Batch crystallization processes control

Batch crystallization process control using the first-principles and direct design approaches were discussed. The first-principles approach utilizes crystallization process models, which require the associated determination of crystallization kinetics. The optimal seed characteristics and/or supersaturation profile to obtain the desired product characteristics are then computed. The direct design approach involves feedback control of a state measurement, in this case the solution... 

Although cooling crystallization is the most common method of inducing supersaturation in batch crystallization processes, other methods can be used, as discussed in Chapter 10. For example, evaporation can be used, in which case the profile of the rate of evaporation through the batch can also be optimized7. Indeed, the profiles of both temperature and rate of evaporation can be controlled simultaneously to obtain greater control over the level of supersaturation as the batch proceeds7. However, it should be noted that there is often reluctance to use evaporation in the production of fine, specialty and pharmaceutical products, as evaporation can concentrate any impurities and increase the level of contamination of the final product. 

J.X. Shen, M.S. Chiu, Q.G. Wang, A comparative study of model-based control techniques for batch crystallization process, J. Chem. Eng. Jpn. 32 (4) (1999) 456 164. 

It is important to recognize the similarity of the mathematical form of Eqs. (10.54) and (10.60), and Eqs. (10.55) and (10.61). One may conclude that all batch crystallization processes may be controlled by a properly designed time-profile of the supersaturation-indu-cing quantity (e.g., the cooling or evaporation rate, in the cases discussed in this chapter and the reagent addition rate, in Chapter 6 on precipitation). 

Seeding is commonly used as a control technique, especially in batch crystallization processes. The nucleation rate depends on the total surface area of the crystals. If the total crystal surface is relatively small in relation to the level of supersaturation, small crystals or crystals of undesired shapes may be formed in the early stage of crystallization. Otherwise, if the crystal surface area is sufficient, supersaturation will cause the existing crystals to grow. 

Determination of the optimal temperature (or supersaturation) trajectory for a seeded batch crystallizer is a well studied problem. This is a dynamic optimization or optimal control problem. The process performance is determined by the crystal size distribution and product yield at the final time. For uniformity of shape and size in the crystals in a seeded batch crystallization process, it is essential to ensure that the nucleation phenomena occurs to the minimum and mostly the seeded crystals grow to the desired size at a certain rate. If nucleation occurs in the initial phase, then there is a possibility that the nucleated crystal will compete with the seeded ones, thus if the phenomena is of late growth, then nucleation in the earlier phase is preferred. Thus, depending upon the process operation, many types of objective functions have been proposed [4]. 

For an example of a control chart see Fig. 1.31 and Sections 4.1 and 4.8. Control charts have a grave weakness the number of available data points must be relatively high in order to be able to claim statistical control . As is often the case in this age of increasingly shorter product life cyeles, decisions will have to be made on the basis of a few batch release measurements the link between them and the more numerous in-process controls is not necessarily straight-forward, especially if IPC uses simple tests (e.g. absorption, conductivity) and release tests are complex (e.g. HPLC, crystal size). 

Both experimental and theoretical work has demonstrated that growth rate dispersion exists, and has a measurable effect on the CSD in both batch and continuous crystallization processes. Further understanding of this phenomenon on a fundamental level will be required to develop methods to make use of or control growth rate dispersion and make it a tool in control of particle size and shape. 

The design and operation of industrial crystallizers is where developments in the laboratory are confirmed and their practical significance determined. In recent years, crystallization processes involving specialty chemicals and pharmaceuticals have increased. This has led increased interest in batch crystallization operation, optimization and desigrt At the same time, the advent of powerful computers and their routine avaUabilily has stimulated interest in the area of on-line control of crystallization process (both batch and continuous). Progress in batch crystallization is surrunarized in a number of recent papers and reviews 173-801. In this section I will discuss two areas which I think will have an impact in the next decade. 

Gron, H. Borissova, A. Roberts, K.J. In-process ATR-FTIR spectroscopy for closed-loop supersaturation control of a batch crystallization producing monosodium glutamate crystals of defined size. Ind. Eng. Chem. Res. 2003, 42 (1), 198-206. 

Doki, N. Seki, H. Takano, K. Asatani, J. Yokota, M. Kubota, N. Process control of seeded batch cooling crystallization of the metastable a-form glycine using an in-situ ATR-FTIR spectrometer and an in-situ FBRM particle counter. Crystal Growth Design 2004, 4 (5), 949-953. 

Chang, C. Epstein, M.A. Identification of batch crystallization control strategies using characteristic curves. In AIChE Symposium Series, Nucleation, Growth, and Impurity 47. Effects in Crystallization Process Engineering, AIChE ... 

Nagy, Z.K. Chew, J.W. Fujiwara, M. Braatz, R.D. Advances in the modeling and control of batch crystallizers. Proceedings of the I FAC Symposium on Advanced Control of Chemical Processes, Jan 11-14, 2004 7th International Symposium on Advanced Control of Chemical Processes Hong Kong, 83-90. 

In the pharmaceutical industry most solid compounds are crystallized in batch operations, i.e., the crystalline product is isolated at the end of the operating cycle. Many bulk chemicals, such as table sugar, are prepared through continuous processes, in which the product is collected throughout the crystallization cycle. Batch crystallization produces a narrower range of particle size and may afford better control for the efficient crystallization of molecules from complex mixture [14]. 

In addition, scale-up of a nucleation-dominated process is difficult to predict, unless the generation of supersaturation is well controlled. The difficulties associated with stirred-batch crystallization scale-up relying on nucleation were highlighted by Nyvlt (1971, p. Ill) ... 

In the design of a crystallization process, therefore, the balance that is achieved between nucleation and growth rates is critical to particle size under the operational constraints of equipment and facilities. The supersamration ratio can be controlled to limit nucleation in order for growth to predominate. This becomes increasingly difficult at lower inherent growth rates since it will extend the batch time cycle substantially and because the nucleation rate becomes more critical at lower growth rates. 

The entire cycle for such equipment may be automated however, in most small processing plants, this expense is not justified. There is some variation in the product size made in batch equipment since the quantity of initial nuclei is difficult to control from one batch to the next. With those materials that are very prone to grow on the walls of the crystallization equipment in continuous equipment, a possible solution is the batch crystallizer that is inherently self-cleaning. 

Based on the approaches proposed for batch crystallization— which employed cooling/evaporation rates to control supersaturation, and on the specifics of the batch precipitation process—the reactant addition rate was chosen as the controlling variable. 

Once prenucleation, nucleation, and post-nucleation processes have resulted in a stable nuclei population—or if seeding material is available—the final size of the precipitate from a batch precipitation process can be controlled via manipulation of the reactant feed rate versus time profile. This concept is described in details and illustrated with an example for submicron crystals of AgBr. 

Three additional considerations are required to successfully control a crystallization process. Local conditions rather than bulk, and instantaneous rates of change rather than mean values control the relative rates of nucleation and crystal growth. Further, the response of the system to control changes is history dependent. Spontaneous nucleation that accompanies an excursion beyond the supersaturation metastable limit dramatically affects the surface area available for crystal growth and influences the product CSD of a continuous process for several residence times or the final CSD of a batch process. 

---