(95a) Switching Between Hydrogenolysis and Bifunctional Hydrocracking on An Unsulfided Co/MoO3/SiO2-Al2O3 Catalyst

In previous work using hexadecane as a feed molecule, we found that under typical catalytic hydrocracking conditions, an unsulfided Co/MoO₃/SiO₂-Al₂O₃ catalyst exhibited primarily hydrogenolysis.  Two types of hydrogenolysis were observed.  One type involved terminal cleavage of methyl groups, in which the feed n-C₁₆ underwent progressive cleavage to first n‑C₁₅, then, to a lesser extent to n-C₁₄ and so on.  This resulted in high quantities of product methane.  The second type of hydrogenolysis reaction resulted in the cleavage of interior C-C bonds with near equal probability.  This second pathway yielded ethane in molar quantities similar to other cracked products in the C₃ to C₁₃ range.  Further, the overall yield of isomerized cracked products was quite low, typically less than 10 mole percent.  However, when feeding a mixture of n-C₂₆/n-C₂₈ to the same catalyst under similar conditions, methane yields were low, ethane was barely detectable, and isomer content of the cracked products was in excess of 80 mole percent.  These latter observations are consistent with a bifunctional hydrocracking mechanism.  Further, a comparison of the product selectivities as a function of conversion in these two cases are consistent with hydrogenolysis in the n-C₁₆ case and bifunctional hydrocracking in the n-C₂₆/n-C₂₈ case.  In an effort to understand the reasons for this feed dependent reaction behavior, we have investigated the cracking of n-C₁₈ under similar conditions.  We have found that we are able to switch between hydrogenolysis and bifunctional hydrocracking by altering reaction conditions so as to either favor or suppress the relative olefin content of the reacting C₁₈.  While holding the feed rate of hydrocarbon constant, hydrogenolysis is nearly exclusively observed at 300° C when H₂ is fed at a molar ratio of 10:1 H₂:n-C₁₈.  However, at the same absolute H₂ and n-C₁₈ feed rate, but with a supplementary flow of Ar at twice the H₂ feed rate, nearly exclusively bifunctional hydrocracking behavior is observed.  In contrast, when only H₂ is fed at a 30:1 H₂:n‑C₁₈ rate, an even higher rate of exclusively hydrogenolysis is observed.  Similarly, we found that when n-C₁₆ is fed with the supplementary Ar flow, it also exhibits predominantly bifunctional hydrocracking behavior.  We believe that when the olefin content of the hydrocarbon reactant is sufficiently high, bifunctional hydrocracking predominates.  Hence, since the equilibrium olefin content of a C₂₆/C₂₈ feed is considerably higher than in a C₁₆ feed, at equivalent molar flux, bifunctional behavior dominates in the C₂₆/C₂₈ case.  For this reason, when seeking to process a typical Fischer-Tropsch wax, where the dominant molecular sizes are in the C₂₆ and larger range, quite typical bifunctional hydrocracking is exhibited by this catalyst.