Predictive studies, compaction

Despite the diligent use of predictive studies, compaction simulation, pilot scale work, the state of the art of tablet scale-up still provide opportunities for problems to arise. The issues typically encountered are those that have multiple factors involved, and are difficult to predict and simulate at smaller scale. Of these problems, tablet mixture flow, including weight uniformity and segregation, and sticking and picking of the tablet mixture to the tools and tablet specks, are common occurrences. 

To address the issue of the scale-up of the tablet compaction process, this chapter will review the following (1) compaction, (2) predictive studies, (3) scale-up/validation process, (4) case studies, and (5) process analytical technologies. 

Stress-strain type equations have been developed for the compaction process, which help provide an understanding of the mechanisms involved in forming a tablet, as well as allowing for the prediction of compaction results. This predictive power of the compaction process is the basis for many scale-up approaches. However, there are compression and consolidation process aspects which are dependent on manufacturing scale, e.g., speed-sensitive materials, and this results in many problems encountered in transferring a technology to production scale. Unfortunately, these scale-sensitive processes have not been as extensively studied, and are less understood. 

It is particularly significant that no evidence is found for localized melting at particle interfaces in the inorganic materials studied. Apparently, effects commonly observed in dynamic compaction of low shock viscosity metals are not obtained in the less viscous materials of the present study. To successfully predict the occurrence of localized melting, it appears necessary to develop a more realistic physical model of energy localization in shock-compressed powders. 

Myelin in situ has a water content of about 40%. The dry mass of both CNS and PNS myelin is characterized by a high proportion of lipid (70-85%) and, consequently, a low proportion of protein (15-30%). By comparison, most biological membranes have a higher ratio of proteins to lipids. The currently accepted view of membrane structure is that of a lipid bilayer with integral membrane proteins embedded in the bilayer and other extrinsic proteins attached to one surface or the other by weaker linkages. Proteins and lipids are asymmetrically distributed in this bilayer, with only partial asymmetry of the lipids. The proposed molecular architecture of the layered membranes of compact myelin fits such a concept (Fig. 4-11). Models of compact myelin are based on data from electron microscopy, immunostaining, X-ray diffraction, surface probes studies, structural abnormalities in mutant mice, correlations between structure and composition in various species, and predictions of protein structure from sequencing information [4]. 

Resistance functions have been evaluated in numerical compu-tations15831 for low Reynolds number flows past spherical particles, droplets and bubbles in cylindrical tubes. The undisturbed fluid may be at rest or subject to a pressure-driven flow. A spectral boundary element method was employed to calculate the resistance force for torque-free bodies in three cases (a) rigid solids, (b) fluid droplets with viscosity ratio of unity, and (c) bubbles with viscosity ratio of zero. A lubrication theory was developed to predict the limiting resistance of bodies near contact with the cylinder walls. Compact algebraic expressions were derived to accurately represent the numerical data over the entire range of particle positions in a tube for all particle diameters ranging from nearly zero up to almost the tube diameter. The resistance functions formulated are consistent with known analytical results and are presented in a form suitable for further studies of particle migration in cylindrical vessels. 

In the early thirties of the last century Baade and Zwicky conjectured in their studies of supernova explosions that supemovae represent a transition from ordinary stars to compact objects, whose size is an order of magnitude smaller than the size of a white dwarf. At that time it was already known that the atomic nucleus consists of neutrons and it was clear that the density of the remnant objects must be of the same order as the nuclear density. Baade and Zwicky predicted that a supernova explosions will result in objects composed of closely packed neutrons (neutron stars). Prior to the beginning of the second World War (1939) a number of theoretical works by Landau, Oppenheimer, Volkoff and Snider showed, that indeed objects could exist with sizes about 10 km and masses about a solar mass. The density in these objects is about the nuclear saturation density and they basically consist of neutrons with a small amount of protons and electrons. The studies of neutron stars were subsequently stopped most likely due to the engagement of the nuclear scientists in the development of the nuclear bomb both in the West and the East. 

Gereg and Capolla developed process parameters determined by a model laboratory bench scale Carver press, model C (Carver Inc. Savannah, Georgia, U.S.A.), which were translated to production scale compactor parameters (6). Their study provided a method to predict whether a material is suitable for roller compaction. Their study objectives were to characterize properties of the material to identify process parameters suitable to achieve the necessary particle size and density using the dry granulation process and then translate laboratory information to a production scale roller compactor. Actually, information developed from a Carver press was correlated and scaled-up to a production scale Fitzpatrick roller compactor. Model IR 520 (Fitzpatrick Co., Elmhurst, Illinois, U.S.A.) The compactor produced very similar powder granule characteristics as the Carver press. Various lactose materials, available as lactose monohydrate or spray dried lactose monohydrate, were used as the model compounds. Results indicated that a parametric correlation could be made between the laboratory bench Carver press and the production scale compactor, and that many process parameters can be transferred directly. 

For example, in the specificity subsite S3 the phenyl rings of Phe P3 occupy almost identical positions in both renin inhibitor complexes. Modeling studies have predicted the specificity subsite S3 to be larger in renins than in other aspartic proteinases [4] due to substitution of smaller residues, Pro 111, Leul 14, and Alai 15, in place of larger ones in mammalian and fungal proteinases. However, a compensatory movement of a helix (hN2) makes the pocket quite compact and complementary to the aromatic ring as shown in Figure 5. Thus, the positions of an element of secondary structure differ between renin and other aspartic proteinases with a consequent important difference in the specificity pocket. 

Since aromatic amino acids and cysteine are absent, there is no protein absorption above 270 nm. Metallothioneins exhibit a broad absorption peak, with the maximum at 190 mn. Absorptions due to the metal-thiolate complexes show as shoulders at 250 nm (Cd), 220 nm (Zn) and 270 nm (Cu).1458,1459 Theoretical predictions based on the amino acid sequence of the peptide chain indicate that the or-helical conformation is forbidden, and /3-structure is almost impossible to attain. CD and NMR studies on both the metal-containing and metal-free protein confirmed the predictions.1459 1460 However, metallothioneins are stable to tryptic digestion and the slow exchange of many peptide hydrogens of metallothionein with those of the solvent suggest that the protein has a compact and well-defined tertiary structure. 

In science, there is a hierarchy of questions (i) what , (ii) how , and (iii) why . The report of a given fact, e.g., the determination of a series of products and their yields, only answers the question what . Additional kinetic studies raise our level of understanding, as it answers the question how . The ultimate scientific question, why , has as yet rarely been answered, but this level of knowledge is a prerequisite for being able to predict a certain reaction without too many flanking experiments. Thus, it will be one of the main goals of future research to strive for an in-depth theoretical understanding. This, of course, has to be based on our present (and future) experimental data, and it is one of the intentions of this book to provide the necessary information in a compact form. 

Winkelmann et al. (54) have studied air-water flows in a corrugated heat exchanger. Flow visualization and two-phase pressure drop measurements have been performed. The flow visualizations have shown that the flow pattern is complex and that a wavy or a film flow occurs in most cases (Figure 29). The two-phase pressure drop depends on the total flow rate and vapor quality, and Chisholm-type correlation is proposed. More work is required to characterize the flow structure in compact heat exchangers and to develop predictive methods for the frictional pressure drop and the mean void fraction. 

---