Laboratory data are presented on the effect of constanttemperature aging on the apparent thermal conductivity of polyurethane foam insulation for refrigerators and freezers. The foam specimens were blown with HCFC-141b and with three of its potential replacements -HFC-134a, HFC245fa, and cyclopentane. Specimens were aged at constant temperatures of 90°F, 40°F, and -10°F. Thermal conductivity measurements were made on two types of specimens: full-thickness simulated refrigerator panels containing foam enclosed between solid plastic sheets, and thin slices of core foam cut from similar panels. Results are presented for the first two years of a multi-year aging study. Preliminary comparisons of measured data with predictions of a mathematical aging model are presented.

INTRODUCTIONPolyurethane foam insulation denves tts good thermal performance from the low thermal conductivity gases wrth which tt is blown. However, there is a tendency for the thermal conductivrty to increase over time, or age. as air diffuses into the cells of the foam and as the blowing agent diffuses out. In July of 1991. a Cooperative Research and Development Agreement (CRADA) was signed between the Appliance Industry-Government CFC Replacement Consortium (also known as the Appliance Research Consortium, or ARC) and the Oak Ridge National Laboratory. One phase of that CRADA involved the aging charactenstics of polyurethane foam msulation produced with second-generation blowing agents. While that CRADA was concluded m 1996. studies of the aging characteristics of second-generation blowng agents have continued. With support from the U. S. Department of Energy, the U. S. Environmental Protection Agency, and the Appliance Research Consortium. a similar, but more detailed. study has been started on the aging charactenstlcs of polyurethane foam msulatlon blown wrth third-generation blowing agents. Results to date on both of these studies are presented in this paper. SPECIMENSTwo types of specimens have been studied. Specimens of one type were fabricated as panels that simulate the construction of a door or wall of a refrigerator. The panels were about two inches thick and had lateral dimensions of 24 by 24 inches. The faces of the panels were bounded by solid sheets. For the study of second-generation blowing agents, the solid sheets consisted of 24 gauge (0.024 in. thick) steel on one side and 0.12 in. thick acrylonitrile-butadiene-styrene (ABS) plastic on the other side. For the study of third generation blowing agents, 0.040 in. thick plastic sheets were used on both faces, with separate sets of panels being made with ABS and high-impact-polystyrene (HIPS) plastic. The thinner plastic was considered to be more representative of current refrigerator production, and plastic was used on both sides to accelerate the aging experiments by allowing gases to permeate through both sides of the panels. It was felt that the effect of a steel sheet on one face in a real refrigerator could be simulated using models that are being developed. The edges of the panels were sealed with aluminum foil tape to simulate the configuration in a refrigerator where there are no cut, exposed foam edges.The panels for the second-generation study were foamed with CFC-11 (to provide a base case), HCFC-141 b, and a blend of HCFC-142b and HCFC-22. The panels were fabricated by three foam suppliers, with each supplier using a different blowing agent. The panels were fabricated around the end of 1993 and since foams blown with second-generation blowing agents had not been optimized at that time, results with these blowing agents are not necessarily representative of current capabilities.The panels for the third-generation study were foamed with HCFC-141 b (to provide the base case), HFC-lMa, HFC-245fa. and cyclopentane. At the present time, t...