Pilot Plants

Abstract

While advances in modeling and predictive methods have greatly expanded the range and utility of theoretical research, there still can remain a significant degree of uncertainty in the final results before a major investment is justified. Small-scale laboratory tests help minimize this uncertainty but are fraught with risk due to the small scale and manual batch simulation of most of the process steps. Hence, the concept of piloting the process in a miniature unit is a way which more closely simulates the actual process plant. Pilot plants have evolved from multistory semi-works units, designed for one-tenth of final process scale, to microunits, which fit in larger laboratory hoods. Their cost and complexity often belie their small size and great utility. The piloting process often occurs several times during the life of a research program. Table 1 shows a sequence of this piloting that might be part of a research program. Depending on the complexity of the process, the results of subsequent steps, and the organization’s experience and comfort with the process and results, some of these steps may be eliminated. Recycle and feedback between each step is also possible when problems or concerns arise. As each step along this path involves an exponential increase in resources, time, and money required, there is a strong incentive to progress to the next stage as soon as it is practical. Conversely, there is a need to minimize the amount of recycling as the project progresses through the various steps to the commercial unit. While this rapid progress is desirable, the consequences of skipping a necessary preliminary stage or failing to fully understand the results of the previous stage can be devastating. Significant time and resources can be wasted at the next stage if it is focused along an incorrect or at least nonoptimum path. A substantial amount of time and money has progressively been invested, and any desire to change the process to make minor improvements must be resisted. At some point the decision must be made that the remaining unanswered process questions are acceptable risks. In today’s climate of rapid change, timeliness can be as important as or more important than minimizing the risks associated with a new product or process. Examples include securing a new market with a totally novel product or attempting to secure a patent position before a competitor. In this situation, the decision may be made to proceed to commercialization earlier than desirable, prior to satisfactorily resolving all design concerns. This often usually requires a more conservative (and more expensive) design approach, results in a significantly longer start-up, or produces a first unit plagued with operational difficulties. Resolving significant problems during start-up or in an operating unit, while feasible, is risky, expensive, and time-consuming. Rarely is it more effective than additional research. A pilot plant is a collection of equipment designed and constructed to investigate some critical aspect(s) of a process operation or to perform basic research. It is a tool rather than an end in itself. A pilot plant can range in size from a laboratory bench-top unit to a facility approaching the size of a commercial unit.