A concise synthesis of 1-substituted-2-tetralones by selective diol dehydration leading to ketone transposition

Article abstract of DOI:10.1016/S0040-4039(00)01014-5

Dehydration of 1-substituted-1,2-tetralindiols with zinc iodide afforded the corresponding 2-tetralones in excellent yields. This procedure was found to be superior to the more conventional BF₃-catalyzed rearrangement of 1-substituted-1,2-epoxytetralins. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01014-5

Tetrahedron Letters 41 (2000) 6029±6033
A concise synthesis of 1-substituted-2-tetralones by selective
diol dehydration leading to ketone transposition1
Bruce L. Jensen* and Sergei V. Slobodzian
Department of Chemistry, University of Maine, Orono, ME 04469, USA
Received 4 April 2000; revised 6 June 2000; accepted 7 June 2000
Abstract
Dehydration of 1-substituted-1,2-tetralindiols with zinc iodide a€orded the corresponding 2-tetralones
in excellent yields. This procedure was found to be superior to the more conventional BF3-catalyzed
rearrangement of 1-substituted-1,2-epoxytetralins. # 2000 Elsevier Science Ltd. All rights reserved.
Keywords: ketones; rearrangements; epoxides; diols.
The tetralone family serves as an important source of synthetic precursors for a wide range of
compounds, including steroids,2 heterocycles,3 and pharmaceuticals.4 While the 1-tetralones are
inexpensive, easily prepared and commercially available, the 2-tetralones are often very expensive5
and much more dicult to synthesize. This is especially true of the 1-substituted-2-tetralones. The
literature describes but two routes for their preparation, neither of which is convenient and/or
high yielding. Alkylation of the metal enolate of 2-tetralone su€ers from polyalkylation and poor
yield.6 The enamine alkylation developed in our laboratory is an excellent method; however, it is
limited by the availability of the 2-tetralone and requires multiple steps.7 Prompted by a recent
report by Meyers and co-workers,8 who demonstrated that epoxides derived from 1-tetralones
can rearrange in good yield to a€ord 2-tetralones, we report herein a complementary method
leading to the title compounds.
Epoxides have long been known for their ability to undergo acid-catalyzed rearrangement by
1,2-oxygen shift a€ording ketones.9 Over the years, numerous catalysts have been investigated for
epoxide isomerization, including BF3, H2SO4, and transition metal complexes.10 During the
syntheses of tetralin-containing heterocyclics, we studied the feasibility of using an epoxide
isomerization step in the formation of 1-alkyl-, 1-aryl-, and 1-arlyalkyl-2-tetralones. Our study
showed that three di€erent, yet complementary, intermediates could be obtainedÐany one of
which would give the desired rearrangement product. Furthermore, the scheme is simple, uses
* Corresponding author. E-mail: bjensen@maine.edu
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01014-5

6030
inexpensive reagents, a€ords the title compounds in high yield, and can be done in only four steps
(Scheme 1).
Scheme 1.
In our initial study we chose to study the acid-catalyzed isomerization of 1-benzyl-1,2-epoxy-
tetralin (3c). The requisite alkene 2c was prepared in a two-step sequence involving Grignard
addition of benzylmagnesium bromide to 1-tetralone (R0=H). The resulting alcohol, without
puri®cation, was dehydrated by p-toluenesulfonic acid in re¯uxing toluene for 1 h to give 1-benzyl-
3,4-Á1-tetralin (2c) in 90% yield. This alkene was then treated with m-chloroperbenzoic acid (m-
CPBA) in methylene chloride for 1.5 h to furnish the corresponding epoxide 3c in 96% yield.
Upon rearrangement with either boron tri¯uoride etherate or zinc iodide, 1-benzyl-1,2-epoxy-
tetralin (3c) gave 1-benzyl-2-tetralone (6c) in 98±99% yield. This four-step sequence gave the
desired product in 85% overall yield without resorting to expensive reagents or tedious work-up
procedures. Encouraged by these results we next turned our attention to the 1-aryl- and 1-alkyl-2-
tetralones. These compounds, however, gave us some unexpected results.
As before, methylmagnesium bromide and phenylmagnesium bromide reacted with 1-tetra-
lones and underwent subsequent dehydration to furnish 1-methyl- or 1-phenyl-Á1-tetralin (2a, 2b)
in 75±90% yield. However, all attempts to isolate the epoxides after oxidation with m-chloro-
perbenzoic acid failed. We discovered that epoxide ring opening had occurred under acidic
conditions resulting in the formation of the a-hydroxy benzoate esters 4. The regiochemistry of
epoxide opening was proven by the substantial resonance deshielding (5.38 ppm) of the C-2
proton of the m-chlorobenzoate ester 4a, whereas the C-2 proton in the diol 5a appeared at 3.83
ppm.11 The problem of epoxide opening was easily overcome by performing this reaction in a
heterogeneous water/methylene chloride medium, where the m-CPBA was converted into its
sodium salt by the addition of sodium bicarbonate. The desired compounds 3a±b were isolated in
86±90% yield. Here too, acid-catalyzed isomerization with BF3 or ZnI2 dehydration gave the
desired 1-substituted-2-tetralones 6a±b in 94±99% yield.
The unexpected isolation of a-hydroxy benzoate esters during the course of our work prompted
us to examine their chemistry as an alternate route to the title compounds. Upon base hydrolysis, the
esters were converted to 1-methyl- and 1-phenyl-1,2-tetralindiol (5a±b). We then decided to examine
their behavior toward acid conditions. When treated with ZnI2 or BF3, yields of 6a±b ranging
from 93±97% were realized. The rami®cations of these results demonstrated that an epoxide

6031
intermediate was not necessary in order to accomplish the desired 1,2-carbonyl transposition. In
order to demonstrate this fact, the 1-substituted-Á1-tetralins 2 were dihydroxylated with 2.5%
osmium tetraoxide in aqueous t-butyl alcohol containing pyridine and using trimethylamine N-
oxide as a co-oxidant a€ording the corresponding 1,2-tetralindiols 5 in 88±91% yield. Spectral
comparison showed them to be identical with the diols formed by oxidation with m-chloro-
perbenzoic acid. Subsequent acid-catalyzed isomerization with zinc iodide or boron tri¯uoride
gave the 1-substituted-2-tetralones 6a±b in 93 to 97% yields.
We were able to extend this study to tetralone rings possessing electron-donating methoxyl
groups. The 6,7-dimethoxytetralin series proved to be extremely sensitive to epoxidation by m-
chloroperbenzoic acid, a€ording low yields of the epoxides and numerous by-products. Unlike
their unsubstituted analogs, diols 5d±e were found to dehydrate most smoothly in a benzene
solution containing a catalytic amount of zinc iodide;12 see the results shown in Tables 1 and 2.
The dehydration method provided high yields of the 2-tetralones with no decomposition or
colored by-products. The complete four-step sequence via the tetralindiols 5d±e a€orded the
desired 2-tetralones 6d±e in overall yields of 79±85%.13
Table 1
1->2-Tetralone transposition via epoxide method
Table 2
1->2-Tetralone transposition via diol method
In conclusion, two ecient high-yielding methods for the preparation of 1-substituted-2-tetra-
lones have been developed using well-known reactions. The `epoxide' method was shown to be
highly ecient but restricted to the synthesis of 1-benzyl-2-tetralone and 1-methyl-2-tetralone. On
the other hand, the `diol' method was universally more convenient for the preparation of all types
of 2-tetralones, including the more sensitive 6,7-dimethoxytetralone series. Zinc iodide was
uniquely suited for the selective dehydration of diols 5a±c providing high yields of the desired 2-
tetralones with neither decomposition nor colored by-products. In each case, the four-step
sequences described here a€orded the desired products in greater than 85% overall yields without

6032
resorting to expensive reagents or tedious workup procedures. Mechanistically, it appears that
diol dehydration and epoxide isomerization proceed through a common enol intermediate leading
to the 2-tetralone (Scheme 2).
Scheme 2.
Acknowledgements
The authors wish to thank the General Research Fund at the University of Maine for support
of this research.
References
1. A preliminary account of this work was presented at the 217th ACS National Meeting, Anaheim, CA, USA,
March 1999, Abstract No. 235. Slobodzian, S. V. Ph.D Thesis, University of Maine, 1999.
2. Lednicer, D.; Mitscher, L. A. In The Organic Chemistry of Drug Synthesis; John Wiley & Sons: New York, 1977;
Vol. 1, Chapter 10, pp. 155±211.
3. (a) Jensen, B. L.; Chockalingam, K. J. Heterocyclic Chem. 1986, 23, 343. (b) Covarrubias-Zuniga, A.; Cantu, F.;
Maldonado, L. A. J. Org. Chem. 1998, 63, 2918.
4. (a) Ye, B.; Yao, Z. J.; Burke, T. R. J. Org. Chem. 1997, 62, 5428. (b) Brewster, W. K.; Nichols, D. E.; Riggs, R. M.;
Mottola, D. M.; Lovenberg, T. W.; Lewis, M. H.; Mailman, R. B. J. Med. Chem. 1990, 33, 1756.
5. Current Aldrich Chemical Co. price lists show that 6,7-dimethoxy-1-tetralone costs $22.50/g while 6,7-dimethoxy-
2-tetralone costs $272.40/g. The conversion of 6,7-dimethoxy-1-tetralone to 6,7-dimethoxy-2-tetralone can be done
for $47.66/g using the methodology described in this paper.
6. Andersag, H.; Salzer, W. US Patent 2271674 (1942).
7. Jensen, B. L.; Michaud, D. P. Synthesis 1977, 848.
8. Pryde, D. C.; Henry, S. S.; Meyers, A. I. Tetrahedron Lett. 1996, 37, 3243.
9. Rickborn, B. In Comprehensive Organic Synthesis; Trost, B. M., Ed.; Pergamon: Oxford, 1991; Vol. 3, Chapter
3.3, pp. 733±775.
10. Kulasegaram, S.; Kulawiec, R. J. J. Org. Chem. 1994, 59, 7196, and references cited therein.
11. Lambert, J. B.; Shurvell, H. F.; Lightner, D. A.; Cooks, R. G. In Organic Structural Spectroscopy; Prentice Hall:
Upper Saddle River, NY, 1998; Chapter 3, p. 42.
12. Zinc iodide dehydration gave the tetralone in exceptional purity and nearly quantitative yield. For a typical
procedure using zinc iodide, see: Nichols, D. E.; Cassidy, J. M.; Persons, P. E.; Yeung, M. C.; Clemens, J. A.;
Smalstig, E. B. J. Med. Chem. 1989, 32, 2128.
13. Representative procedures. Epoxide method: Solid zinc iodide (100 mg, 0.30 mmol) was heated under aspirator
pressure at 120ꢀC for a period of 1 h. After cooling, a solution of epoxide 3c (210 mg, 0.89 mmol) in benzene (5
mL) was added to the zinc iodide. The mixture was stirred under re¯ux for 1 h, cooled to room temperature, and
washed with water. After drying over magnesium sulfate, the solvent was removed a€ording 6c (208 mg, 99%).

6033
Diol method: Boron tri¯uoride etherate (2 drops) was added to a solution of diol 5c (333 mg, 1.31 mmol) in ether
(3 mL). The solution was stirred at 30ꢀC for 20 min, washed with water and dried over magnesium sulfate.
Evaporation of the solvent a€orded a colorless oil which distilled at 127ꢀC (750 mm) giving 6c (301 mg, 97%): 1H
NMR (300 MHz, CDCl3) ꢀ 2.40±2.54 (m, H-3, 2H), 2.62 (m, H-4, 1H), 2.83 (m, H0-4, 1H), 3.21 (ddd, J=14.9,
10.1, 6.2 Hz, H-Bn, 2H), 3.74 (t, J=6.2 Hz, H-1, 1H), 6.85±6.96 (m, Ar-H, 3H), 7.13±7.22 (m, Ar-H, 6H); 13C
NMR (75 Hz, CDCl3) ꢀ 27.48, 38.51, 39.27, 55.19, 126.67, 126.89, 127.04, 127.91, 128.37, 128.80, 129.63, 136.61,
137.10, 138.39, 212.59; MS (m/z) 236 (M+); IR (neat) 1717 cm^1 (CˆO). Semicarbazone: mp 199±200ꢀC. Anal.
calcd for C18H19N3O: C, 73.69; H, 6.53. Found: C, 73.43; H, 6.57. Similarly, 6f was produced in 91% yield by the
1
diol method using zinc iodide: H NMR (300 MHz, CDCl3) ꢀ 2.46 (ddd, J=17.3, 9.2, 6.2 Hz, H-3, 1H), 2.59 (dt,
J=17.3, 4.2 Hz, H0-3, 1H), 2.65 (ddd, J=15.6, 9.6, 4.9 Hz, H-4, 1H), 2.78 (ddd, J=15.6, 6.4, 4.2 Hz, H0-4, 1H),
3.06 (dd, J=13.2, 8.3 Hz, H-Bn, 1H), 3.20 (dd, J=13.2, 4.9 Hz, H-Bn, 1H), 3.61 (dd, J=8.3, 4.9 Hz, H-1, 1H),
3.63 (s, OMe, 3H), 3.88 (s, OMe, 3H), 6.21 (s, H-5, 1H), 6.63 (s, H-8, 1H), 6.88 (m, Ar-H, 2H), 7.17 (m, Ar-H,
3H); 13C NMR (75 MHz, CDCl3) ꢀ 27.20, 38.80, 40.05, 54.86, 55.99, 56.10, 110.98, 112.09, 126.65, 128.24, 128.83,
129.80, 138.44, 147.60, 147.78, 212.92; MS (m/z) 296 (M+); IR (neat) 1700 cm^1 (CˆO). Semicarbazone: mp 189±
191ꢀC (dec). Anal. calcd for C20H23N3O3: C, 67.97; H, 6.56. Found: C, 67.46; H, 6.65.