Thermal conduction behavior and prediction model of scrap tire rubber-sand mixtures

Abstract

Increasing stockpile of scrap rubber tires imposes a serious threat to the safety of surrounding ecological environment. Rubber particle of excellent thermal insulating properties with comparison of granular soils makes it an ideal candidate for developing sustainable construction materials. While thermal conduction behaviors of rubber-sand mixtures have not been clearly revealed. Several series of thermal probe tests were conducted on scrap rubber tire-sand mixtures with varied rubber contents, moisture contents, and dry densities. A predictive model was proposed by resorting to the artificial neural network technology to capture the thermal conductivity data. The results showed that an obvious decrease in thermal conductivity is taken after the addition of rubber, and the decreasing rate is related to moisture content of mixtures. The presence of pore water is beneficial to the improvement of thermal conductivity. The critical moisture content of investigated rubber-sand mixtures is approximately 8 %, further increase in moisture content leads to a faint increment of thermal conductivity. Rubber-sand mixtures of high dry density have good particle contact behaviors, exhibiting a high thermal conductivity value. The influences of rubber content, moisture content, and dry density on thermal conductivity are intertwined together that should be comprehensively analyzed. The developed model exhibits satisfactory accuracy for thermal conductivity prediction, as reflected by correlation coefficient higher than 85 % and relative error being controlled under 10 %. Additional study is recommended to evaluate the thermo-mechanical behaviors and durability of rubber-soil mixtures. The outcomes of this study provide one of available answers for reusing scrap rubber tire in engineering fields.