Thermodynamics and transport properties have always influenced the engineering of chemical products and in the processes that manufacture them. Although use of experimental data in product-process development is desirable, they are not always available. Therefore, property models of various forms, types and scales are commonly employed to supplement the missing data (or knowledge). The most typical use of property models is in the service role, where given the details of the system under study, a specific set of properties are calculated. The use of property model packages through process simulators is an example of this service role. Here, the choice of the appropriate property model is very important because the accuracy of the calculated property values will directly or indirectly influence decisions that will make the product-process feasible or infeasible. A more interesting use of the property models is in the targeted design of chemical products-processes where a set of property values define the product-process design targets and the property models are used in the reverse of the service role. That is, given a set of property (target) values, for which chemicals and/or process conditions, the property targets can be matched? The use of property models in the design/selection of chemical products such as solvents, refrigerants and polymers is an example of the reverse service role. Here, property models need to be truly predictive and at least, also qualitatively correct. From the above examples, it is clear that the same property model will usually not satisfy the needs – because the property models that are quantitatively accurate do not usually have a wide application range. Also, since more than one property model may be available for the calculation of the same property, a set of selection criteria needs to be established for their appropriate selection.

The presentation will highlight the important issues and needs with respect to the use of property models in product-process development, through illustrative examples. It is important to note that the use of any model-based technique depends on the application range and reliability of the available property models. Questions related to predictive versus correlative, speed versus accuracy, simple versus complex, etc., need to be analyzed from the point of view of product-process design and/or evaluation. Ability to provide fast and reliable values for the needed properties without the need for additional experimental data is also an important issue for systematic development of property models for product-process design and/or evaluation. Finally, the gaps in the available property models and difficulties in their implementation/use will be highlighted as opportunities for property model developers.