FERMENTATION SCALEUP: INDUSTRIAL EXPERIENCE WITH A TOTAL ENVIRONMENTAL APPROACH

The scaleup problem most often faced by practicing industrial biochemical engineers in the pharmaceutical industry today is different from that of our predecessors who first carried penicillin, streptomycin, and a host of other new antibiotics from shake flasks to deep tank fermenters. In those “good old days” the scaleup problem encountered was largely mechanical or equipmental: How to sterilize large volumes of fermentation media and air, how to control broth pH and temperature, and how to design a mechanical agitator for adequate mixing and oxygen transfer. In the 1950s and 1960s fermentation plants were built around the world principally for manufacture of antibiotics. Processes employed in these plants were young, unoptimized. Profits were high, and new proprietary fermentation based antibiotics were commonplace so that incentives for process optimizations were somewhat limited. Today, while there are occasional new fermentation plants on the drawing board, the pace of new plant proliferation has slowed. Government regulatory pressures and other factors have slowed emergence of new fermentation-based antibiotics. Processes for production of major antibiotics are mature and well tuned. As a consequence, the scaleup problem most commonly encountered in our industry today is the translation of a new process for an existing product or an improvement on an existing process, from the laboratory into an existing fermentation plant or plants. The total environmental (TE) approach to fermentation process scaleup that I will describe to you is certainly not an original concept. In fact it is as old as biochemical engineering and might well derive from one of Elmer Gaden’s early publications’** back in 1955 in which fermenter volumetric and average cellular specific rates are analyzed for various fermentations. The term total environment was chosen in response to a great deal of biochemical engineering literature, which has been concerned with various subsets of environment in relation to scaleup (usually oxygen transfer) but not often with all subsets a t once. Furthermore, it would be less than candid for me to claim that we industrial types have perfected the application of this so-called T E approach and now have a foolproof method of process scaleup. We have found it to be at least as effective as voodoo dolls, rosaries, and crystal balls; however, and if experience teaches anything of value, it is perhaps worthwhile to relate to this distinguished group how T E is applied in industry today. I will give some examples of success and failure and some thoughts on how the approach will be improved in the future. I will begin with a brief description of the TE approach.