162
G. Onyestyák et al. / Catalysis Communications 27 (2012) 159–163
AA, intermediate aldehyde can be detected and ester formation is ob-
servable, however, the course of main reaction, i.e., the selective hydro-
genation of carboxylic acids to alcohols, is very similar. The conversions
are practically the same when mass rates are the same but in this case
the molar feed rate of AA is at least 2 times higher than that of octanoic
acid; it means that the conversion rate in moles/s×gcat is at least 2 times
higher for AA than for OA.
Ni,Mo(P,Si)/Al2O3 commercial hydrotreating catalyst was tested
in the AA hydroconversion with or without indium doping under
the same conditions as that with the other catalysts (Fig. 3). In spite
of the lower Ni content the industrial NiMo/Al2O3 catalyst shows
hardly lower activity than the novel InNi(P,Si)/Al2O3 composite (cf.
Figs. 3a and 2b). However a whale of difference can be observed in
the product distributions supposing to be controlled by the different
co-catalysts (MoO3 or In2O3 added approx. in the same concentra-
tion). From acetic acid the commercial Ni,Mo(P,Si)/Al2O3 catalyst pro-
duces mainly ethyl acetate. It seems likely that AA hydroconversion is
not followed by Fischer esterification (reaction 3 in Scheme 1) but by
direct formation of the ester from aldehyde over the molybdena as a
heteropolyacid catalyst according to Tishchenko reaction (reaction 4
in Scheme 1) [9]. (Understanding of this mechanism is not a subject
of the present work studying influences of indium admission.). Fi-
scher esterification is less presumptive since initial formation rate of
ethyl acetate is higher than that of ethanol. The molybdenum compo-
nent of the 9Mo2.5Ni(P,Si) catalyst seems to be more efficient and at
the low nickel content the influence of indium doping is weak.
It was found lately that, at a fixed Ni-loading (9 wt.%) decreasing
the amount of added indium oxide in a wide range down to 10 wt.%
hardly affects the selectivity of the catalyst composites in octanoic
acid hydroconversion [1]. Hence it seems likely to apply in low con-
centration the expensive, but strongly efficient In2O3 co-catalyst. It
was estimated that both active metal components can be used in
about 10 wt.% to attain high carboxylic acid conversion with high
and stable alcohol yield. Therefore, the conversion of the practically
more important acetic acid on the same catalyst is studied only in
this low In2O3 co-catalyst range. In Fig. 4 product distributions as a
Fig. 5. Hydroconversion of AA over 9Ni/Al2O3+10% In2O3 catalysts as a function of
space time at 21 bar total pressure and 300 °C.
function of the admixed indium content can be compared at high con-
versions at 280 and 300 °C. Acetaldehyde concentration is negligible
in the full range of indium loadings. The first step of acetic acid reduc-
tion should give acetaldehyde, as a highly reactive intermediate,
which can produce ethanol through reduction or directly ethyl ace-
tate in the bimolecular Tishchenko reaction (Scheme 1). The last
route seems to be more dominant at lower reaction temperature
and lower indium content. At higher indium loading the reduction
of acetaldehyde can be faster on the active sites of increased number
or/and the hydrogenation of ethyl acetate to two ethanol molecules
can be more complete. Around 300 °C and 10 wt.% In2O3 loading
ethanol can be produced with high selectivity.
The increasing space time does not favor the accumulation of the
acetaldehyde intermediate (Fig. 5) which can be the reactant of two
parallel reaction routes: ester formation or reduction.
The AA conversion to ethanol and ethyl acetate increases roughly
linearly with increasing hydrogen partial pressure over 9Ni/Al2O3+
6% In2O3 composite catalyst. The acetaldehyde selectivity is the
highest at lowest hydrogen partial pressures (at low conversions).
The hydrogenation activity is suppressed by an increased AA coverage
(higher AA partial pressures). The found activity dependence on the
reactant partial pressures is the same as in our previous work [1]
and it is in line with the Langmuir–Hinshelwood type kinetics and
mechanism.
a
4. Conclusions
Hydroconversion of aliphatic carboxylic acids over a Ni/Al2O3
catalyst and the yield of selectively produced alcohols (octanol or
ethanol) can be increased drastically by In2O3 doping. Appearance
of metallic indium near nickel can effectively direct the step by step
catalytic reduction to alcohol formation over partly reduced Ni catalysts
instead of hydrodecarbonylation. Indium doping has been found to be
highly efficient also for the reduction of acetic acid with hydrogen to
ethyl alcohol.
b
On comparing the hydroconversion of octanoic and ethanoic acids
over InNi/Al2O3 composite catalyst under similar conditions the con-
version rate of ethanoic acid is twice as high that of octanoic acid. It
suggests that the molecular reduction rate of the carboxylic group
on the bimetallic surface (perhaps only to a formyl group) is higher
for lighter carboxylic acid than for heavier one. However, the product
compositions are different: after hydroconversion of octanoic acid ei-
ther aldehyde or ester cannot be detected but there are some alcohol
dehydration, while for acetic acid, ester formation is considerable and
aldehyde formation can be observed but no dehydroxylation.
The activity dependence on the reactant partial pressures denotes
the rate-controlling Langmuir–Hinshelwood kinetics.
Fig. 4. Comparison of the yields of main products (■) and the conversions (●) on 9Ni/Al2O3
catalysts with various amounts of In2O3 admission in the hydroconversion of AA at 280 and
300 °C at 21 bar. Reaction conditions as in Fig. 1.