Process Fingerprints
Explain how your proposed measurement does (or doesn't) satisfy the criteria outlined by Dr. Shih during his presentation at the Polymer Processing Society Meeting. Probably the most feasible candidates for a process fingerprint of the polyurethane reaction injection process are: (1) a measurement of material temperature as a function of time during curing or (2) a measurement of rheology as a function of time during curing. 1. Due to the exothermic nature of the polymerization reaction, the material temperature as a function of time may be used to follow the chemical reaction. This is the basis for the adiabatic reactor method [Camargo et al. Rubb. Chem. Tech., 56, p. 774 (1982)] and differential scanning calorimetry methods [Malkin et al., Polymer, 23, p. 385 (1982)] used for quantifying reaction kinetics. In a RIM process heat transfer to and from the mold is also critical to the temperature evolution and this must also be taken into account [for example, Castro and Macosko, AIChE J., 28, p. 250 (1982)]. A thermocouple or other temperature measuring device could be surface mounted on the mold or within the part (which may not be feasible for obvious reasons.) The temperature history would thus be on-line. It would also be sensitive to many of the process and material input variables such as: the initial monomer temperature, mold temperature, catalyst concentration, stochiometric balance, quality of mixing in the impingement mixer, etc. The results are clearly linked to the fundamentals of the process such the heat transfer and chemical kinetics. Some of the key elements of product quality, particularly those that depend on the extent of reaction and phase separation will be functions of this proposed fingerprint. 2. During the reaction injection molding process the material rheology changes dramatically from a low viscosity liquid to a solid. This provides ample opportunity to follow the reaction using the rheological changes [Richter and Macosko, Polym. Engr. Sci., 20, p. 921 (1980); Osinski, Polym. Engr. Sci., 23, p. 756 (1983); Ng et al., Polym. Engr. Sci., 34, p. 921 (1994)]. The primary challenge with this approach is developing an on-line capability for a rheological measurement. However, dielectric measurements which are correlated to the rheology could be made with surface mounted sensors. As above, this would be sensitive to most of the process and material parameters and related to the process fundamentals. As usual, the weakest point would be the connection between the product properties and this fingerprint. The main practical difficulty here is likely to be the life of the dielectric sensor under the rigorous conditions it would be exposed to. |
