Cyclopropane-shift type reaction of diaryl(2- halogenocyclopropyl)methanols promoted by Lewis acids

Article abstract of DOI:10.1016/S0040-4039(00)00979-5

Novel cyclopropane-shift type reaction of diaryl(2- halogenocyclopropyl)methanols 2b and 7 proceeded by using BF₃·OEt₃ and TiCl'4, respectively. Utilizing these reactions, 1-aryl-3-chloro-1,2- methanoindans 5 and 9, 1-aryl-3-methylnaphthalenes 6, and 1,2-methano-1- arylindens 10 were constructed. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00979-5

Tetrahedron Letters 41 (2000) 5937±5942
Cyclopropane-shift type reaction of
diaryl(2-halogenocyclopropyl)methanols
promoted by Lewis acids
Kazunori Wakasugi,a Yoshinori Nishiib and Yoo Tanabea,*
aSchool of Science, Kwansei Gakuin University, 1-1-155 Uegahara, Nishinomiya, Hyogo 662-8501, Japan
bThe Physical Chemical Research Institute (RIKEN), 2-1 Hirosawa, Wako, Saitama 351-0198, Japan
Received 15 May 2000; revised 5 June 2000; accepted 9 June 2000
Abstract
Novel cyclopropane-shift type reaction of diaryl(2-halogenocyclopropyl)methanols 2b and 7 proceeded
.
by using BF3 OEt2 and TiCl4, respectively. Utilizing these reactions, 1-aryl-3-chloro-1,2-methanoindans 5
and 9, 1-aryl-3-methylnaphthalenes 6, and 1,2-methano-1-arylindens 10 were constructed. # 2000 Elsevier
Science Ltd. All rights reserved.
Utilization of cyclopropylmethyrinium cationic intermediates has attained versatile synthetic
transformations, for example, for the precursor of homoallylic compounds and cyclobutanes.1
Concerning readily available gem-dihalogenocyclopropanes, the Nazarov-type annulation1d and
benzannulations have been exploited.1e,f The driving force of these transformations is derived
from the variation from the unstable cyclopropanes into thermodynamically stable products.
Therefore, the reconstruction of cyclopropanes is quite rare and to the best of our knowledge
there has only been a single study on cationic cyclopropane-shift reactions of humulene
derivatives.2 Along with our continuing interest in the synthetic utilization of gem-dihalocyclo-
propanes from many sided cationic,1e radical3 and anionic4 type approaches, we report here a
novel, unique Lewis acid-promoted cyclopropane-shift reaction using monohalogenocyclo-
propanes 2b to construct 1-aryl-3-halogeno-1,2-methanoindans 5, which are the precursors of
1-aryl-3-methylnaphthalenes 6.
Other cyclopropane-shift reactions of 3-methyl monochloro substrates 7a and 7b were
performed to give 1-aryl-3-chloro-1,2-methanoindans 9, which are the precursors of 1,2-methano-
1-arylindens 10.
Monohalogeno substrates [(1R*,2R*)-2a and its diastereomer (1R*,2S*)-2b] were prepared by
the reduction of esters 1 using t-BuMgCl/cat. CoCl2(dppe),4 followed by addition of 2LiPh-(p-R)
* Corresponding author. Fax: +81-798-51-0914; e-mail: tanabe@kwansei.ac.jp
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00979-5

5938
(Scheme 1).5 Diastereomers 2a and 2b were easily separated by column chromatography.
Precursors 2b were prepared with moderate to good selectivities, particularly in the cases of
bromo analogs 2b-5 to -8.
Scheme 1.
.
Benzannulation of chloro compounds 2a-1 to -4 using BF3 OEt2 proceeded through normal
mode to give the corresponding 1-aryl-2-methylnaphthalenes 3 along with open chain compounds
4 (Scheme 2). In clear contrast, the reaction of 2b-1 to -3 resulted in the main formation of
Scheme 2.

5939
tricyclic compounds 5-1 to -3 at 50ꢀC, i.e. the cyclopropane-shift reaction occurred except for p-
Cl analog 2b-4.6 Reactions of bromo substrates 2b-5 to -8 proceeded in higher yields compared
1
with chloro substrates 2b-1 to -3. The structure of 5-1 was determined by H, 13C NMR, IR,
GC±MS and NOE experiments.7 Despite a number of acid-catalyzed ring openings of the gem-
dihalogenocyclopropylmethylinium cation, this is the ®rst example of bond C ®ssion across the
cation (Scheme 3).
Scheme 3.
Treatment of tricyclic compounds 5-1 to -3 (X=Cl) with a strong Lewis acid, SnCl4, furnished
1-aryl-3-methylnaphthalenes 6-1 to -3,8 which is regarded as the cyclopropane-shift type benz-
.
annulation (Scheme 4). BF3 OEt2 did not promote the benzannulation of 5-1 to -3 (X=Cl); in
clear contrast, 5-5 to -7 (X=Br) underwent smooth cyclopropane-shift type benzannulation at
80ꢀC. Thus, isomeric naphthalenes were alternatively prepared from diastereomers 2a and 2b.
Encouraged by the results, direct cyclopropane-shift type benzannulation of 2b-5 to -7 (X=Br)
were performed in excellent yields (Scheme 5).
Scheme 4.
On the other hand, TiCl4-promoted reaction of 3-methyl monochloro substrates 7a and
diastereomers 7b showed another di€erent mode. Compounds 7a were exclusively transformed
into open chain product 8, whilst compounds 7b gave exclusively cyclopropane-shift tricyclic
product 9 as diastereomeric mixtures in good yields (Scheme 6).9 Several attempts to transform 9
into naphthalenes failed (no reaction); however, treatment with DBU was found to give uniquely
1
fused cyclopropanes 10 in good yields. The structure of 10-1 was determined by H NMR and
GC±MS measurements.10 It should also be noted that open chain product 8 was converted to
cyclopropane-shift tricyclic product 9 using TiCl4 at 80ꢀC. Eventually, both substrates 7a and 7b
were converted into 9 in good yields.

5940
Scheme 5.
Scheme 6.
The mechanistic reasoning for the present cyclopropane-shift reaction would be rationally
supported by the calculation.11
Acknowledgements
This work was partially supported by a Grant-in-Aid for Scienti®c Research from the Ministry
of Education, Science, Sports and Culture (Japan). One of the authors (Y.N.) acknowledges the
JSPS fellowship for Japanese Junior Scientists.

5941
References
1. (a) Houben-Weyl Methods of Organic Chemistry, 4th ed.; Georg Thieme: Stuttgart, 1997; Vol. E17. (b) Patai, S.;
Rappoport, S. Z., Eds. The Chemistry of the Cyclopropyl Group; Wiley, London, 1987. (c) Wong, H. N. C.; Hon,
M.-Y.; Tse, C.-W; Yip, Y.-C.; Tanko, J.; Hudlicky, T. Chem. Rev. 1989, 89, 165. (d) Hiyama, T.; Tsukanaka, M.;
Nozaki, H. J. Am. Chem. Soc. 1974, 96, 3713. (e) Tanabe, Y.; Nishii, Y. J. Synth. Org. Chem. Jpn. 1999, 57, 170. A
recent work: Nishii, Y.; Yoshida, T.; Tanabe, Y. Tetrahedron Lett. 1997, 38, 7195.
2. (a) Fujita, T.; Ohtsuka, T.; Shirahama, H.; Matsumoto, T. Tetrahedron Lett. 1982, 23, 4091. (b) Hayasaka, K.;
Ohtsuka, T.; Shirahama, H.; Matsumoto, T. Tetrahedron Lett. 1985, 26, 873. Related cyclopropane formation was
reported: Nagasawa, T.; Handa, Y.; Onoguchi, T.; Suzuki, K. Bull. Chem. Soc. Jpn. 1996, 69, 31.
3. Tanabe, Y.; Wakimura, K.; Nishii, Y. Tetrahedron Lett. 1996, 37, 1837. Tanabe, Y.; Nishii, Y.; Wakimura, K.
Chem. Lett. 1994, 1757.
4. Nishii, Y.; Tanabe, Y.; Wakasugi, K. Synlett 1998, 67.
5. Compound 2a-1: colorless crystals; mp 115.5±117.0ꢀC; 1H NMR (400 MHz) ꢀ=0.89 (1H, dd, Jgem=6.1 Hz,
J=7.8 Hz), 0.92 (3H, s), 1.95 (1H, dd, Jgem=6.1 Hz, J=4.6 Hz), 2.82 (1H, s, OH), 3.40 (1H, dd, J=7.8 Hz, 4.6
Hz), 7.22±7.42 (8H, m), 7.57±7.62 (2H, m); ꢁmax (KBr)/cm^1 3536, 3084, 1447, 972, 957, 762, 702. Compound 2b-1:
colorless crystals; mp 79.5±80.5ꢀC; H NMR (400 MHz) ꢀ=0.63 (1H, dd, Jgem=6.1 Hz, J=4.4 Hz), 1.23 (3H, s),
1
1.49 (1H, dd, Jgem=6.1 Hz, J=7.8 Hz), 2.06 (1H, s, OH), 3.52 (1H, dd, J=7.8 Hz, 4.4 Hz), 7.26±7.41 (10H, m);
ꢁmax (KBr)/cm^1 3569, 3084, 1445, 949, 752, 700, 679. Compound 2b-5: colorless crystals; mp 60.0±61.0ꢀC; H
1
NMR (400 MHz) ꢀ=0.70 (1H, dd, J=4.6 Hz, Jgem=6.1 Hz), 1.28 (3H, s), 1.58 (1H, dd, Jgem=6.1 Hz, J=8.3 Hz),
2.05 (1H, s, OH), 3.48 (1H, dd, J=4.6 Hz, 8.3 Hz), 7.25±7.48 (10H, m); ꢁmax (KBr)/cm^1 3578, 3057, 1445, 1005,
770, 702, 644.
.
6. Typical procedure is as follows. BF3 OEt2 (0.046 ml, 0.37 mmol) was added to a stirred solution of 2a-1 (100 mg,
0.37 mmol) in 1,2-dichloroethane (0.5 ml) at 50ꢀC under Ar atmosphere, and the mixture was stirred at the same
temp. for 30 min. Aqueous satd NaHCO3 solution was added to the mixture, which was extracted with ether. The
organic phase was washed with water, brine, dried (Na2SO4) and concentrated. The obtained crude oil was
subjected to ¯ash column chromatography on SiO2 using hexane to give the products 3-1 (R=H; 11 mg, 14%)
and 5-1 (R=H; 66 mg, 71%).
7. Compound 5-1: colorless crystals; mp 35.5±36.5ꢀC; H NMR (400 MHz) ꢀ=1.52 (3H, s), 3.55 (1H, d, Jgem=10.7
1
Hz), 3.77 (1H, d, Jgem=10.7 Hz), 6.49 (1H, s), 7.11±7.62 (9H, m); 13C NMR (100 MHz) ꢀ=20.21, 52.02, 52.85,
121.06, 122.44, 125.73, 127.66, 127.95, 128.07, 129.19, 133.93, 135.23, 139.00, 142.48, 150.04; ꢁmax (KBr)/cm^1
3061, 1445, 833, 781, 756, 698. Relative stereochemistry was determined by NOE measurement between 1-H and
2-CH3. GC±MS (70 eV) m/e 254 (M+). Compound 5-5: colorless crystals; mp 42.5±43.5ꢀC; H NMR (400 MHz)
1
ꢀ=1.54 (3H, s), 3.44 (1H, d, Jgem=9.8 Hz), 3.68 (1H, d, Jgem=9.8 Hz), 6.47 (1H, s), 7.12±7.61 (9H, m); 13C NMR
(100 MHz) ꢀ=21.24, 41.25, 52.06, 121.03, 122.19, 125.67, 127.51, 127.61, 127.90, 128.54, 135.09, 139.63, 142.32,
143.37, 149.98; ꢁmax (neat)/cm^1 3061, 1445, 1238, 947, 779, 758, 698.
8. Compound 6-1 (R=H): colorless oil; 1H NMR (400 MHz) ꢀ=2.72 (3H, s), 7.28-7.54 (9H, m), 7.91 (1H, d, J=8.5
Hz), 8.04 (1H, d, J=8.1 Hz); ꢁmax (neat)/cm^1 1443, 980, 910, 833, 764, 702.
9. Typical procedure: TiCl4 (1.0 M solution in CH2Cl2, 1.00 ml, 1.00 mmol) was added to a stirred solution of 7b-1
(273 mg, 1.00 mmol) in CH2Cl2 (2.0 ml) at 0±5ꢀC under Ar atmosphere, and the mixture was stirred at the same
temp. for 1 h. After usual work up and ¯ash column chromatography on SiO2 using hexane, the products 9a-1
.
(R=H; 139 mg, 55%) and 9b-1 (R=H; 96 mg, 38%) were obtained. Reactions using BF3 OEt2 and SnCl4 were a
1
little sluggish. Compound 9a-1: brown oil; H NMR (400 MHz) ꢀ=1.23 (3H, d, J=6.6 Hz), 4.01 (1H, dd, J=1.9
Hz, 5.1 Hz), 4.63 (1H, m), 6.65 (1H, d, J=1.9 Hz), 7.15±7.65 (9H, m); ꢁmax (neat)/cm^1 3063, 1447, 867, 774, 747,
700. Compound 9a-2: brown crystals; mp 39.5±40.5ꢀC; 1H NMR (400 MHz) ꢀ=1.54 (3H, d, J=6.6 Hz), 3.92 (1H,
dd, J=2.2 Hz, 4.2 Hz), 4.57 (1H, m), 6.49 (1H, d, J=2.2 Hz), 7.22±7.75 (9H, m); ꢁmax (neat)/cm^1 3063, 1443, 810,
774, 743, 696. Relative stereochemistries of 9a-1 and 9b-1 were determined by NOE measurements. The reaction of
aryl(gem-dihalocyclopropyl)-methanols without the 1-Me group gave substantial amounts of the open chain
product.1e This fact coincides with the reaction of 9a.
10. Compound 10-1: yellow oil; 1H NMR (400 MHz) ꢀ=2.22 (3H, d, J=7.3 Hz), 6.73 (1H, m), 6.91 (1H, s), 7.20-7.70
(9H, m); ꢁmax (neat)/cm^1 3061, 3027, 1445, 748, 737, 700. GC±MS (70 eV) m/e 218 (M+). Halton, B.; Banwell,
M. G. In The Chemistry of the Cyclopropyl Group; Patai, S.; Rappoport, Z., Eds.; Wiley: London, 1987; p. 1223.
11. We propose two mechanistic speculations. One is that bond C not of 2a-1 but of 2b-1 is located in a transannular
position toward the leaving OH group and bond C of 2b-1 is longer than bonds A and B; therefore, this bond C is

5942
weakened and so inclined to cleave. The other is that cationic intermediate 11 after the cyclopropane-shift is
signi®cantly stable compared with normal open chain intermediate 12. These are supported by calculations
[SPARTAN version 5.0 (Wavefunction, Inc., Irvine, CA), semiempirical, AM1].