[pullquote]
In managing a chemical plant, we invariably work with people with strongly different opinions and agendas. My early exposure to solids processing involved a group of people who had preconceived ideas on why solids behaved as they did. These people had developed rules not based on scientific principles but rather on their experiences. This wasn’t because they were unaware of the science — far from it. Actually, little science about particulate behavior existed at that time. So, they were forced to develop their own rules of thumb. One “rule” I’ve heard about particle blending is that increased blending time results in a better blend. However, the frictional and electrical characteristics of every solid or mixture of solids actually determine the best blending time. Too much time can give a poorer blend, resulting in poor utilization of the equipment, as I’ve already covered in “Beware of Blending Myths.”
I’ve developed rules to tide me over when I’m unable to resolve a discrepancy between science and experience. (Hopefully, advances in science will make such rules unnecessary in the future.) Crystallization primarily is science but, for years, nucleation and growth went unexplained. Our experience told us that some fouling eventually would happen unless we took some production time off to clear deposits. It was easier to defrost than to clean out the crystallizer after it crashed. So, we would heat the solution up enough to clear these deposits.
To prevent flash-crystallization as the crystallizer cooled, we would rap on the tank to speed up the process. The operators had tons of experience from rapping at the wrong temperature and, so, had figured out when rapping works. We continued to use this approach until in-situ nucleation devices were developed. These units went hand-in-hand with an accurate solubility curve and an understanding of the width of the meta-stable zone of the product to provide a more-reliable method to forestall flash crystallization.
Drying of solids is an area riddled with experiential learning, most of it bad. In theory, drying is a two-step process. The first step is removal of surface moisture. The science is well established for this step, which is heat-transfer limited. The books tell us there’s a critical moisture at which the second step — carrying away of capillary or internal moisture — starts. However, in reality, this step is treacherous. Most people expect to see a distinct transition at the critical moisture but fail to account for the cooling of the solids due to evaporation prior to this point. Slight variations in initial moisture can drastically change the final moisture content of the product.
To compensate for these variations, we recommend setting the drying time for a product based on when the bulk solids started to increase in temperature — treating that as an indicator of reaching the critical moisture. This isn’t foolproof, though, as one of our plants unfortunately found out. Research had shown that 30 minutes at a specific temperature gave a dry product without overheating it. During a campaign, the filter ahead of the dryer yielded an ever-increasing moisture content. The final product from the dryer varied over the course of the campaign, which resulted in clumping. By itself, clumping wasn’t a problem. However, it led to a polymorph on storage that ruined the product.
Many people relied on another old rule of thumb that, to maintain flow of solids, chutes should be designed based on the poured angle of repose. That rule was abused or modified over the years to try to understand why particles never seemed to follow it. The advent of shear testing has brought science back into the picture. Now it’s clear that the finer particles dominate the flowability of a bulk solid — and that wall materials of construction have a large influence on chute design.
Another old rule that dominated the pneumatic conveying industry was to use long-radius elbows to reduce attrition. We now know that T-bends and even short radius elbows of special configurations often outperform the long radius elbow in attrition and even pressure drop. While long radius elbows intuitively would seem to be better, the particles have a different agenda. This is why I’m still convinced that particles have a mind of their own.
