Gordon H. Fricke
 |
|
||||||||||||||
Loren T. Jones Award for Excellence in Teaching, 1977; Lola L. Loshe Award for Outstanding Service, 1978; Purdue School of Science Counseling Award, 1992; Glenn W. Irwin Jr. Experience Excellence Award, 1993; Indiana University Teaching Excellence Recognition Award, 1997, 1998.
Research
We are developing useful, user-friendly, educationally
sound expert systems which teach people how to initiate and solve
complex chemical problems. Expert systems are part of the broad
field of artificial intelligence. They are computer programs designed
to rival a human expert. Data bases of empirical rules, or heuristics,
are being developed to search for best answers to problems which
the user presents to the expert system. Unlike conventional computer
programs, which most often deal with numeric data and algorithms,
expert systems are primarily symbolic and deal with complex, uncertain,
and ambiguous situations. Currently, one expert system helps the
user set-up and solve competing chemical equilibria problems; we
solve for unknown concentrations of many species using symbolic
manipulation. It is anticipated that new ways of solving systems
of nonlinear equations applied to chemical equilibria will be revealed
by this program as the expert system searches for the best route
to the answers. A second program is in the initial stages; it will
be used to help interpret infrared and nuclear magnetic resonance
data. We are building the expertise of many scientists into this
expert system.
We apply mathematics, statistics, and computer science to extract
vital and obscure information and patterns from chemical data. With
the increased use of computer-coupled data acquisition and computer-controlled
chemical instrumentation, comes massive amounts of data and the
ability to guide experimentation in new ways. These methods and
data are being used to help design efficient experiments. For example,
we have used computer aided design to optimize solvent composition
to separate mixtures in thin-layer chromatography (TLC) and high-performance
liquid chromatography (HPLC). We have also applied mathematics and
statistics to derive equations to separate and quantitate sources
of errors in an experimental design so attention can be focused
on problem areas in the design.
Publications
W.C. Stagner, B.J. Cerimele, and G.H. Fricke, "Content Uniformity: Separation and Quantitation of Sources of Dose Variation," Drug Development and Industrial Pharmacy, 17, 233-244 (1991)
G.H. Fricke, P.G. Mischler, F.P. Stafferi, and C. Housmyer, "Sample Weight as a Function of Particle Sizes in Two-Component Mixtures," Analytical Chemistry, 59, 1213-1217 (1987)
R.E. Tecklenburg, G.H. Fricke, and D. Nurok, "Overlapping Resolution Maps as an Aid in Parallel Development Thin-Layer Chromatography," Journal of Chromatography, 290, 75-81 (1984)
G.H. Fricke, "Ion-Selective Electrodes," Analytical Chemistry, 52, 259R-275R (1980)
