Molecular Interactions Probed by MD Simulation

Typically such amorphous dispersions are prepared by melt extrasion (i.e., cooling from a melt) [79], spray drying [80], or lyophilization [81]. Although the reliable 

FIGURE 13.10 Probabilities for various types of hydrogen bonds (HBs) between an IMC-COOH donor and varions fnnctional gronps in simnlated IMC (white) and IMC-PVP (58% 41%, w/w) (black) glasses at 298 K. A simnlated stmctnre of the IMC-PVP HB is also shown. Source Xiang and Anderson [32]. Reprodnced with permission of John WUey Sons. 

The total solubility parameter is often expressed in terms of separate contributions including nonpolar dispersive interactions (5 ) and electrostatic components 5) including both polar and hydrogen bonding interactions as developed by Hansen [88]  

Regular solution theory assumes that specific interactions such as hydrogen bonding are absent, and therefore Hildebrand solubility parameters are generally applicable only to systems containing relatively nonpolar constiments. It is important to remember that neither of these approaches are theoretically justified for mixtures in which specific solvation interactions are important. 

Gupta et al. [84a] generated solubility parameters for indomethacin (IMC), polyethylene oxide (PEO), sucrose (SUC), and glucose (GLU) using MD simulations with the goal of predicting whether or not blends of indomethacin with each of these excipients would be miscible or immiscible. The solubility parameter values obtained (in MPa° 0 were 23.9 0.3,22.2 0.2,29.9 0.5, and 34.8 0.2 for IMC, PEO, SUC, and GLU, respectively. The solubilities were then predicted from the differences in solubility parameters  

---