'l'iclcling and decoupling (triple resonance) techniques show that the ortho, meta, and para proton coupling constants have the same sign in 2-bromo-5-chlorotoluene. If the ortho coupling in the ring is positive, then decoupling experiments show that the methyl proton couplings to the ring protons in the ortho and para position to the methyl group are negative, while that to the meta position is positive. These signs are in agreement with the n-r exchange polarization mechanism. INTRODUCTION Long range coupling between the methyl and ring protons in substituted benzenes is lcnown (1-3). Hoffman (1) attributed such coupling in mesitylene to hyperconjugation. Calculations by McConnell (4, 5) invoked a u-a exchange polarization mechanism for long range couplings like the para coupling of ring protons. Acrivos (6) found that the magnitudes of the ring-methyl proton couplings in mesitylene were predicted satisfactorily by the same mechanism, as did Cohen and McLauchlan (3) for 2-carbomethoxy-5,6-dimethylbenzofuran. Theory (5, 6) also predicts the signs of the couplings. While the relative signs of methyl proton couplings with protons on the furan (7) and thiophene rings are known (8, 9), such is not the case for the benzene ring, although there have been a few careful determinations of the magnitudes of these couplings (3, 10).T h e ortho coupling between the protons on the benzene ring is positive (11-13) as are the meta and para couplings in all lcilown cases (14-17). We describe experiments here u~hich determine the sign of the ring-methyl proton couplings and of the meta and para couplings relative to the ortho coupling in 2-bromo-5-chlorotoluene.
EXPERIMENTALMeasurements were made on a degassed, 10 mole yo solution of 2-bromo-5-chlorotoluene in carbon disulhde, with tetra~nethylsila~le (TIMS) as internal reference. Decoupling and tickling experiments, some of them involving triple resonance, were performed with a DA-60-1 Varian spectro~neter according t o standard procedures (18)(19)(20)(21)(22). The temperature of the sample was 28.5 "C as determined using an ethylene glycol s a n~p l e .~ RESULTS