A DRAMATIC SOLVENT EFFECT DURING AROMATIC HALOGEN METAL EXCHANGES - DIFFERENT PRODUCTS FROM LITHIATION OF POLYFLUOROBROMOBENZENES IN ETHER AND THF

As part of a synthetic program in quinolone antiinfec-tives, we wished to prepare 2,4,5-trifluorobenzoic acid (1) and 3-bromo-2,5,6-trifluorobenzoic acid (2) from 1-bromo-2,4,5-trifluorobenzene (3). Although acid 1 can be prepared via cyanation (CuCN) of 3 followed by acid hydrolysis,1 it appeared that bromine-lithium exchange followed by carboxylation would be operationally simpler. Halogen-metal exchange is a very well-established technique for the regiospecific generation of aromatic anions.2 Because aromatic bromine-lithium exchange is very rapid, it allows for lithiation to be carried out at positions in the molecule that would not normally be available due to the presence of kinetically or thermodynamically more acidic sites.3 Conversely, simple metalation of 3 should occur at the most acidic site, and carboxylation should yield acid 2. Although lithiation between two fluorines with butyllithium (BuLi) is known,4 in this case halogen-metal exchange must be avoided. In this note we wish to report that bromide 3 is acidic enough to be metalated by LDA, that acid 1 can also be prepared from 3 via halogen-metal exchange in ether, and that this latter reaction shows a remarkable solvent effect, such that changing the solvent to THE leads to a mixture of acid 2 and 2,3,6-trifluorobenzoic acid (4). The mechanism of this latter process was shown to be compatible with autometalation.5. © 1990, American Chemical Society. All rights reserved.