkylation was then attempted on biaryl 11a
and led to the formation of two inseparable regioisomers 11b
and 12 in a good overall yield (Scheme 3). The formation of
11b as the major isomer indicates that the reduction occurs
on the most electron-rich aromatic ring, in good agreement
with the observed relative reduction rates on simple arenes,
which follow the order ArOMe>ArH>ArOH.7a,17 This
suggests that the relative rates of reduction of biaryls
(conjugated arenes) do not differ from the order observed
for the reduction of isolated (nonconjugated) arenes. It also
confirms the hypothesis of the formation of 11b from 10 via
11a (Scheme 2). Birch reductive alkylation of 13 having two
methoxy groups meta to the biaryl linkage led to arylcyclo-
hexadienes 14a,b in good yield and complete regiocontrol.
Likewise, symmetrical biaryl 15 provided diene 16 in excel-
lent yield.
The previous experiments thus demonstrate that the pre-
paration of arylcyclohexadienes of type I, using Birch re-
ductive alkylation of biaryls, is a viable route but still limited
in scope to biaryls having the OMe group meta to the biaryl
linkage. As illustrated below, when the reaction was carried
out on a tetrasubstituted but unsymmetrical biaryl such as
17, the reaction provided a diene 18 and recovered starting
material (Scheme 4). This can be rationalized by invoking a
selective reduction of the 3,4-dimethoxyphenyl leading, after
the loss of a OMe group, to a trimethoxy-substituted biaryl
(with consumption of 2 equiv of Li per trimethoxy-substi-
tuted biaryl formed), which upon further reduction and
alkylation afforded diene 18. Direct reduction of the 3,5-
dimethoxyphenyl fragment apparently did not occur as the
(13) (a) Lebeuf, R.; Robert, F.; Landais, Y. Org. Lett. 2005, 7, 4557–4560.
(b) Lebeuf, R.; Berlande, M.; Robert, F.; Landais, Y. Org. Synth. 2009, 86,
1–10.
(14) Beniazza, R.; Dunet, J.; Robert, F.; Schenk, K.; Landais, Y. Org.
Lett. 2007, 9, 3913–3916.
(15) Kuehne, M. E.; Lambert, B. F. J. Am. Chem. Soc. 1959, 81, 4278–
4287.
(16) (a) Thornton, T. A.; Woolsey, N. F.; Bartak, D. E. J. Am. Chem. Soc.
1986, 108, 6497–6502. (b) Dewald, R. R.; Conlon, N. J; Song, W. M. J. Org.
Chem. 1989, 54, 261–262.
(17) For instance, anisole is reduced 3.3 times faster than benzene; see:
Krapcho, A. P.; Bothner-By, A. A. J. Am. Chem. Soc. 1959, 81, 3658–3666.
J. Org. Chem. Vol. 74, No. 17, 2009 6471

Lebeuf et al.
SCHEME 5. BRA of Biaryls Bearing OH Groups
JOCFeatured Article
SCHEME 4. BRA of Biaryls Bearing m-OMe Groups
expected alkylated tetramethoxydiene was not observed in
the crude reaction mixture.
Birch Reductive Alkylation of Biaryls with Phenol Substit-
uents (Meta and Para Positions). In order to overcome the
limitation mentioned above, we then focused our attention
on precursors bearing a phenol moiety as it is well-known
that phenols are generally inert under Birch reduction con-
ditions.7,18 As a consequence, it was foreseen that the pre-
sence of a phenol fragment would direct the reduction onto
the other aromatic ring, therefore circumventing the pro-
blem of regioselectivity and the loss of o- and p-OMe
substituents. This proved to be partially correct as illustrated
by the examples below (Scheme 5). We observed that the
reaction carried out on biaryls 19 and 20 afforded modest
yields of the alkylated product 9b or only the reduced
product 21, accompanied by over-reduced byproducts that
could not be purified. It is, however, worthy of note that in
biaryl 19, the alkylation takes place on the “non-phenolic”
ring, leaving a cyclohexadiene without resident methyl enol
ethers, which proved to be useful for the subsequent devel-
opments of the BRAD strategy.4d Although the amount of
lithium was raised to 3.6 equiv, the presence of over-reduced
products seems to indicate that either the phenol is not
deprotonated under the present conditions or more likely
that the ammonium phenolate (formed by reaction between
the phenol and NH3) acts as a proton source, eventually
competing with the alkylation process.
SCHEME 6. BRA with Added Base of Biaryls Bearing OH
Groups
The involvement of the phenol as a proton source could be
easily prevented by performing the Birch reduction after
prior deprotonation of the phenol with n-BuLi. Under these
conditions, we were pleased to find that phenol 19 could now
be transformed into the diene 9b in an improved 68% yield
(Scheme 6). Polysubstituted biaryls 20 and 22 led to the
corresponding dienes 23a,b in excellent yields indicating that
a careful tuning of the nature of the substituents on the
aromatic ring allowed the alkylation to take place selectively
on the substituted arene (as in 14a-b, Scheme 3) or on the
unsubstituted one (as in 9b, 25a and 25b, Scheme 6). It is
important to add that the temperature is a crucial parameter
in this reaction. When the temperature was raised above
-50 °C, a large amount of the reduced product was formed
at the expense of the desired alkylated compound, indicating
that at higher temperature, protonation is favored over
alkylation. In addition, we have observed that the attempt
to scale up the reaction (reduction of 20 to 23a) led to a drop
in yield (from 80% to 50%), further suggesting that local
raising of the temperature (for instance during addition of
the alkylating agent) affects the reduced/alkylated product
ratio (vide infra).
Based on these results, we then focused our attention on
the Birch reductive alkylation of biaryls bearing a phenol and
OMe groups in the ortho and para positions. Arylcyclo-
hexadienes with an ortho-oxygenated substituent would be
particularly attractive as in our general strategy, they are
potent precursors of morphine alkaloids and analogues. The
first attempts were unfortunately not successful. As above,
BRA on precursors 26 and 27, bearing both a phenol and a
p-OMe or an o-O-i-Pr substituent, invariably led to loss of
these substituents (Scheme 7). In the case of 26, demeth-
oxylation likely occurred first leaving the resulting phenol 22
which was then partially reduced and alkylated into 23b as
indicated by the presence of these two compounds in the
mixture, along with recovered starting material in large
quantities. Similarly, biaryl 27 bearing the bulkier O-i-Pr
substituent in the ortho position, which was suggested to slow
down the elimination,19 led to an even larger amount of the
elimination product 23b.
(18) Birch, A. J.; Subba Rao, G. S. R. Tetrahedron Lett. 1967, 857–858.
6472 J. Org. Chem. Vol. 74, No. 17, 2009
(19) Cambie, R. C.; Missen, A. W. Aust. J. Chem. 1972, 25, 973–983.

Lebeuf et al.
JOCFeatured Article
SCHEME 7. BRA of Biaryls Bearing m-OH Groups and Ortho
and Para Substituents
SCHEME 8. BRA of Biaryls Bearing o-OH Substituents
SCHEME 9. BRA of Biaryls Bearing o-OSiR3 Substituents
Birch Reductive Alkylation of Biaryls with Phenol Substi-
tuents (Ortho Position). The results above indicate that OR
groups ortho to the biaryl linkage are labile even in the
presence of a lithium phenolate moiety on the same ring.
We therefore attempted a series of Birch reductive alkylation
on precursors 28a,b having a OH group ortho to the biaryl
bond, using as above the n-BuLi procedure, as it was
anticipated that OLi would be a poor leaving group
(Scheme 8). Surprisingly, while the reduction proved to be
very efficient under these conditions, affording dienes 29a
and 29b, no alkylation was observed. The regioselectivity of
the reduction is high, but apparently the protonation of the
resulting anion is too fast, preventing the alkylation from
occurring. This was attributed to electrostatic repulsion
between the lithium phenolate and the benzylic carbanion
species resulting from the Birch reduction (e.g., II, Scheme 8).
This repulsion leads to an increase of the basicity of the
carbanion which is then protonated by ammonia, thus
preventing the alkylation. The presence of the ortho pheno-
late substituent may also prevent the conjugation of the
benzylic carbanion with the neighboring aromatic ring thus
raising its basicity.
Our attempts to metalate (with n-BuLi in THF at 0 °C) and
alkylate the benzylic position of 29a,b after protection of the
OH group (as a OMe) having failed, we turned our attention
to the reductive alkylation of various O-protected analogues
of 28a,b. While o-acetate (OCH2CO2Me) and o-amide
(OCH2CONMe2) substituents were cleaved under the re-
ductive conditions, more encouraging results were obtained
with silylated analogues 30 and 31 (R=TIPS and TBDMS,
respectively) (Scheme 9).19 TIPS-protected phenol 30 thus
afforded a separable 2/1 mixture of the desired alkylated
diene 32a and its regioisomer 32b (66% corrected overall
yield). Similarly, the TBDMS analogue 31 led to the forma-
tion of a 3/7 mixture of regioisomers 33a,b (86% overall
yield). In contrast with OMe substituents, OSiR3 groups are
thus poor leaving groups under Birch reaction conditions.
They also induce much lower regioselectivity. The higher
stability of o-OSiR3 substituents as well as their influence on
the regioselectivity of the process under the above conditions
remain unclear but follow previous observations reported by
Cambie et al. on related bulky ether groups.19 Silyl ethers are
known to be less basic than their carbon analogues due to
oxygen lone pairs overlapping with silicon, which is a good
π-acceptor. The unexpected amount of dienes 32b and 33b
formed might then be tentatively explained by the better
stabilization of the radical anion generated in the prelimi-
nary stage of the process by the siloxy substituent, as
compared to the OMe group.
Birch Reductive Alkylation of Biaryls with Amino Substi-
tuents (Ortho Position). Referring again to the nature of the
alkaloid targets depicted in Figure 1, it came into view that
an access to arylcyclohexadiene I having an ortho amino
group on the remaining arene would be particularly helpful
in the context of the synthesis of strychnos (e.g., strychnine 4)
and aspidosperma alkaloids. The first study in this direction
was carried out starting from precursor 34a, having a 3,5-
dimethoxyphenyl group and a Boc protected aniline moiety,
which afforded, following the n-BuLi procedure, the desired
diene 35a in a satisfying 60% yield, along with over-reduced
products (Table 1, entry 1). It is worth noting that in contrast
with the phenolic analogue 28a, the presence of the Boc
substituent probably reduces (through delocalization) some
of the anionic repulsion illustrated above (e.g., II, Scheme 8),
allowing the alkylation to compete with the protonation by
NH3. Reduction-alkylation was then tested on precursor
34b, having a simple phenyl group and a Boc-protected
amino group on the second arene (entry 2). Deprotonation
with n-BuLi and reduction with lithium led, after alkylation,
to the desired diene 35b, albeit in modest yield, due to the
formation of over-reduced products which could not be
purified. Based on the above hypothesis, it was reasoned
that better acceptors on nitrogen might improve the yield of
J. Org. Chem. Vol. 74, No. 17, 2009 6473

Lebeuf et al.
JOCFeatured Article
TABLE 1. BRA of Biaryls Bearing o-Amino Substituents
SCHEME 10. Isomerized 1,3-Dienes in Birch Reductive Alkyl-
ation of Biaryls
entry
biaryl
R
R0
diene
yield (%)
60a
1
2
3
4
5
6
34a
34b
34c
34d
34e
34f
Boc
Boc
H
Piv
EtSO2
EtSO2
OMe
H
H
H
H
Me
35a
35b
35c
35d
35e
35f
26a
b
40c
70d
75
aOver-reduced products were also observed. bOnly over-reduced
products were formed. Reduction and alkylation on the aniline ring
c
was formed in ∼20% yield. d1,3-Cyclohexadiene 38 is also formed (8%).
the alkylation by limiting the N-Li-carbanion repulsion.
This was also supported by the absence of alkylated products
when starting from the unprotected amine 34c (entry 3),
whose reduction only afforded a complex mixture of reduced
products. Amide 34d and sulfonamide 34e (entries 4-5) were
subjected to the n-BuLi procedure, giving improved yields of
the corresponding dienes 35d-e, supporting the above rea-
soning. Biaryl 34f possessing a 3,5-dimethylphenyl ring
(entry 6) was also converted into the desired alkylated
product 35f in a satisfying 75% yield.
TABLE 2. BRA of Biaryls Varying the Nature of the Electrophiles
We also extended the reduction to more electron-rich
analogues such as 36 having an additional OMe group in
the meta position. Surprisingly, we observed the formation
of diene 37 as the sole product, resulting from the reduction
and alkylation of the aminophenyl ring (Scheme 10). This
result is indicative of the strong electronic effect of the m-
OMe group on the regioselectivity of the process (vide infra).
Finally, we observed in several cases a mixture of over-
reduced products (cyclohexenes) and 1,3-cyclohexadienes
which were obtained along with the desired 1,4-dienes. For
instance, Birch reductive alkylation of biaryl 34e with ethyl
R-bromoacetate afforded no trace of the alkylated product,
but a quantitative yield of 1,3-diene 38, whose structure was
tentatively assigned through iodo amination (under kinetic
control),20 affording sensitive allylic iodide 39 in modest
yield (Scheme 10). This surprising result suggests that ethyl
R-bromoacetate may act as a better proton source than NH3,
providing the 1,3-diene upon protonation at C2 instead of
the usually favored 1,4-cyclohexadiene.21 Isomerization of
the “kinetic” 1,4-diene isomer by LiNH2, present in the
medium under the above conditions, was ruled out on the
basis of observations made by Rabideau and Harvey who
showed that deprotonation of a 1,4-cyclohexadiene by
LiNH2 in NH3 at -50 °C does not occur.11a,22 Therefore,
this result may simply reflect the relative protonation rate at
C2 and C4 in this precursor 34e, suggesting an implication of
aCyclohexa-2,5-dienylbenzene was formed exclusively. bYield esti-
mated by 1H NMR. Starting biaryl 7 was the only other product present
in the crude reaction mixture. c84% based on recovered start-
ing material. dn-BuLi is used to deprotonate the sulfonamide eOver-
reduced products were also observed. 1,3-Cyclohexadiene 38 is also
formed (15%).
f
the NLiSO2Et group in the regiocontrol. Finally, the easy
formation of 1,3-cyclohexadiene 38 also explains the pre-
sence of over-reduced products in certain cases as conjugated
1,3-dienes are more easily reduced than the nonconjugated
ones in Li/NH3 medium.
Birch Reductive Alkylation of Biaryls: Nature of the Electro-
phile. In the above studies, R-chloro-nitrile and esters were
used as electrophiles, providing a straightforward manner
to install the ethylamino group found in alkaloids 2-4
(Figure 1). We generalized the approach to other useful
electrophiles, as summarized in Table 2 and Scheme 11. Birch
reductive alkylation was carried out first with biphenyl 7 and
then extended to substituted biaryls. It is worthy of note that
the nature of the leaving group on the electrophile may have
a strong effect on the reaction outcome as suggested by the
alkylation of 7 with R-haloacetamide (Table 1, entries 1 and 2).
(20) (a) Cardillo, G.; Orena, M. Tetrahedron 1990, 46, 3321–3408.
(b) Fiorelli, C.; Marchioro, C.; Martelli, G.; Monari, M.; Savoia, D. Eur.
J. Org. Chem. 2005, 3987–3993.
(21) The conjugation in 1,3-diene provides little stabilization (0.41 kcal/
mol) as compared to the 1,4-diene: (a) Bates, R. B.; Carnighan, R. H.;
Staples, C. E. J. Am. Chem. Soc. 1963, 85, 3032–3033. (b) Turner, R. B.;
€
Mallon, B. J.; Tichy, M.; Doering, W. V. E.; Roth, W. R.; Schroder, G.
J. Am. Chem. Soc. 1973, 95, 8605–8610.
(22) Rabideau, P. W.; Huser, D. L. J. Org. Chem. 1983, 48, 4266–4271.
6474 J. Org. Chem. Vol. 74, No. 17, 2009

Lebeuf et al.
JOCFeatured Article
SCHEME 11. BRA with Esters and Aziridines As Electrophiles
agents. However, the attempts to establish unambiguously
the mode of alkylation of our substrates led to contrasting
results, as shown by the alkylation of biaryl 7 in the presence
of a diphenylcyclopropane (entry 6, Table 2), which led to the
alkylated diene 40f in a satisfying yield, without a trace of the
cyclopropane ring-opening product. Similarly, Birch reduc-
tive alkylation of 20 followed by alkylation with cyclopro-
pylmethyl bromide led to 43, albeit in moderate yield
(Scheme 11). This was intriguing as cyclopropyl bromide is
frequently used as a radical clock, with cyclopropyl ring-
opening rate constant as high as 6.7 ꢀ 107 M-1 s-1 24
.
Cyclopropane from bromomethyl-2,3-diphenylcyclopro-
pane (entry 6, Table 2) exhibits an even higher ring-opening
rate constant, e.g., >2 ꢀ 1010 M-1 s-1 25
. Cyclopropylmethyl
bromides may thus react through an SN2-type mechanism,
while tertiary alkyl electrophiles react through a radical
mechanism as the polar mechanism is not accessible due to
steric hindrance. A single electron transfer followed by
radical recombination in a solvent cage may also be envi-
sioned as an alternative pathway to explain the formation of
40f and 43.26
Birch Reductive Alkylation of Biaryls: Mechanistic Con-
siderations. A mechanism is finally proposed (Figure 2) that
tentatively rationalizes the observations made during our
investigations. The Birch reductive alkylation of arenes
involves an initial electron-transfer process from the metal
to the arene followed by the alkylation of the resulting
anion.27 The first electron transfer from Li in NH3 to the
biaryl provides the radical anion A.28 At this stage, two
different pathways may be envisioned. Protonation may
occur at the site of highest electron density (C4) as to provide
radical B. It is well described that the cyclohexadienyl anion
is kinetically more reactive at the central position, leading to
the radical intermediate B which is both benzylic and bis-
allylic. In precursors 13, 15, 20, 22, 28a,b, and 36, this radical
would be further stabilized by OMe substituents in meta
positions, rationalizing the high regioselectivity of the pro-
cess in these cases, with the predominant reduction of the
methoxy-substituted arene in good agreement with literature
precedent on simple arenes,17 as well as with kinetic studies
and calculations.29 A second electron transfer from lithium
to B may then provide the benzylic anion C which is finally
protonated by NH3 (at C1) or alkylated depending on its
basicity. This second protonation (or alkylation) generally
occurs at the benzylic position leading to the 1,4-cyclohex-
adienyl system, thus controlling the regioselectivity of the
whole process. Subtle substituent effects were shown to alter
The better leaving group ability of bromine relative to
chlorine but also the higher acidity of R-chloroacetamide
may explain these results. Birch reduction of 7 then alkyla-
tion with R-chloroacetate led to diene 40e in an improved
yield as compared to that reported in the literature8a,b (entry
5, Table 2). In contrast, the same reaction carried out on
biaryl 34e did not provide the corresponding alkylated
product but instead the reduced product 38 in high yield,
indicating that R-chloro- and R-bromoacetates may also
behave as protonating agent in this context, depending on
the nature of the starting biaryl (Scheme 10). Other electro-
philes including allyl bromide (entries 3 and 9) and R-
chloroamide (entry 4) provided the desired alkylated pro-
ducts in satisfying yields. Epoxides and acetals are also
valuable electrophiles in the Birch reductive alkylation
(entries 8 and 10). We also observed that increasing the steric
bulk on the electrophile modified the regioselectivity of the
alkylation process. For instance, alkylation of biaryls 34e,
using tert-butyl R-chloroacetate led to the alkylated product
at the 3-position, as indicated by the formation of 41
(Scheme 11). Finally, aziridines were found to be potent
electrophiles under the Birch conditions as shown by the
reduction-alkylation of biaryls 7 and 34e. These led respec-
tively to dienes 42a and 42b having two orthogonally pro-
tected amino groups. This one-pot formation of a precursor
of aspidosperma alkaloids is worthy of note and shortens
to a significant extent the access to this class of alkaloids.14
Interestingly, it was also possible to introduce a tert-butyl
group at the benzylic position (entry 7). Nucleophilic sub-
stitution at a tertiary center in t-BuBr seems inconsistent with
a SN2 process and is probably better interpreted as an
electron-transfer (ET) mechanism,23 occurring from a
benzylic anion intermediate to tert-butyl bromide. The oc-
currence during the process of a radical species issued from
the alkylating agent may be assessed by the use of radical
clock agents such as cyclopropyl containing alkylating
(24) (a) Newcomb, M. In Radical in Organic Synthesis; Renaud, P., Sibi,
M. P., Eds.; Wiley-VCH: Weinheim, 2001; Vol. 1; pp 317-336. (b) Newcomb,
M.; Choi, S. Y.; Horner, J. H. J. Org. Chem. 1999, 64, 1225–1231.
(25) (a) Castellino, A. J.; Bruice, T. C. J. Am. Chem. Soc. 1988, 110, 1313–
1315. (b) Castellino, A. J.; Bruice, T. C. J. Am. Chem. Soc. 1988, 110, 7512–
7519. (c) Adam, W.; Heil, M.; Castellino, A. J.; Bruice, T. C. J. Am. Chem.
Soc. 1991, 113, 1730–1736.
(26) (a) Ashby, E. C. Acc. Chem. Res. 1988, 21, 414–421. (b) Houmam, A.
Chem. Rev. 2008, 108, 2180–2237.
(27) (a) Harvey, R. G. Synthesis 1970, 2, 161–172. (b) Harvey, R. G.;
Arzadon, L. Tetrahedron 1969, 25, 4887–4894. (c) Rabideau, P. W.;
Burkholder, E. G. J. Org. Chem. 1978, 43, 4283–4288. (d) Rossi, R. A.;
Camusso, C. C.; Madoery, O. D. J. Org. Chem. 1974, 39, 3254–3258.
(e) Rabideau, P. W.; Harvey, R. G. J. Org. Chem. 1970, 35, 25–30.
€
(28) (a) Mullen, K.; Huber, W.; Neumann, G.; Schnieders, C.; Unterberg,
H. J. Am. Chem. Soc. 1985, 107, 801–807. (b) Mullen, K. Angew. Chem., Int.
Ed. Engl. 1987, 26, 204–217.
(29) Zimmerman, H. E.; Wang, P. A. J. Am. Chem. Soc. 1993, 115, 2205–
2216 and references cited therein.
€
(23) (a) Garst, J. F. Acc. Chem. Res. 1971, 4, 400–406. (b) Gawley, R. E.;
Low, E.; Zhang, Q.; Harris, R. J. Am. Chem. Soc. 2000, 122, 3344–3350. (b)
Hazimeh, H.; Mattalia, J. M.; Marchi-Delapierre, C.; Barone, R.;
Nudelman, N. S.; Chanon, M. J. Phys. Org. Chem. 2005, 18, 1145–1160.
J. Org. Chem. Vol. 74, No. 17, 2009 6475

Lebeuf et al.
JOCFeatured Article
FIGURE 2. Mechanism of the Birch reductive alkylation of biaryls.
the basicity of C (vide supra) and modify its long lifetime in
the medium, leading in certain cases to larger amount of
reduced products E through protonation of C by ammonia,
but also ammonium phenolate or alkylating agents such as
R-haloesters. Among the alternative pathway, the occur-
rence of a dianion has been proposed several times for the
reduction of biaryls under Birch-type conditions (NH3) or
during lithium reduction in THF.9d,22 A second electron
transfer from Li to radical-anion A may thus provide dianion
D. Protonation of D at C-4 would then return monoanion C
that can be protonated or alkylated as above. Deeply colored
solutions are observed upon mixing the biaryl with Li and
NH3. For instance, in our hands, reduction of 34e led to a
solution having a green coloration during addition of lithium
wire, which turned deep red after 2-3 min. This coloration
rapidly vanished after several minutes, the alkylating agent
being generally added after 5 min at -78 °C. We noticed that
addition of the alkylating agent after the deep red coloration
had disappeared resulted in a higher amount of reduced
products. Literature precedent9d suggests that in our case the
green coloration is due to the radical anion A, a stable
intermediate in the medium when only 1 equiv of lithium is
added, which may give rise, upon addition of 2-2.5 equiv of
Li, to dianion D or radical B. Both B and D are nonpersistent
species in the medium and thus are readily transformed
into the monoanion C which is then protonated or alkylated.
The red solution may thus be attributed to anion C, whose
coloration vanishes as a result of the protonation by
NH3. The persistence of C in the medium may thus be
qualitatively assessed by rapid changes in the coloration
of the solution prior to addition of the electrophile. During
our attempt to scale up the Birch reductive alkylation
SCHEME 12. Dialkylation of Biaryl 20
of biaryl 20, we have observed the formation of the dialky-
lated product 45 as a 1:3 cis/trans30 mixture of two diaster-
eomers27b (along with the desired product 44, Scheme 12).
Compound 45 might result from a regioselective 1,4-dialkyla-
tion of dianion such as D. Dialkylation of lithium dianion of
biphenyl 7 has been observed by Yus et al.31 during their
studies on lithiation of polyaromatic compounds in THF.
High concentration of strongly basic dianion D, however,
seems incompatible with the presence of NH3 as a solvent.
Dianion D is rapidly protonated in NH3 leading to C, which
is likely the only stable intermediate species in this media.9b
(31) (a) Melero, C.; Guijarro, A.; Yus, M. Tetrahedron Lett. 2006, 47,
6267–6271. (b) Melero, C.; Guijarro, A.; Baumann, V.; Jimenez, A. J. P.;
ꢀ
Yus, M. Eur. J. Org. Chem. 2007, 5514–5526. (c) Melero, C.; Perez, H.;
Guijarro, A.; Yus, M. Tetrahedron Lett. 2007, 48, 4105–4109 and references
cited therein.
(30) The 1:3 cis/trans mixture could be partially separated with the first
eluting cis isomer isolated pure. The relative stereochemistry of each isomer
was assigned on the basis of 2D 1H NMR experiments.
6476 J. Org. Chem. Vol. 74, No. 17, 2009

Lebeuf et al.
JOCFeatured Article
SCHEME 13. Iodoetherification of Dienylphenol 46
our regioselective Li/NH3 reduction of biaryls complements
the well-known Birch reductive alkylation of benzoic acid and
heterocyclic esters and amides, which is central to many total
syntheses.33
Experimental Section
General Procedure for Birch Reductive Alkylation with De-
protonation. A dried, three-necked, round-bottomed flask was
fitted with a coldfinger condenser connected to a vacuum-
nitrogen line. Another neck was fitted with a gas inlet valve
connected to an ammonia tank. A solution of biaryl in anhy-
drous THF (0.3 M) was introduced under nitrogen. The flask
and the coldfinger condenser were cooled -20 °C, n-BuLi (2.5 M
solution in hexane, 1.1 equiv) was added dropwise, and the
solution was stirred for 10 min. Then ammonia was condensed
(twice more than THF) at -78 °C. Finely cut lithium (2.2 equiv)
was added portionwise at -78 °C under nitrogen pressure. The
solution turned rapidly blue during lithium addition, then
brown, and finally brick red. After 45 min of stirring at this
temperature, a cold solution of electrophile (3 equiv) in anhy-
drous THF (4 M) was added dropwise over 5 min keeping the
temperature below -78 °C. The mixture turned immediately
brown. After 20 min of stirring, the reaction was quenched by
addition of solid ammonium chloride (5 equiv). The cooling
bath and condenser were removed, and ammonia was allowed to
evaporate under air. When the reaction mixture reached room
temperature, a 1:1 solution of aqueous saturated ammonium
chloride and water was added. The aqueous phase was extracted
three times with ether or ethyl acetate. The combined organic
phases were washed with brine and dried over sodium sulfate.
The brown paste, obtained after filtration and removal of the
solvents, was purified by flash chromatography (petroleum
ether/EtOAc mixtures).
[1-(3-Hydroxy-5-methoxyphenyl)cyclohexa-2,5-dienyl]acetic
Acid Methyl Ester (25b). Compound 25b was synthesized ac-
cording to the general procedure from 3-hydroxy-5-methoxybi-
phenyl 24 (2.25 g, 11.25 mmol, 1 equiv), THF (50 mL), n-BuLi
(2.5 M in hexane, 5.0 mL, 12.5 mmol, 1.1 equiv), ammonia
(approximately 100 mL), lithium (0.173 g, 24.71 mmol, 2.2
equiv), methyl chloroacetate (3.1 mL, 34 mmol, 3 equiv), and
THF (10 mL). Purification by flash chromatography (silica gel,
petroleum ether/EtOAc 75/25) afforded 25b as a yellow oil (1.51
g, 49%): IR (film, NaCl) νmax (cm-1) 3404, 2952, 1715, 158,
1434, 1195, 1159, 1059, 967; 1H NMR (CDCl3, 300 MHz) δ =
6.45-6.44 (m, 1H), 6.42-6.41 (m, 1H), 6.27-6.25 (m, 1H),
5.87-5.76 (m, 4H), 3.73(s, 3H), 3.59 (s, 3H), 2.81 (s, 2H), 2.63
(broad s, 2H); 13C NMR (CDCl3, 75.5 MHz): δ = 172.0, 160.9,
157.1, 149.4, 131.1, 124.2, 106.2, 105.1, 99.2, 55.4, 51.7, 45.6,
42.7, 25.9; HRMS (EI) [M]þ• C16H18O4 calcd 274.1205, found
274.1212 (2 ppm).
The formation of dialkylated product 45 was only observed
during large-scale synthesis where temperature is more diffi-
cult to control due to the exothermicity of the process
and when the amount of LiNH2 present in the medium
would be sufficient to deprotonate F, allowing the alkylation
in the presence of an excess of electrophile. This is in
good agreement with the observations of Harvey,9d who
reported a higher amount of polyalkylated products when
reactant concentration was high and attributed that to a
larger amount of lithium amide and cyclohexadiene in the
medium.
Conclusion
In summary, we described a procedure for rapid access to
symmetrical arylcyclohexadienes through a regioselective
Birch reductive alkylation of biaryls. Reduction of a series of
biaryl precursors bearing electron-donating substituents, fol-
lowed by alkylation of the resulting anion with various elec-
trophiles, has been investigated, demonstrating that high levels
of regiocontrol could be attained through a careful choice of
substituents on the arene moieties. Biaryls having an aromatic
ring with two methoxy groups in the meta position relative to
the biaryl linkage were reduced selectively. Arenes with OH
and NHR substituents were not reduced if deprotonation
(with n-BuLi) of these functions was carried out prior to the
reaction with Li/NH3. We also observed that a slight mod-
ification of the position and the nature of substituents on the
arenes profoundly modified the protonation vs alkylation
ratio. This study establishes the scope and limitation of the
Birch reductive alkylation of biaryls and demonstrates that it
may be a valuable tool for organic synthesis, offering a
straightforward entry toward cyclohexa-2,5-dienyl arene sys-
tems bearing a quaternary center. These dienes are valuable
synthons that can be elaborated further into highly functiona-
lized intermediates. For instance, removal of the silyl protect-
ing group from ortho-substituted diene 33a led to phenol 46,
which was then engaged in a iodoetherification process,6,32
affording, in good yield, a benzofuran 47, having two qua-
ternary stereogenic centers set up in a single step (Scheme 13).
Such a synthon should find utility en route to morphinan and
Amaryllidaceae alkaloids. Some utilization of these dienyl
precursors have been reported recently4d,14 and other meth-
odologies relying on this approach are currently under scrutiny
in our laboratories and will be reported in due course. Finally,
N-(2-(1-(Cyanomethyl)cyclohexa-2,5-dienyl)phenyl)ethane-
sulfonamide (35e). Compound 35e was synthesized according to
the general procedure from biaryl 34e (3 g, 11.5 mmol, 1 equiv),
THF (60 mL), n-BuLi (5.06 mL, 12.65 mmol, 1.1 equiv),
ammonia (approximately 120 mL), lithium (200 mg, 28.7 mmol,
2.5 equiv), chloroacetonitrile (2.2 mL, 34.5 mmol, 3 equiv), and
THF (15 mL). Purification by flash chromatography (silica gel,
petroleum ether/EtOAc 80/20) afforded 35e as a yellow solid
(2.4 g, 70%): mp 124.5-125.1 °C; IR (solid, KBr) νmax (cm-1
)
2953, 1718, 1522, 1431, 1346, 1196, 975; 1H NMR (CDCl3,
300 MHz) δ (ppm) 7.60-7.49 (m, 1H), 7.35-7.25 (m, 1H),
(33) (a) Schultz, A. G. Chem. Commun. 1999, 1263–1271 and references
cited therein. (b) Donohoe, T. J.; House, D. J. Org. Chem. 2002, 67, 5015–
5018. (c) Donohoe, T. J.; Johnson, D. J.; Mace, L. H.; Bamford, M. J.;
Ichihara, O. Org. Lett. 2005, 7, 435–437. (d) Gopalan, A.; Magnus, P. J. Org.
Chem. 1984, 49, 2317–2321.
(32) Butters, M.; Elliott, M. C.; Hill-Cousins, J.; Paine, J. S.; Walker,
J. K. E. Org. Lett. 2007, 9, 3635–3638.
J. Org. Chem. Vol. 74, No. 17, 2009 6477

Lebeuf et al.
JOCFeatured Article
7.18-7.12 (m, 3H), 6.31-6.18 (m, 2H), 5.70-5.58 (m, 2H), 3.13
(q, J = 7.5 Hz, 2H), 3.01-2.95 (m, 2H), 2.93 (s, 2H), 1.31 (t, J =
7.1 Hz, 3H); 13C NMR (CDCl3, 75.5 MHz) δ (ppm) 137.2, 130.1,
129.3, 128.8, 128.3, 125.3, 124.2, 119.8, 117.0, 46.9, 41.2, 30.6,
26.0, 8.1; MS (ESI) m/z 325 [M þ Na]þ (100); HRMS (ESI) calcd
for C16H18N2O2NaS [M þ Na]þ 325.0987, found 325.0977.
Anal. Calcd for C16H18N2O2S (302.1): C, 63.55; H, 6.00; N,
9.26; S, 10.60. Found: C, 63.56; H, 6.03; N, 9.12; S, 10.24.
Ethanesulfonic Acid (6-Cyanomethyl-2-methoxy-6-phenyl-
cyclohexa-1,4-dienyl)amide (37). Compound 37 was synthesized
according to the general procedure from biaryl 36 (0.5 g, 1.71
mmol, 1 equiv), THF (10 mL), n-BuLi (1 mL, 1.881 mmol, 1.1
equiv), ammonia (approximately 20 mL), lithium (29.8 mg,
4.27 mmol, 2.5 equiv), chloroacetonitrile (387 mg, 5.13 mmol,
3 equiv), and THF (5 mL). Purification by flash chromatography
(silica gel, petroleum ether/EtOAc 80/20) afforded 37 as a yellow
oil (0.349 g, 60%): IR (film, NaCl) νmax (cm-1) 3263, 2941, 2361,
1687, 1494, 1415, 1316, 1238, 1135, 1038, 888; 1H NMR (CDCl3,
300 MHz) δ (ppm) 7.38-7.28 (m, 2H), 7.23-7.19 (m, 2H),
6.89-6.85 (m, 1H), 5.94-5.88 (m, 1H), 5.64-5.61(m, 1H), 3.62
(s, 3H), 3.27-3.10 (m, 2H), 3-2.91 (m, 2H), 2,48 (q, J = 4.5 Hz,
2H), 1.24 (t, J =7.5 Hz, 3H); 13C NMR (CDCl3, 75.5 MHz) δ
(ppm) 149.9, 141.1, 130.8, 129.1, 127.9, 127.3, 122.50, 117.9,
112.8, 55.1, 48.5, 48.4, 27.2, 26.7, 8.4; MS (ESI) m/z 355 (100)
[M þ Na]þ, 200 (20) [C13H13NO þ H]þ; HRMS (ESI) calcd for
C17H20N2O3NaS [M þ Na]þ 355.1086, found 355.1082.
chromatography (silica gel, petroleum ether/EtOAc 90/10)
afforded 40j as a yellow oil (2.45 g, 60%): IR (solid, KBr) νmax
(cm-1) 2940, 1718, 1664, 1486, 1342, 1153, 996; 1H NMR (CDCl3,
300 MHz) δ (ppm) 7.58-7.56 (m, 1H), 7.55-7.50 (m, 1H),7.36-
7.33 (m, 1H), 7.25-7.22 (m, 1H), 7.11-7.08 (m, 1H), 6.09-6.02
(m, 2H), 5.56-5.53 (m, 2H), 4.41 (t, J = 4.5 Hz, 1H), 3.30 (s, 6H),
3.10 (q, J = 7.5 Hz, 2H), 2.95-2.87 (m, 2H), 2.23 (d, J = 4.1 Hz,
2H), 1.29 (t, J = 7.5 Hz, 3H); 13C NMR (CDCl3, 75.5 MHz) δ
(ppm) 137.8, 132.5, 131.0, 128.5, 126.3, 125.3, 124.0, 119.3, 102.7,
53.1, 46.4, 43.6, 41.1, 25.7, 8.1; MS (ESI) m/z 374 [M þ Na]þ (100);
HRMS (ESI) calcd for C18H25NO4NaS [M þ Na]þ 374.1402,
found 374.1399.
N-(2-(1-(2-(Ethylsulfonamido)phenyl)cyclohexa-2,5-dienyl)-
ethyl)-4-methylbenzenesulfonamide (42b). Compound 42b was
synthesized according to the general procedure from biaryl
34e (500 mg, 1.91 mmol, 1 equiv), THF (7 mL), n-BuLi
(0.95 mL, 2.1 mmol, 1.1 equiv), ammonia (approximately
14 mL), lithium (33 mg, 4.77 mmol, 2.5 equiv), 1-tosylaziridine
(942 mg, 4.77 mmol, 2.5 equiv), and THF (4 mL). Purification
by flash chromatography (silica gel, petroleum ether/EtOAc
80/20) afforded 42b as a pale pinky solid (341 mg, 40%): mp
140.3-141.6 °C; IR (film, NaCl) νmax (cm-1) 3283, 1598, 1441,
1318, 1145, 1084, 894, 809; 1H NMR (300 MHz, CDCl3) δ
(ppm) 7.74 (d, J=7.14 Hz, 2H), 7.52 (d, J=7.89 Hz, 1H), 7.32-
7.20 (m, 4H), 7.04 (t, J=7.71 Hz,1H), 6.02 (d, J=8.47 Hz, 2H),
5.39 (d, J=9.78 Hz, 2H), 3.11-3.01 (m, 4H), 2.76 (q, Jab
=
4.20 Hz, 2H), 2.42 (s, 3H), 2.06 (t, J=7.53 Hz, 2H), 1.27 (t, J= 7.33
Hz, 3H); 13C NMR (75 MHz, CDCl3) δ (ppm) 143.6, 137.6, 137.1,
132.3, 130.1, 129.9, 128.6, 127.2, 126.8, 126.1, 124.0, 119.3, 46.5,
41.6, 39.6, 39.0, 25.8, 21.6, 8.1; MSMS (ESI) m/z 183 (39), 192 (32)
[M þ Na - CH2CH2NHTs-SO2Et]þ, 206 (43), 235 (52) [M þ Na
- Ts-SO2Et]þ, 284 (100) [M þ Na - CH2CH2NHTs]þ, 352 (19)
[M - SO2Et]þ, 389 (65) [M þ Na - HSO2Et]þ, 483 (59) [M þ
Na]þ; HRMS (ESI) calcd for C23H28N2O4S2Na [M þ Naþ]
483.1382, found 483.1394.
1-(tert-Butylcyclohexa-2,5-dienyl)benzene (40g). Compound
40g was synthesized according to the general procedure from
biaryl 7 (2 g, 12.97 mmol, 1 equiv), THF (40 mL), ammonia
(approximately 80 mL), lithium (209 mg, 29.83 mmol, 2.3
equiv), tert-butyl bromide (3.6 mL, 32.42 mmol, 2.5 equiv),
and THF (18 mL). Purification by flash chromatography (silica
gel petroleum ether 100%) afforded an analytic sample of 40g as
a colorless oil: IR (film, NaCl) νmax (cm-1) 3034, 2965, 2873,
1
1597, 1495, 1391, 1362, 948. H NMR (300 MHz, CDCl3) δ
(ppm) 7.41-7.17 (m, 5H), 6.47 (d, J = 10.72 Hz, 2H), 5.95 (dd,
J = 3.16, 7.32 Hz, 2H), 2.61 (AB, Jab = 36 Hz, 2H), 0.96 (s, 9H);
13C NMR (75 MHz, CDCl3) δ (ppm) 146.2, 130.1, 128.2, 127.4,
125.5, 124.8, 47.4, 37.7, 26.6, 26.4; MS (ESI) m/z 319.06 (100)
[M þ Ag]þ; HRMS (ESI) calcd for C16H20Ag [M þ Ag]þ
319.0610, found 319.0615.
N-(2-(1-(2,2-Dimethoxyethyl)cyclohexa-2,5-dienyl)phenyl)-
ethanesulfonamide (40j). Compound 40j was synthesized accord-
ing to the general procedure from biaryl 34e (3 g,
11.5 mmol, 1 equiv), THF (60 mL), n-BuLi (6.02 mL, 12.65 mmol,
1.1 equiv), ammonia (approximately 120 mL), lithium (200 mg,
28.7 mmol, 2.5 equiv), 2-bromo-1,1-dimethoxyethane (5.83 mg,
34.5 mmol, 3 equiv), and THF (15 mL). Purification by flash
Acknowledgment. We thank the CNRS for financial sup-
port, MNERT for grants to R.L., J.D., and G.B., as well as
UNESCO for a Ph.D. grant to D.I. We gratefully acknow-
ledge J.-C. Lartigue and C. Vitry for NMR and mass
spectrometry experiments.
Supporting Information Available: Experimental proce-
dures and spectroscopic data for compounds not described in
the Experimental Section and copies of 1H and 13C NMR
spectra for all new compounds. This material is available free
of charge via the Internet at http://pubs.acs.org.
6478 J. Org. Chem. Vol. 74, No. 17, 2009