Introduction - Bioprocess Engineering: Kinetics, Biosystems, Sustainability and Reactor Design - page 14

involved the growth of the mold on top of a quiescent medium. The surface method used
a variety of containers, including milk bottles, and the term “bottle plant” indicated such
a manufacturing technique. The surface method gave relatively high yields but had a long
growing cycle and was very labor intensive. The first manufacturing plants were bottle
plants because the method worked and could be implemented quickly. However, it was clear
that the surface method would not meet the full need for penicillin. If the goal of the War
Production Board was met by bottle plants, it was estimated that the necessary bottles would
fill a row stretching from New York City to San Francisco. Engineers generally favored
a submerged tank process. The submerged process presented challenges in terms of both
mold physiology and tank design and operation. Large volumes of absolutely clean, oil-
and dirt-free sterile air were required. What were then very large agitators were required,
and the mechanical seal for the agitator shaft had to be designed to prevent the entry of
organisms. Even today, problems of oxygen supply and heat removal are important
constraints on antibiotic fermenter design. Contamination by foreign organisms could
degrade the product as fast as it was formed, consume nutrients before they were converted
to penicillin, or produce toxins.
In addition to these challenges in reactor design, there were similar hurdles in product
recovery and purification. The very fragile nature of penicillin required the development
of special techniques. A combination of pH shifts and rapid liquid e liquid extraction proved
useful.
Soon processes using tanks of about 10,000 gal were built. Pfizer completed in less than 6
months the first plant for commercial production of penicillin by submerged fermentation
(Hobby, 1985). The plant had 14 tanks each of 7000-gal capacity. By a combination of good
luck and hard work, the United States had the capacity by the end of World War II to produce
enough penicillin for almost 100,000 patients per year. A schematic of the process is shown in
Fig. 1.5 .
This accomplishment required a high level of multidisciplinary work. For example, Merck
realized that men who understood both engineering and biology were not available. Merck
assigned a chemical engineer and microbiologist together to each aspect of the problem. They
planned, executed, and analyzed the experimental program jointly, “almost as if they were
one man” (see the chapter by Silcox in Elder, 1970).
Progress with penicillin fermentation has continued, as has the need for the interaction of
biologists and engineers. From 1939 to now, the yield of penicillin has gone from 0.001 g/L to
over 50 g/L of fermentation broth. Progress has involved better understanding of mold phys-
iology, metabolic pathways, penicillin structure, methods of mutation and selection of mold
genetics, process control, and reactor design.
Before the penicillin process, almost no chemical engineers sought specialized training in
the life sciences. With the advent of modem antibiotics, the concept of the bioprocess engi-
neering was born. The penicillin process also established a paradigm for bioprocess develop-
ment and biochemical engineering. This paradigm still guides much of our profession's
thinking. The mindset of bioprocess engineers was cast by the penicillin experience. It is
for this reason that we have focused on the penicillin story, rather than on an example for
production of a protein from a genetically engineered organism. Although many parallels
can be made between the penicillin process and our efforts to use recombinant DNA, no
similar paradigm has yet emerged from our experience with genetically engineered cells.