PSI - Issue 67

Jorge S. Dolado et al. / Procedia Structural Integrity 67 (2025) 23–29 Jorge S. Dolado/ Structural Integrity Procedia 00 (2024) 000–000

26 4

4. Thermoelectrical concrete Thermoelectric (TE) technology has gained attention as a promising solution to this issue, enabling the direct conversion of heat into electricity. This process relies on the Seebeck effect, which states that a voltage is generated when there is a temperature difference between the two ends of a TE material. The resulting Seebeck voltage is expressed as ΔV = ΔT * S, where ΔV represents the voltage generated in response to the temperature difference (ΔT) between the hot and cold sides, and S is the Seebeck coefficient of the TE material. The thermoelectric (TE) properties of plain ordinary Portland cement (OPC) are relatively limited, with a Seebeck coefficient of around 2 μV/K (Wen and Chung (1999)), and thus require enhancement. The main approach to improving the TE performance of OPC cement-based materials is to incorporate functional additives. These additives can generally be divided into six categories: carbon fibers, graphene/graphite, carbon nanotubes (CNT), steel fibers, metal oxides, doped metal oxides, and graphene, see for instance Liu et al (2021). A recent paper by Barzegar el al. (2024) has shown that the structures of geopolymer or alkali-activated cements are much more appropriate for this application, because the zeolitic nano-cavities enhance the ionic mobility in the porous water with respect to the ultra confined of OPCS (see for instance Monasterio et al. (2013)). Though high Seebeck coefficient values can be achieved (order of mV/K ), just with the use of Copper Slag aggregates (Barzegar et al. (2024), current challenge is the simultaneous improvement of the electrical conductivity. Indeed, the so-called Figure of Merit (ZT) is a dimensionless parameter that ranks the thermoelectric capacity. It reads as ZT= (S 2 s /k) T, where S is the Seebeck coefficient, s the electrical conductivity and k the thermal conductivity. Excellent thermoelectric materials have ZT of the order of unity, while the concrete proposed by Barzegar et al. (2024) peaked at about 10 -7 at 80º C, in spite of its high Seebeck coefficient. This extremely low performance can be attributed to the a priori intrinsic low electrical conductivity (about 10 -4 S/m) and is the reason why several groups try to increase the electrical conductivity without reducing the Seebeck coefficient too much. The best attempt so far is the one of Ghosh et al. (2020), who reported a cement composite with a remarkably high figure of merit of ~10 -2 thanks to an extraordinary electrical conductivity of 1390 S/m and a good Seebeck coefficient of 142 μV/K. The followed strategy combined additions that improve the electrical conductivity (wt 10% graphene) and additions that favor high Seebeck coefficients (wt 10% ZnO 2 ). While this strategy seems promising, recent atomistic simulations (Agbaoye et al. (2024)) have explored the intrinsic thermoelectric properties of Portlandite and Tobermorite, the most important mineral phases appearing upon the hydration of OPCs. Interestingly, the electronic conductivities, Seebeck coefficients and the resulting Figure of merits can be tuned by changing the electron or hole carrier concentrations, something attenable by appropriate doping. In particular, the impact of hole concentration on the mentioned thermoelectric parameters (ZT, S and s ) of Tobermorite is displayed in Figure 2, where the cases of Tobermorite doped with one or two Al atoms are shown by dashed brown and violet lines respectively. As can be seen, values for the Figure of Merits around 0.25-0.5 are predicted within the studied temperature range. These values clearly indicate that cementitious materials can be used as thermoelectric materials if appropriately doped. In this sense, it is worth noting that the Al by Si substitutions are typical in the cement chemistry and provide a solid basis for future improvements.