Trace organosulfur compounds present as natural impurities in oleochemical
feedstocks may lead to deactivation of copper-containing catalysts applied
for hydrogenolysis of esters toward fatty alcohols. In this paper, the sulf
ur deactivation of Cu/SiO2 and Cu/ZnO/SiO2 catalysts was studied in the liq
uid-phase hydrogenolysis of methyl palmitate. The rate of deactivation is f
ast and increases as a function of the sulfur-containing compound present:
octadecanethiol approximate to dihexadecyl disulfide < benzyl isothiocyanat
e ( methyl p-toluene sulfonate < dihexadecyl sulfide ( dibenzothiophene. Th
e rapid deactivation is caused by the fact that sulfur is quantitatively re
moved from the reaction mixture and because mainly surface sulfides are for
med under hydrogenolysis conditions. The life time of a zinc-promoted catal
yst is up to two times higher than that of the Cu/SiO2 catalyst, most likel
y due to zinc surface sulfide formation. The maximum sulfur coverage obtain
ed after full catalyst deactivation with dibenzothiophene and dihexadecyl s
ulfide-the sulfur compounds that cause the fastest deactivation-may be as l
ow as 0.07. This is due to the fact that decomposition of these compounds a
s well as the hydrogenolysis reaction itself proceeds on ensembles of coppe
r atoms. For the most reactive sulfur compounds, surface coverage near the
maximum value of theta(Cu) = 0.5 or-in the presence of zinc-formally in exc
ess of this quantity may be reached at full catalyst deactivation. At that
point, still some sulfur uptake occurs. Decomposition of such compounds is
even possible in the absence of hydrogen and sulfur is not laid down in a d
ispersed fashion, as in the case of dibenzothiophene and dihexadecyl sulfid
e. Catalyst regeneration studies reveal that activity cannot be regained by
reduction or combined oxidation/reduction treatments. XRD, TPR, and TPO re
sults confirm that no distinct bulk copper or zinc sulfide or sulfate phase
s are present. (C) 1999 Academic Press.