Three-Dimensionally Printed Gypsum Located Within Micro-Encapsulated Phase Change Material: Thermal Conductivity Benefits of Selective Activation Technique

Abstract

The widespread occurrence of encapsulated phase change materials (PCMs) within a mineral matrix has been demonstrated to improve the thermo-physical properties of final products. The upscaling of such materials has not yet been achieved, as traditional onsite mixing and casting processes could damage the capsule, leading to a leakage of the active content and then a deterioration of the final element. The aim of this paper is to evaluate the influence of selectively depositing a layer of PCM on plaster, through a powder bed 3D printing process, on its density and thermal conductivity. A home-made selective-binding 3D printer has been used to assess samples of composites of calcium sulfate and encapsulated PCM. Thermal conductivity and Scanning Electron Microscope measurements were carried out on pure calcium sulfate as well as on a mix design containing a 5% mass ratio of PCM. The SEM measurements highlight that the PCM shells are undamaged by the selective-binding 3D printing process compared to the traditional mixing and casting process. Also, the 3D-printed composite material demonstrates a thermal conductivity reduction of 39%, which is linked to the 17% decrease in density. This applicative study validates the idea of designing functionally composite construction materials with phase change materials inserted as a thin layer between printed plaster layers and also demonstrates the great potential of this innovative selective-binding 3D printing technique.

1. Introduction

The built environment is responsible for more than a third of global energy-related carbon emissions, while the construction materials (concrete, aluminum and steel) themselves represent 6% [1]. In Europe, the building sector represents 40% of these emissions, and this energy is mostly generated by fossil fuels (80%). Operational energy demand is still high, while demand for heating is declining due to energy efficiency improvements. Meanwhile, the cooling demand continues to grow, as recently reported by the United Nations [2]. The impact of material choices on both embodied carbon and cooling requirements is a key ingredient to reducing our global carbon footprint.

Thus, innovative and immediate research studies are needed to decrease our energy consumption. Building fabrics need insulation and glazing upgrades alongside heating and cooling system improvements, and the adoption of passive building solutions could reduce service demands. As reported by [3], incorporating a phase change material (PCM) helps to optimize the energy requirement for space cooling. This kind of “intelligent material” can store heat and release it in buildings when needed. The phase change materials market is going to increase from USD 423 million in 2020 to USD 890 million in 2025, as reported by [4,5].

The development of innovative insulating materials is already largely underway, but the integration of phase change materials (PCMs) into building structures is still a niche technique even though it has produced the desired results. These materials store thermal energy in a latent form, while conventional building materials store thermal energy in a sensible form. Combining PCMs with a mineral-based component used in buildings, such as plaster or cement, helps to store thermal energy through latent heat to improve their thermal performance and to reduce interior temperatures [6,7,8,9].

Several experimental and numerical studies devoted to the direct incorporation of PCMs into building materials [6,10,11] have emerged over the last decade. PCMs can be integrated into civil engineering applications by adding them into building components such as wall panels [10,11,12], concrete and mortar [6,7,9]. The most commonly used phase change materials are based on the solid–liquid transition phenomenon, due to their availability on the thermal storage market [9,10,11,12,13,14].

However, the leakage of these materials presents a real physical constraint that limits their application in the building field [15]. The process of incorporating a PCM in a construction material must ensure no leakage during phase its transition along with a maximum utilization of latent heat storage, as reported by [16]. Several studies have pointed out that this important leakage could be avoided by using micro-encapsulation before the incorporation of the PCMs [3]. Encapsulating the PCM within a suitable coating or shell material has several benefits; see [17]. Nevertheless, the coating or the shell of these materials can be damaged during the mixing process by the mineral binder and by millimetric or centimetric aggregates of construction materials.

With the emergence of additive manufacturing as an option for shaping concrete, the use of system formwork and traditional onsite mixing procedures can be substituted [18,19,20,21,22]. The digital fabrication of construction materials implies that we will be able to aim for freeform architecture and precision material placement in the near future. While traditional construction techniques using formwork could lead to a homogenization of the PCM material with the cementitious matrix, the paradigm of targeting a location and accurately placing the PCM inside the final element is promising.

On the one hand, the commonly used extrusion technique involves a traditional pumping and mixing process that could damage the encapsulation of the PCM. On the other hand, the selective binding of a loose particle bed by a fluid is an innovative 3D printing process that could be adapted to the integration of a PCM into building components. This additive manufacturing technique consists of a particle bed and the local application of a fluid used to selectively bind the particles [23,24,25]. This selective mineral activation technique consists of a mixture of very fine aggregates and mineral powder, which are selectively bound by the deposition of a water-based activator. As reported by Lowke et al. [24], this is a layer-by-layer process where the deposition of the powder layer and the dropping of the binder through the print head continue until the required part geometry is achieved.

This technique is expected to be more efficient at achieving a high thermal performance as it can place the phase change material only where it is needed and without damaging the encapsulating shell. Thus, this technique also allows us to eliminate the large amounts of waste generated due to leakage during the mixing process. The reduction in waste and the decrease in costs could be promising.

To the best of our knowledge, the studies published on the use of PCMs in construction materials have been limited to classical mixing and casting into formworks to produce their samples or simple elements in which the PCM is homogenized and dispersed in a mineral matrix. In this work, we propose the implementation of the PCM using additive manufacturing and a selective activation technique.

This study focuses primarily on the overall thermal performance of PCM-integrated composites. Optimizing their sensible and latent heat is essential to further optimizing these materials’ energy storage capabilities.

In the current work, we study the influence of the content and location of a PCM in calcium sulfate powder beds on the energy storage of the created composites. Thermal conductivity measurements and SEM observations were carried out on conventional molded samples and on 3D-printed samples containing the same amount of PCM located in a unique layer. Encapsulation damage to the PCM is not expected to occur when using the developed 3D printer. Our results suggest that the use of a 3D-printed selective activation technique is of great interest for the integration of PCM into construction materials.

2. Materials and Methods

2.1. Calcium Sulfate and Phase Change Materials

Figure 1 illustrates the observations made with a Scanning Electron Microscope of the calcium sulfate and the PCM used in this study. The raw material is a natural plaster, Extha Ibérica (cf. Figure 1a), which is a β-calcium sulfate hemihydrate (CaSO₄·0.5H₂O, containing 5% CaCO₃ and some traces of SiO₂) in powder form, with a median particle diameter of 15.5 μm and a density of 2300 kg·m⁻³. As observed in the literature [22,23], its heterogeneous microstructure is easily characterized by the acicular morphology of its crystals. For the 3D printing process, cellulose ether, at 2% of the plaster mass, was added as an additive to the plaster. This additive allows us to control the naturally uncontrolled capillary-driven penetration of water into the powder bed [26,27].

A commercial micro-encapsulated phase change material 28-S50, supplied by MikroCaps (Ljubljana, Slovenia), was used in this work because of its high thermal energy storage capacity and phase change temperature, which occurs within a comfortable temperature range for the building. The MikroCapsPCM28-S50 is made of a paraffin wax and has a spherical membrane of polyurethane, as observed in Figure 1b. As specified in the product data sheet, its phase change temperature is around 23–27 °C (melting at 27 °C, crystallization at 23 °C) and its latent heat is between 140 and 180 J·g⁻¹. Its apparent density is between 880 g·L⁻¹ and 950 g·L⁻¹ and its equivalent diameter ranges between 1 and 20 μm (see

Table 1. Thermal and physical properties of PCM.

PCM Melting Area (°C)	Heat Storage Capacity (J·g−1)	Density (g·L−1)	Average Particle Size (μm)
26–29	180	880–950	1–20

).

2.2. 3D Printer

The home-made 3D printer used includes a frame that creates the working area (support for the particle bed and its feeding) and supports a 6-axis robot (ABB) and a CNC (Computer Numerical Control).

A pressurized nozzle injects the binder, following a pre-programmed trajectory (Figure 2b). With the aim of controlling the water-to-plaster mass ratio, several parameters, listed in

Table 2. Parameters used for three-dimensional printing.

Printing Parameters	Values
Mass of water deposited per layer (g)	1.68
Nozzle velocity displacement (mm·s−1)	5
Droplet deposition frequency of the nozzle (Hz)	2.01
Number of droplets	131
Mass of a droplet of water (g)	0.015
Mass of plaster per layer (g)	3.84
W/P mass ratio	0.6

, were previously studied, such as the mass of the droplet, which is tuned trough the capillary size of the nozzle; the speed of the articulated arm (nozzle); the drop deposition frequency; and the duration of the trajectory (Figure 2a).

2.3. Suspension Preparation

A schematic illustration of the samples is shown in Figure 3. For the homogenized samples (Figure 3a), the PCM, represented by small white circles, was incorporated directly into the gypsum in powder form. Mixtures of gypsum-based composites with a W/P (water/plaster) ratio of 0.65 and a PCM mass ratio of 5% are referred to as P/PCM. For the 3D-printed samples creating using the selective activation method, the same amount of phase change material was deposited as one layer in the middle of the sample (Figure 3b). Samples without PCM were also 3D-printed as references (Figure 3c). The printing process is an iterative process in which a layer of plaster is deposited and compacted with a roller to obtain a final thickness of 3 mm. Finally, the water is injected through the nozzle, following a pre-programmed path (Figure 2a). The reliability of the developed procedure ensures an adequate local water content with to respect the projected W/P mass ratio (see Table 2 for values). All samples were first recovered from the printer several hours after printing to allow for completion of the setting period, and the samples were then stored in a controlled atmosphere at a room temperature of 20 °C and a relative humidity of 50% until mass stabilization. After curing, the apparent density was determined for all the samples created in this study.

2.4. Thermal Conductivity Measurements

Thermal conductivity quantifies the ability of a material to transfer heat under a temperature gradient. It plays a crucial role in evaluating the thermal insulation properties of construction materials, where lower values are desirable for improved energy efficiency.

Thermal conductivity measurements were performed at room temperature using a hot disk thermal constant analyzer (TPS 1500 Hot-Disk from Thermo-concept, Hot Disk AB, in Göteborg, Sweden). The conductivity measurement is based on the analysis of the temperature evolution of the sample, induced by a constant heat power, over time. A Kapton-type probe was placed between two identical samples with flat surfaces (Figure 4). All experiments were conducted with a heat power of 40 mW and an 80 s duration. This method is the one most widely used in the literature [28,29] because of its several advantages, such as its fast characterization and adaptability to several types of materials. All tests were performed three times to ensure the reliability of the measurements.

The thermal conductivity measurements were conducted using the hot disk method. The associated uncertainty of the instrument is ±5%, which corresponds to ±0.01 W·m⁻¹·K⁻¹ for the range of values measured in this study [30]. To ensure accuracy, all measurements were performed under controlled conditions, including a stable environmental temperature of 20 °C. The uncertainty in the thermal conductivity values is primarily influenced by sample homogeneity and surface flatness, which were carefully addressed during sample preparation.

The hot disk method is validated for a broad range of thermal conductivities (0.005–500 W·m⁻¹·K⁻¹), as reported in [30]. The materials investigated in this study fall within this range, ensuring accurate measurements.

3. Results

3.1. Influence of the Mixing Procedure

In Figure 5, SEM images illustrate the microstructure of the plaster and P/PCM composite containing a 5% mass ratio of phase change materials. The spherical shape of the PCM is observed, with the diameter of its particles ranging from 1 to 20 μm (see Figure 5b). Please note that the needle structure (acicular morphology of the gypsum crystals) becomes more and more coarse with the addition of P/PCM. In Figure 5b, the single capsule observed is damaged. These results confirm that the PCM microcapsules have been damaged during the mixing procedure. The encapsulating shell can break during the mixing process and this causes the PCM to leak into the plaster matrix.

3.2. Observations of the Printed P/PCM Samples

Figure 6a,b show, respectively, a 3D-printed calcium sulfate sample, labeled P3D, and a printed sample with a single layer of PCM deposited in the middle of the sample, labeled P3D/PCM. Figure 6c shows the cast homogenized composite. These photographs demonstrate the ability of the selective printer to create the expected cubic shape, with the addition of cellulose ether into the mixture to better control the water diffusion in the absence of a formwork.

Figure 7 illustrates, at two different scales, the organization of the PCM layer in the printed sample, where the 5% mass is highly concentrated. Clear contact between the microparticles is observed, with most remaining intact, though some show signs of damage. Compared to the non-homogenized composite, it is evident that, for the same amount of MCP added to the plaster, the number of damaged capsules is significantly lower when 3D printing is used.

3.3. Comparison Between Homogenized and 3D-Printed Samples

Figure 8 shows the thermal conductivity as a function of the apparent density of all the samples studied. Relative to the case of the plaster without PCM, the measurements show that the apparent density was clearly influenced by the printing process. A density of 900 kg·m⁻³ was recorded for the reference printed plaster, and 860 kg·m⁻³ for the P3D/PCM. The incorporation of the phase change material reduced the apparent density by 4.5% compared to the plaster without PCM. The values for the 3D-printed samples decreased by 29% compared to the reference homogenized plaster. The use of such powder bed techniques increases the porosity of the samples, as already observed by [24].

It can be seen that the conventionally mixed reference plaster has a higher thermal conductivity than the printed plaster. A difference of 0.1 W·m⁻¹·K⁻¹ is observed between the two plaster samples without PCM, a reduction of 19%, which is related to the air content of the samples. Concerning the P/PCM composites, a decrease of 0.18 W·m⁻¹·K⁻¹ is noted for the case of the P3D/PCM compared to the P/PCM composite. Thermal insulation was improved by the use of 3D printing to form the samples and by the incorporation of the PCM, which induces a significant reduction in thermal conductivity. Thus, a 33% reduction in thermal conductivity for the case of the composite P3D/PCM, in comparison with P3D, was measured.

This study suggests that the difference in thermal conductivity values depends on the method used to create the materials: additive manufacturing is the one that significantly reduces the thermal conductivity of the tested samples. On the other hand, the method of localizing the PCM to within the middle of the sample highlights the possibility of minimizing the damage to the microcapsules due to the absence of a mixing step.

This study reveals that the additive manufacturing process not only lowers thermal conductivity but also achieves substantial reductions in density, especially for the P3D/PCM composite. These improvements can be attributed to the increased porosity and the localized incorporation of PCM, which synergistically enhance the material’s thermal insulation and efficiency. This approach underscores the potential of 3D printing to simultaneously optimize the structural and thermal properties of construction materials.

4. Conclusions

This study investigated the impact of incorporating phase change materials (PCMs) into plaster composites and evaluated how their thermal and physical properties changed under different production processes, including 3D printing and conventional mixing. In the first part of this paper, the results demonstrated significant improvements in thermal insulation performance, particularly for the 3D-printed samples with localized PCM, highlighting the benefits of additive manufacturing in enhancing material properties.

In the second part of this paper, we assessed the sensitivity of these micro-encapsulated materials to the mixing procedure used via SEM observations. The potential applications of this technique, incorporating the PCM using selective-activation 3D printing, were then studied. The results of this investigation demonstrate the full potential of the selective binding process for creating functionally graded construction materials. Their thermal conductivity decreases by 39% when this technique is used, compared to the traditional casting process, while their apparent density only decreases by 17%.

This work paves a way for the design of functionally composite construction materials that contain phase change materials and also highlights the great potential of this selective-binding 3D printing technique. For instance, there are positive economic aspects to working with a slightly lower number of insulating materials or phase change materials, which is easier with this 3D printing technique. The formation of such composites allows for the control of and an increase in the hygro-thermal properties of construction materials in the near future.