2Department of Energy Systems Engineering, Gazi University, Ankara, TURKEY
3Department of Civil Engineering, Gazi University, Ankara, TURKEY
4Natural and Applied Science Institute, Gazi University, Ankara, TURKEY
5Department of Mechanical and Manufacturing Engineering, Düzce University, Düzce, TURKEY
6Department Civil Engineering, Antalya Bilim University, Istanbul, TURKEY
Abstract
Thermal resistance can be increased by using proper heat insulation materials. Traditional heat insulation materials do not stand all desired properties. Thus, developing new heat insulation materials is very important. In this study, expanded perlite based heat insulation material was developed as an alternative to the traditional insulation materials. The composition of the developed material was designed and prepared using the theoretical thermal conductivity prediction models. The prepared material was molded in a rectangular shape panel. Thermal conductivities of panels were measured experimentally and the results were compared with the calculated results. Also, the results showed that the developed panels can be used for heat insulation applications. On the other hand, the closest model to the experimental results is the parallel model whose average deviation is 4.22% while the farthest model is the Cheng and Vachon model whose average deviation is 12.43%. It is obtained that parallel and series models are generally in good agreement with the experimental results. Nevertheless, it is seen some deviations between experimental and theoretical calculation results. The theoretical prediction models do not include any processing conditions such as molding and curing. It is thought that these deviations have originated because of the missing processing parameters in theoretical prediction models. As a result of experimental studies, the lowest thermal conductivity value of expanded perlite based panels was obtained 43.5 mW/m.K. Consequently, the heat transfer coefficient of the panels containing expanded perlite can be calculated nearly by the parallel method.