Molecular architectures, computer modeling

There is increased emphasis throughout Part B on the representation of transition structures to clarify stereoselectivity, including representation by computational models. The current practice of organic synthesis requires a thorough knowledge of molecular architecture and an understanding of how the components of a structure can be assembled. Structures of enantioselective reagents and catalysts are provided to help students appreciate the three-dimensional aspects of the interactions that occur in reactions. 

Using computer modeling, jointly with x-ray fiber diffraction data, the molecular architectures of two different gel-forming polysaccharides have been examined. Preliminary results indicate that the neutral and doubly branched capsular polysaccharide from Rhizobium trifolii can form a 2-fold single helix of pitch 1.96 nm or a half-staggered, 4-fold doublehelix of pitch 3.92 nm. The molecules are likely to be stabilized by main chain — side chain interactions. 

Client-Server Architecture. A computer architecture in which a "server" computer (usually a larger and faster machine at a central location) runs programs that communicate over a network with numerous workstations or "client" machines that reside in offices and laboratories. The server computer performs heavy duty computing tasks such as database searching and molecular and data modeling, in response to commands from the users of the client comvuters. It then communicates the results back to the client machines. There, depending on whether the client is "thick" (a... 

The earliest models of molecules were constructed of wire, wood, or plastic. The most famous of these, the Corey-Pauling-Koltun (CPK) models, were commercially available in plastic and consisted of a large variety of atom types representative of average atoms with, say, sp, sp, and sp hybridization. The models were helpful because they enabled one to see the three-dimensional relations between atoms more accurately than could be portrayed on the printed page. The extent of the utility of the models was just that they helped to convey molecular architecture. With the advent of high speed graphical computing, physical models have been replaced with computerized pictures that enable one to visualize and move molecules in many different ways. 

There are a few areas where more research would be immediately helpful. The first is the large scale off-lattice computer simulation of realistic polymer blends. This will help test the approximations in the liquid state theories. The second is the development of equations of state for hard chain mixtures where the components have different molecular architectures. This will be useful because integral equations are not expected to be quantitatively accurate for the hard chain pressure, and attractions are easily incorporated via a perturbation theory. Finally, accurate intermolecular potential models for more polymers will, of course, be invaluable. 

---