Similarities and differences in catalytic hydrocarbon conversions by liquid and solid acids

The acidities of solids and liquids differ in an essential way, because aci d groups on solid surfaces cannot cooperate in the transfer of hydrons to b ases. Solids are intrinsically much weaker than liquid acids of similar str ucture. Claims of superacidity in solids, particularly zeolites and sulfate d metal oxides, are shown to be incorrect. The inadequacies of the electron egativity equalization principle and of the representation of zeolites as p seudoliquids are discussed. The activation of saturated hdrocarbons by soli ds for carbocation-type reactions, like isomerization and cracking, is base d oil bifunctional, rather than superacid catalysis. In some cases, like su lfated metal oxides, the bifunctional catalysis consists of one-electron ox idation and strong acid (but not superacid) catalysis. Conversion of hexane isomers on zeolite HZSM-5 at 120-160 degrees C shows an inversion of the r elative reactivity of hexane (n-H) and 3-methylpentane (3MP) as liquids und er pressure (3MP reacts faster) from the reaction in the gas phase (n-H rea cts faster). The weak superacid trifluoromethane-sulfonic acid (TFMSA, H-o- 14.2) cleanly catalyzes the isomerizations 3MP --> 2MP and n-H --> (2MP + 3 MP) at room temperature, if unsaturated organic species formed in the acid layer near the interface are dispersed. The relative reactivity 3MP/n-H (ca 1000) is similar to that in HF-BF3 and HF-SbF5. The activation parameters for 3MP --> 2MP (Delta H++ = 19 kcal/mol, Delta S++ >> -16 cal/mol.deg) sug gest that the rate-determining step is the ionization of an ester intermedi ate. If the acid layer is not homogenized an accelerated conversion, consis ting mostly of disproportionation and cracking, occurs after an induction p eriod. The acceleration is much higher for n-H, such that under these condi tions the overall rate ratio 3MP:n-H (ca 10) is similar to that on HZSM-5 f or the reaction of the liquid hydrocarbonss. The initiation by TFMSA of the "cracking mode" appears to be oxidizing in nature, as shown by the fact th at small amounts of Fe3+ ions suppress the induction period.