Use of aromatic radical-anions in the absence of THF. Tandem formation and cyclization of benzyllithiums derived from the attack of homo- and bishomoallyllithiums on α-methylstyrenes: Two-pot synthesis of cuparene

Article abstract of DOI:10.1021/ja004353+

When a homo- or bishomoallyllithium, generated by reductive lithiation of the corresponding phenyl thioether by the radical anion lithium 1-(dimethylamino)naphthalenide (LDMAN), is added to α-methylstyrene, a tandem addition/cyclization to a phenyl-substituted five- or six-membered-ring occurs. The yields are compromised by polymerization of the α-methylstyrene, a process favored by tetrahydrofuran (THF), the solvent used to generate lithium aromatic radical anions. Thus, a new method of generating LDMAN (unsuccessful for other common radical anions) in the absence of THF has been developed. The radical anion can be generated and the reductive lithiation performed in dimethyl ether at -70 °C. After the addition of diethyl ether or other solvent, and evaporation of the dimethyl ether in vacuo, the α-methylstyrene is added and the solution is warmed to -30 °C. When the unsaturated alkyllithium is primary, no adduct forms in THF due to polymerization of the α-methylstyrene, but moderate yields are attained in a solvent containing mainly hexanes. It was also found that the cyclized organolithiums, which would have become protonated in the presence of THF, can be captured by an electrophile, even at ambient temperature. A two-pot synthesis, the most efficient reported, of the sesquiterpene (±)-cuparene in 46% yield, using this technology is reported.

Full text of DOI:10.1021/ja004353+

3478
J. Am. Chem. Soc. 2001, 123, 3478-3483
Use of Aromatic Radical-Anions in the Absence of THF. Tandem
Formation and Cyclization of Benzyllithiums Derived from the Attack
of Homo- and Bishomoallyllithiums on R-Methylstyrenes: Two-Pot
1
Synthesis of Cuparene
Theodore Cohen,* Thanapong Kreethadumrongdat, Xiaojun Liu, and
Vithalanand Kulkarni
Contribution from the Department of Chemistry, UniVersity of Pittsburgh, Pittsburgh, PennsylVania 15260
ReceiVed December 27, 2000
Abstract: When a homo- or bishomoallyllithium, generated by reductive lithiation of the corresponding phenyl
thioether by the radical anion lithium 1-(dimethylamino)naphthalenide (LDMAN), is added toR-methylstyrene,
a tandem addition/cyclization to a phenyl-substituted five- or six-membered-ring occurs. The yields are
compromised by polymerization of the R-methylstyrene, a process favored by tetrahydrofuran (THF), the solvent
used to generate lithium aromatic radical anions. Thus, a new method of generating LDMAN (unsuccessful
for other common radical anions) in the absence of THF has been developed. The radical anion can be generated
and the reductive lithiation performed in dimethyl ether at -70 °C. After the addition of diethyl ether or other
solvent, and evaporation of the dimethyl ether in vacuo, the R-methylstyrene is added and the solution is
warmed to -30 °C. When the unsaturated alkyllithium is primary, no adduct forms in THF due to polymerization
of the R-methylstyrene, but moderate yields are attained in a solvent containing mainly hexanes. It was also
found that the cyclized organolithiums, which would have become protonated in the presence of THF, can be
captured by an electrophile, even at ambient temperature. A two-pot synthesis, the most efficient reported, of
the sesquiterpene (()-cuparene in 46% yield, using this technology is reported.
2
Since its discovery in 1978, the reductive lithiation of phenyl
cessful for the generation of the aromatic radical-anions with
lithium counterions. We and others have often found THF to
be detrimental to the optimum utilization of reductive lithiation.
thioethers with aromatic radical-anions in tetrahydrofuran (THF)
has become recognized as one of the most general methods of
organolithium production (Scheme 1).3 Its great versatility has
been demonstrated repeatedly. For example, it is one of the rare
methods, and a particularly general one, for the production of
tertiary organolithiums5 and, unlike deprotonation and related
,4
(4) For some recent examples, in addition to the ones referred to below,
see: Foubelo, F.; Yus, M. Tetrahedron Lett. 2000, 41, 5047-51. Christo-
pher, J. A.; Kocienski, P. J.; Kuhl, A.; Bell, R. Synlett 2000, 463-466.
Strohmann, C.; L u¨ dtke, S.; Ulbrich, O. Organometallics 2000, 19, 4223-
27. Miyazaki, H.; Honda, Y.; Honda, K.; Inoue, S. Tetrahedron Lett. 2000,
,6
7
3
methods involving electrophile removal, sp organolithiums are
generated substantially more rapid than the more stable sp
41, 2643-47. Ireland, T.; Perea, J. J. A.; Knochel, P. Angew. Chem., Int.
2
Ed. Engl. 1999, 38, 1457-60. Chen, F.; Mudryk, B.; Cohen, T. Tetrahedron
1999, 55, 3291-3304. Huang, P. Q.; Zheng, X.; Wang, S. L.; Ye, J. L.;
Jin, L. R.; Chen, Z. Tetrahedron: Asymmetry 1999, 10, 3309-17.
Hollingsworth, G. J.; Dinnell, K.; Dickinson, L. C.; Elliot, J. M.;
Kalugowski, J. J.; Swain, C. J.; Thomson, C. G. Tetrahedron Lett. 1999,
organolithiums.8,9
A number of somewhat less versatile10 synthetic methods also
depend on reductive lithiation with aromatic radical-anions.
1
1
1
2
13
These include the reduction of organic halides, sulfones,
4
0, 2633-36. Kessar, S. V.; Singh, P.; Singh, K. N.; Singh, S. K. J. Chem.
14
15
16
17,18
sulfates, nitriles, selenides; allylic and benzylic ethers,
Soc., Chem. Commun. 1999, 1927-28. Rychnovsky, S. D.; Muckmelter,
A. J.; Dahanukar, V. L.; Skalitzky, D. J. J. Org. Chem. 1999, 64, 6849-
19
19
20
21
sulfides, amines, and acetals; small ring heterocycles such
60. Foubelo, F.; Guti e´ rrez, A.; Yus, M. Synthesis 1999, 503-514. Liu, H.;
as epoxides,22 oxetanes, and tetrahydrofurans,
23
18,24
and their
Shook, C. A.; Thiruvazhi, M.; Jamison, J. A.; Cohen, T. J. Am. Chem. Soc.
1998, 120, 605-606. Amano, S.; Fujiwara, K.; Murai, A. Chem. Lett. 1998,
409-410. Cohen, T.; Zhu, S. Tetrahedron 1997, 53, 17607-17624. Cohen,
T.; Tong, S. Tetrahedron 1997, 53, 9487-9496. Sammtleben, F.; Nolte-
meyer, M.; Br u¨ ckner, R. Tetrahedron Lett. 1997, 38, 3893-96. Coldham,
I.; Holman, S.; Lang-Anderson, M. M. S. J. Chem. Soc., Perkin Trans. 1
2
1
25
26
amine and thioether analogues; and carboxylic acids, esters,
and certain ketones.27
One of the few weaknesses of these methods is the necessity
of using THF as the solvent. No other solvent has been suc-
1
997, 1481-85. Gibson, C.; Noltemeyer, M.; Br u¨ ckner, R. Tetrahedron
*
To whom correspondence should be addressed: cohen@pitt.edu; 412-
Lett. 1997, 38, 2933-36. Foubelo, F.; Guti e´ rrez, A.; Yus, M. Tetrahedron
Lett. 1997, 38, 4837-4840.
6
1
24-8220.
(
1) Taken in part from the M.S. thesis of X. Liu, University of Pittsburgh,
999. A preliminary report of some of this work has appeared: Liu, X.;
Kulkarni, V.; Cohen, T. Book of Abstracts; 219th National Meeting of the
American Chemical Society, San Francisco, March 26-30, 2000; American
Chemical Society: Washington, DC, 2000; ORGN-630.
(5) Mudryk, B.; Cohen, T. J. Am. Chem. Soc. 1993, 115, 3855-65.
(6) (a) Cherkauskas, J. P.; Cohen, T. J. Org. Chem. 1992, 57, 6-8.
(b) Cohen, T.; Zhang, B.; Cherkauskas, J. P. Tetrahedron 1994, 50,
11569-84.
(7) Gray, M.; Tinkl, M.; Snieckus, V. In ComprehensiVe Organo-
metallic Chemistry II; McKillop, A., Ed.; Pergamon: Oxford, 1995; Vol.
11, Chapter 1.
(8) Cohen, T.; Doubleday: M. D. J. Org. Chem. 1990, 55, 4784-86.
Doubleday, M. D., Ph.D. Thesis, University of Pittsburgh, 1991.
(9) Chen, F.; Mudryk, B.; Cohen, T. Tetrahedron 1999, 55, 3291-3304.
(
2) Screttas, C. G.; Micha-Screttas, M. J. Org. Chem. 1978, 43, 1064-
1
7
1
071. Screttas, C. G.; Micha-Screttas, M. J. Org. Chem. 1979, 44, 713-
19. Cohen, T.; Daniewski, W. M.; Weisenfeld, R. B. Tetrahedron Lett.
978, 4665-4668.
(
3) Cohen, T.; Bhupathy, M. Acc. Chem. Res. 1989, 22, 152-161.
1
0.1021/ja004353+ CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/23/2001

Use of Aromatic Radical-Anions in the Absence of THF
J. Am. Chem. Soc., Vol. 123, No. 15, 2001 3479
The ability of organolithiums to remove a proton from the
Scheme 1
2-position of THF is sometimes a major problem especially for
the more basic organolithiums or at temperatures above about
5
,28-30
0
°C.
In other cases, as in that described below of the
(10) The superior versatility of compounds containing the phenylthio
group as substrates for reductive lithiation arises from their almost unique
ease of construction, particularly by methods involving C-C bond formation
but also by the ability of the phenylthio group to enter a molecule as a
nucleophile, electrophile, or radical. In addition, the substrates are almost
always able to withstand the powerful nucleophiles/bases that are present
in the reductive lithiation conditions. For example, alkyl halides, sulfates,
sulfonates, etc. (see below) are subject to ready nucleophilic substitution,
but most seriously to base-induced elimination, thus limiting their use largely
to the preparation of primary alkyllithiums unless an aryl or vinyl group is
present to increase the rate of the reductive lithiation and favor it over
competing processes.
Scheme 2
(11) Reviews of those in which the aromatic is present in only catalytic
amounts: Yus, M. Chem. Soc. ReV. 1996, 155-161. Ram o´ n, D. J.; Yus,
M. Eur. J. Org. Chem. 2000, 225-237. In some of the many cases described
in these reviews, it is not always certain that the organolithium is actually
produced since the electrophile, usually a carbonyl compound, is present
during the reductive lithiation. It seems likely in these cases that a fairly
stable ketyl is produced first and that it reacts with the radical intermediate
in the reductive lithiation of the substrate. However, it appears likely that
many of the substrates could be converted to alkyllithiums under the
stoichiometric conditions that are probably necessary to take advantage of
the new reductive lithiation procedure introduced in the present article.
(12) Sargent, G. D.; Browne, M. W. J. Am. Chem. Soc. 1967, 89, 2788.
Wilson, S. E. Tetrahedron Lett. 1975, 4651-4654. Caine, D.; Frobese, A.
S. Tetrahedron Lett. 1978, 883-86. Barluenga, J.; Fl o´ rez, J.; Yus, M. J.
Chem. Soc., Perkin Trans. 1 1983, 3019-26. Beau, J.-M.; Prandi, J.
Tetrahedron Lett. 1989, 30, 4517-20. Ram o´ n, D. J.; Yus, M. Tetrahedron
Lett. 1990, 31, 3763-3766. Rawson, D. J.; Meyers, A. I. Tetrahedron Lett.
cyclization of benzyllithiums which were generated by attack
of homo- and bishomoallyllithiums on R-methylstyrene, the
presence of THF encourages an unwanted side reaction. Thus,
it was deemed desirable to find an alternative solvent in which
radical-anions could be generated and which could either be
used for the subsequent reactions or be readily replaced by
suitable solvents.
1991, 32, 2095-98. Wittman, V.; Kessler, H. Angew. Chem., Int. Ed. Engl.
993, 32, 1091-93. Oku, A.; Ose, Y.; Kamada, T.; Yoshida, T. Chem.
1
Lett. 1993, 573-76. Vlaar, C. P.; Klumpp, G. W. Angew. Chem., Int. Ed.
Engl. 1993, 32, 574-76. Barluenga, J.; Montserrat, J. M.; Fl o´ rez, J. J. Org.
Chem. 1993, 58, 5976-80. Kondo, Y.; Murata. N.; Sakamoto, T. Hetero-
cycles 1994, 37, 1467-68. Lesimple, P.; Beau, J.-M. Bioorg. Med. Chem.
994, 2, 1319-30. Tamao, K.; Kawachi, A. Organometallics 1995, 14,
Generation of a Lithium Aromatic Radical-Anion in the
Absence of THF
In preliminary experiments, the formation of several lithium
31
radical-anions, lithium p,p′-di-tert-butylbiphenylide (LDBB),
32
(13) Beau, J.-M.; Sina y¨ , P. Tetrahedron Lett. 1985, 26, 6185-88, 6189-
lithium 1-(dimethylamino)naphthalenide (LDMAN), and lithium
2,6-di-tert-butylnaphthalenide (LDBN), in a variety of solvents
other than THF was studied in an attempt to find such a solvent
and the proper radical-anion. Bis(phenylthio)methane was used
as a reductive lithiation substrate to estimate the approximate
yields of radical-anions in a specified solvent by noting when
during addition of the thioacetal the color of the radical-anion
disappeared (Scheme 2).
(14) Guijarry, D.; Mancheno, B.; Yus, M. Tetrahedron Lett. 1992, 33,
5
597-5600. Guijarro, D.; Guillena, G.; Mancheno, B.; Yus, M. Tetrahedron
1
994, 50, 3427-36.
(15) Zeller, E.; Sajus, H.; Grierson, D. S. Synlett 1991, 44-46.
Buckmelter, A. J.; Powers, J. P.; Rychnovsky, S. D. J. Am. Chem. Soc.
998, 120, 5589-90.
16) Agawa, T.; Ohshiro, Y. Synthesis 1980, 933-935. Hoffmann, R.;
1
Attempted generation of LDBB in other solvents, by stirring
a suspension of lithium wire cuttings and DBB together in the
selected solvent, was unsuccessful (Table 1). When diethyl ether
was used, no radical-anion was formed even at ambient
(
Br u¨ ckner, R. Chem. Ber. 1992, 125, 1957-63. Krief, A.; Hobe, M.; Badaoui,
E.; Bousbaa, J.; Dumont, W.; Nazih, A. Synlett 1993, 707-9. Krief, A.;
Nazih, A.; Hobe, M. Tetrahedron Lett. 1995, 36, 8111-14. Krief, A.; Nazih,
A. Tetrahedron Lett. 1995, 36, 8115-18.
(
17) Eisch, J. J.; Jacobs, A. M. J. Org. Chem. 1963, 28, 2145-46.
Lansbury, P. T.; Caridi, F. J. J. Chem. Soc., Chem. Commun. 1970, 714-
5. Ireland, R. E.; Smith, M. G. J. Am. Chem. Soc. 1988, 110, 854-60.
Sa a´ , J. M.; Ballester, P.; Dey a´ , P. M.; Cap o´ , M.; Garc ´ı as, X. J. Org. Chem.
(23) Mudryk, B.; Cohen, T. J. Org. Chem. 1989, 54, 5657-59. Mudryk,
B.; Cohen, T. J. Org. Chem. 1991, 56, 5760-61. Hashemzadeh, M.; Howell,
A. R. Tetrahedron Lett. 2000, 41, 1855-58. Hashemzadeh, M.; Howell,
A. R. Tetrahedron Lett. 2000, 41, 1859-62.
(24) Oikawa, M.; Oikawa, H.; Ichihara, A. Tetrahedron 1995, 51, 6237-
54. Dvorak, C. A.; Rawal, V. H. J. Chem. Soc., Chem. Commun. 1997,
2381-82.
(25) Karaman, R.; Fry, J. L. Tetrahedron Lett. 1989, 30, 6267-70.
(26) Karaman, R.; Fry, J. L. Tetrahedron Lett. 1989, 30, 4935-38.
(27) Rawal, V. H.; Dufour, C.; Iwasa, S. Tetrahedron Lett. 1995, 19-
22.
1
1
1
2
996, 61, 1035-46. Azzena, U.; Demartis, S.; Melloni, G. J. Org. Chem.
996, 61, 3-19. Azzena, U.; Demartis, S.; Pilo, L.; Piras, E. Tetrahedron
000, 56, 8375-82.
(
18) Mudryk, B.; Cohen, T. J. Am. Chem. Soc. 1991, 113, 1866-67.
(19) Block, E.; Guo, C.; Thiruvazhi, M.; Toscano, P. J. J. Am. Chem.
Soc. 1994, 116, 9403-04.
20) Azzena, U.; Melloni, G.; Pisano, L.; Sechi, B. Tetrahedron Lett.
(
1
994, 35, 6759-62.
(28) Rychnovsky, S. D.; J. Org. Chem. 1989, 54, 4982-84. Jung, M.
E.; Blum, R. B. Tetrahedron Lett. 1977, 3791-94.
(29) Krief, A.; Barbeaux, P. Synlett 1990, 511-514.
(30) (a) Chen, F.; Mudryk, B.; Cohen, T., Tetrahedron 1994, 50, 12793-
12810. (b) Cheng, D.; Zhu, S.; Liu, X.; Norton, S. H.; Cohen, T. J. Am.
Chem. Soc. 1999, 121, 10241-42.
(
21) Yus, M.; Foubelo, F. ReV. Heteroat. Chem. 1997, 17, 73-107.
22) Bartmann, E. Angew. Chem., Int. Ed. Engl. 1986, 25, 653. Cohen,
(
T.; Jeong, I.-H.; Mudryk, B.; Bhupathy, M.; Awad, M. M. A. J. Org. Chem.
1
990, 55, 1528-36. Mudryk, B.; Cohen, T. Org. Synth. 1995, 72, 173-9.
Behnke, D.; Hennig, L.; Findeisen, M.; Welzel, P.; M u¨ ller, D.; Thormann,
M.; Hofmann, H.-J. Tetrahedron 2000, 56, 1081-95. Falvello, L. R.;
Foubelo, F.; Soler, T.; Yus, M. Tetrahedron: Asymmetry 2000, 11, 2063-
(31) Freeman, P. K.; Hutchinson, L. L. J. Org. Chem. 1980, 22, 1924-
30.
6
6.
(32) Cohen, T.; Matz, J. R. Synth. Commun. 1980, 10, 311-17.

3480 J. Am. Chem. Soc., Vol. 123, No. 15, 2001
Cohen et al.
Table 1. Radical-Anion Formation in THF and in Diethyl Ether
The LDMAN case thus appeared to be promising. In a series
of experiments it was found that the yield of LDMAN in
dimethyl ether could reach as high as 80%. Furthermore, the
subsequent reductive lithiation reaction was successful in this
solvent. After reductive lithiation, if the subsequent reaction
required the absence of dimethyl ether, then a desired solvent
was added and a vacuum was applied to remove dimethyl ether
at a very low temperature: -60 °C or lower was used to the
prevent organolithiums from being protonated by dimethy ether
or rearranging at higher temperatures.
SM
radical-anion
in THF
in diethyl ether
DBB
LDBB 1
generated at 0 °C,
dark blue solution
no reaction, even at
ambient temperaturea
DMAN LDMAN 2 generated at -45 °C, no reaction until 0 °C at
dark green solution
which temperature
LDMAN decomposes
generated at 0 °C, a
deep purple solution;
only 20-30% yield.
DBN
LDBN
generated at 0 °C,
dark green solution
a
LDBB was also prepared in 1,2-dimethoxyethane in less than 10%
yield.
The color of LDMAN in dimethyl ether is dark green. It is
always contaminated by a red color that is assumed to be the
color of the decomposition product 1-lithionaphthalene. A lower
temperature was used in dimethyl ether (-70 °C) than in THF
(-45 °C) for the generation of LDMAN to suppress the
decomposition of the resulting radical-anion. Surprisingly,
prolonged reaction times did not seem to be necessary: usually
5-6 h appeared to be long enough even at that low temperature.
When vacuum is applied to remove dimethyl ether, great care
must be taken to avoid air being sucked into the reaction system.
Organolithiums in a variety of solvents could thus be generated
by the following typical procedure: (1) generation of LDMAN
in dimethyl ether; (2) reductive lithiation of thioethers at -78
°C; (3) addition of the desired solvent for the subsequent
reaction; and (4) removal of dimethyl ether under reduced
pressure.
temperature. In 1,2-dimethoxyethane, the radical-anion LDBB
did indeed form (the solution was dark-blue, a color indicative
of a radical-anion ), but the yield was very low. The dark-
blue color disappeared when less than 10% of 1 equiv of bis-
3
3
(phenylthio)methane was added.
DBN turned out to be a better radical-anion precursor. When
it was treated with metallic lithium in diethyl ether at 0 °C, a
deep-purple solution was obtained, but once again the yield (20-
3
0%) was far from being practically useful.
DMAN also failed to produce radical-anion 2 in diethyl ether.
A reddish solution was formed by mixing lithium and DMAN
together in diethyl ether at 0 °C (below this temperature, there
was no reaction as indicated by the color of the solution). The
red color was completely different from the dark-green color
of LDMAN formed in THF. We interpret this experiment as
follows: the radical-anion LDMAN was produced at 0 °C in
An example of the utility of this procedure is given below.
For another example, see ref 34.
diethyl ether, but it was unstable at this temperature and
decomposed rapidly, as found previously.32 The decomposition
Tandem Formation and Cyclization to Five- and
Six-Membered Rings of Benzyllithiums
product, 1-lithionaphthalene, is very probably the source of the
red color.
Unlike allylic organolithiums,30b,35 benzylic organolithium
After attempts to generate radical-anions in common solvents
failed, we turned our attention to dimethyl ether which has not
been widely used in synthesis. This solvent was expected to
display behavior similar to that of THF since, as in THF, its
oxygen atom is more exposed than that of diethyl ether and
thus dimethyl ether should be more effective at solvation of
the lithium ion. An added advantage of dimethyl ether was that
because of its low boiling point it could be readily removed
after reductive lithiation and be replaced by the desired solvent.
All three aromatic compounds were tried and the results are
listed below:
compounds have found relatively few synthetic applications.
This is certainly due in part to the difficulties encountered in
their preparation by the usual methods of organolithium
36
production. Krief’s group has prepared benzylic organolithiums
37
by Se-Li exchange. A potentially useful method of production
of benzyllithiums, because it is connective, is the addition of
38
organolithiums to styrenes. However, the anionic polymeri-
zation of styrene and its derivatives with organolithiums is a
39
well-known process. Nevertheless, there are some successful
cases in which styrene and its alkenyl-substituted derivatives
can undergo synthetically useful organolithium addition reac-
tions, although it is usually necessary to optimize the choice of
Li + DBB: There was no reaction in dimethyl ether at - 30
C. LDBB is universally generated in THF at ambient temper-
°
40
reaction conditions to minimize oligomerization. Fraenkel
ature and its formation in dimethyl ether may require a higher
temperature than is allowed by its boiling point, -22 °C.
Li + DBN: A deep-purple solution was generated at -30
found that a variety of R-methylstyrenes can undergo nearly
(
34) Cheng, D.; Zhu, S.; Yu, Z.; Cohen, T. J. Am. Chem. Soc. 2001,
23, 30-34.
35) Yamamoto, Y. Acc. Chem. Res. 1987, 20, 243-249. Schlosser, M.;
Desponds, O.; Lehmann, R.; Moret, E.; Rauchschwalbe, G. Tetrahedron
993, 45, 10175-10203. McCullough, D. W.; Bhupathy, M.; Piccolino,
E.; Cohen, T. Tetrahedron 1991, 47, 9727-9736.
36) For some of the problems, see: Kulkarni, V.; Cohen, T. Tetrahedron
1997, 53, 12089-12100 and citations therein.
37) Krief, A.; Kenda, B.; Barbeaux, P.; Guittet, E. Tetrahedron 1994,
1
°
C, but the color disappeared after the addition of one drop of
(
thioacetal solution. Again, it is conceivable that if a higher
temperature were possible in dimethyl ether the radical-anion
would be generated.
1
(
Li + DMAN: A dark-green solution was formed at -70 °C
and initial experiments indicated a 60-70% yield in its
formation. In the original work on LDMAN,32 it was determined
that it could be formed at -45 °C and that indeed it could not
be used very much above that temperature due to decomposition
to 1-lithionaphthalene. In that work, it was also found that if it
was desired to use DMAN at a higher temperature, it could be
successfully done by using a catalytic quantity of the aromatic
and a stoichiometric amount of lithium. Presumably, under these
conditions, the concentration of the putative unstable intermedi-
(
2
7, 7177-92. Krief, A.; Barbeaux, P. J. Chem. Soc., Chem. Commun. 1987,
1
214-16. Krief, A.; Remacle, B.; Dumont, W. Synlett 1999, 1142-44.
Krief, A.; Bousbaa, J. Tetrahedron Lett. 1997, 38, 6291-94. Review: Krief,
A.; Laboureur, J.-L.; Hobe, M.; Barbeaux, P. In Studies in Natural Product
Synthesis; Atta-ur-Rahman, Ed.; Elsevier: Amsterdam, 1991; pp 3-14.
(38) Recent review of carbometalations of unsaturated molecules: Marek,
I.; Normant, J.-F. In Metal Catalyzed Cross Coupling Reactions; Diederich,
F., Stang, P., Eds.; Wiley VCH: New York, 1998; pp 269-335.
(39) Merton, M. Anionic Polymerization: Principles and Practice;
Academic Press: New York, 1983.
40) (a) Fraenkel, G.; Estes, D.; Geckle, M. J. J. Organomet. Chem. 1980,
(
3
2
ate, the aromatic dianion, never becomes significant.
185, 147-155. (b) Fraenkel, G.; Russell, J. G.; Chen, Y. J. Am. Chem.
Soc. 1973, 95, 3208-15. (c) Wei, X.; Johnson, P.; Taylor, R. J. K. J. Chem.
Soc., Perkin Trans. 1 2000, 1109-1116.
(33) Holy, N. L. Chem. ReV. 1974, 74, 243-277.

Use of Aromatic Radical-Anions in the Absence of THF
J. Am. Chem. Soc., Vol. 123, No. 15, 2001 3481
Scheme 3
Scheme 6
Scheme 4
Scheme 5
Scheme 7
Scheme 8
quantitative 1:1 addition with tert-butyllithium in isooctane-
4
0b
N,N,N′,N′-tetramethylethylenediamine (TMEDA) (eq 1).
solvent, rapid polymerization was observed upon addition of
the alkyllithium reagents. No reaction was observed with hexane
as solvent from -78 °C to ambient temperature.
When n-butyllithium was added to R-methylstyrene, both 1:1
adduct and polymer were detected.40 This result is not surprising
since polymerization of the R-methylstyrene is the competing
side reaction. For n-butyllithium, the rate of the addition step
We eventually took advantage of these discoveries, adding
homo- and bishomoallyllithiums directly to R-methylstyrene to
obtain benzyllithium systems (Scheme 3). Krief had previously
shown that such organolithiums are capable of undergoing
(known as the initiation step) is slower than that of the
3
7
carbanionic cyclizations.
propagation step and hence formation of oligomers results along
with the 1:1 adduct. Since with secondary and tertiary organo-
lithium compounds, the initiation step is faster than propagation,
The homo- and bishomoallyllithiums were all prepared by
reductive lithiation of the corresponding phenyl thioethers. The
phenyl thioether precursors of 4 (R ) H; n ) 1), 4 (R ) Me;
n ) 1), and 4 (R ) Me; n ) 2) as well as of 10 (Scheme 4)
under certain conditions, it is possible to obtain 1:1 adducts in
high yield.41
5,30a
were prepared as in the earlier reports.
3 (n ) 1 and 2) were prepared as in Scheme 4.
In our initial experiments, we used our preferred reagent
The phenyl thioethers
In addition to the nature of organolithium substrates, solvents
also play a key role in the addition of organolithiums to styrene
5
,6a
1
and R-methylstyrene. When using pure hydrocarbon solvent,
Fraenkel40b observed virtually no reaction between organolithi-
LDBB in THF because this generally gives the best yields of
reductive lithiation product. However, it turned out that the
byproduct DBB co-eluted with the product 8. As usual in such
situations, we turned to LDMAN because the byproduct DMAN
is readily separable from the hydrocarbon products by a dilute
ums and R-methylstyrene. However, when mixed solvents
(
isooctane-ether, isooctane-THF, isooctane-TMEDA) were
40c
used, the reaction proceeded smoothly. Taylor found that in
diethyl ether at -78 to -25 °C, styrene underwent efficient
addition reactions with a variety of primary, secondary, and
tertiary alkyllithium reagents and the intermediate benzyllithiums
could be trapped with electrophiles. In contrast, with THF as
3
2
acid wash. The solvent was initially a mixture of THF and
ether (1:2) because preliminary studies indicated that the ratio
of addition/cyclization to polymerization was more favorable
than in THF alone, as might be anticipated from the results in
ref 40 as described above. The reactions of the unsaturated
(41) Glaze, W. H.; Jones, P. C. Chem. Commun. 1969, 1434-35.

3482 J. Am. Chem. Soc., Vol. 123, No. 15, 2001
Cohen et al.
Scheme 9
organolithiums with R-methylstyrene were performed by adding
the latter as a THF solution at -78 °C very slowly, using a
syringe pump, into a solution of the organolithium in THF/ether.
The resulting mixture in some cases was warmed and then an
electrophile, either methanol or 4,4′-dimethoxyphenyl disulfide,
was added. The reaction conditions shown in Scheme 5 were
the best that could be found after numerous experiments. The
provide the fused ring hydrocarbon 21 as a single diastereomer
of undetermined configuration.
While attaining any tandem addition/cyclization product in
the case of the primary organolithium from 16 was gratifying,
the poor yield of 19 did not bode well for the use of this
technology in a planned highly efficient synthesis of the
sesquiterpene (()-cuparene (see below) which also requires the
use of a primary organolithium. However, we were gratified to
discover that the organolithium from 16 could be used in a still
less polar solvent consisting mainly of hydrocarbon, but
containing TMEDA, with the result of doubling the yield of
product 19 (Scheme 8) as a mixture of two diastereomers. As
indicated above, even primary alkyllithiums do not survive in
THF. However, in the solvent used in this experiment, the
precursor of 19 does survive as indicated by the deuteriation
experiment shown. Thus, it is highly likely that the new method
of generating and using the aromatic radical-anion LDMAN will
be very useful in preventing unwanted side reactions caused
by the presence of THF.
1
stereochemistry assigned to 9 was determined by the H NOESY
42
spectrum. The yields of 8 and 9 were unsatisfactory, apparently
due to polymerization of the R-methylstyrene. Furthermore, in
the case of the primary alkyllithium derived from 16 only
oligomerization was observed. In the case of the tertiary
5
homoallyllithium derived from 17, it was clear that the known
rearrangement to a primary homoallyllithium was more rapid
than addition to the R-methylstyrene; as expected, the resulting
primary organolithium caused polymerization of the R-meth-
ylstyrene rather than addition.
In light of the recent reports of efficient addition of even
primary organolithiums to styrene in diethyl ether, with very
little concomitant oligomerization, it may be possible to carry
out the tandem addition/cyclizations in higher yields by using
diethyl ether as the solvent. Traditional reductive lithiation,
which is the only general method to prepare homo- and
bishomoallyllithiums, would not be appropriate since it involves
the use of THF. However, by taking advantage of the above
discovery that LDMAN can be generated in dimethyl ether, it
was possible to avoid the use of THF and to replace the dimethyl
ether by diethyl ether to attain a higher temperature than would
be possible with dimethyl ether present. After LDMAN had been
prepared in dimethyl ether at -70 °C, a solution of the reductive
lithiation substrate in diethyl ether was added followed by
precooled diethyl ether. The dimethyl ether was then removed
at -70 °C under vacuum. As can be seen in Scheme 6, con-
siderably improved yields are achieved in the tandem addition/
cyclization reactions conducted in diethyl ether. Reaction of
homoallyllithium 4 (R ) Me; n ) 1) derived from 14 with
R-methylstyrene in diethyl ether resulted in the formation of
cyclized product 8 in 56% yield, while the six-membered-ring
product 9 was produced from bishomoallyllithium 4 (R ) Me;
n ) 2), derived from 15, in 52% yield. Even primary organo-
lithium 4 (R ) H; n ) 1), derived from 16, which failed to
give any cyclized product in THF with R-methylstyrene,
afforded cyclized product 19, albeit in low yield. However, only
rearrangement of 10, the reductive lithiation product of 17,
occurred as in the experiments in which the solvent contained
THF.
Finally, the new technology was applied to the total synthesis
of the sesquiterpene (()-cuparene 27, a frequent target in the
development of synthetic methodology, in part because of the
presence of the two adjacent quarternary carbon atoms. It has
2
9
already been shown by Krief and Barbeaux, in the seminal
study of this type of cyclization, that 26, which was prepared
by an apparent four-step route from commercially available
materials, cyclizes to cuparene in good yield. Since 26 is readily
43
produced from commercially available 22 and 25 by the
methodology revealed in the present work, Scheme 9 constitutes
a two-pot synthesis of the racemic natural product and the most
44
efficient of those reported. The mass and NMR spectra of 27
were identical with those reported (see Supporting Information).
In conclusion, for the first time, a method of generation of
lithium aromatic radical-anions in a solvent devoid of THF has
been developed. This finding should considerably enhance the
utility of the widely used reductive lithiation for the preparation
of organolithium compounds. Using this technology, a tandem
addition of homoallyl- and bishomoallyllithiums to R-methyl-
styrenes and cyclization of the resultant benzyllithiums has been
developed as an effective way to build certain five- and six-
membered rings, including annulated systems, containing a
pendant aryl ring. Limitations due to polymerization of the
(43) 23 would also be conveniently produced in one pot by reaction of
thiophenol in base on commercial 4-bromo-2-methyl-1-butene.
(44) (a) Other total syntheses of (()-cuparene in addition to the one by
Krief:29 Mandelt, K.; Fitjer, L. Synthesis 1998, 10, 1523-26. Bailey, W.
F.; Khanolkar, A. D. Tetrahedron 1991, 47, 7727-38. Nakatani, H.; So,
T. S.; Ishibashi, H.; Ikeda, M. Chem. Pharm. Bull. 1990, 38, 1233-37.
Kametani, T.; Kawamura, K.; Tsubuki, M.; Honda, T. J. Chem. Soc., Perkin
Trans. 1 1988, 193-200. Ishibashi, H.; So, T. S.; Nakatani, H.; Minami,
K.; Ikeda, M. J. Chem. Soc., Chem. Commun. 1988, 12, 827-8. Bovicelli,
P.; Mincione, E. Synth. Commun. 1988, 18, 2037-50. Kametani, T.;
Kawamura, K.; Tsubuki, M.; Honda, T. J. Chem. Soc., Chem, Commun.
This method also provided a strategy for the formation of
annulated hydrocarbon systems (Scheme 7). Reaction of R-
methylstyrene with the cyclic homoallyllithium derived from
1
3 (n ) 1) resulted in formation of 20 as a mixture of two
1
985, 19, 1324-5. Reetz, M. T.; Westermann, J.; Kyung, S.-H. Chem. Ber.
diastereomers. The cyclic homoallyllithium, produced by reduc-
1
985, 118, 1050-7. Posner, G. H.; Kogan, T. P. J. Chem. Soc., Chem.
tive lithiation of 13 (n ) 2), reacted with R-methylstyrene to
Commun. 1983, 24, 1481-2. Wu, C.-L.; Liu, S. Tetrahedron 1983, 39,
2657-61. (b) Enantiomeric syntheses of (+)-cuparene: Fuganti, C.; Serra,
(
42) The spectrum exhibited strong NOE values between the protons on
methyl-1 and those of methyls-2 and -5. The proximity of methyls-1 and
5 is due to their expected diaxial orientation.
S. J. Org. Chem. 1999, 64, 8728-30. Abad, A.; Agullo, C.; Arno, M.;
Cunat, A. C.; Garcia, M. T.; Zaragoza, R. J. Org. Chem. 1996, 61, 5916-
19 and citations therein.
-

Use of Aromatic Radical-Anions in the Absence of THF
J. Am. Chem. Soc., Vol. 123, No. 15, 2001 3483
R-methylstyrenes and protonation of the cyclized organolithium
by the medium have been largely overcome.
tion Systems and DuPont Pharmaceutical for generous software
and database support.
Acknowledgment. We are grateful to the National Science
Foundation and the Petroleum Research fund, administered
by the American Chemical Society, for financial support,
Dr. Fu-Tyan Lin for help in NMR spectroscopy, Dr. Kasi
Somayajula for help with the mass spectra, and MDL Informa-
Supporting Information Available: Experimental proce-
dures and compound characterization (PDF). This material is
available free of charge via the Internet at http://pubs.acs.org.
JA004353+