Expert systems catalyst design

The system ESKA (Expert System for Selection and Optimization of Catalysts [20]) was designed at BASF specifically for hydrogenation reactions. The main component of a catalyst is proposed on the basis of activity patterns which describe the applicability of catalysts for different types of hydrogenations. The system also is able to propose secondary catalyst components and, if necessary, a support material which is stable under reaction conditions and does not have any undesired catalytic properties. Based on heuristics for required as well as undesired side reactions and for different catalytically active components, the system also proposes reaction conditions as temperature, pressure, the solvent or the pH. 

A Data Procurement for Knowledge-based Systems Progress in analytical characterization of catalysts plays an important role in their further development and improvement. Synergistic effects of complimentary characterization tools by which different properties of the catalytic materials arc determined are claimed to be beneficial in catalyst design. If this is so, then an expert system for assisting in catalyst selection should be designed in such a way that it accounts for different chemical and physico-chemical properties and their relation to catalytic performance of solid materials. 

The more surface- and bulk-related properties of solid materials and their relevance to catalysis are incorporated into an expert system for assisting in catalyst design, the more effective such a system will be. Results and experience may be preserved in the knowledge base. Since such information may be applicable to other reaction types, if some reaction steps or relevant material properties arc common, knowledge bases describing particular fields of interest may even be licensed to third parties, and so become tradable items. 

In proportion to the importance of the catalyst materials, experimental information on their synthesis, characterization and catalytic properties are accumulating over the years. Databases of such information could be generated and further expert system approach can be applied to optimally utilize this information for the catalyst design. Relations between the observations and inferences are expressed as mathematical expressions. These independent and interdependent mathematical expressions which relate observations to the inferences are used in the decision-making steps for predictions. Finally the recent achievements in the human interfacing technology with computers - for e g., one can talk to the molecules simulated in computer, command their motions just by waving his hands, etc. have to be seen to be believed. 

Computerizing the information on zeolite catalysts have been attempted successfully and different databases concentrate on specific properties[25-28]. The information in several databases are shown in Table 3. Our approach[29] involves retrieval of information from the database and additionally an expert system approach is followed to derive a set of conditions to achieve one s goal in the synthesis of zeolites. The structure of the system is designed to perform three salient fiinctions as shown in Fig. 7. The first function is to provide access to a large database of physico-chemical properties and crystallographic information of all reported zeolite[30]. The second function provides for the synthesis of zeolites - the most logical route for the synthesis of a desired zeolite structure is provided. The third function is a graphic tool application to simulate X-ray powder diffraction patterns for zeolite phases with different amount and nature of purity. 

There are also reports in the literature where a database on Ci catalytic chemistry is developed by Ito et al[31], computer aided design of oxidation catalysis using expert system approach[32, 33] and an expert system approach for the general design of catalysts[34], Thus processing the information by computational methods proves to be a efficient approach. 

The use of computer databases and computer-controlled plants facilitates daily laboratory work. The recently set up expert systems for catalysts [79] help to reduce the number of experiments by rapid preselection, and accelerate catalyst development. In the structure determination of zeolites or non-zeolite molecular sieves (APOs, SAPOs), computers provide a valuable and fast service in that zeolite models are designed graphically on the screen, and the corresponding X-ray diffraction... 

Artificial Neural Network (ANN) is another assistant method for the catalyst design, which is used more often presently. Compared with the expert system, ANN is under the case of a rather obscure catal3ftic mechanism based on the experimental data, and is able to establish the reflecting relationship between the catalyst compositions, preparing conditions and catalytic properties (including selectivity and conversion) and then obtain such parameters as the proportions through with the optimizations. This kind of method is simple and universally applicable, and is especially suitable for the design of the multi-component catalysts. 

The performance of a catalyst is determined by its lifetime, activity and selectivity in converting raw materials into end products. Therefore, the design objectives are to satisfy the requested parameters for these properties. From the computational chemistry point of view, there are several different approaches in the research of catalyst design applic ion of databases and expert systems, chemistry of the catalytic process, in the nonmolecular field, and the many aspects of molecular science. 

The objectives of catalyst design are to improve the catalytic performance, defined by lifetime, selectivity, and activity, Utilization of data bases and expert systems will continue to be the first step and will develop towards the application of new information sciences. Chemistry of the catalytic process is playing an important practical role in industry. The integration with molecular and material science is a natural extension. The molecular science approach is now able to provide insight into structures and mechanisms at the atomic level. 

---