(697b) The Virtual Product-Process Design Laboratory
AIChE Annual Meeting
2008
2008 Annual Meeting
Computing and Systems Technology Division
Integrated Product and Process Design
Thursday, November 20, 2008 - 3:40pm to 4:05pm
In design of chemical products and the processes that can make them, one first tries to find a candidate product that exhibits certain desirable or targeted behavior and then tries to find a process that can manufacture it with the specified qualities. The candidate may be a single chemical, a mixture, or a formulation of active ingredients and additives. For the later product type, additives are usually added to an identified active ingredient (molecule or mixture) to significantly enhance its desirable (target) properties. The common practice is experiment-based trial and error approach, supplemented sometimes with model-based computer-aided tools to speed-up some of the steps.
The objective of this paper is to present a ?virtual? laboratory for chemical product-process design where the developer can test their ideas on model-based computer-aided tools rather than perform experiments. As the name suggests, the virtual lab contains methods and tools to allow the modeling and simulation of the needed experimental scenarios. For this to work the models need to be reliable and efficient and the architecture of the software needs to include the work-flows related to many different product-process design. Also, interfaces for efficient data-flow between different tools used in the work-flow needs to be defined. The architecture needs to be flexible to allow changes in work-flow, additions of new models and the addition of new data for future extension of the application range.
A prototype virtual PPD-lab has been developed using EXCEL as the simulation environment with proven and tested computer aided methods and tools developed by the authors. It includes databases of properties of chemicals, chemical product characteristics, of property models, and many more. It includes tools for molecular and mixture design to generate product alternatives. It includes tools for modeling to generate models for evaluation of process-product performance. It includes recipes for design of formulated products and mixture design. In principle, it can perform many of the virtual experiments that the product-process developer would need to perform, on a trial and error basis. Alternatively, the developer can ask for a generation of the short list of the best (or feasible) alternatives. The objective here is to quickly and efficiently generate a list of promising candidates for final testing (and selection) by experiments. In this way, rather than use the experiment-based trial and error approach from the start, the experimental resources are reserved for the final selection and verification step while the virtual PPD-lab is used for identifying the promising feasible candidates, thereby achieving faster to market at reduced cost for any potential chemical product-process.
The presentation will highlight the main features of the virtual PPD-lab, providing details of the available chemical product design recipes and the resident models, methods and tools that have been included. The applicability of the virtual PPD-lab will be highlighted through several chemical product-process design case studies from the pharmaceutical, agrochemical and petrochemical industries.