(97d) Original Methodology for the Screening of Catalysts Presenting High Reaction Rates at High Conversions

The heterogeneous catalysts screening is the
first and one of the most important steps in catalysts development. The main
aim of screening is to select the catalyst that provides the best performance
in terms of activity and selectivity. Thus, the experimental tests should be
carried out without hydrodynamics and mass transfer limitations. The challenge
appears in the case of industrial applications coupling very high reaction
rates with high conversions, for which the influence of those phenomena cannot
be neglected.

Objective and methodology

The aim of this work was to develop a global
methodology allowing the screening of catalysts, presenting high reaction rates,
at high conversion. This methodology includes both an experimental and a
modeling approach, and it was validated by using the benzene hydrogenation as
case study. The following figure illustrates the different steps of the
methodology.

Figure 1 – Methodology description.

The first step of the methodology consist on
determining the kinetics of the chemical system. Thus, experiments at different
operating conditions must be carried out, and a simulation tool must be develop
and coupled to an optimizer in order to regress the kinetics parameters. The
simulation tool must take into account both hydrodynamics and mass transfer
phenomena, and it must give access to the different species concentration along
the reactor in each phase. The second step consist on analyzing the different
concentration profiles in order to deduce the impact from both hydrodynamics
and mass transfer phenomena on the results. If the impact is not negligible,
modifications, based on numerical simulations, are made in the experimental
protocol to reduce the impact from those phenomena. In case a good fit exists
between model predictions and experiments, the modifications are validated and
the catalyst screening can be carried out in good conditions. Otherwise, the
model hypothesis must be reviewed and a new iteration must be done. This
process is repeated until “convergence” achievement.

Results and discussion

The methodology described above was applied to
the benzene hydrogenation reaction, which is interesting since the industrial
process operates at very high conversions.

The initial experiments at different
temperatures allowed the determination of an activation energy of about 6 kcal/mol for two different catalysts, which is quite low
comparing to the values available in literature of about 12 kcal/mol. This may
be a prove of existence of reactor limitations, internal diffusion limitations
or both, even if a dilution factor of three (three volumes of inert per volume
of catalyst) was used during those experiments.

A simulation tool coupling all the relevant
phenomena within the reactor was developed. The kinetic law was extracted from
literature and the kinetic parameters were regressed and compared with those determined
in the literature. A very good agreement was found, which allows the simulation
tool to be validated. The following figure presents the evolution of the
different species concentration along the reactor in the gas and in the liquid phase
for both catalysts.

Figure 2 – Species concentration evolution
along the reactor in both the gas and liquid phases at 140 °C and 30 bar for a
diluted catalyst bed. BZ – benzene; CH – cyclohexane. CL – liquid
concentration; CI – liquid side interface concentration

The gas-liquid mass transfer limitation is
negligible with the used experimental protocol, and the liquid-solid mass
transfer is unlikely to be a limiting phenomenon since other experiments
carried out at different liquid rates support this hypothesis. The simulation
tool considers the diffusion within the catalyst, which allows the evolution of
the different species concentration within the catalyst particle to be
determined at different reactor heights. This is represented in figure 3. It is
possible to verify that a strong internal limitation exists at these
conditions, since the value for catalyst effectiveness is between 0.3 and 0.5.
This strong limitation explains the low activation energy found for both
catalysts, which is around a half from the one found in literature. This result
consolidates the fact that a strong limitation coming from internal mass transfer
exists for those experiments. This limitation is not critical since it can be
transposed to the commercial scale. In fact, this is a limitation due to the
catalyst itself, and as a result, it is independent from reactor features.

Figure 3 – Species concentration evolution
within the catalyst at the reactor inlet, outlet and middle at 140 °C and 30
bar for a diluted catalyst bed. BZ – benzene; CH – cyclohexane.

The experimental protocol consist on testing
those catalysts near industrial conditions in small fixed-bed reactors and by
using a dilution factor of three. This allows a better performance from the
thermal and from the mass transfer point of view, since the reactant converted
quantity per unit of reactor length is lowered when comparing to the
non-dilution case, while the transfer phenomena is kept constant. The
simulation tool was used in order to generate results for the non-dilution
case. The results are presented in the following figures.

Figure 4 – Species concentration evolution
along the reactor in both the gas and liquid phases at 140 °C and 30 bar for a
non-diluted catalyst bed. BZ – benzene; CH – cyclohexane. CL – liquid
concentration; CI – liquid side interface concentration

Figure 5 – Species concentration evolution
within the catalyst at the reactor inlet, outlet and middle at 140 °C and 30
bar for a non-diluted catalyst bed. BZ – benzene; CH – cyclohexane.

It is possible to verify that the dilution
effect is very effective mainly for catalyst B, which is actually the one
presenting the highest activity. In fact, an important gas-liquid mass transfer
limitation exists for benzene and hydrogen. This result is very important, since,
even if both catalyst can be differentiated under undiluted conditions, the
performance of the best catalyst might not be sufficiently high to justify a
further development if tested under undiluted conditions. Thus, it is important
to couple both the experimental and the numerical approaches in order to
perform the catalyst screening in an efficient way. 

The simulation tool was finally used to analyze
the results at 90 °C. The results are shown in figures 6 and 7. At these
conditions the gas-liquid mass transfer limitation is mainly observed for
hydrogen. The results between the diluted and the non-diluted modes are quite
similar, and are slightly better in the case of the undiluted bed. This result
was expected since hydrogen is a reaction inhibitor under these conditions.
Thus, a stronger gas-liquid mass transfer limitation on hydrogen allows the
reaction to be accelerated, which was not observed at 140 °C because in this
case, the benzene was affected by the gas-liquid limitation as well. The internal
diffusion limitation is lower in this case, since the catalyst effectiveness is
between 0.6 and 0.8. The diffusion model was validated by comparing the
catalyst effectiveness of catalyst B to the experimental value of 0.85 at 80
°C. A simulated value of 0.86 was determined, which allows to consolidate the
simulation tool validation.

Figure 6 – Species concentration evolution
along the reactor in both the gas and liquid phases at 90 °C and 30 bar for
both a diluted and a non-diluted catalyst bed. BZ – benzene; CH – cyclohexane.
CL – liquid concentration; CI – liquid side interface concentration

Figure 7 – Species concentration evolution
within the catalyst at the reactor inlet, outlet and middle at 90 °C and 30 bar
for both a diluted and a non-diluted catalyst bed. BZ – benzene; CH –
cyclohexane.

Conclusion

An original methodology allowing the screening
of catalysts, presenting high reaction rates, at high conversion was developed
and successfully applied to benzene hydrogenation. The results show that it is
quite important to perform catalyst dilution at these conditions, even if
special care must be taken in order to avoid some reactor by-passing and other
hydrodynamics problems. Another important conclusion is the critical character
of the numerical approach, that must be used in order to avoid high
experimental costs or erroneous interpretations from experiments. In fact, the
numerical approach permits a better understanding of what actually happens
during the experiments. A good example is what happened when two different
catalysts were numerically tested by using both the dilution and non-dilution
experimental protocols at the same operating conditions. Even if the results
are quite similar, it does not mean that gas-liquid mass transfer limitations
are negligible in the non-diluted catalytic bed.