Experimental Study on Fire Resistance of Phase Change Energy Storage Concrete Partition Walls

Abstract

Phase change material (PCM) concrete walls represent a new type of energy storage wall. It is of great significance to study the fire resistance of PCM concrete walls to ensure the safety of these kinds of components in service. For this reason, fire resistance tests on eight PCM concrete partition wall specimens under the conditions of the ISO-834 standard fire curve were carried out. The tested wall structures included a solid wall and a double-layer wall with an air gap. The PCM used was paraffin phase change microcapsules, which were replaced with a fine aggregate according to the principle of equal volumes, at replacement proportions of 0%, 7%, 10%, and 14%. The test results showed that explosive spalling of the PCM concrete occurred when the double-layer wall specimen with a 10% replacement proportion was heated for 31 min, and the other seven specimens met the integrity requirements after heating for 90 min. The 100 mm thick ordinary concrete solid partition wall specimen did not meet the thermal insulation requirements after 90 min. The addition of PCM and the use of a double-layer structure with an air gap can both improve the wall’s thermal insulation performance; however, it is not the case that, the greater the amount of PCM used, the better the thermal insulation performance of the wall. The reasons that the PCM concrete spalled in the double-layer wall specimen with a 10% replacement proportion are discussed. This study provides critical insights into optimizing the PCM content and wall design for fire-safe energy-efficient buildings, offering practical guidance for sustainable construction practices.

1. Introduction

Energy and climate change issues are the biggest challenges facing mankind in the 21st century, so energy conservation and emission reduction are being highly prioritized by the international community and national governments. According to statistics, building energy consumption accounts for about 35–40% of all energy consumption and its associated greenhouse gas emissions account for 33% of all energy-related emissions. Achieving energy conservation and emission reductions in buildings has important economic and social benefits [1].

The thermal insulation performance of building envelopes has an important impact on buildings’ energy consumption [2,3]. In recent years, with the continuous improvement of building energy efficiency standards, the performance requirements for building envelopes have also become more demanding, not only in terms of the performance of thermal insulation [4] but also the performance thermal storage performance [3], so the application of phase change energy storage materials (hereinafter referred to as PCMs) in building envelopes has become a topic of interest for many scholars. Phase change energy storage technology uses phase change materials’ ability to absorb/release energy during the phase change process to change the material’s form and stabilize its temperature to sensitively regulate the surrounding environment’s temperature. Using phase change energy storage technology to construct concrete buildings helps maintain the ambient temperature within a comfortable range for the human body over extended periods. This approach reduces reliance on air conditioning and heating, ultimately lowering the energy consumption of building operations and achieving the goals of energy conservation and emission reduction [5,6,7,8].

There are various types of PCMs, among which paraffin is the most commonly used for civil building walls. Paraffin is a solid–liquid phase change material that exhibits minimal volume changes during phase transition [8]. Its phase change temperature typically ranges from 10 °C to 80 °C, making it suitable for civil applications. Additionally, paraffin boasts a high latent heat of phase change (200–300 J/g), a stable energy storage density, and chemical properties. For this study, octadecane paraffin was selected as the PCM, with a phase change temperature set at 28 °C to meet the functional requirements of the partition walls.

Currently, PCM concrete can be produced through direct mixing [9], impregnation [10], microencapsulation [11], and photocuring [12]. The advantages of the direct mixing and impregnation methods are the simplicity of the preparation process and ease of operation [9,10]. The disadvantages of both methods are also notable: as the corrosive phase change materials are in direct contact with the building materials for a long time, the durability of the concrete is not guaranteed. The microcapsule encapsulation method uses a film-forming material to cover the phase change material to form phase change energy storage microcapsules, which avoids direct contact between the phase change material and the building material and reduces the corrosion of the building material by the phase change material, thus ensuring the concrete’s durability [11]. Similarly, the microcapsule encapsulation method has obvious disadvantages, such as its complex preparation process, high cost, and difficulties with large-scale applications. Liu et al. [12] encapsulated PCMs (e.g., paraffin) via photocuring technology, and the materials’ thermal stability and energy storage density were significantly enhanced. This approach provided valuable insights into improving encapsulation techniques for PCMs in concrete applications.

Due to the incorporation of unstable phase change materials in PCM concrete, unlike Portland cement concrete, its mechanical properties [13,14,15,16] and striking heat transfer properties and energy consumption [16,17,18] have attracted considerable interest from scholars. For instance, Hadjieva et al. [13] investigated the mechanical properties of PCM concrete in terms of its tensile and compressive resistance, as well as the effect of phase change material leakage when it is in the liquid state on building material corrosion and reductions in the material’s service life. Meshgin et al. [14] conducted a mechanical study on concrete mixed with phase change materials. The results of the study showed that the compressive and flexural strengths were significantly reduced when 5–20% of fine aggregate was replaced with an equal amount of PCM [11]. Figueiredo et al. [15] also reported similar discoveries, in which concrete with PCM yielded losses of resistance and compression and bending strengths of up to 66% and 52%, respectively, compared with the reference concrete specimens without PCM. Min et al. [16] tested the mechanical properties of concrete components mixed with a shape-stabilized phase change material (SSPCM). It was found that the compressive strength and elastic modulus of the concrete mixed with SSPCM decreased with increased SSPCM content, and the reduction rate was linearly proportional to the mass fraction of SSPCM to concrete.

Min et al. [16] also tested the thermal properties of concrete components mixed with SSPCM. From the tests for thermal behaviors, it was observed that the thermal conductivity decreased and specific heat increased as the SSPCM content increased [16]. Cabeza et al. [17] and Eddhahak-Ouni et al. [18] investigated the thermal properties of PCM concrete. The results showed that PCM concrete had a greater energy storage capacity, improved thermal inertia, and reduced internal temperature compared to conventional concrete without PCM. Essid et al. [19] also explicitly highlighted the role of PCM in decreasing the indoor temperature maintained by different PCM wallboards and also calculated thermal fluctuations using numerical approaches. Cao et al. [20] incorporated microencapsulated phase change materials (MPCMs) into Portland cement concrete (PCC) and geopolymer concrete (GPC) to prepare concrete with a high energy storage capacity. The effect of the MPCMs on the thermal properties of PCC and GPC was investigated. The results showed that the MPCM substituted with sand had a lower thermal conductivity and higher energy storage capacity, while the specific heat capacity of the concrete essentially remained constant when the phase change material (PCM) was in a liquid or solid state [20].

In summary, as new types of concrete walls, PCM concrete walls exhibit outstanding performances in terms of thermal insulation, energy conservation, and emission reduction, and have great prospects for development in the future. However, the research on them so far has mainly focused on their mechanical properties and energy storage capacity, and research on the fire resistance performance of PCM concrete walls is still very limited. In this paper, tests of the fire resistance performance of PCM concrete single solid wall and double-layer wall specimens under the ISO-834 standard [21] fire curve were carried out in order to provide a reference for this material’s fire safety properties during service.

2. Materials and Methods

2.1. Specimen Design

The tests included eight specimens, with four designated as solid walls and four as double-layer walls with air gaps. The main parameter examined was the amount of PCM used. The dimensions of the solid wall are shown in Figure 1a. The thickness of the wall was 100 mm, which satisfied the minimum thickness requirement of 60 mm for non-load-bearing reinforced concrete walls specified in the Code for Fire Protection Design of Buildings (GB 50016-2014) [22], and the steel bar arrangement was a double-layer two-way HRB400 reinforcement; reinforcements with a diameter of 8 mm were employed at a spacing of 200 mm. The outer cover of the reinforcements was 25 mm. The dimensions of the double-layer wall are shown in Figure 1b; this structure consisted of a 50 mm thick inner and outer PCM concrete wall leaf and an intermediate air interlayer, with the same reinforcement configuration as the solid wall. This design intentionally isolated the influence of the air interlayer on fire resistance by maintaining the total concrete volume equivalent to the solid wall. The inner and outer wall leaves were connected by steel connectors. The steel connectors were fabricated from steel plates with dimensions of 5 mm (thickness) × 30 mm (width) × 80 mm (length). Each specimen was equipped with three connectors, spaced at 500 mm intervals (as illustrated in Figure 1b). The connectors were welded to pre-embedded steel components cast within the concrete leaves, ensuring the integration of the double-layer wall system.

2.2. Specimen Production

The first step was to carry out PCM concrete formulation. The tests were designed according to the strength of ordinary C40 concrete, in which the sand was replaced with PCM in equal volumes, and the replacement proportions used were 0, 7%, 10%, and 14%. The composition of the PCM concrete is shown in

Table 1. PCM concrete composition.

PCM Replacement Proportion	Cement (kg/m3)	Sand (kg/m3)	Gravel (kg/m3)	Water (kg/m3)	PCM (kg/m3)
0%	575	455.96	1121.72	308.32	0.00
7%	575	433.16	1121.72	308.32	9.96
10%	575	410.36	1121.72	308.32	13.95
14%	575	392.12	1121.72	308.32	19.49

. PCMs selected for the tests were cetyl paraffin phase change microcapsules, which had the appearance of a white waxy solid; their density was 800 kg/m³, the diameter of microcapsules was 1–2 µm, and the phase transition temperature was 28 °C. The capsule core material was cetyl paraffin, and the capsule wall material was polymethyl methacrylate (PMMA), which is a kind of polyacrylic acid polymer. The cement used was 42.5 MPa ordinary Portland cement, compliant with common Portland cement requirements (GB 175-2023) [23]. The fine aggregate used was river sand with a maximum grain size of 2.5 mm, and the coarse aggregate was graded 5–16 mm limestone gravel.

For each PCM replacement ratio (0%, 7%, 10%, 14%), three 150 mm cubic specimens were cast and cured for 28 days. Compressive strength tests were conducted using a universal testing machine at a loading rate of 0.5 MPa/s. The measured values of the compressive strength of the PCM concrete are shown in Table 1. If the maximum or minimum value within a group deviates from the median value by more than 15%, the median value is adopted as the representative result. If both the maximum and minimum values deviate from the median by over 15%, the entire dataset is deemed invalid. Otherwise, the average value of the three specimens is used. The average/representative values of compressive strength of the PCM concrete are given in the last column of

Table 2. Compressive strength of the PCM concrete.

PCM Replacement Proportion	Measured Value 1 (N/mm2)	Measured Value 2 (N/mm2)	Measured Value 3 (N/mm2)	Average/Representative Value (N/mm2)
0%	36.0	40.2	40.7	39.0
7%	36.6	33.0	29.6	33.1
10%	31.7	29.8	17.0	29.8
14%	41.0	35.5	33.8	35.5

. While prior studies generally report reduced compressive strength with higher PCM content, the observed increase at 14% PCM remains unexplained. Further investigation, including expanded replicates and microstructural analysis, is needed to explain this phenomenon.

The PCM concrete specimens were cast using custom-made steel formwork. Prior to pouring, reinforcing bars were placed in the formwork and then the thermocouples were tied on the bars. Finally, the concrete was poured. After various procedures were completed, such as vibration, smoothing, demolding and, curing for 28 days, the tests were started. It should be noted that during the actual fabrication process, one row of longitudinal reinforcement bars was omitted due to worker negligence, resulting in a discrepancy between the as-built reinforcement arrangement and the original design scheme. Such deviations must be strictly avoided in practical engineering applications. The process of fabricating PCM concrete walls is shown in Figure 2.

2.3. Thermocouple Layout

In order to measure the temperature change inside the specimens, five thermocouples were arranged on each specimen. Measurement points 1~4 were in the middle section of the specimen and measurement point 5 was 300 mm away from measurement point 4; the distribution of the thermocouples is shown in Figure 3.

2.4. Test Device and Loading Scheme

The experiment was conducted at Shanghai Normal University, and the experimental equipment consisted of a heating furnace and data acquisition systems. The dimensions of the electric heating furnace were 800 mm × 800 mm × 1200 mm, as shown in Figure 4a. In all the experiments, the heating patterns followed the ISO-834 standard fire curve, which was achievable as the maximum heating value of the heating furnace was 1200 °C and the rated power was 90 kW. The temperature of the furnace was measured by two S-scale thermocouples arranged in the furnace, and the rising temperature was controlled by intelligent PID (Proportional–Integral–Derivative), which can achieve ISO-834 standard temperature heating. Meanwhile, to make accurately recording the time–temperature curve more convenient, data acquisition systems were installed in each wall being tested.

Due to the limitations of our laboratory furnace, which requires the entire specimen to be placed inside the heating chamber, direct one-sided fire exposure could not be physically achieved. To simulate one-sided fire conditions, a fireproof insulation method was adopted by wrapping the unexposed surfaces and edges of the specimen with 50 mm-thick fire-resistant ceramics fiber blankets (thermal conductivity: 0.04 W/m·K at 800 °C). This method aligns with ISO 834-1:1999, which permits the insulation of non-exposed surfaces to maintain boundary conditions representative of real-world fire scenarios. The contact surfaces were bonded with high-temperature-resistant adhesive and bound with steel wire to prevent the ceramic fiber blankets from falling off during the tests. The integrity and thermal insulation of building separation components should meet the corresponding requirements of the ISO-834 standard for the fire resistance of components. The scheduled heating time was 90 min. If the specimen reached any criterion of loss of integrity or thermal insulation within 90 min, the test was terminated early.

3. Results

3.1. Test Results

After the test furnace was preheated, the actual warming test was commenced. Within the first 15 min, water vapor escaped from the specimens. With the increase in the furnace temperature, the specimens’ temperatures also increased gradually, but most of the specimens maintained good integrity. Four of the single solid wall specimens and three of the double-layer wall specimens maintained integrity after 90 min of heating, and the test was terminated. The double-layer wall specimen with a 10% proportion of PCM replacement emitted a loud noise at 31 min. After the furnace temperature cooled down, it was found that PCM spalling occurred and the specimen had lost integrity. The specimens’ damage patterns are shown in Figure 5.

3.2. Time–Temperature Curves

The temperature during each test was monitored, and then, corresponding time–temperature curves were drawn. Through comparing the time–temperature curve recorded in the furnace with the predefined ISO-834 standard fire curve, it was found that the furnace was capable of simulating standard fire conditions. The hysteresis of the temperature inside the furnace could not be completely removed, which is reflected by the generally lower temperatures in the first 8 min. After 8 min, the temperature deviations were negligible, as shown in Figure 6.

According to the ISO-834 standard’s requirements for the thermal insulation performance of components, the difference between the average temperature of the backfire surface of the specimen and the initial average temperature cannot be higher than 140 °C, and the difference between the temperature at any point on the backfire surface of the specimen and the initial average temperature cannot be higher than 180 °C. According to the thermocouple layout, shown in Figure 3, thermocouples 4# and 5# can be used to judge the thermal insulation of the specimens. As described in Section 3.1, the double-layer wall specimen with a 10% replacement proportion burst after 31 min of heating, and the specimen’s integrity was destroyed. Therefore, the thermal insulation of the specimen did not meet the ISO-834 requirements. Thus, only the specimens that met the integrity requirements were analyzed for thermal insulation.

The temperature curves measured by thermocouples 4# and 5# for the seven specimens are shown in Figure 7. For comparative purposes, the maximum and average temperature rise in each specimen relative to the initial average temperature is also given in Table 2. According to Figure 7 and

Table 3. Parameters and test results for specimens.

ID	Type	PCM Replacement Proportion	Integrity	Backfire Surface		Insulation
				Maximum Temp. Rise (°C)	Average Temp. Rise (°C)
Solid-0	Solid wall	0	Yes	161	145	No
Solid-7	Solid wall	7%	Yes	133	131	Yes
Solid-10	Solid wall	10%	Yes	126	120	Yes
Solid-14	Solid wall	14%	Yes	156	147	No
Doublelayer-0	Double-layer wall with air gap	0	Yes	141	137	Yes
Doublelayer-7	Double-layer wall with air gap	7%	Yes	82	79	Yes
Doublelayer-10	Double-layer wall with air gap	10%	No	--	--	No
Doublelayer-14	Double-layer wall with air gap	14%	Yes	130	125	Yes

, it can be seen that the average temperature rise for the backfire surfaces of the solid wall specimen without PCM (Solid-0) exceeded 140 °C, and so did not meet the thermal insulation requirement. After adding 7% and 10% PCM, the insulation performance of the specimens met the requirements. However, when the proportion of PCM replacement was 14%, the specimen did not meet the thermal insulation requirement either. The above test results show that the proper admixture of PCM can reduce the thermal conductivity of the concrete, but the amount of admixture should not be too large. When the hollow sandwich structure was adopted, all of the walls (except Doublelayer-10) met the thermal insulation requirements, which indicates that the air sandwich structure improved the thermal insulation performance of the specimens.

4. Discussion

In this paper, an experimental study on the fire resistance of phase change energy storage concrete partition walls was conducted, with the wall structures including solid walls and double-layer walls with air gaps. Both the integrity and thermal insulation of these structures should have met the corresponding requirements of the ISO-834 standard for the fire resistance of partition walls. According to the test results, it was found that the average temperature rise in the backfire surfaces of the solid wall specimen without PCM (Solid-0) was 145 °C, exceeding 140 °C, so specimen Solid-0 did not meet the thermal insulation requirements. After adding PCM at replacement ratios of 7% and 10%, the thermal insulation of the solid wall specimens met the requirement. However, the thermal insulation did not meet the requirement when the PCM replacement ratio was 14%. Therefore, adding an appropriate proportion of phase change energy storage material can improve the thermal insulation performance of walls.

The thermal insulation performance of the double-layer wall specimens with air gaps met the requirements, unlike that of the single solid concrete walls. Therefore, besides adding PCM, the double-layer wall structure with an air gap can be considered able to improve the thermal insulation performance of the wall when the space allows. It can be seen from Table 2 that specimen Doublelayer-7 had the best thermal insulation among all the specimens. However, as an exception to this, the double-layer wall specimen with a 10% replacement proportion burst after 31 min of heating, and PCM concrete spalling occurred so the specimen lost its integrity.

Concrete spalling exhibits a high level of randomness, and its mechanism is complex. According to the measured compressive strength of the PCM concrete, provided in Table 1, the concrete with a 10% replacement proportion of PCM had the lowest compressive strength, so its corresponding tensile strength was also the lowest. Supposing that the internal pore pressure of concrete is similar at high temperatures, concrete with a low tensile strength is more likely to display non-stationary crack propagation and spalling. Figure 8 compares the time–temperature curves at measuring point 1# for specimens Solid-10 and Doublelayer-10. As the wall leaf thickness of the specimen Doublelayer-10 was only 50 mm, compared with the solid wall of specimen Solid-10 with 100 mm thickness, the temperature rise in the fire surface of the Doublelayer-10 specimen was significantly faster than that for specimen Solid-10 after 20 min of heating. Therefore, the temperature gradient inside the Doublelayer-10 specimen was larger, resulting in greater temperature stress.

When the specimen Doublelayer-10 burst, the temperature of thermocouple 1# was 270 °C (shown in Figure 8). In order to study the effect of the PCM on the internal pore structure of concrete at different temperatures, the open porosities of the four types of PCM concrete at room temperature, after being heated to 200 °C and 300 °C, were measured, respectively. The testing method for open porosity was as follows: First, the mass and volume of the specimen were measured. Then, the specimen was immersed in water for 24 h and its saturated mass was recorded. The open porosity was calculated by dividing the difference between the saturated mass and the initial mass by the product of the specimen’s volume and the density of water, which can be expressed as follows:

$$P={\displaystyle \frac{m_{\mathrm{s}\mathrm{a}\mathrm{t}}-m_{\mathrm{d}\mathrm{r}\mathrm{y}}}{V.{\rho }_w}}\times 100\%$$

(1)

where P is the open porosity (%), m_sat and m_dry are the saturated and dry masses (g), ${\rho }_w$ is the density of water (1.0 g/cm³), and V is the specimen volume (cm³). The measured values are given in

Table 4. PCM concrete open porosity (%).

PCM Replacement Ratio	At Room Temperature	After Heating to 200 °C	After Heating to 300 °C
0%	2.81	14.89	20.76
7%	2.59	15.74	21.13
10%	1.97	14.00	15.16
14%	1.53	12.03	13.55

.

According to the data in Table 4, it can be seen that PCM incorporation leads to a decrease in the open porosity of concrete at room temperature compared to that of ordinary concrete, and the higher the PCM content incorporated, the greater the decrease in the open porosity. The open porosity of the concrete after exposure to high temperatures increased significantly, in which the open porosity of the PCM concrete with a 7% replacement proportion exceeded that of ordinary concrete, but the opening porosity of PCM concrete with a 10% and 14% replacement proportion was still lower than that of the ordinary concrete. Therefore, the capillary joint voids of the specimen Doublelayer-10 were relatively small when subjected to high temperatures, and water vapor did not easily diffuse outward, which, generally, tends to form a higher pore pressure inside, thus increasing the probability of spalling. In summary, when the pore pressure value generated by the combined effect of vapor pressure and thermal stress exceeds the tensile strength of PCM concrete, the water vapor will diffuse non-stationarily along the initial microcrack, which makes the crack expand rapidly and show non-stationary development, thus may leading to concrete spalling. It should be noted that the occurrence of explosive spalling in concrete is characterized by significant randomness and complexity. In this study, since spalling was observed in only one specimen, the analysis provided herein specifically investigates the causes of spalling for this individual case. Consequently, the results should not be considered representative of broader scenarios. Further experimental studies are required to comprehensively analyze the underlying mechanisms of explosive spalling.

5. Conclusions

Based on the experimental and analytical findings of this study, the following conclusions can be drawn:

(1) The optimal PCM proportion enhances the thermal insulation of concrete. The incorporation of a paraffin-based PCM at replacement proportions of 7% and 10% significantly improved the thermal insulation performance of concrete walls under ISO-834 fire conditions. These proportions reduced the average temperature rise on the unexposed surface to 131 °C and 120 °C, respectively, meeting the thermal insulation criterion (ΔT ≤ 140 °C). However, a higher replacement proportion (14%) compromised the concrete’s performance due to reduced thermal conductivity and increased pore pressure risks.

(2) The design of a double-layer wall with air gaps exhibits improved fire resistance. The double-layer wall with a 50 mm air interlayer demonstrated superior fire resistance compared to that of the solid walls. The specimens with 7% PCM achieved the lowest temperature rise (ΔT = 79 °C), highlighting the synergistic effect of PCM integration and air gap insulation. This structural configuration effectively delays heat transfer, making it a viable solution for energy-efficient building envelopes.

(3) The explosive spalling observed in the double-layer wall with a 10% phase change material (PCM) replacement ratio may be attributed to the relatively lower compressive strength of this specific specimen. However, since only one specimen exhibited spalling in this experiment, the underlying mechanisms necessitate further investigation to unravel the complex interactions among material properties, thermal loading, and pore structure evolution.

This study validates the feasibility of PCM-enhanced concrete walls for fire-safe, energy-efficient buildings. Future research should focus on (i) optimizing the PCM dosage and encapsulation techniques to mitigate spalling risks; (ii) evaluating the long-term durability under cyclic thermal loading; and (iii) exploring hybrid PCM systems for diverse climatic conditions.