Crystal global minimization

Ultimately the aim of molecular modeling is the design (before synthesis) of novel materials. In molecular crystal chemistiy, this requires the ab initio prediction of solid-state structures. Although much progress has been made and the potential of such an ability is enormous, progress in this field has been hindered by problems with global minimization and force-field accuracy. 

The advantage of the crystal structure solution from powder diffraction data over the ab initio crystal structure prediction method is that in the former approach the accuracy of prediction can be validated by matching the simulated powder pattern with the experimental powder diffraction pattern. In retrospect, concerns over global minimization algorithms, force field, and charge distribution accuracies are reduced with this method. 

Crystal structure calculations have been successfully employed to obtain the structures of DPPI [46], BTI [47] (Figure 8.8), the a-form of quinacridone [11, 48], and the a-form of copper phthalocyanine [49] and a dioxazine pigment (see Chapter 20). Because of low crystallinity and excessive peak overlap, the powder patterns of these compounds could not be indexed. In all cases global minimization techniques yielded structure models, which subsequently could be refined by Riet-veld methods to solve the structures. 

Molecular modeling using either Monte-Carlo simulations or molecular dynamics is used to apply molecular mechanics energy minimizations to very complex systems [348]. In complex flexible molecules such as proteins or nucleic adds, the number of variable parameters, i.e., bond torsion angles, is such that the global search for energy minima becomes impossible The same problem occurs with theoretical calculations of water structure in aqueous solutions or in heavily hydrated crystals. 

Control of both local (point of addition) and global supersaturation is essential, as in all crystallizations, if a satisfactory balance between nucleation and growth is to be achieved. This is particularly relevant to reactive crystallization because of the creation of local high supersaturation of these low-solubility compounds that is unavoidable at the point of reaction. In addition to the mixing issues outlined above, the critical variables in minimizing supersaturation are as follows ... 

Fig. 3 The success rate of searches for a given crystal structure by lattice energy minimization, as derived from a survey of published studies.The results are distinguished as to whether the known stracture was found as the global or local minimum or not found. The number of searches significantly exceeds the 189 molecules categorized in Fig. 2. because of searches for polymorphic systems (where one structure is necessarily a local minimum) and where different groups have published a search for the same molecule.

FIGURE 5.1 Schematic representations of the steps of crystal strucmre prediction (CSP) by global lattice energy minimization. 

---