The urban heat island (UHI) effect describes the tendency of urban areas to be significantly hotter than nearby rural areas. One of the leading causes of this effect is due to the thermal properties of concrete. Concrete and asphult, cornerstones of urban design, absorb lots of thermal energy and slowly radiate it into the environment (Zafra, 2025). Not only does this make urban areas hotter and much worse to live in during warm seasons, but it is also highly wasteful energy-wise. Of course, building solar panels into sidewalks and roads would be impractical. Thus, by utilizing the Seebeck effect, a phenomenon in which a temperature difference between semiconductors creates voltage, concrete can be modified into modules for harnessing thermoelectric energy. Thermoelectric devices are suitable for this need, since they emit no carbon dioxide emmissions and have no moving parts (Singh et al., 2021).
Cement-based thermoelectric materials (CTEMs) are a rapidly expanding field of scientific research. Increases in energy demand and the introduction of carbon nuetrality goals have made CTEMs a popular field of research in recent years (Li et al., 2025). This research will also allow remote area to have off-grid clean energy generation with zero moving parts. Additionally, the conversion from thermal energy to electrical will aid in mitigating the UHI effect and reduce heat waste in urban areas.
We will be investigating the effect of different fillers on the thermoelectric properties and efficiency of cement.
Our main goal in this research is to observe which fillers are both cost-effective and efficient. We will try to find which materials most effectively satisfy two conditions: Generating the most electrical energy, and being cost effective. From our research, we expect the fillers to increase the Seebeck Coefficient of the material and generate more thermoelectric power. We predict that graphite will perform best, industrial waste-based fillers the worst, and metal oxides in between the two. However, the better-performing materials will be more costly. We will be measuring two main dependent variables, thermoelectric power factor and cost per Watt. Our hypotheses are as follows: If carbon-based fillers are added into cement , then it will increase the power factor compared to plain cement more significantly than that of metal oxide and industrial waste fillers. Additionally, if industrial waste fillers are added, then the resulting cement sample will have the lowest cost per watt of power produced. These hypotheses support our rationale of making use of wasted thermal energy as well as allowing cheap alternatives for hands-free power generation in remote areas.
The materials needed for our experiment are:
First, the cement samples must first be mixed with a filler and left to cure and dry over time. Since moisture content has a significant effect on the sample’s conductivity (Jani et al., 2022), all samples will be dried for an equal amount of time. Each sample will be a small cube, exceeding no more than a few centimetres in side length. The apparatus in our experiment will be a cement sample heated on one end and cooled on the other. It will be insulated to ensure accurate measurements, and conductive plates placed on the hot and cold faces to act as electrodes. The voltmeter will measure the voltage throughout the temperature gradient, and thermometers placed on both faces to measure the temperature difference. First, a current will be passed through the sample, and along with the resulting voltage, will be used to calculate the internal resistance R (R = V/I) of the sample. The difference in temperature will not exceed 10 K in order to minimize error in voltage measurements (Wei et al., 2018). The heat lamp will be turned on until the temperature difference stabilizes, afterwards, the measurements for temperature (ΔT) and voltage (ΔV) differences will be taken. This process will be repeated multiple times for different values of ΔT for each filler type. Percent weight of filler, sample volume, and curing time will all be kept constant.
From our measurements, many other important figures can be extracted. First, the resistance and the sample’s geometry will be used to first calculate resistivity (p = R * Length / Area) , then the electrical conductivity (sigma = 1/p) of the sample. Power generated can also be found using P = V2/R. For each measurement, ΔV will be plotted against ΔT in a scatter plot. The slope of the line formed by this graph is the Seebeck coefficient S = ΔV/ΔT, measured in μV/K. Using σ and the mean value of S, the thermoelectric power factor PF = S2σ can be found. This is one of the main dependent variables of our experiment. The other is Cost per Watt ($/W). The total cost of the module, plus the total cost of the cement sample and filler, each divided by their respective masses, will be divided by the value of power generated that was derived earlier. Regarding analysis of this data, the values of σ, S, PF, and $/W will all be mapped onto a table for each sample. PF and $/W will be charted onto a bar graph for each filler type. This is to illustrate our predicted trends of these values for filler type, most notably, the inverse relationship between the two values.
In our experiment there remains the possibility of risk. Some potential risks include burns from the heat lamp and exposure to hazardous chemicals during the mixing of the cement. The first of which can be mitigated by turning off the lamp when not in use and surrounding the cement sample in insulation. To prevent exposure to hazardous chemicals, we will wear personal protective equipment (PPE) including gloves, safety goggles, lab coats, close-toed shoes, and respirators during the cement mixing process. In addition to this, we will store chemicals properly based on the chemical’s packaging.
Mohajerani, A., Bakaric, J., & Jeffrey-Bailey, T. (2015, July 15). Science Direct.https://doi.org/10.1016/j.jenvman.2017.03.095
Jani, R., Holmes, N., West, R., Gaughan, K., Liu, X., Qu, M., Orisakwe, E., Stella, L., Kohanoff, J., Yin, H., & Wojciechowski, B. (2022). Characterization and Performance Enhancement of Cement-Based Thermoelectric Materials. Polymers, 14(12), 2311–2311. https://doi.org/10.3390/polym1412231
Li, W., Du, C., Liang, L., & Chen, G. (2025). Cement-Based Thermoelectric Materials, Devices and Applications. Nano-Micro Letters, 18(1). https://doi.org/10.1007/s40820-025-01866-2
Luo, Y., Liu, C., Mazlan, D., & Kumar, S. S. N. (2025). Sustainable thermoelectric energy harvesting in fly ash bamboo fiber reinforced concrete for smart infrastructure. Energy and Buildings, 343, 115927. https://doi.org/10.1016/j.enbuild.2025.115927
Singh, V. P., Kumar, M., Srivastava, R. S., & Vaish, R. (2021). Thermoelectric energy harvesting using cement-based composites: a review. Materials Today Energy, 21, 100714. https://doi.org/10.1016/j.mtener.2021.100714
Wei, J.-J., Zhang, Q., Zhao, L., Hao, L., & Nie Zhengbo. (2017). Effect of moisture on the thermoelectric properties in expanded graphite/carbon fiber cement composites. Ceramics International, 43(14), 10763–10769. https://doi.org/10.1016/j.ceramint.2017.05.088
Wei, T.-R., Guan, M., Yu, J., Zhu, T., Chen, L., & Shi, X. (2018). How to Measure Thermoelectric Properties Reliably. Joule, 2(11), 2183–2188. https://doi.org/10.1016/j.joule.2018.10.020
Zafra, M. (2025, July 1). The Floor Is Lava. Reuters. https://www.reuters.com/graphics/CLIMATE-CHANGE/URBAN-HEAT/byvrejywdpe/
Tong, S., Prior, J., McGregor, G., Shi, X., & Kinney, P. (2021). Urban heat: an increasing threat to global health. BMJ (Clinical research ed.), 375, n2467. https://doi.org/10.1136/bmj.n2467