Scale-up of continuous reactors using phenomenological-based models

Abstract

A methodology for scaling up continuous reactors using a Phenomenological-Based Semiphysical Model (PBSM) of the process is presented. First, a review of the most popular scale-up methods is made, finding that these methods do not guarantee an adequate commercial unit design and that a PBSM is a fundamental tool when scaling up chemical reactors. Taking into account these facts, a novel methodology is presented in which a process PBSM and its Hankel matrix are used for analysing the process dynamic behaviour and scaling it up, including the effect of the design variables over each state variable. Finally, the proposed methodology is applied to a polymerization reactor, comparing the scaled unit design when using this approximation and a traditional method, finding the scale factors for keeping the same polymer molecular weight at the new scale and, demonstrating that traditional scale-up methods do not always lead into the best commercial unit design.

Although the scale-up activity is one of the major tasks for chemical engineers and represents a fundamental step in the design and optimization of industrial plants (1), since the 1960s there has not been significant progress in the methods for scaling up chemical reactors (use of rules of thumb, similarity criteria and dimensional analysis) (2-7). Industrial scale-up is dominated by empirical criteria that require geometrical similarity fulfilment, leading to drawbacks from keeping a single parameter constant (8, 9) and originating changes in other important variables. This fact results in an erroneous commercial unit design that demands additional costs and time to be corrected (10-12).