Metastable zone measurement

Trends in the crystallization process development in the pharmaceutical industry is to carry out measurements at a small scale in addition to utilizing automation and high throughput systems as exemplified by the use of automated metastable zone measurement for 1 mL samples. It is expected that the future batch crystallization recipes will be designed based on the data collected from much smaller scale crystallizers than what is currently used in industry. 

The rate of primary nucleation and width of the associated metastable zone are difficult to measure with precision in the laboratory, because of their dependence on environmental factors. Dust particles contaminating a solution, and imperfections on the surface of the crystallizer and agitator are often... 

Parsons, A.R., Black, S.N., Colling, R., 2003, Automated Measurement of Metastable Zones for Pharmaceutical Compounds, Trans IChemE, 81, 700-704. 

The measurement of the width of the metastable zone is discussed in Section 15.2.4, and typical data are shown in Table 15.2. Provided the actual solution concentration and the corresponding equilibrium saturation concentration at a given temperature are known, the supersaturation may be calculated from equations 15.1-15.3. Data on the solubility for two- and three-component systems have been presented by Seidell and Linkiv22 , Stephen et alS23, > and Broul et a/. 24. Supersaturation concentrations may be determined by measuring a concentration-dependent property of the system such as density or refractive index, preferably in situ on the plant. On industrial plant, both temperature and feedstock concentration can fluctuate, making the assessment of supersaturation difficult. Under these conditions, the use of a mass balance based on feedstock and exit-liquor concentrations and crystal production rates, averaged over a period of time, is usually an adequate approach. 

Figure 15.10. Simple apparatus for measuring metastable zone widths 36 ...

Nucleation kinetics are experimentally determined from measurements of the nucleation rates, induction times, and metastability zone widths (the supersaturation or undercooling necessary for spontaneous nucleation) as a function of initial supersaturation. The nucleation rate will increase by increasing the supersaturation, while all other variables are constant. However, at constant supersaturation the nucleation rate will increase with increasing solubility. Solubility affects the preexponential factor and the probability of intermolecular collisions. Furthermore, when changes in solvent or solution composition lead to increases in solubility, the interfacial energy decreases as the affinity between crystallizing medium and crystal increases. Consequently, the supersaturation required for spontaneous nucleation decreases with increasing solubility, ° as shown in Fig. 7. 

Accounts of nucleation inhibition in the pharmaceutical literature are sometimes confusing because the dependence of the nucleation event (nucleation rate, metastability zone width, or induction time) on supersaturation is not considered. In search of additives that inhibit nucleation, induction times are often measured as a function of additive concentration, while the dependence of the nucleation event on supersaturation is neglected. Results from such studies possibly lead to the erroneous conclusion that the additive inhibited nucleation when indeed the additive decreased the supersaturation and frequently led to an undersaturated state. Hence, the system is under thermodynamic control instead of kinetic control. 

The use of ATR-FTIR spectroscopy is not a requirement for the determination of the metastable zone. If the goal is the determination of metastable zone or solubility curve alone, then there are less technically complicated methods, such as the gravimetric method for solubility measurement and observation by eye for detection of the metastable limit. Even for the automation of the system, ATR-FTIR spectroscopy is not a requirement. Automated determination... 

Similar to solubility, the metastable zone width and induction time of a supersaturated solution are affected by various factors, including temperature, solvent composition, chemical structure, salt form, impurities in the solution, etc. Therefore, although the spinodal point is a thermodynamic property, it is very difficult to measure the absolute value of the metastable zone width experimentally. Regardless, understanding the qualitative behavior of the metastable zone width and the induction time can be helpful for the design of crystallization processes. 

Figure 8-6 A concentration profile for uncontrolled crystallization by evaporation (initial procedure in Example 8-1. The seed was charged to a highly supersaturated solution which may exceed the metastable zone boundary. The metastable zone width was not measured, so this figure is used only for illustration purposes.

Since the inherent growth rate of many organic compounds is relatively slow, addition times may be long in order to achieve supersaturation control within the metastable zone. Higher addition rates can result in nucleation and the creation of a bimodal distribution. Experimentation to determine acceptable addition rates can be evaluated by focused beam reflectance measurement (FBRM) and other in-sim, online methods (Chapter 2) or microscopic observation of the crystal slurry which could reveal the presence of fines. These issues are highhghted in the examples below. 

Using this formula for a measured solubility at low temperatures immediately gives an idea about potential yield and about the feasibility of a cooling crystallization process. In the metastable zone no nucleation is observed even though the solution is supersaturated. Spontaneous nucleation will only occur when the supersolubility curve is reached. 

Measurement of the metastable zone width and values for the metastable zone width obtained by a variety of methods for inorganic materials can be found in the work of Nyvlt et al. (1985). In general there are two types of methods for the measurement of the metastable limit. In the first method, solutions are cooled to a given temperature rapidly and the time required for crystallization is measured. When this time becomes short then the effective metastable limit has been approached. A second method is to cool a solution at some rate and observe the temperature where the first crystals form. The temperature at which crystals are first observed will vary with the cooling rate used. Measured metastable limits for a number of materials are given in Table 1.10. 

The introduction of seed crystals to a solution that is saturated or within the lower portion of the metastable zone prevents spontaneous nucleation (Karpinski et al. 1980). In most industrial cases, seeding is a manual operation. The indication of when to seed is derived from an indication of the process temperature, typically provided by the control system, and knowledge of the product solubility and current solution concentration, whether measured on-line, off-line, or calculated from charge amounts. Obviously, accurately calibrated temperature sensors in the laboratory, where the solubility relationship was established, as well as in the crystallizer, where the solubility relationship will be utilized, are necessary. 

Experimental techniques for determining the metastable zone width, the amount of undercooling that a solution will tolerate before nucleating, are described in section 5.3. The significance of the metastable zone and the interpretation of metastable zone width measurements are somewhat contentious subjects. Experimental values depend very strongly on the method of detection of the onset of nucleation, but it is still possible to extract kinetic information on the nucleation process as well as on the growth behaviour of very small crystals. These topics are discussed in some detail in section 5.3. 

Janse and de Jong (1978) have warned that attempts to evaluate crystallization kinetics from metastable zone width evidence should be treated with caution, while Kubota, Kawakami and Tadaki (1986) have suggested that the cooling rate dependenee of A max can reasonably be explained by a random nueleation model. Other detailed analyses of metastable zone width measurements and their relationship to nueleation and growth kinetics have been made by Mullin and Jancic (1979) and Sohnel and Mullin (1988b). 

Measurements of the metastable zone width should be made in a manner relevant to the expected working conditions, if possible using the actual liquors to be processed, but most importantly in the presence of crystalline phase (section 5.3). 

The choice of the working level of supersaturation should be based on reliable measurements of the metastable limits of the system which may be made in the laboratory under carefully controlled conditions. Metastable zone widths depend on many factors including the temperature, cooling rate, agitation, presence of impurities, etc., but the most important requirement is that they must be determined in the presence of the crystalline phase and, if possible, with the actual liquor to be processed. 

Smodls, M., Livk, I., Pohar. C. Measurements of solubility, metastable zone width and conductivity of aqueous SPB solutions for the precipitation design purpose. In CMSA 93, F2.46,3196, Praha 1993... 

The metastable control can be achieved if the crystalline compound has a detectable metastable zone width represented by AT et- One of Nyvlt s methods (1985) is used to measure the metastable limit. Saturated solutions of known concentrations containing a few large size seeds are cooled down at a steady cooling rate until the first nuclei appeared. The difference between saturation and nucleation temperatures AT et represents the metastable zone width at a given concentration. Nyvlt s methods are not applicable for slow growth compounds at low temperatures. In this case alternative methods, which increase concentration at a specified temperature instead of reducing temperature at a given concentration, must be employed. Details of these methods will be discussed in another paper due to space limitation. As the metastable limit and the solubility curves respectively serve as upper and lower constraints in a dynamic optimisation problem, they should be estimated beforehand if unavailable in the literature. 

Polythermal methods have in common that a suspension containing known amounts of solvent and solute in excess is heated and the temperature where last particles dissolve is detected. For detection, visual observation (e.g., under a microscope), turbidity measurements, particle-detecting inline probes (e.g., FBRM probe (Lasentec , Mettler Toledo GmbH)), or calorimetry may be used. Since it is a dynamic method, the results depend on dissolution kinetics of the particular system. In general, polythermal measurements are easier to automate since just a temperature has to be followed and no special analytical technique is required. The above-mentioned Crystall6 multiple reactor system can also be used to perform such kind of measurements. To detect both the dissolution process for derivation of saturation temperatures (clear points) and the formation of particles (cloud points) for determination of the metastable zone width, the... 

Any decrease in nucleation rate leads to an increase in the metastable zone width [9] and the induction time, both of which can be measured using suitable techniques, for example, ultrasonic probe [10]. Figure 6.2 illustrates the impact of Fe on the metastable zone width of ammonium sulfate the width increases from 1 K to more than 4 K. 

The supersaturation before the addition of seeds should be adjusted according to the solubility curve and the supersolubility curve (cf. Figure 10.2). Typically, seeding at 4—5 K below saturation temperature is fine. Of course, the metastable zone width has to be considered here and the seeding point should be closer to the solubility curve than to the supersolubility curve. It should be kept in mind that the metastable zone width is not thermodynamically determined, but strongly depends on plant properties and process parameters, such as cooling rate. If the metastable zone width is very narrow, for the sake of process robustness temperature control has to be improved or even an inline measurement of the supersaturation (e.g., by NIR) may have to be used to detect supersaturation close to the solubility curve and to avoid spontaneous nudeation or unwanted dissolving of the seed crystals. In such cases, special care has to be taken that no crystals are present in the crystallizer from the previous batch. 

---