Vinyl hydrogen more reactive than benzyl hydrogen toward base in significantly twisted styrenes

Article abstract of DOI:10.1016/S0040-4039(01)00308-2

The novel example of vinyl hydrogen more reactive than benzylic hydrogen was found by treatment of a twisted styrene derivative with a strong base followed by D₂O quenching. The characteristic nature of the vinyl hydrogen, which is activated by

Full text of DOI:10.1016/S0040-4039(01)00308-2

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 3093–3095
Vinyl hydrogen more reactive than benzyl hydrogen toward base
in significantly twisted styrenes
Hajime Mori,a,† Takafumi Matsuo,a Yasunori Yoshiokab and Shigeo Katsumuraa,*
aSchool of Science, Kwansei Gakuin University, Uegahara 1-1-155, Nishinomiiya, Hyogo 662-8501, Japan
bDepartment of Chemistry, Graduate School of Science, Osaka University, Toyonaka, Osaka 560-0043, Japan
Received 9 February 2001; revised 21 February 2001; accepted 22 February 2001
Abstract—The novel example of vinyl hydrogen more reactive than benzylic hydrogen was found by treatment of a twisted styrene
derivative with a strong base followed by D2O quenching. The characteristic nature of the vinyl hydrogen, which is activated by
s–p* orbital interaction, was apparently demonstrated. © 2001 Elsevier Science Ltd. All rights reserved.
Recently, we have found that singlet oxygen selectively
abstracts the vinyl hydrogen rather than the allyl
hydrogen in significantly twisted 1,3-dienes and that the
inherent reactivity of the vinyl hydrogen toward singlet
oxygen in these novel systems is nearly equal to or
higher than that of the allyl hydrogen.1 This unique
observation prompted us to investigate how the vinyl
hydrogens in twisted 1,3-dienes behave toward a base.
In this paper, we report the novel example of vinyl
hydrogens more reactive than benzyl hydrogens toward
a base, with discussion on the relationship between the
tendency of the twist in styrenes and the reactivity of
the vinyl hydrogen.
2c were synthesized by LAH reduction4 of the acetylene
moiety of the corresponding coupling products fol-
lowed by methylation.5 The maximum absorption val-
ues of the K-band of 1a and 1b in their electronic
spectra were not distinguishable from those of the
substituted benzene ring, whereas that of 1c clearly
appeared at 251 nm (m=12989). Those of E isomers,
2a–2c, appeared at 237 nm (m=8246), 248 nm (m=
As a new variety of unusual twisted 1,3-dienes, we
designed 2,6-dimethylphenyl, 2-methylphenyl, and 4-
methylphenyl substituted Z styrene derivatives 1a–1c
(Fig. 1). They were synthesized by the Pd-catalyzed
cross-coupling method as shown in Scheme 1. Thus,
aryl triflate 3, prepared from 2,6-dimethylphenol, was
coupled with 2-methyl-3-butyn-2-ol by Pd catalyst to
give 4. Tetrabutylammonium iodide is essential for this
cross-coupling reaction.2 Partial hydrogenation of the
acetylene moiety in 4 and then methylation of the
hydroxy group gave 1a.3 The 2-methylphenyl and 4-
methylphenyl derivatives, 1b and 1c, were also synthe-
sized by the same method starting from the
corresponding iodides, respectively. The E isomers 2a–
Figure 1. Various designed styrene derivatives.
b, c
1a
OH
OTf
a
4
d, e
3
2a
Keywords: carbanions; twisted 1,3-dienes; vinyl hydrogens.
* Corresponding author. Fax: +(81)79851-0914; e-mail: katumura@
kwansei.ac.jp
Present address: Industrial Research Center of Wakayama Prefec-
ture, Ogura 60, Wakayama 649-6261, Japan.
Scheme 1. Synthesis of styrene derivatives 1 and 2. (a) 2-
Methyl-3-butyn-2-ol, PdCl2(PPh3)2, CuI, Bu4NI, Et2NH
(47%); (b) Pd–C/H2 (86%); (c) NaH, Bu4NI, MeI (98%); (d)
LAH/THF (72%); (e) NaH, Bu4NI, MeI (75%).
†
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00308-2

3094
H. Mori et al. / Tetrahedron Letters 42 (2001) 3093–3095
n-BuLi, TMEDA
t-BuOK/BuLi/TMEDA system7 at −43 and −95°C as
1a
(ca. 60%)
2a-D
the kinetic conditions. Treatment of 1a with n-BuLi (5
equiv.) in the presence of TMEDA (2.5 equiv.) in
hexane at room temperature for 1 h followed by
quenching with D2O yielded the corresponding E-iso-
mer 2a-D in approximately 60% yield based on NMR,
in which the C4 vinyl hydrogen was deuterated, along
with the starting material (Scheme 2). The obtained
results clearly demonstrated that the vinylic proton Ha
was selectively abstracted by the base in preference to
the benzylic proton Hb. The vinyl anion produced
would be followed by Z–E isomerization and then
quenching with D2O. The vinyl anion isomerization
pathway is strongly supported by the study of Paneck
and co-workers.8 Next, methyl ether 1a was treated
with 1 equiv. of t-BuOK/BuLi/TMEDA at −43°C in
hexane and THF (3:1) for 1 h followed by quenching
with D2O. After the usual work-up, the NMR spectrum
of the crude products showed that 2a-D and 2a were
selectively produced in 41 and 14% yield, respectively,
along with the starting compound (42%), as shown in
Scheme 3.9 None of the compound deuterated at the
benzylic methyl hydrogen was detected. Even at −95°C,
under the same reaction conditions, the selective forma-
tion of compounds 2a-D and 2a occurred in a relatively
lower yield (40% yield in a 1:1 ratio). The results
obtained under both thermodynamic and kinetic condi-
tions revealed that the vinyl proton at the C4 position
was kinetically more favorable for the abstraction by
the base than the benzyl proton,10 and the produced
vinyl anion was easily transformed into the thermody-
namically favorable E-isomer, which was stabilized by
coordination of the lithium with the etherial oxygen.
Furthermore, the anion produced from 1a, under the
same kinetic conditions, was treated with 2 equiv. of
benzaldehyde and n-propyl aldehyde in the presence of
LiBr (9 equiv.)7c to yield the corresponding adducts (2a:
R=PhCH(OH) and R=PrCH(OH)) in 53 and 45%
yield, respectively. Adducts at the benzylic position
were not detected once again.
r. t., 1h
then D2O
D2O
Me
Li
Li
O
Isomerization
OMe
Scheme 2. Plausible mechanism for the formation of 2a-D.
D
1a
2a +
+
OMe
a)
(14%) (42%)
OMe
2a-D
(41%)
1a
D
a)
a)
+
+
+
1b-D
2b 1b
OMe
OMe
OMe
(6%) (14%) (36%)
2b-D
(44%)
1b
1c
+
2c-D
(30%)
1c
(30%)
+
OMe
1c-D
(40%)
D
Scheme 3. Proton abstraction of various styrene derivatives
by a strong base. (a) t-BuOK, n-BuLi, TMEDA, hex-
ane:THF=3:1, −43°C, 1 h, then D2O.
π* orbital
π* orbital
R3
R3
H
R1
R2
R1
R2
σ orbital
H
H
R4
R4
structure B
The relation between the magnitude of twist in styrenes
and the reactivity of the vinyl hydrogens was clearly
demonstrated by the following experiments. Reaction
of 1b under the same kinetic conditions at −43°C
produced a mixture of the products, the NMR spec-
trum of which showed that the formation of 2b-D and
2b were clearly observed in 44 and 14% yields, respec-
tively, along with 1b (36%). In this case, a small amount
of the compound, 1b-D, in which the benzylic methyl
hydrogen was deuterated, was detected (approximately
6% yield). In the reaction of 1c under the same kinetic
conditions, 2c-D and 1c-D were obtained in approxi-
mately 30 and 40% yield, respectively, along with 1c
(30%). In this case, the deuterated compound at the
benzylic position was the major product. The results
obtained apparently showed that the reactivity of the
vinyl hydrogen at the C4 position evidently increased
accompanying with the increasing magnitude of twist at
the central single bond of styrenes, although, from a
steric effect view point, the surrounding of the vinyl
hydrogen of 1a is the most crowded of all Z-styrene
derivatives. The vinyl hydrogen must be highly acti-
structure A
Figure 2. Orbital interaction of a twisted styrene.
13776), and 254 nm (m=20176), respectively. These
obtained results obviously showed that the benzene ring
of 1a and 1b was significantly twisted toward the cis-
olefin plane. In 1c, 2a, and 2b, their benzene ring would
be slightly twisted resulting from the comparison of
their electronic spectra with that of 2c. The geometries
of Z compounds 1a–1c and E compounds 2a–2c were
fully optimized at the Hartree–Fock/6-31G* level using
the Gaussian 94 program package. The torsion angles
between the benzene ring and the C3ꢀC4 double bond
of 1a–1c were 85, 84, and 53°, while those of 2a–2c
were 65, 39, and 21°, respectively.
We then tried to carry out a base treatment of 1a–1c.
Among various type of bases that can abstract benzyl
protons, we chose a BuLi/TMEDA system6 at room
temperature as the thermodynamic conditions and a

H. Mori et al. / Tetrahedron Letters 42 (2001) 3093–3095
3095
1
vated by the novel orbital interaction resulting from the
twist described below. On the contrary, the same treat-
ment of E isomers, 2a–2c, with 2.5 equiv. of the base
gave the respective compounds deuterated at the ben-
zylic position in 80–95% yield. The vinyl hydrogen at
C4 of the E isomers might be hindered from the attack
by the base complex.
1078; H NMR (400 MHz, CDCl3) l 7.06–6.98 (3H, m),
6.41 (1H, d, J=13.2 Hz), 5.72 (1H, d, J=12.8 Hz), 3.22
(3H, s), 2.26 (6H, s), 1.03 (6H, s); 13C NMR (100 MHz,
CDCl3) l 137.81, 137.30, 135.22, 128.54, 126.91, 126.54,
76.18, 50.64, 25.36, 20.69; EI HRMS m/e calcd for
C14H20O (M+) 204.1514, found 204.1513.
4. Grant, B.; Djerassi, C. J. Org. Chem. 1974, 39, 968.
5. Compound 2a: IR (neat, cm−1) 2976, 1470, 1170, 1076;
UV (heptane) umax 237.6 nm (m=8380); 1H NMR (400
MHz, CDCl3) l 7.05 (3H, m), 6.45 (1H, d, J=16.8 Hz),
5.67 (1H, d, J=16.8 Hz), 3.27 (3H, s), 2.30 (6H, s), 1.40
(6H, s); 13C NMR (100 MHz, CDCl3) l 140.16, 136.94,
135.78, 127.67, 126.94, 126.49, 75.36, 50.66, 26.00, 20.91.
6. Klein, J.; Medlik-Balan, A. J. Am. Chem. Soc. 1977, 99,
1473.
7. (a) Schlosser, M.; Strunk, S. Tetrahedron Lett. 1984, 25,
741; (b) Schlosser, M. Pure Appl. Chem. 1988, 60, 1627;
(c) Brandsma, L.; Verkruijsse, H. D.; Schade, D.;
Schleyer, P. R. J. Chem. Soc., Chem. Commun. 1986, 260.
8. (a) Panek, E. J.; Neff, B. L.; Chu, H.; Panek, M. G. J.
Am. Chem. Soc. 1975, 97, 3996; (b) Zwaifel, G.; Rajago-
palan, S. J. Am. Chem. Soc. 1985, 107, 700.
The quite unique results obtained could be rationalized
by considering the orbital interaction as shown in Fig.
2. In the significantly twisted styrenes, the vinylic CꢀH
bond must be activated by the s–p* orbital interaction,
because that is closely parallel to the p* orbital of
benzene ring (structure A), and the vinyl anion gener-
ated by a strong base must be stabilized by delocaliza-
tion to the p orbital of benzene ring (structure B) in a
manner similar to an allylic CꢀH bond.11 The vinylic s
bond activation, based on s–p orbital interactions,
which leads to a remarkable acceleration of the reaction
rates, can be found in the chemistry of stabilized vinyl
cations reported about 30 years ago.12 Furthermore, it
is well-known that acidity of the hydrogen on an sp2
carbon is generally higher than that on an sp3 carbon.
Thus, the vinylic CꢀH bond of significantly twisted
styrenes may be regarded as a novel allylic CꢀH bond
attaching at an sp2 carbon.
1
9. In the H NMR spectra, the C3 vinyl proton of 2a-D (l
5.65, t, J=2.4 Hz) was clearly distinguishable from those
of 2a (l 5.67, d, J=16.8 Hz) and 1a (l 5.72, d, J=12.8
Hz).
10. The intramolecular rearrangement of the anion from the
vinylic position to the thermodynamically stable benzylic
position of tolylstilbene derivatives was extensively inves-
tigated by Knorr and co-workers. In the present case, the
rearrangement pathway from the kinetically generated
benzylic anion to the thermodynamically stable C4
vinylic anion was excluded because (1) in the significantly
twisted styrene derivatives, the benzylic anion, if gener-
ated, is not situated close enough to abstract the vinyl
hydrogen, and (2) the anion formation at the benzylic
position of 1a was not detected by quenching experiments
with D2O, even under kinetically controlled conditions.
See: (a) Broaddus, C. D.; Muck, D. L. J. Am. Chem. Soc.
1967, 89, 6533; (b) Knorr, R.; Lattka, E.; Rapple, E.
Chem. Ber. 1981, 114, 1581; (c) Knorr, R.; Lattke, E.;
Ruf, F.; Reissig, H.-U. Chem. Ber. 1981, 114, 1592; (d)
Lattke, E.; Knorr, R. Chem. Ber. 1981, 114, 1600; (e)
Knorr, R.; Lattke, E. Chem. Ber. 1981, 114, 2116.
11. We also measured the coupling constant of C4ꢀH4 in
order to examine the s character of the most ‘twisted’
styrene 1a, and found that it was the normal value of an
sp2 carbon (JCꢀH=157.5 Hz).
Acknowledgements
The authors would like to thank Mr. Tamotsu
Yamamoto of the Institute for Life Science Research at
Asahi Chemical Industry Co. Ltd for the high-resolu-
tion mass spectroscopy measurements. This work was
supported by a Grant-in-Aid for Scientific Research on
Priority Areas (No. 283, ‘Innovative Synthetic Reac-
tions’) from the Ministry of Education, Science, Sports,
and Culture of Japan.
References
1. (a) Mori, H.; Ikoma, K.; Isoe, S.; Kitaura, K.; Kat-
sumura, S. Tetrahedron Lett. 1996, 37, 7771; (b) Mori,
H.; Ikoma, K.; Katsumura, S. Chem. Commun. 1997,
2243; (c) Mori, H.; Ikoma, K.; Isoe, S.; Kitaura, K.;
Katsumura, S. J. Org. Chem. 1998, 63, 8704; (d) Mori,
H.; Matsuo, T.; Yamashita, K.; Katsumura, S. Tetra-
hedron Lett. 1999, 40, 6461.
12. (a) Hanack, M. Acc. Chem. Res. 1976, 9, 364; (b) Rappo-
port, Z. Acc. Chem. Res. 1976, 9, 265; (c) Apeloig, Y.;
Schleyer, P.; PoPle, J. A. J. Org. Chem. 1977, 42, 3004;
(d) Hanack, M.; Sprosser, L. J. Am. Chem. Soc. 1978,
100, 7066.
2. Powell, N. A.; Rychnovsky, S. D. Tetrahedron Lett. 1996,
37, 7901.
3. Compound 1a: IR (neat, cm−1) 2980, 1468, 1376, 1166,
.