The mistake was obvious. Most experienced engineers saw it in a few minutes. The safety manager quizzed me as I entered the trailer. I nodded my head in disbelief. The problem was the pencil-style fluid bed reactor. Instead of designing the production-scale reactors from prototype production runs, the client built the production units without really testing the prototype. The “startup” concluded in complete failure after a few hours. I spent the remainder of the contract documenting equipment operations and writing a report on how the bed carryover would affect downstream equipment and production capacity.
Fluid and fixed beds exemplify technologies that require serious attention to scale-up. I already can hear the simulation crowd howling about computational flow dynamics; it certainly is helpful but not definitive — at best, providing answers without explanations.
Some technologies don’t demand as much attention to scale-up because of insights gained over many years of industrial development — more than a century in the case of distillation columns. (However, estimating tray efficiency may call for testing.) Others rely on a well-defined relationship — for centrifuges, this is the so-called “sigma factor” — to provide a rough scale-up. Unfortunately, many other technologies, including crystallizers, dryers, fluid beds, batch reactors, gravity settlers and solvent extractors, require labor-intensive scale-up. As a general rule, the faster the fluid flows and the lower its viscosity, the easier the scale-up is; scale-up seldom poses an issue where dry gases are involved.
Scalability of equipment should receive thought early on. A reactor six stories high going through a 100-psid filter may seem practical in the laboratory but will pose nightmares for the engineer sizing pumps. The pump that served well in the pilot plant may not come in a size large enough for the production unit. Moreover, regulatory standards tighten once a process leaves the laboratory, so seal selection becomes tougher. These challenges often haunt prototype design.
Pumps aren’t the only problem. Once I had to explain to a research chemist that I would need a 300-hp agitator to duplicate what he was doing in the lab; he changed his batch schedule.
Agitation is a good example of technology that has advanced but still requires some attention to scale-up. (That’s why it’s one of the most popular areas of CP’s online “Ask the Experts Forum,” www.ChemicalProcessing.com/experts/mixing/.) Mixing becomes less homogenous during scale-up from the lab: poor agitation reduces heat transfer and mass transfer, can ruin the product and even cause runaway reactions. My first assignment after college was investigating an explosion in a reactor, prompted in part by scaling-up a solid rocket propellant from a 1-gal. batch to a 5-gal. batch. Poor heat dissipation caused the explosion; we changed the ingredients and batch schedule. Certainly, agitation, or lack of it, at the production scale can affect the order in which ingredients are added and, perhaps, necessitate additional equipment or precursor steps. Other influences that aren’t easily visible at the lab scale may come into play at the pilot, prototype and production scales; these include corrosion, impurity accumulation, side reactions, unknown catalysts caused by changes in materials and, of course, poor communication among staff.
In addition, for instance, a solvent may work well in the lab but not in production because controlling its contact with water is harder in the plant. My suggestion is to keep the components the same or start over.
Chemists prefer glass and plastic; engineers like shiny new stainless steel. This can be a problem, especially with side reactions that corrode metal. Cleaning usually becomes more difficult at larger scales. Be especially careful about safety tests: the auto-ignition temperature for benzene is 200°F higher in a steel container compared to a glass one. So, it’s a good idea to repeat safety testing as development progresses.
Also keep in mind that lab tests usually don’t consider static electricity, flash points, etc. Such factors affect not only production batch reactors but also downstream equipment.
Separation equipment typically represents 30–70% of the capital investment in a new process. A serious mistake in separation can destroy the viability of a new process. It’s far better to catch these problems at the pilot scale than waste vast sums on prototype-scale equipment.
DIRK WILLARD is a Chemical Processing contributing editor. He recently won recognition for his Field Notes column from the ASBPE. Chemical Processing is proud to have him on board. You can e-mail him at firstname.lastname@example.org