Phase change materials in passive solutions for maintaining thermal comfort in buildings

This study uses the ANSYS Workbench Transient Thermal module to analyze phase change heat storage and release in a PCM integrated with underfloor heating under transient room temperature fluctuations. The ANSYS Transient Thermal module represents an advanced numerical tool for simulating transient thermal processes in engineering materials and structures. Employing the finite element method (FEM), it enables the modeling of dynamic heat transfer phenomena; including convection, conduction, and radiation, while accounting for nonlinear material dependencies and time-varying boundary conditions. The Transient Thermal module in Ansys enables the modeling of variable heat conduction temperatures, taking into account the phase change material (PCM) effects as well as the heat transfer of the underfloor heating system.

Phase transition in PCMs occurs at a particular temperature, resulting in the absorption or release of latent heat. This phenomenon can be explained by the fact that, during the process of heating, the material absorbs heat more slowly due to the energy supplied being partly used for phase transition. Conversely, during cooling, the material releases heat more gradually because latent energy is released. The value of latent heat, in conjunction with the melting range, is indicative of a particular phase change material. In this study, Rubitherm RT21 was utilised as the phase change material in the analyses.

The adopted equation for thermal conductivity in solids was as follows: (1) ρCp(T)∂T∂t=∇(k(T)∇T)+Q \rho {C_p}(T){{\partial T} \over {\partial t}} = \nabla (k(T)\nabla T) + Q where k(T) is thermal conductivity depending on temperature, t is time, T is temperature, Q is the rate of heat flux inside the volume, ρ is the density of material and C_p(T) is specific heat of the material depending on its temperature.

The material properties of Rubitherm RT21 were adopted on the basis of manufacturer specifications (Rubitherm GmbH, 2024). According to this document, the latent heat of the analysed PCM material is L = 165 kJ/kg, while melting occurs in the temperature range of 19 °C – 24 °C, with the main peak occurring at 21.5 °C. Outside the phase transition range, the specific heat is 2 kJ/(kg·K). The thermal conductivity value was assumed to be 0.2 W/(m·K). The phase change effect is simulated by artificially increasing the heat capacity over the melting/freezing temperature range. When the material absorbs heat (heating), the temperature rises more slowly because part of the energy is consumed by the phase transition. When the material releases heat (cooling), the temperature decreases more slowly because latent heat is released. During the definition of material parameters, a temperature-dependent specific heat capacity (C_p) was introduced. Within the phase change temperature range, a large increase in C_p is applied.

The dependence of the material's specific heat on temperature (Fig. 1) graph is triangular in shape, a choice that is motivated by the desire to ensure stability and numerical simplicity. The area of the triangular part above the specific heat value of 2 kJ/(kg·K) is equal to the latent heat of the PCM material. In the preparation of this graph, it was hypothesised that the peak specific heat value would be attained at a temperature of 21.5 °C. The value was calculated based on a triangular area formula, thus height of a triangle equals 66 kJ/(kg·K). C_p,base, is the specific heat outside the phase transition range, L is latent heat, and Δt is the difference in temperatures at which phase transition occurs. Utilising this data, a maximum value of C_p,max = 68 kJ/(kg·K). The values for temperatures of 19 °C, 20 °C, 21 °C, 22 °C, 23 °C and 24 °C were obtained by means of linear interpolation.

Fig. 1.

Chart of specific heat constant pressure to temperature. View from engineering data – Ansys Workbench (own research)

In order to verify the above assumptions regarding PCM material modeling, a simple numerical simulation was performed. A cuboid solid with dimensions of 5.0 m × 1.0 m × 0.04 m was modelled. The Quad Dominant meshing method was adopted, with a basic element size of 3 cm. The finite elements were implemented along the shortest edges. The total number of elements created was 17,640 with 90,440 nodes. The side walls of the solid were modeled as perfectly insulated. The initial temperature of the model was set at 14 °C. On the bottom plane of the model (direction of the normal axis along the Y-axis of the global coordinate system), a time-varying temperature was set.

Fig. 2.

Temperature on the top surface of the model (own research)

The change in temperature (Fig. 2) on the upper surface was analysed. The temperature graph demonstrates that upon reaching the temperature at which phase transition begins on the lower surface of the model, the temperature on the upper surface increased more slowly than when the model temperature was lower. The first peak of maximum temperature on the upper surface was attained after 133,200 seconds, whereas the maximum temperature on the lower surface was achieved at 79,200 seconds, i.e. approximately 15 hours earlier. Despite the decline in temperature on the lower surface, an increase in temperature could still be observed on the upper surface. Furthermore, it is evident that the precipitous decline in temperature during the cooling process occurred exclusively when the surface temperature was outside the phase transition range. This behaviour is consistent with that of PCM materials, thus indicating that the material has been modeled correctly.

Initial temperature was assigned individually using APDL Commands in Ansys Mechanical. Boundary conditions applied to the models: temperature 0 °C on the outer walls surfaces, temperature 9 °C on the bottom surface of the sand bedding, temperature of 18 °C and 35 °C applied to the layer representing the floor heating system (a variable used to determine the optimal heating schedule). For the Heat Flow – all outer surfaces were perfectly insulated, excluding outer walls and bottom faces of the numerical model. Total analysis time was set to 96 hours with a time step every 3600 seconds. This duration was selected to capture multiple diurnal heating cycles and allow for equilibration of transient effects.

The temperature time histories obtained from transient simulations of underfloor heating were analyzed for 27 different heating schedules. Each simulation covered a period of 96 hours, with hourly recording of the temperature measured at a height of 1 m above the floor surface. For each case, both the REF and PCM models were considered (Fig. 3). To facilitate the presentation of temperature variations, the results were plotted in the form of the function: L = T_av − 15 °C, where T_av is the average temperature measured over one hour at the measurement point (e.g., 1 m above the floor surface), and 15 °C was adopted as the reference temperature. Figure 4 presents the time evolution of the function L for a representative heating schedule. Each set of 27 variations was compared with the function L35(24) (continuous heating at 35 °C-maximum baseline condition). In the REF model, temperature rises and drops over time are observed. In contrast, in the PCM model, the high maximum and low minimum temperature values are noticeably flattened. Owing to the presence of the phase change material, the model responds more slowly to rapid changes in the imposed heating power. Similar effects are observed in all comparisons of the L function for the REF and PCM models.

Fig. 3.

a) Model 1 (REF), b) Model 2 (PCM) (own research)

The analysis allowed the formulation of the following characteristics of the results: Clear connection between changes in heating schedule and PCM material work was observed, the PCM material provided pronounced time delays in temperature response on the altitude of 1 meter. The integration of the PCM material with under-floor heating systems and different heating schedules can provide meaningful improvements in maintaining thermal comfort without the use of additional energy. With heating schedules correctly matched, there can be better thermal management.

Fig. 4.

Example heating schedule used on REF and PCM model compared to baseline condition L(35) (own research)

The difference in the average temperature (Fig. 5) between the models was calculated by formula: ΔT_av (t) = T_av,PCM − T_av,REF and presented as a function of time (Fig. 4). Heating schedules characterized by high amplitude indicate a high level of phase change material activity. This results from the evident absorption and release of heat stored in the material, which is reflected by pronounced peaks and troughs in the plots. Figure 4 also shows results for poorly selected heating schedules. These curves are flattened, and no clear PCM activity correlated with the heating periods can be observed. In such cases, the heating system operates outside the optimal range for the PCM cycle, resulting in limited thermal storage benefit and wasted potential energy gain.

Fig. 5.

The difference in average temperatures between the PCM and REF models as a function of time (own research)

The findings in this study have major implications for contemporary design: Passive – Active Integration: Hybrid approaches between PCM and active under-floor heating systems have proved to be superior to either technology on its own. Ground – Contact Floors: Represents an often overlooked opportunity for improving thermal performance. Typical outdoor temperature amplitudes are too limited to fully activate PCM layers. By combining PCM with an underfloor heating system designed to generate controlled temperature oscillations, a stable and repeatable operating environment for the PCM can be ensured.

Several limitations have been noted during the study: PCM material selection: The study was run on a specific compound. Other PCMs, with different melting temperatures, may give different outcomes. Numerical simulation simplification: The simulation of phase change was run on a basis of a simplified model. Using more sophisticated methods (enthalpy) may improve the accuracy of the study. Climate scope: Simulations represents limited outdoor scenarios. Application in variable outdoor temperatures needs to be studied. Technical limitations: Long term material degradation, form of encapsulation of PCM material and limitations in connecting underfloor heating with PCM layer were not considered.

Considering future research: Other expected day – and night – time temperature profiles should be analysed in rooms such as bedrooms and offices, taking into account typical occupancy patterns and comfort requirements. The performance of alternative underfloor heating systems should be examined, particularly under different outdoor temperature conditions. The use of AI-based methods for enhancing and optimising heating schedules should be investigated, for example through data-driven control and predictive algorithms.

Phase change materials, when properly integrated with underfloor heating systems in ground – contact floors, provide measurable improvements in thermal stability. The numerical simulations in ANSYS Transient Thermal validated the chosen PCM modeling approach and analysed benefits of the study.

Fig. 6.

Charts presenting the functions R (the ratio of the REF and PCM temperature difference to the REF difference) and K (the difference between REF and PCM maximum temperatures), in time, for example schedules (own research)

Key findings include: Temperature stabilization: PCM integration reduces indoor temperature fluctuations by 10 % – 25 % (Fig. 6 top chart), with peak reductions up to 2.5 °C (Fig. 6 lower chart) when heating schedules are optimally designed. Heating schedules: PCM effectiveness depends on coordinated heating system scheduling. Material properties validation: The triangular specific heat capacity model captures PCM thermal behavior during phase transitions, providing a computationally simplified yet efficient approach for transient thermal analysis. Climate adaptation: The approach is well-suited for climates that are heating dominated; ground temperatures are significantly cooler therefore maximizing PCM thermal cycling potential.

Integration of PCM with underfloor heating presents a realistic strategy for enhancing energy efficiency and thermal comfort without significant additional complexity, particularly in renovation projects where existing underfloor heating systems can be integrated with thin PCM layers.