As any process development chemist knows, there is motivation to optimize a chemical process to produce maximum output in the minimum of reaction space. In the context of this essay, I’m referring to batch or semi-batch processes. Most multipurpose fine chemical production batch reactors have a capacity somewhere between 25 and 5000 gallons. These reactors are connected to utilities that supply heat transfer fluids for heating and cooling. These vessels are connected to inerting gases- nitrogen is typical- and to vacuum systems as well.
Maximum reactor pressure can be set as a matter of policy or by the vessel rating. Organizations can, as a matter of policy, set the maximum vessel pressure by the selection of the appropriate rupture disk rating. Vessel pressure rating and emergency venting considerations are a specialist art best left to chemical engineers.
Reactor temperatures are determined by the limits of the vessel materials and by the heat/chiller source. Batch reactors are typically heated or chilled with a heat transfer fluid. On heating, pressurized steam may be applied to the vessel jacket to provide even and controlled heating. Or a heat transfer fluid like Dowtherm may be used in a heating or chilling circuit.
Process intensification is about getting the maximum space yield (kg product per liter of reaction volume) and involves several parameters in process design. Concentration, temperature, and pressure are three of the handles the process chemist can pull to increase the reaction velocity generally, but concentration is the important variable in high space yield processes. Increasing reaction temperatures or pressures might increase the number of batches per week, but if more product per batch is desired and reactor choices are limited, then eventually the matter of higher concentration must be addressed.
The principle of the economy of scale says that on scale-up of a process, not all costs scale continuously or at the same rate. That is, if you double the scale, you double the raw material costs but not necessarily the labor costs. While there may be some beneficial economy of scale in the raw materials, most of the economy will be had in the labor component of the process cost. The labor and overhead costs in operating a full reactor are only slightly greater than a quarter full reactor. So, the labor component is diluted over a greater number of kg of product in a full reactor.
The same effect operates in higher space yield processes. The labor cost dilution effect can be considerable. This is especially important for the profitable production of commoditized products where there are many competitors and the customer makes the decision solely on price and delivery. Low margin products where raw material costs are large and relatively fixed and labor is the only cost that can be shaved are good candiates for larger scale and higher space yield.
But the chemist must be wary of certain effects when attempting process intensification. In general, process intensification involves increasing some kind of energy in the vessel. Process intensification through increased concentration will have the effect of increasing the amount of energy evolution per kilogram of reaction mixture.
Energy accumulation in a reactor is one of the most important things to consider when attempting to increase space yield. It is crucial to assure that process changes do not result in the accumulation of hazardous energy.
Energy accumulation in a reactor occurs in several ways. The accumulation of unreacted reagents is a form of stored energy. The danger here is in the potential for a runaway reaction. Accumulated reagents can react to evolve heat leading to an accelerated rates and eventually may open further exothermic pathways of decomposition. As the event ensues, the temperature rises, overwhelming the cooling capacity of the reactor. The reactor pressure rises, accelerating the event further. At some point the rupture disk bursts venting some of the reactor contents. Hopefully the pressure venting will result in cooling of the vessel contents and depressurizing the vessel. But it may not. If the pressure acceleration is greater than the deceleration afforded by the vent system, then the reactor pressure will continue to a pressure spike. This is where the weak components may fail. Hopefully, nobody is standing nearby. Survivors will report a bang followed by a rushing sound followed by a bigger bang and BLEVE-type flare if the system suffers a structural failure.
Energy accumulation can manifest in less obvious ways. Here is an example. Assume a spherical reaction volume. As the radius of the sphere increases, the surface area of the sphere increases as the square of the radius. The volume increases as the cube of the radius. So, on scale-up the volume of reaction mixture (and heat generation potential) will increase faster than the heat transfer surface area. The ratios are different for cylindrical volumes, but the principle is the same. Generally the adjustment of feed rates will take care of this matter in semi-batch reactions. Batch reactions where all of the reagents are added at once are where the unwary and unlucky can get into big trouble.
Process intensification via increased concentration may have deleterious effects on viscosity and mixing. This is especially true if slurries are produced and is even worse if a low boiling solvent is used. Slurries result in poor mixing and poor heat transfer. Low boiling solvents may be prone to cavitation with strong agitation, exacerbating the heat transfer problem. Slurry solids provide nucleation sites for the initiation of cavitation. Cavitation is difficult to detect as well. The instinct to increase agitator speed to “help” the mixing may only make matters worse by increasing the shear and thus the onset of cavitation.
Denser slurries resulting from process intensification are more problematic to transfer and filter as well. Ground gained from higher concentrations may be lost in subsequent materials handling problems. Filtration is where the whole thing can hang up. It is important for the process development chemist to pay attention to materials handling issues before commiting to increased slurry densities. Crow is best eaten while it is still warm.