Evaluating the effective thermal conductivity of geothermal pavements constructed using demolition wastes by DEM and 3D printing techniques

This paper studies the effect of various parameters on the effective thermal conductivity (lambda(eff)) of the geothermal pavement system when constructed using construction and demolition (C&D) materials. Recycled asphalt pavement (RAP) and recycled concrete aggregates (RCA) were utilized to experimentally model the geothermal pavement system, consisting of copper tubes to circulate water in the pavement and harness thermal energy. Numerical computational fluid dynamics (CFD) and discrete element method (DEM) methods were used to simulate the geothermal pavement system. Six samples of RAP and RCA aggregates with different particle size distributions were utilized to study the effect of important parameters in the geothermal pavement system. According to the results, coupled CFD-DEM approach could simulate the heat transfer mechanism in geothermal pavement compared to the results of the CFD simulation. The results indicated that increasing porosity would decrease lambda(eff). The decrease rate depended on the gravel content of the sample where samples with a higher gravel content experienced a higher rate of decrease. Also, simulation results showed that increasing the gravel content could increase the value of lambda(eff). Increasing the axial stress resulted in an increase in the value of coordination number (CN) and lambda(eff). Moreover, comparing the thermal power and contact distributions indicated that the morphology of the heat conduction was the same as the contact network distribution. To study the effect of the particle shape on lambda(eff), a novel approach was used that comprised of 3D scanning and printing real particle shape of RCA aggregates by thermochromic material. Simulated assemblies with DEM were verified using 3D printed particles. The effect of aspect ratio (AR) on lambda(eff) was studied which indicated that increasing AR would increase lambda(eff). Decreasing AR to one, on the other hand, increased the time needed to reach the steady state.