X X X OH OHOH OH X Bu t Bu t Bu t Bu t C1 T1 : X = CH 2 : X = S O O OH OH Bu t Bu t Bu t Bu t 2 Si Me Me TETRAHEDRON LETTERS Tetrahedron Letters 43 (2002) 621–625 Pergamon Proximal O,O -capped calix[4]arenes with a disiloxane bridge as highly efficient synthetic intermediates for 1,2-dialkylation at the lower rim Fumitaka Narumi, a, * Naoya Morohashi, b Nobuji Matsumura, b Nobuhiko Iki, b Hiroshi Kameyama a and Sotaro Miyano b, * a Department of Basic Sciences, School of Science and Engineering, Ishinomaki Senshu University, 1 Shinmito, Minamisakai, Ishinomaki 986 -8580, Japan b Department of Biomolecular Engineering, Graduate School of Engineering, Tohoku University, Aramaki -Aoba 07, Aoba -ku, Sendai 980 -8579, Japan Received 2 October 2001; revised 14 November 2001; accepted 16 November 2001 Abstract—Treatment of calix[4]arenes with 1,3-dichloro-1,1,3,3-tetraisopropyldisiloxane using imidazole as a base yielded O,O- bridged calixarenes in excellent yields, which were subsequently treated with an alkyl halide/base to give O,O-dialkylated products in high yields. These were, in turn, amenable to fluoride-promoted quantitative desilylation, thus providing an efficient method for the net proximal dialkylation of calix[4]arenes at the lower rim. © 2002 Elsevier Science Ltd. All rights reserved. In the last two decades, calix[4]arenes have been one of the most extensively investigated molecular platforms for the construction of synthetic receptors in molecular recognition and supramolecular chemistry. 1,2 This is mainly due to the synthetic availability of large quanti- ties of the parent macrocycles, especially p -tert -butyl- calix[4]arene (C1), 3 and their ready modifications at the phenolic hydroxy groups (lower rim) as well as at the para positions (upper rim) to introduce a variety of substituents to develop varying functions. 4,5 However, during the course of our study on the development of new functions of thiacalix[4]arenes (e.g. T1), 6–8 we real- ized that regioselective O,O -difunctionalization at the neighboring phenolic hydroxy groups of the calix[4]arene class compounds has not yet been suffi- ciently explored, and development of a general method for such transformations would be highly desirable because they would provide very useful building blocks for the design of synthetic receptors. 9–12 One plausible approach to accomplish the above objec- tive may be to block two neighboring hydroxy groups out of the four by bridging them with a proper protect- ing unit. Although Reinhoudt and co-workers proposed O,O -bridging by reacting C1 with phthaloyl chloride, the yield of the bridged product was only fair (30%). 13 Furthermore, the ester functionality of the protecting unit may not always be suitable for application to proximal diesterification of the remaining hydroxy groups. In this context, a paper reported by Lattman et al. attracted our attention. They reacted C1 with dichlorodimethylsilane/triethylamine to give O,O - dimethylsilyl-bridged product 2 in 60% yield, which, as they had expected, should be an ideal starting material for the synthesis of 1,2-disubstituted calix[4]arenes. 14 However, treatment of 2 with BuLi/CF 3 SO 3 Me gave not the desired O ,O -dimethylated product but dou- bly 1,2- and 3,4-silyl-bridged calix[4]arene accompany- ing the formation of a tetramethoxy derivative of C1, seemingly via the attack of the phenoxide on one 2 on the silicon on another 2. 14 We supposed that such kind of disproportionation might have been brought about Keywords : O,O-bridged calix[4]arenes; proximal dialkylation; disy- loxane. * Corresponding authors. Tel.: +81-225-22-7716; fax: +81-225-22- 7746 (F.N.); tel./fax.: 81-22-217-7262 (S.M.); e-mail: fnarumi@ isenshu-u.ac.jp; miyano@orgsynth.che.tohoku.ac.jp 0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII:S0040-4039(01)02202-X