Page 60 - Principles of Catalyst Development

CATALYST DEVELOPMENT 47
trends become apparent. Longer periods are desirable but are seldom
justified economically.
4. To develop scale-up data for commerical plant design. Economic
evaluations based on projected plant configurations and unit construction
are speculative up to this point. Pilot units studies provide more realistic
data so that refinement of earlier estimates is possible. Also, any new reactor
designs or innovations can be tested.
5. To obtain product samples. Large quantities of products may be
accumulated for specification assessment. This is useful not only for testing
effectiveness of the catalytic process but also in providing data for design
of downstream components, safety, or market development, and so on.
6. To train personnel. Operators and engineers for the commercial
units may be trained on pilot units, where the range of "adjustable" condi-
tions is much wider and the result of misjudgments less disastrous.
Pilot units are investments and commitments of time and money. They
are expensive and should not be misused. As much technical information
as possible should be obtained during the laboratory work and a clear
realization of the objectives of the pilot program formalized.(87) The size of
the pilot unit is a very important factor in construction and operating costs.
Keeping in mind that testing commerical formulations is the key task, the
reactor unit must be sized so as to simulate the final design but yet be within
reasonable limits. Sometimes simulating the performance of larger beds is
necessary, e.g., to determine temperature profiles in adiabatic reactors.
Heat-transfer-limited processes are best tested with commercially sized tubes
since heat-transfer characteristics are difficult to estimate.
Every application has a different set of constraints. The degree of
uncertainty and risk factors in the final design are important in determining
the size of intermediate pilot units. It is useful at this point to simulate
performance with computer models that may then be "calibrated" with
pilot unit data. Such models provide very effective tools for parametric
studies and assessment. (88)
The catalysts used in pilot units should be manufactured with com-
merical procedures. Large, consistent quantities of the catalyst are required.
If the organization conducting the pilot program does not have access to
manufacturing facilities, it is wise to engage the cooperation of catalyst
manufacturers at this stage.
Automation of pilot units increases construction but reduces operating
costs. Turnkey units are available for almost any size and application.
However, before embarking on this route, care should be taken to ensure
that the critical objectives of the program are achieved.
We hope, at this point, that pilot unit studies justify catalyst design.
Wrong decisions in defining objectives will certainly be obvious. Unexpected