An Overview of Sandbox Experiment on Ground Heat Exchangers

Abstract

As an energy-efficient and low-carbon technology, ground-source heat pumps are promising to contribute to carbon neutrality in the building sector. A crucial component of these systems is the ground heat exchanger, which has been extensively studied through sandbox experiments. These experiments play a vital role in understanding heat transfer characteristics and validating simulation results. In order to facilitate the improvement of ground heat exchangers and the development of ground-source heat-pump systems, this article provides a comprehensive summary of existing sandbox experiments. The borehole sandbox experiments are classified into the single borehole experiment, borehole group experiment, seepage experiment, and multi-layer soil experiment. It was observed that the heat transfer efficiency of a single spiral tube is only 80% compared to that of a double spiral tube. Moving on to energy-pile sandbox experiments, they are further divided into mechanical performance, thermal performance, and thermal-mechanical coupled performance tests. It was revealed that the heat transfer distance of a single U-shaped energy pile in the radial direction is three times greater than in the vertical direction. For the mentioned sandbox experiments, the sandbox design, experiment conduction, testing conditions, and result analyses are summarized. To improve the sandbox experiments, there are still some difficulties in building a similarity experiment, testing the temperatures in a small error, controlling the boundary conditions accurately, and testing the thermophysical properties of soil accurately. Furthermore, the perspectives of sandbox experiments of ground heat exchangers are also proposed. The sandbox experiments under complex environment conditions or with novel composite energy geo-structures or ground heat exchangers with new materials and new technologies would be further investigated. By addressing these aspects, this review aims to provide guidelines for the design, construction, operation, and optimization of sandbox experiments for different ground heat exchangers, ultimately promoting the wider adoption of ground-source heat pumps in achieving carbon neutrality.

1. Introduction

Building energy consumption typically accounts for approximately 20–40% of the total energy consumption in different regions, with space heating and cooling alone contributing to over 40% of this consumption [1,2,3]. As a result, reducing energy consumption in air conditioning systems is of utmost importance for improving building energy efficiency and mitigating emissions [4,5,6]. With the improvement in living standards, the demand for indoor thermal comfort has significantly increased [7,8]. Therefore, it is crucial to develop and implement efficient air-conditioning technologies that can simultaneously maintain low building energy consumption and enhance indoor thermal conditions [9,10].

Ground-source heat pumps (GSHPs) [11,12,13,14] utilize the shallow ground as a low-grade heat source (or a heat sink) for heating (or cooling). During the winter, the heat pump unit absorbs low-grade heat from the soil to provide space heating for buildings. In the summer, the heat within the building is transferred to the soil through the system, cooling the building and storing the heat for subsequent use during the winter [15,16,17]. The soil temperature remains relatively stable throughout the year, making it an advantageous heat source [18,19,20]. This stability allows GSHPs to achieve a high heating coefficient of performance (COP) during winter and a high cooling COP during summer, highlighting their significant energy-saving potential [21,22,23].

In a GSHP system, the ground heat exchanger (GHE) is any critical component that exchanges heat with the soil [24,25,26]. The heat transfer performance of GHE affects the design and operation of the system [27,28]. Commonly used forms of GHEs include vertical boreholes [29,30,31] and energy piles [32,33,34,35]. Despite the energy efficiency of GSHP systems, some practical projects have experienced underperformance primarily due to suboptimal GHE designs [36]. With transient heat transfer or coupled heat/mass transfer processes in long-term operation, an accurate GHE model plays a key role in improving the GSHP system design. The heat transfer mechanism of GHEs can be investigated by two methods: simulation (e.g., numerical simulation [37,38] by using two model order reduction (MOR) methods [39], COMSOL environment [40], FORTRAN language [41] and analytical calculation [42,43] by considering thermophysical parameters and temperature distribution [44], non-uniformity of the ground [45], and temperature response [46]) and experiment (e.g., sandbox experiment [47,48,49,50] field test about energy piles [51,52,53] under groundwater flow [54], equipped with steel pipe heat exchangers [55]). Among these methods, the simulation is relatively easily performed under different working conditions, but it has errors caused by calculation assumptions. In addition, although the field test can reflect the practical behaviors of GHEs during operation, the cost is high, and there are many uncontrollable factors. In contrast to the field test, the parameters of the sandbox experiment are controllable and repeatable. Therefore, it provides a convenient and effective verification approach for simulation methods and also reveals the heat transfer mechanism of GHEs experimentally.

Numerous researchers have conducted sandbox experiments for different forms of GHEs to reveal the heat transfer mechanisms and validate the heat transfer models. According to the types of GHEs, sandbox experiments include borehole sandbox experiments [47,48,49,50] and energy pile sandbox experiments [56,57,58], considering various factors including conventional single borehole [47], circulating water flow in tubes [48], tube group thermal disturbance [59], underground seepage [60], and layered soil [61], etc. For the sandbox experiments [62] that do not use the similarity principle, the measurements can be utilized to verify the heat transfer mechanisms of GHEs under specific assumptions. As for the sandbox experiments following the similarity principle [63,64], the results can be directly referred to by the full-scale GHEs. The sandbox experiment tests the heat transfer mechanism of GHEs and provides an important experimental basis for the heat transfer model establishment. With increasing studies on diversified types of GHEs serving various GSHP systems, a comprehensive overview of GHE sandbox experiments is of great necessity to facilitate theoretical investigations as well as promote practical applications.

This paper presents an overview of GHE sandbox experiments by discussing the sandbox design, experimental procedures, and test results of both borehole and energy pile experiments. The research status of GHE sandbox experiments is also summarized to provide valuable insights into the current state of the field. Regarding borehole sandbox experiments, the focus lies on various aspects such as single borehole experiments, borehole group experiments, seepage experiments, and experiments considering multi-layered boreholes under complex geological conditions. In the case of energy pile sandbox experiments, the research primarily focuses on mechanical performance, thermal performance, and coupled thermal-mechanical performance experiments. By providing guidelines for the design, construction, operation, and optimization of sandbox experiments for different GHEs, this paper offers valuable insights for researchers and practitioners. The ultimate goal is to promote the development of reasonable GHE designs and facilitate wider adoption of ground source heat pump (GSHP) technology, thereby contributing to the pursuit of carbon neutrality and sustainable energy solutions.

2. Borehole Sandbox Experiment

The buried borehole sandbox experiment typically consists of scaled boreholes, a sandbox, soil, the seepage system, and a temperature testing system. As for the boundary conditions, the sandbox wall can be isothermal, adiabatic, or adiabatic-isothermal (adiabatic walls and bottom but isothermal top). The adiabatic boundaries can be achieved by coating insulation materials. The isothermal boundaries can be achieved by maintaining a constant indoor air temperature. The sandboxes can have cuboid or cylindrical shapes, and the soil inside can be clay, sand, or layered soil. The scaled borehole is typically created using simplified heating wires to represent a group of boreholes or metal tubes with circulating water, effectively mimicking the actual heat transfer process within the boreholes (including heat convection in the tube, heat conduction in the tube wall, and heat transfer through the surrounding soil). The seepage systems can emulate the seepage in vertical or horizontal directions. Wang et al. [47], Diao et al. [48], Xu et al. [49], and Li et al. [50,65] carried out conventional sandbox experiments with a single borehole. Chen et al. [66], Aydan et al. [67], and Zhang et al. [59] performed sandbox experiments with multiple buried borehole groups. Liu et al. [60] performed a sandbox experiment with seepage. Wang et al. [61] conducted a sandbox experiment with multi-layered soil. The heat transfer performances of boreholes considering different borehole forms, seepage conditions, thermal interference inside the tube group, and different soil layers have been mainly studied. Next, we will categorically introduce some research conducted on the borehole sandbox experiment.

2.1. Single Borehole Experiment

The single borehole sandbox experiment aims to reflect the heat transfer mechanism and validate the mathematical heat transfer model for an individual borehole. Some researchers [47,48,67,68] set up experiments based on the similarity principle and used the test results to reflect the heat transfer characteristics of a borehole in a practical project. The test rig built by Wang et al. [47] adopted a stainless-steel cylinder with a diameter of 0.8 m and a length of 5 m as a sandbox, as shown in Figure 1. The boundary condition is isothermal. In order to ensure the homogeneity of the soil within the sandbox, the entire cylinder is equipped with a rotary mechanism, enabling a 90° rotation from side to side. The U-shaped tube can be placed horizontally in the horizontal borehole to avoid the influence of the height difference between the inlet and outlet tubes. Two kinds of U-tubes (D32 and D25, nominal wall thickness of 3 mm) are adopted for comparison. In addition, the cylinder is separated into three sections by stainless steel plates along the axial direction to emulate the effects of different soil types and moisture contents at different depths on the heat transfer performance.

Wang et al. [47] obtained reasonable boundary conditions of the U-tube heat exchanger for numerical simulation through experimental results. Moreover, they conducted simulations to evaluate the heat transfer characteristics of vertical double-helix tubes and investigated the effects of tube pitch, inlet velocity, and inlet temperature on the heat transfer performance. For the U-tube, the single spiral tube, and double spiral tubes with different pitches, the temperature distributions, heat transfer efficiencies, and heat capacities were compared. The results showed that the heat transfer performance of the spiral tube was better than that of the single U-tube with the same drilling depth. Additionally, the heat transfer performance was improved with the increase in the spiral pitch. The total heat transfer of the single spiral tube could only reach 80% of that of the double spiral tube. The heat transfer performance could be enhanced when the tube spacing was increased, and the pitch of the double helix tube was reduced appropriately.

Different from the cylindrical experimental devices constructed by Wang et al. [47], experimental devices assembled by Diao et al. [48] used two cuboid sandboxes of 2 × 1.2 × 0.8 m and 4 × 1 × 1 m, as is shown in Figure 2. The sandbox was filled with fine dry sand and had isothermal boundary conditions. A U-tube was buried in the center of the box with circulating water flowing through it.

From the experiment of Diao et al. [48], it can be concluded that: Figure 3a shows the layouts of temperature measuring points in Section B. During operation, during the experiment, point 21 exhibited a consistently steady temperature, indicating that the boundary of the sandboxes remained unaffected by the heat transfer originating from the U-tube. As shown in Figure 3b, in the first few hours of operation under the inlet temperature of 20 °C, the soil temperatures near the U-tube (points 25, 24) rose rapidly, and the soil temperatures in other areas (e.g., points 23, 22, 21) had little increase. After a few hours, the temperature rises of all measuring points decreased, the temperature field tended to be stable, and the temperature gradient of soil close to the U-shaped tube was large.

Different from the two experimental devices for horizontally placing a U-tube mentioned above, You et al. [62] developed a square sandbox device with a size of 1 × 1 × 1 m, as is shown in Figure 4a. The device was covered with insulation cotton to establish an adiabatic boundary condition, and a U-tube was centrally positioned within the sandbox. The U-tube had an inner diameter of 8 mm and a length of 950 mm. For measuring the water flow velocity within the U-shaped tube, a glass rotameter was used in the experiment. A water bath was employed to maintain a constant inlet water temperature, and six T-type thermocouples were used to test the soil temperature at typical locations to verify the mathematical model of the ground heat exchanger (GHE) proposed by the authors. When the excess temperature of inlet water temperature is −10 °C, the change of soil temperature at the typical points is shown in Figure 4b. Point 1, positioned on the outer wall of the heat exchange tube, displayed a decrease of 5.3 °C in soil temperature after 24 h. On the other hand, point 6, located 500 mm from the heat exchange tube, experienced a decline of only 0.5 °C in soil temperature during the same period.

Beier et al. [69] used different materials to construct the experimental device and create an adiabatic boundary condition. Beier et al. [69] built a large wooden frame sandbox with a size of 18 × 1.8 × 1.8 m, which was lined with plastic and filled with saturated sand, as shown in Figure 5. The aluminum tube with an inner diameter of 12.6 cm was selected as a borehole, which had a length of 18 m and was located in the center of the sandbox. The U-shaped tube was made of high-density polyethylene (HDPE) with an outer diameter of 3.340 cm and an inner diameter of 2.733 cm. It was horizontally placed in the center of the borehole setup. The borehole was filled with cement slurry (20% solid, thermal conductivity 0.73 W/(m·K)) inside.

The author conducted a thermal response test on the borehole, obtaining temperature measurements for the borehole wall and the soil at different distances (24 cm, 44 cm, 65 cm, and 85 cm) from the borehole wall, as shown in Figure 6. After approximately 0.7 h of testing, the temperature of the borehole wall began to rise steadily, exhibiting a linear increase after around 1.1 h. Roughly 7 h later, the soil temperature at a distance of 24 cm from the borehole wall started to rise, while the temperatures of the soil at 44 cm, 65 cm, and 85 cm from the borehole wall began to increase after 10 h. At the end of the test, the temperature rise at the sandbox wall (at 85 cm) was merely around 0.3 °C, indicating that the circulating water had little thermal effect on the sandbox boundary during the test. The temperatures of the borehole wall and the fluid increased at the same rate, and their temperature difference could reach 10 °C after about 10 h. This experiment provides detailed information regarding the working conditions, parameters, and results, enabling the validation of the mathematical model for borehole heat transfer.

In the single borehole sandbox experiment, researchers commonly employ a scaled-down model that mimics an actual borehole. This model allows them to replicate the conditions of practical engineering projects in a laboratory setting by applying the principle of similarity. The experiment entails measuring the temperature distribution surrounding the borehole in various time intervals, with the aim of investigating the heat transfer characteristics.

The obtained experimental results serve to verify the precision and viability of the mathematical model. By meticulously analyzing and contrasting the experimental data, researchers can evaluate the extent to which the mathematical model is capable of accurately predicting the heat transfer behavior of boreholes in practical engineering projects.

2.2. Borehole Group Experiment

In practical projects, GSHP usually adopts borehole groups to exchange more heat with the soil. Different from a single borehole, heat transfer among different boreholes in a group can influence each other. To study the heat transfer characteristics of the borehole group, the heating wires or the tubes with water flowing inside were used as heat exchangers to conduct sandbox experiments. The use of heating wires in the experiments facilitated the emulation of boreholes with a constant heat transfer rate, making the apparatus setup and operational procedures relatively straightforward. In contrast, the configuration involving tubes with flowing water in the boreholes, although more complex than the heating wire setup, enabled not only the emulation of a constant heat flow condition but also the investigation of the impact of different parameters (such as inlet water temperature and circulating water flow rate) on heat transfer. Zhang et al. [59] set up a test rig, employing a stainless steel sandbox with dimensions of 0.6 × 0.6 × 1 m, as shown in Figure 7. The sandbox was constructed using 2 mm thick stainless steel plates and covered with 3 cm thick insulation material to ensure adiabatic conditions. The waterproof electric heating rods were adopted with a length of 0.4 m and a diameter of 6 mm. One electric heating rod was positioned upstream to conduct the single borehole experiment, while three additional electric heating rods were placed on the left, right, and back sides of the former rod. Together, these four electric heating rods were employed to investigate the heat transfer within borehole groups.

The soil temperature distributions of the single borehole and the boreholes group with seepage and non-seepage were tested. With the same heating capacity of the heating rod, the temperature rise and thermal storage capacity of the saturated soil are higher and larger than those of the dry soil. Figure 8 shows the variations in soil temperature under different seepage velocities, assuming each heating rod has a heat capacity of 20 W. Due to the movement of heat within the soil towards the seepage direction, the temperatures in the middle and downstream regions are relatively high, and the thermal exchange is intensified by the presence of seepage. Therefore, the borehole spacing in the seepage direction should be increased as much as possible to alleviate the heat accumulation in the borehole groups.

Different from Zhang et al. [59], who conducted a borehole tube group experiment using electric heating rods, You et al. [70] conducted their experiment using a borehole group consisting of five U-shaped tubes with circulating water. The test rig consisted of a sandbox, a water bath with a constant temperature supply, thermocouples, a glass rotameter, and a data logger. The sandbox was designed with a size of 1 × 1 × 1 m, with the five U-tubes buried inside (Figure 9b). The initial temperature of the sand was 21 °C, while the U-shaped tube’s inlet water temperature was maintained at 31 °C. The total water flow rate was constant at about 4 L/min. Four T-type thermocouples were placed near the central tube at a depth of 0.5 m to monitor soil temperature variations at typical locations, as shown in Figure 9b. Figure 9c illustrates the soil temperature changes at typical points after 24 h of operation, with the tube wall temperature (P1) decreased by about 6 °C. The soil temperature of P4, which is located 25 mm away from the tube, decreased by about 1.5 °C after the 24 h operation. These results can verify the heat transfer calculation model developed for the borehole group.

In conclusion, the research on borehole group experiments offers significant insights into the heat transfer behavior within borehole groups. By employing heating wires or tubes with flowing water as heat exchangers and strategically arranging the relative positions of the boreholes within the group, different operating conditions can be simulated, and the impact of different parameters on heat transfer can be analyzed. These findings are essential for gaining a deeper comprehension of heat transfer in borehole groups and for investigating parameter optimization to maximize energy efficiency.

2.3. Seepage Experiment

In practical projects, it is common to encounter horizontal seepage flows in vertically drilled boreholes. The seepage can take away the cold or heat upstream of the borehole group and consequently slow down the cold or heat accumulation inside the borehole group. For the sandbox experiments, the seepage in the radial direction of the borehole is modeled by the seepage flowing in a vertical or horizontal direction, which is called the sandbox experiment with vertical or horizontal seepage, respectively.

2.3.1. Sandbox Experiments with Vertical Seepage

In the sandbox experiment with vertical seepage, the radial seepage of the boreholes flows downwards due to gravity, and the boreholes themselves are positioned horizontally within the sandboxes. In order to regulate the velocity and distribution of seepage across the entire cross-section, a series of filter nets is introduced between the water chamber and the sand. By adjusting the number of filter nets and the water level in the high-level water tank, the seepage velocity can be effectively controlled. Next, we will discuss two sandbox experiments that specifically examine vertical seepage flows.

The first sandbox experiment involved regulating the seepage rate through the manipulation of the water tanks and valves. Liu et al. [60] set up a sandbox test rig with vertical seepage, as shown in Figure 10. The sandbox size was 0.8 × 0.6 × 0.77 m, and insulation materials were used to establish the adiabatic boundary condition. A heater with adjustable powers of 0~300 W was placed horizontally in the middle of the sandbox to emulate the borehole. The sandbox was connected to a fixed upper water tank and a movable lower water tank. By adjusting the relative height of the water tanks and corresponding valves, the different water head levels for seepage were achieved. As a result, the regulating valves could be employed to control the seepage velocities within the sandbox. The water in the sandbox flows out from the overflow tube of the moving water box, and the water seepage velocity in the sandbox can be obtained by measuring the flow rate. The research findings indicate that: (1) Seepage can improve the heat transfer capacity of the GHEs, resulting in smaller soil temperature variations around the borehole compared to saturated soil conditions. (2) Higher seepage velocities lead to reduced temperature fluctuations in the surrounding soil. (3) The soil temperature is no longer distributed in a circular shape, and the heat transfer increases along the seepage direction. Furthermore, the distance of influence expands with the growth of flow velocity.

In contrast to the first sandbox experiment, which controlled the seepage rate by adjusting the water tanks and valves, another sandbox test rig with vertical seepage was built by Zhang et al. [59], as shown in Figure 11. The upper and lower sections of the sandbox were equipped with a 50 mm high water tank, and an overflow hole was positioned 40 mm above the liquid level of the upper water tank. To facilitate seepage, a steel plate with small drilled pores, covered with gauze to prevent sand loss, was used to separate the water tank from the sandbox. A glass rotor flowmeter was installed at the outlet of the water tank to monitor seepage velocities. To simulate the flow of vertical seepage, a horizontally positioned waterproof electric heating rod was employed as a substitute for the vertical borehole in this experiment. The schematic diagram of the electric heating rod and layout of temperature sensors in the experiment are shown in Figure 10a. Additionally, a water bath with a constant temperature was connected to the upper and lower water tanks to ensure that the sandbox attained the desired initial temperature. By adjusting the controlled temperature of the water bath, seepage at various temperatures could be achieved.

The experiment conducted by Zhang et al. [59] yields the following conclusions: Figure 12a shows the temperature variation of point 3 downstream of the electric heating rod under different seepage velocities when the heat dissipation rate of the electric heating rod was 15 W. It can be observed that the greater the seepage velocity was, the shorter the time needed to reach stability, and the lower the stability temperature was. Figure 12b shows the temperature variation of point 3 under different heat dissipation rates with a seepage velocity of 1 × 10⁻⁵ m/s. As the heat dissipation rate increases, the temperature at point 3 reaches stability in less time, and the temperature growth is higher.

Based on the aforementioned sandbox experiments involving vertical seepage, it can be concluded that achieving downward water flow driven by gravity is relatively straightforward, and the experimental operation is easily controllable. However, it should be noted that the simplified vertical seepage experiment may not fully capture the precise heat transfer characteristics of boreholes with horizontally radial seepage in practical projects.

2.3.2. Sandbox Experiments with Horizontal Seepage

In sandbox experiments involving horizontal seepage, water can flow horizontally by utilizing the pressure difference between the left and right sides of the sandbox. This mechanism is achieved by placing a water tank at a higher level on one side of the sandbox and a water tank at a lower level on the other side. The height disparity between the two water tanks facilitates the lateral flow of water from the higher tank to the lower tank, resulting in horizontal seepage. Additionally, the seepage velocity can be controlled by adjusting the height difference between the high and low tanks.

Li et al. [71] conducted a sandbox experiment with horizontal seepage, as shown in Figure 13. The rectangular sandbox had dimensions of 2.5 × 2.5 × 1 m, and the sandbox boundaries were maintained at a constant temperature by regulating the ambient room temperature. To ensure good thermal conductivity, a copper U-shaped tube was utilized as the heat exchanger, measuring 0.9 m in length, 0.02 m in diameter, and with a leg spacing of 0.04 m. The U-shaped tube received flowing water with the desired temperature from a constant-temperature water tank. Plastic plates were positioned on each side of the sandbox to create two water tanks, and eight tubes were installed to measure water pressures along the side wall at 20 cm intervals. To control the water levels, an overflow hole was positioned 0.95 m above the bottom of the tank on the right side of the sandbox. On the left side, there were 11 overflow holes with the same distance between them, facilitating control of the upstream water tank’s water level. The horizontal interval between adjacent overflow holes was 0.05 m, and the water level could vary from 0.90 m to 0.40 m. The difference in water level between the upstream and downstream tanks ranged from 0.05 m to 0.55 m, with intervals of 0.05 m. By manually manipulating the overflow hole switch, the seepage velocity could be adjusted according to requirements.

Based on the sandbox experiment on horizontal seepage conducted by Li et al. [71], the following conclusions can be drawn: Figure 14 presents the soil temperature distribution in the sandbox under different heat sources with a water level difference of 5 cm. The experimental results show that: (1) In the presence of seepage in the soil, the isothermal line is no longer circular and deviates downstream. (2) The soil closer to the heat source yielded a greater temperature gradient. (3) With an increase in the heat source temperature, the peak value of temperature variation shifts downward in the direction of seepage. The area with temperature change ∆T ≥ 0.5 °C is defined as the area with obvious temperature change. For heat source temperatures of 30 °C, 35 °C, and 40 °C, the areas exhibiting discernible temperature changes upstream in the sandbox measured 0.43 m, 0.37 m, and 0.44 m, respectively. On the other hand, the areas with noticeable temperature changes downstream measured 0.55 m, 0.45 m, and 0.70 m, respectively.

Different from the approach adopted by Li et al. [71], which involved using a U-shaped tube as the heat exchanger and constant-temperature water tanks on both sides of the sandbox to facilitate horizontal seepage flow, Wang et al. [72] developed a 1.5 × 1.5 × 1.0 m cuboid adiabatic sandbox with horizontal seepage, with the schematic diagram shown in Figure 15. To facilitate water seepage, two water chambers with a width of 0.15 m were constructed on opposite sides of the sandbox. The test rig consisted of three main parts: experimental sandbox, auxiliary water system, and data acquisition system. In the middle of the sandbox, a 200 W electric heating rod was positioned as the heat source. The operation of the heating rod was regulated by a timer, while its power was controlled using a voltage stabilizer. The auxiliary water system consisted of hot and cold water tanks, as well as an overflow tank, which were utilized to achieve the desired water temperature and pressure. A wired hose connected the overflow tank to the water chambers, allowing for water flow. The seepage velocity was found to vary in accordance with the height of the overflow tank.

Wang et al. [72] performed sandbox experiments encompassing dry, unsaturated, saturated, and seepage sand conditions. Figure 16 presents the heating temperature and capacity of the electric heating rod under dry, unsaturated, saturated, and seepage sand conditions. For the electric heating rod with a heating power of 30 W/m, the heating temperatures could be 76.4 °C, 34.1 °C, and 28.7 °C under the dry, unsaturated, saturated, and seepage conditions, respectively. In addition, for a certain soil temperature, the heating capacity was proportional to the soil moisture content. As shown in Figure 16a, upon activation of the electric heating rod, the soil temperature initially increased rapidly before gradually stabilizing. Approximately 15 min later, the temperature growth rate remained below 0.1 °C/min and gradually stabilized with soil temperature increases of 1.5 °C and 2.0 °C under the seepage and saturated conditions. At a heat release rate of 10 W/m, a heat balance was quickly achieved between the electric heating rod and the soil in both seepage and saturated conditions. Following 8 h of heat release, the soil temperatures increased by 19.8 °C, 6.5 °C, 3.7 °C, and 3.2 °C under the dry, unsaturated, saturated, and seepage conditions, respectively. These results highlight the significant influence of soil water content on heat transfer within the soil. Soils with higher water content, particularly in seepage scenarios, exhibit a greater heat dissipation capacity and enable faster attainment of stable heat transfer.

Based on the aforementioned experiments, it can be inferred that the horizontal seepage experiment provides a relatively accurate representation of groundwater seepage in real-life projects. The seepage velocity can be regulated by manipulating the water levels in the high and low water tanks. However, in comparison to the vertical seepage experiment described in Section 2.3.1, the construction, operation, and measurement requirements of the test rig in the case of horizontal seepage are more intricate.

In conclusion, these seepage experiments offer valuable insights into the heat transfer characteristics of borehole systems in the presence of seepage. By conducting laboratory simulations of vertical or horizontal seepage, the study establishes that seepage conditions significantly improve the heat transfer efficiency of the borehole group. Moreover, it is observed that heat is preferentially transferred in the direction of seepage, effectively mitigating the cold or heat accumulation within the borehole group.

2.4. Multi-Layer Soil Experiment

In practical projects, boreholes typically have depths ranging from 50 to 120 m, and the soil is commonly composed of multiple layers. Different soil thermal properties also affect the heat transfer of boreholes in each layer of soil. To investigate the heat transfer characteristics in different soil layers, sandbox experiments involving multi-layered soil systems have been conducted.

Wang et al. [61] constructed a cuboid sandbox bench of 8 × 6 × 5.5 m with adiabatic boundaries. The sandbox featured a three-layered soil structure, a sloping bottom equipped with a drainage system, a 0.5 × 0.5 m outlet on the lower side, and a pumping well. The U-type borehole adopted a high-density polyethylene (HDPE) tube with a diameter of 32 mm and was connected to a heating system (a field thermal response tester). The platform was equipped with a monitoring system to measure the moisture and temperature of the soil. As shown in Figure 17, the vertical arrangement incorporated seven temperature sensors situated at distances from 0 m to 5 m near the borehole, while three additional groups of temperature sensors were positioned horizontally at depths of 2 m, 4 m, and 5 m.

Through the sandbox experiment conducted by Wang et al. [61], the characteristics of heat transfer in different soil layers were investigated. The experiment commenced with an initial soil temperature of 13 °C, a heating power of 255 W, and a water flow rate of 9.6 m³/d, lasting for a duration of 10 days. The sandbox is originally filled with homogenous clay with a thermal conductivity of 0.52 W/(m·K). When the low-layer soil, initially having a thermal conductivity of 0.52 W/(m·K), was replaced with soil having a thermal conductivity of 4.23 W/(m·K), the outlet water temperature decreased from 37.07 °C to 35.05 °C, resulting in a notable 16.19% increase in the average heat transfer coefficient. Similarly, substituting the middle-layer soil with soil having a thermal conductivity of 4.23 W/(m·K) caused the outlet water temperature to decrease from 35.11 °C to 31.81 °C, accompanied by a 7% increase in the average heat transfer coefficient. Likewise, replacing the upper-layer soil with soil possessing a thermal conductivity of 4.23 W/(m·K) led to a decrease in the outlet water temperature from 31.82 °C to 28.95 °C, resulting in a 14% increase in the average heat transfer coefficient. Therefore, the soil properties in the lower layer exhibited the most significant influence on the overall heat transfer of GHE, whereas the impact of the middle-layer soil was relatively weaker.

In conclusion, this study highlights the significant role played by various soil layers in determining the overall heat transfer efficiency of GSHPs. When the thermal conductivity of the soil increases, there is an enhancement in the heat transfer capacity of the GSHPs, and the impact of the lower soil layers on heat transfer is particularly noteworthy. These findings demonstrate the importance of understanding and considering the thermal properties of different soil layers in designing and optimizing GSHPs.

2.5. Summary

The borehole sandbox experiments aim to investigate the impact of various factors such as tube type, borehole number, seepage, and multi-layer soil on the heat transfer performance of boreholes. Regarding tube type, the spiral tube demonstrates superior heat exchange performance compared to a single U-tube with the same drilling depth, and the heat transfer performance can be enhanced by enlarging the tube spacing and reducing the spiral pitch. Operation parameters such as flow velocity and inlet temperature also play a role in influencing heat transfer efficiency and soil temperature. The soil near the borehole exhibits rapid heat transfer and significant temperature gradients. In terms of soil parameters, seepage and thermal properties of soil layers also affect the heat transfer performance of boreholes. In different hydrogeological conditions (dry soil, saturated sand, and groundwater seepage), the thermal conductivity increases, and the thermal influential area expands successively, which is proportional to the thermal diffusivity. The thermophysical property of the low-layer soil has a greater impact on the overall heat transfer coefficient of the borehole compared to the up-layer and the middle-layer soils. The detailed research of the borehole sandbox experiment is shown in

Table 1. Research status on the boreholes sandbox experiment.

Type of Study	Years	Researchers	Sandbox Form	Borehole Form	Research Results
Single borehole experiments	2012	Wang et al. [47]	A stainless steel cylindrical sandbox, diameter: 0.8 m, length: 5 m, isothermal boundary, no seepage	Horizontally arranged U-shaped tubes with two different diameters	(1) The heat transfer performance of the spiral tube is better than that of the single U-shaped tube. (2) A larger leg space of the U tube and a smaller tube pitch of the double spiral tube can enhance the heat transfer performance.
	2004	Diao et al. [48]	Two cuboid sandboxes of 2 × 1.2 × 0.8 m and 4 × 1 × 1 m, isothermal boundaries, no seepage	A single horizontally arranged U-shaped tube	(1) The temperature variation near the U-shaped tube is large. As the distance from the U-shaped tube increases, the soil temperature growth rate decreases. (2) The temperature gradient of soil closer to the U-shaped tube is larger.
	2018	You et al. [62]	A cube sandbox of 1 × 1 × 1 m, adiabatic boundary, no seepage	A single vertically arranged U-shaped tube	After 24 h continuous cooling, the soil temperature decrease is about 5 °C and 0.5 °C at positions 0 mm and 500 mm away from the U-shaped tube, respectively.
	2011	Beier et al. [69]	An 18 × 1.8 × 1.8 m cuboid wooden frame sandbox, isothermal boundary, no seepage	A single horizontally arranged U-shaped tube	(1)The soil temperature at positions 0 cm, 24 cm, 44 cm, 65 cm, and 85 cm away from the borehole wall are obtained. (2) The test data can be used for model validation.
Borehole group experiments	2016	Zhang et al. [59]	A cuboid sandbox of 0.6 × 0.6 × 1 m, adiabatic boundary, seepage	4 horizontally arranged electric heating rod groups	(1) The soil temperature rise in the saturated sand condition is lower than that of the dry sand condition, but the heat storage capacity is improved. (2) The increase of seepage velocity accelerates the process to reach a stable state, and the stable temperature is lower. (3) Under the influence of seepage, the heat accumulation phenomenon downstream of the tube group is more obvious.
	2018	You et al. [70]	A 1 × 1 × 1 m cube sandbox (adiabatic boundary), no seepage	5 horizontally arranged U-shaped tubes	(1) After 24 h of cooling, the soil temperature decreases at positions 0 mm and 25 mm away from the heat exchange tube are about 6 °C and 1.5 °C, respectively. Experimental results are used to verify the borehole group model.
Vertical seepage experiments	2008	Liu et al. [60]	A 0.8 × 0.6 × 0.77 m cuboid sandbox (adiabatic boundary)	A single horizontally arranged electric heating rod	(1) Seepage can improve the heat transfer of the borehole, and the soil temperature change is smaller than that in the saturated condition. (2) The temperature field is no longer distributed in a circular shape under the seepage condition. The heat transmission distance increases in the seepage direction. (3) The greater the seepage velocity, the smaller the soil temperature variation and the greater the heat transmission distance.
	2016	Zhang et al. [59]	A 0.6 × 0.6 × 1 m cuboid sandbox, adiabatic boundary	A single horizontally arranged electric heating rod	(1) The water carries heat to move in the seepage direction. (2) A higher seepage velocity brings better soil thermal stability. (3) The larger power of the heating rod leads to faster and higher soil temperature variations.
Horizontal seepage experiments	2017	Li et al. [71]	A 2.5 × 2.5 × 1 m cuboid sandbox, isothermal boundary	A single horizontally arranged U-shaped tube	(1) There is seepage in saturated soil, and the temperature change downstream is larger. (2) The faster seepage velocity will lead to a larger dense area and downstream offset of isotherms.
	2019	Wang et al. [72]	A 1.5 × 1.5 × 1 m cuboid sandbox (adiabatic boundary)	A single longitudinally arranged electric heating rod	(1) The soil with a higher water content has a higher heat dissipation capacity, especially when seepage occurs. (2) The heat transfer rate decreases from unsaturated soil to saturated soil and dry soil.
Multi-layer soil experiments	2019	Wang et al. [61]	A 2 × 1.5 × 2 m cuboid sandbox (isothermal boundary)	A single longitudinally arranged U-shaped tube group	The thermal conductivities of the upper, middle, and lower layers have different effects on the heat transfer of each layer; the lower layer has the most significant effect on the heat transfer performance.

.

3. Energy-Pile Sandbox Experiment

An energy pile refers to a type of building pile foundation that incorporates heat exchange tubes buried within it [73]. Compared to a borehole, an energy pile typically has a larger diameter and shorter length, resulting in a higher heat transfer rate. Consequently, energy piles have gained significant attention in recent years. Figure 18 depicts the schematic diagram of an energy pile [74]. The heat transfer characteristics of energy piles not only impact the mechanical properties of the piles but also have a significant influence on the energy performance of the GSHP system [75,76]. Therefore, the mechanical performance [56,57,58], thermal performance [77,78,79,80,81], and coupled thermal-mechanical performance [82,83,84] are carried out by sandbox experiments.

3.1. Mechanical Performance Testing of Energy Piles

The heat transfer within energy piles has a direct impact on the bearing capacity of the building pile foundation, which is of paramount importance to the structural integrity of buildings. Hence, sandbox experiments on energy piles are conducted to examine the mechanical performance of the building pile foundation. Many experiments have specifically concentrated on investigating the effects of thermal cycles and pile head loads on the settlement of the pile foundation, axial stress in the pile body, and the frictional resistance between the pile and soil.

Nguyen et al. [56] conducted a mechanical test of an energy pile model using a sandbox with a diameter of 584 mm and a height of 900 mm under long-term thermal cycles to investigate the effect of the thermal cycles and the load on the settlement and forces in the pile, as is shown in Figure 19. In the experiment, the piles were subjected to loading and unloading with pressures ranging from 10% to 100% of the maximum load (approximately 50 N) with a 10% interval. Each loading phase lasted for one hour, and the process was repeated until the pile head settlement reached 10% of the pile diameter. Simultaneously, the researchers conducted 30 thermal cycles under four different conditions (10%, 20%, 40%, and 60% of the maximum load), with each cycle lasting for 24 h to examine the bearing characteristics of the piles under thermal cycling. The results demonstrate that the settlement of the pile, both under mechanical loads with and without thermal cycles, increases as the load magnitude increases. The heating process of the pile does not cause significant movement of the pile body, while the cooling process induces pile settlement. Both processes are irreversible, except when the load is zero, as illustrated in Figure 20. Additionally, the settlement rate of the pile experiences substantial changes in the first 10 cycles and gradually stabilizes afterward. Moreover, the axial force along the pile body increases with an increasing number of thermal cycles.

Kalantidou et al. [57] conducted a sandbox experiment to study the mechanical properties of energy piles in 2012, which focused on the effect of applied loads and temperature on the settlement of energy piles. Figure 21 illustrates the schematic diagram of the experiment, which involved the use of a cylindrical steel sandbox with a diameter of 570 mm. The experimental setup consisted of a closed aluminum tube placed outside the model pile, while a metal U tube with a total length of 800 mm was placed inside the pile. The researchers aimed to investigate the relationship between the load and displacement under different head loads. In the experiment, assuming the internal temperature of the pile remained constant, four different loads (0 N, 200 N, 400 N, and 500 N) were applied. Each test underwent thermal cycles, starting from 25 °C, increasing to 50 °C, and then returning to 25 °C, repeating this cycle twice.

Figure 22 demonstrates the pile head settlement under the axial load with and without thermal cycles, and the results show that pile settlement is influenced by the coupling of temperature and load. The experimental results indicate that increasing the head load leads to a larger settlement of the pile body. The relationship between displacement and the temperature of the pile body during the first heating/cooling cycle resembles the thermal expansion and shrinkage curves of the pile body, respectively. Furthermore, the pile body exhibits a normal thermal response when the head load is below 40% of the maximum load. However, if the head load exceeds 40% of the maximum load, the settlement of the pile body accumulates under thermal cycles, potentially leading to safety hazards.

Qian et al. [58] conducted an energy pile sandbox experiment using a cuboid sandbox measuring 2 × 2 × 1.5 m to investigate the soil mechanical properties of energy piles. Figure 23 illustrates the diagram of the pile-soil sandbox experiment. The experiment utilized a single U-shaped energy pile with a diameter of 200 mm and a length of 1200 mm, subjected to a maximum load of 4000 kN. Throughout the study, the room temperature was maintained at 11 °C, and the heating temperature was set at 50 °C. The thermal cycling commenced after the completion of loading, and the settlement of the pile reached a stable state.

The internal forces within the soil and energy piles are primarily influenced by the temperature changes during the thermal cycle and the relative displacement between the soil and the energy piles. The experimental results reveal that the water pressure within the soil increases as the temperature rises during the thermal cycle. Figure 24 displays the displacement of the soil and energy piles over time. During the thermal cycle, the soil initially bulges upwards, followed by settlement as the temperature decreases, eventually reaching a lower height than its initial state. The pile body also experiences settlement, but its final settlement is less than that of the soil. This difference in settlement can lead to negative frictional resistance between the pile and the surrounding soil.

The conducted experimental results on the mechanical performance of energy piles demonstrate that the settlement, internal stress, and frictional resistance of energy piles are influenced by both pile head loads and thermal cycles, thereby exploring the service life and safety of energy piles. However, these experiments have certain limitations as they did not consider the coupled effects of various parameters. Experiments conducted by Nguyen et al. [56] and Kalantidou et al. [57] primarily focused on the influence of load magnitude, while experiments conducted by Qian et al. [58] emphasized the effect of relative displacement between the pile and soil. Therefore, further investigations considering the coupled effects of multiple parameters are necessary.

3.2. Thermal Performance Testing of Energy Piles

The thermal performance of energy piles plays a crucial role in determining the energy consumption of GSHP systems. However, the quantitative heat transfer laws derived from thermal performance experiments conducted on boreholes cannot be directly applied to energy piles due to their distinct size and structural differences. Consequently, numerous researchers have conducted sandbox experiments to analyze the impacts of various parameters, such as heat medium temperature, flow rate, intermittent operation, and geometric factors, on the thermal performance of energy piles. These experiments aim to provide insights into optimizing the thermal performance and energy efficiency of energy pile-based GSHP systems.

Kramer et al. [77] designed a single U-shaped energy pile to investigate its heat transfer characteristics. Figure 25 illustrates the schematic diagram of the sandbox used in the experiment. To achieve a consistent static stress in the soil surrounding the cast-in-place energy pile, sedimentation was employed.

The temperature of the soil around the energy pile is influenced by the pile spacing and the type of energy pile. Figure 26a displays the soil temperature at various radial distances from the pile over a period of seven days. The heat transfer characteristics of a single U-shaped energy pile can be approximated as axisymmetric. Figure 26b demonstrates the impact of the circulating water flow rate on the heat transfer capacity of the pile. The results indicate that as the water flow rate increases, the heat transfer rate of the energy pile improves. However, excessively high flow rates can result in increased power consumption, which may negatively affect the seasonal performance of the system. Therefore, a balance must be struck between achieving optimal heat transfer and avoiding excessive power consumption.

Yang et al. [79] experimented on the factors affecting the heat transfer efficiency of a spiral-coil energy pile, and a schematic diagram of the experimental system is shown in Figure 27. In the experiment, a wooden sandbox measuring 0.8 × 0.8 × 1.2 m was utilized, with 25 mm thick insulation materials covering both the top and bottom of the box. The spiral coil was constructed using a polyurethane tube with an outer diameter of 6 mm and an inner diameter of 5 mm, which was wrapped around the model of the pile foundation. The pile model itself was made of a polyvinyl chloride tube with an outer diameter of 200 mm and an inner diameter of 190 mm.

Figure 28 presents the influences of various factors on the heat transfer rate of the spiral-coil energy pile. It is evident that implementing a reasonable intermittent operation mode and reducing the spiral pitch can effectively enhance the heat transfer efficiency of energy piles, leading to improved performance of GSHP systems. Moreover, increasing the inlet water temperature has a positive effect on the heat transfer rate of the heat exchange tube, demonstrating an approximately linear relationship with the inlet water velocity.

The changes in temperature and moisture content of the soil around the energy pile during the cooling/heating cycles were mainly studied in an energy pile sandbox experiment constructed by Faizal et al. [81], and the layout of the experiment is shown in Figure 29. The rectangular glass sandbox used in the experiment has dimensions of 0.56 × 0.56 × 0.3 m. The energy pile employed in the study has a diameter of 25 mm and a total length of 264 mm. Throughout the experiment, the room temperature was maintained at approximately 20 °C, and the sandbox was designed to be adiabatic, minimizing heat exchange with the surrounding environment.

The moisture content and temperature of the soil around the energy pile varies with circulation, as shown in Figure 30. During the heating phase, water moves away from the energy pile, while during the cooling phase, it moves toward the pile. This movement of water results in the drying and wetting of the soil, respectively. In comparison to monotone heating/cooling, cyclic heating/cooling leads to less variability in soil temperature and volumetric water content, which is advantageous for the long-term operation of energy piles. Figure 30a demonstrates the most significant changes in soil temperatures and volumetric water content near the pile, specifically at a radial distance of 20 mm. These changes decrease as the radial distance from the pile increases. Additionally, the volumetric water content adjacent to the pile exhibits a decrease during continuous thermal cycling. During cyclic heating/cooling, the volumetric water content shows an irreversible response to changes in soil temperature, although this response diminishes with increasing radial distance from the pile.

The thermal performance of energy piles is closely related to their arrangement, types, and water flow rates. The research conducted by Kramer et al. [77], Yang et al. [79], and Faizal et al. [81] has revealed the spatial distribution of heat transfer in energy piles, as well as the heat transfer effectiveness of different energy pile types under various operating conditions.

3.3. Thermal-Mechanical Coupled Performance Testing of Energy Piles

The thermal and mechanical performances of energy piles are interconnected during their operation. Therefore, it is crucial to examine the thermal-mechanical coupled performance of energy piles. This investigation should encompass factors such as the type of heat exchange tube and the temperature of the heat medium, which influence the thermal-mechanical coupling. Additionally, studying the temperature distribution in the pile-soil system and the lateral friction distribution between the pile and soil are essential aspects of understanding the thermal-mechanical behavior of energy piles.

Wang et al. [82] conducted a sandbox experiment to investigate the heat transfer rates and mechanical properties of energy piles using different heat exchange tube configurations, namely single U-shaped, W-shaped, and spiral tubes. The schematic diagrams of the energy piles with these different heat exchange tubes are depicted in Figure 31. The sandbox used in the experiment had dimensions of 3 × 2 × 1.75 m and was placed in a closed environment where the temperature difference was maintained within 2 °C throughout the day. The fluid temperatures during the experiment for heating, indoor air, and cooling conditions were 57 °C, 20 °C, and 5 °C, respectively.

The heat exchange performance and displacement of energy piles exhibit variations depending on the style of tubes used. Figure 32 illustrates the measured heat exchange performance of energy piles with different tube configurations. It is observed that, under heating and cooling conditions, the W-type energy pile exhibits the highest heat exchange rate, while the spiral-type energy pile demonstrates a lower heat exchange rate compared to the W-type energy pile. The single U-shaped energy pile consistently exhibits the lowest heat exchange rate among the three configurations, primarily due to its shorter heat exchange tube. Figure 33 presents the mechanical performances of energy piles with different tube configurations under heating and cooling conditions. It can be seen that, under both heating and cooling conditions, the W-shaped energy pile experiences the largest changes in thermal strain and pile head displacement, while the U-shaped energy pile shows the smallest changes in these parameters.

Bao et al. [83] conducted a steel sandbox experiment with dimensions of 2.5 × 2.5 × 2.0 m to investigate the mechanical and thermal performances of energy piles equipped with double U-shaped tubes. The soil in the sandbox was divided into 18 layers, each with a height of 100 mm. To ensure proper contact between the soil layers, they were carefully trimmed. In the experiment, the inlet water temperature was maintained at a constant value of 55 °C. The initial soil temperature and indoor temperature were both set to 31 °C. The thermal cycle consisted of a 48 h heating process at a constant temperature, followed by a 24 h recovery process during which natural cooling occurred. Three thermal cycles were performed in the sandbox, each with a different water flow rate: 0.11 m³/h, 0.32 m³/h, and 0.51 m³/h, respectively.

The heat transfer performance of energy piles is primarily influenced by water flow and the distance from the pile, whereas pile settlement remains essentially unaffected by temperature variations. Figure 34a illustrates the heat transfer rate of the energy pile under different water flow rates. It is observed that as the water flow rate increases, the heat exchange rate of the energy pile improves, indicating a more efficient heat transfer process. Figure 34b presents the variation in soil temperature at different distances from the pile surface. During the heat transfer process, the soil temperature surrounding the pile primarily changes within the first 48 h and subsequently decreases gradually with an increasing distance from the pile. Figure 34c displays the displacement of the pile head observed in this experiment. The displacement demonstrates a declining trend with fluctuations during the thermal cycle, and no clear relationship is evident between the displacement and the temperature.

Kramer et al. [84] conducted a single U-shaped energy pile sandbox experiment to investigate the heat transfer properties and mechanical properties of energy piles. The experiment was performed using a steel sandbox with a bottom size of 1.83 × 1.83 m. Figure 35 provides the schematic diagram of the experiment setup, where the sandbox had a total height of 2.13 m and a buried depth of 1.22 m. To ensure uniform sand density, the sand was screened using a sand control device measuring 0.76 × 0.76 m. During the experiment, the load on the pile head was monitored using a load cell with a capacity of 24.5 kN, while the displacement of the pile head and bottom was measured using a Linear Variable Differential Transformer (LVDT). The water flow velocity and inlet temperature were set at 0.135 m/s and 40 °C, respectively. The initial average soil temperature in the sandbox was 19 °C, and thermocouples were used with a sampling interval of 30 s for temperature measurements.

The heat transfer in energy piles exhibits distinct spatial characteristics, as depicted in Figure 36, which illustrates the soil temperature distribution around the pile after seven days of heat transfer. The soil temperature near the energy pile rapidly increases and approaches the water temperature. The heat transfer distance in the radial direction is three times larger than that in the vertical direction. Mechanical load tests were conducted on the energy pile to measure the continuous displacement of the pile head with a load increment of 0.1 kN before and after the heat transfer experiment. Figure 37 illustrates the results, showing a slight increase in the ultimate load of the energy piles after seven days of heat transfer. Additionally, the head stiffness of the pile decreases during the initial stage of load application.

Wu et al. [85] carried out a sandbox experiment to investigate the heat transfer and displacement behavior of energy piles. In the experiment, the model piles used were constructed using steel tubes with an outer diameter of 23 mm, an inner diameter of 21 mm, and a length of 550 mm. These model piles were buried into the soil, with only 450 mm of their length submerged. At the center of the model pile, a U-shaped copper tube with an inner diameter of 2 mm was placed, and it was connected to a water tank equipped with temperature control. The remaining space within the pile was filled with epoxy resin to ensure stability and integrity. Additionally, a steel plate measuring 100 × 40 × 3 mm was utilized as the pile cap, which was welded to the pile head. This steel plate provided support and reinforcement for the pile structure.

Temperature variations significantly influence the properties of the soil surrounding energy piles, resulting in a positive correlation between water pressure and soil temperature. Figure 38 compares the pile head displacements for single energy piles during heating and cooling, where the pile head displacements are smaller under heating conditions compared to cooling conditions. The adjacent unheated piles form a shielding effect on the energy piles, resulting in the displacement of adjacent piles being smaller than that of a single energy pile. Furthermore, the pile cap restricts the movement of the energy pile, helping to reduce the differential settlement between the energy pile and the adjacent piles.

The thermal-mechanical coupling sandbox experiments on energy piles have been conducted to explore the interrelationship between the thermal and mechanical performance of energy piles. The current experiments focus on investigating the influence of heat transfer on the displacement, ultimate load capacity, stiffness, and other mechanical properties of energy piles. However, The understanding of the complex interactions between thermal and mechanical behavior in energy piles is still limited. Further research is needed to investigate the coupling mechanisms and the influence of various factors, such as groundwater flow, soil properties, and pile geometry.

3.4. Summary

The aforementioned sandbox experiments focused on studying the mechanical, thermal, and thermo-mechanical performances of energy piles without considering seepage. Regarding the mechanical performance, the settlement of energy piles during thermal cycles increases as the head load of the energy piles grows [56,57], and negative frictional resistance is observed between the soil and energy piles [58]. Concerning thermal characteristics, the heat transfer process of energy piles is approximately axisymmetric, with the soil temperature being higher closer to the energy piles [77,81,84]. The operating conditions and spiral-coil pitch of energy piles [79] are known to impact the heat transfer rate. In terms of thermal-mechanical coupling, experimental results indicate that both soil and pore water pressures increase with higher fluid temperatures inside the tube [83,85]. A comprehensive overview of the research progress on energy-pile sandbox experiments is provided in

Table 2. Research status on the energy piles sandbox experiment.

Type of Study	Year	Researchers	Sandbox Form	Buried Form	Research Result
Mechanical performance experiments of energy pile	2017	Nguyen et al. [56]	A diameter of 0.548 m, height of 0.9 m cylindrical sandbox	Single U-shaped energy pile	(1) The energy pile with head load will undergo irreversible settlement after thermal cycling. (2) The greater the head load, the larger the irreversible settlement. (3) The irreversible settlement caused by the first thermal cycle is the largest. (4) The axial force along the pile body increases with the growth of the number of thermal cycles.
	2012	Kalantidou et at. [57]	A cylindrical steel sandbox with a diameter of 0.57 m and a height of 1 m	Single U-shaped energy pile	(1) The settlement of the pile gets reduced with the decline of the load. (2) When the mechanical load is less than 40% of the maximum load, the energy pile performs a normal thermal response. (3) Once the mechanical load is greater than 40% of the maximum load, the irreversible settlement of the energy pile will gradually accumulate.
	2020	Qian et al. [58]	2 × 2 × 1.5 m cuboid sandbox	Single U-shaped energy pile	(1) The excess static pore water pressure in the soil increases with the growth of temperature within a certain range. (2) After thermal cycling, the soil will settle due to the thermal consolidation effect. (3) During the thermal cycle, the total settlement of the pile body is less than that of the soil around the pile, so the soil will produce negative frictional resistance to the pile body.
Thermal performance experiments of energy pile	2015	Kramer et al. [77]	1.83 × 1.83 × 2.13 m cuboid wooden sandbox	Single U-shaped energy pile	(1) The heat transfer of a single U-shaped energy pile is approximately an axisymmetric process. (2) The heat transfer rate of the energy pile increases with the growth of the water flow velocity.
	2016	Yang et al. [79]	0.8 × 0.8 × 1.2 m cuboid wooden sandbox	Spiral energy pile	(1) Increasing the inlet water temperature and velocity improve the heat transfer rate of the heat exchange tube. (2) The heat transfer rate of the energy pile is higher under intermittent operating conditions. (3) Total heat transfer of the heat exchange tube increases with the decline of the pitch.
	2021	Faizal et al. [81]	0.56 × 0.56 × 0.3 m rectangular glass sandbox	Single U-shaped energy pile	(1) The moisture content and temperature of soil within 20 mm from the pile body change significantly under the influence of the thermal cycle, and the change is more obvious for the soil closer to the pile. (2) The soil around the pile is dry and wet under the heating and cooling cycles, respectively.
Thermal-mechanical coupled performance experiments of energy piles	2017	Wang et al. [82]	3.0 × 2.0 × 1.75 m cuboid sandbox	Single U-shaped, W-shaped, spiral energy pile	The thermal strain change, pile top displacement, and heat exchange rate of the W-shaped energy pile are the largest, while those of the single U-shaped energy pile are the lowest.
	2020	Bao et al. [83]	2.5 × 2.5 × 2.0 m cuboid steel sandbox	Double U-shaped energy pile	(1) The heat exchange rate of the energy pile is inversely proportional to the water flow velocity and temperature difference between the tube inlet and outlet. (2) The temperature of the soil around the pile changes mainly during the first 24 h of the operation. (3) The pressure on the soil is mainly affected by the temperature and is positively related to the temperature.
	2014	Kramer et al. [84]	1.83 × 1.83 × 2.13 m rectangular steel sandbox	Single U-shaped energy pile	(1) The soil far from the energy pile needs a long time to get thermal interaction. (2) The heat transfer distance in the radial direction is three times that in the vertical direction. (3) The ultimate load of the energy pile increased slightly, and the pile head stiffness at the initial stage of load application decreased after thermal applying loading for seven days.
	2018	Wu et al. [85]	0.8 × 0.4 × 0.75 m cuboid sandbox	Single U-shaped energy pile	(1) The pore water pressure is proportional to the temperature. (2) The displacement of the pile top during the heating process is less than that during the cooling process. (3) The shielding effect of the unheated energy pile and the pile cap both restrict the displacement of the energy pile.

, detailing the specific areas of investigation.

4. Summary and Outlook

4.1. Summary

As an energy-efficient and low-carbon technology, GSHP holds significant promise in achieving carbon neutrality within the building sector. One crucial element of GSHP systems is the GHE, which has undergone extensive investigation through sandbox experiments. These experiments serve as an essential approach to studying heat transfer characteristics and validating simulation results. By categorizing existing sandbox experiments, valuable insights can be gained to enhance GHE performance and facilitate the rational development of GSHP systems. The experiments are systematically classified into two categories: borehole sandbox experiments (including single borehole, borehole group, and seepage) and energy pile sandbox experiments (encompassing mechanical performance, thermal performance, and thermal-mechanical coupled performance). A comprehensive summary is provided below:

(1)

Briefly, the borehole sandbox experiments can be categorized into single-tube and tube group experiments, with a significant focus on single-tube experiments, particularly those without seepage. Single-tube experiments are straightforward and primarily investigate the impact of heat exchange tube form, circulating water temperature, and flow rate on the borehole’s heat transfer capacity and the distribution of soil temperature. Tube group experiments, on the other hand, typically utilize heating wires or metal tubes with circulating water as heat exchange elements. The use of electric heating wires allows for the emulation of heating and cooling conditions with constant powers, resulting in simple experimental devices and operations. Metal tubes with internal circulating water can replicate the actual internal heat transfer conditions of boreholes, thereby reflecting the influence of circulating water temperature and flow rate on the heat transfer within the borehole group. However, the necessity for a water-circulating system introduces a relatively complex experimental setup. Tube group experiments aim to study the soil temperature distribution around the tube group under constant heat flow and seepage conditions, as well as the impact of heat disturbance from the tube group with known circulating water temperature on the soil temperature field of the tube group without seepage;

(2)

Seepage experiments and layered soil experiments are conducted to study the heat transfer effect of boreholes in actual complex environments. Seepage experiments are mainly categorized into vertical seepage and horizontal seepage. In the vertical seepage test, the flow of water in the soil occurs naturally due to gravity, making it a relatively straightforward setup. The horizontal seepage test creates seepage pressure by utilizing a height difference between high and low water tanks. These seepage experiments mainly focus on examining the impact of seepage velocity on the heat transfer of a single tube, the thermal disturbance within the tube group, and the distribution of soil temperature. Those layered soil experiments study the influence of the thermal conductivity of geotechnical materials in different layers on the overall heat transfer;

(3)

Energy piles are building pile foundations embedded with heat exchange tubes, of which mechanical performance, thermal performance, and thermal-mechanical coupled performance are the three main research directions. In addition, most of the studies did not consider the seepage factor. Related studies are mainly concerned with sedimentation due to load and thermal cycle. The effect of the heat medium in the tube, the structure of the pile body, and the operating parameters are also discussed to investigate the heat transfer rate and temperature distribution of energy piles under different operation conditions.

4.2. Existing Problems

Although sandbox experiments have been widely conducted, there are several remaining challenges that need to be addressed.

(1)

Establishing similarity experiments for practical projects is difficult due to the long depth and small diameter of vertical boreholes. As a result, the experimental data may not accurately depict the heat transfer characteristics of ground heat exchangers (GHEs) in real-world applications;

(2)

The measurement accuracy of sandbox experiments is limited. The small size of the sandboxes used in model experiments restricts the heat duty of GHEs and hampers accurate temperature measurements. Consequently, the temperature changes and heat duty obtained from sandbox experiments may not be directly applicable to practical projects;

(3)

Controlling the boundary conditions in a sandbox experiment is challenging. Adiabatic boundaries are typically achieved by covering the outer surface of the sandbox with thermal insulation materials, while isothermal boundaries are realized using constant temperature water baths or controlled indoor air temperatures. However, these methods have limitations in precisely maintaining idealized adiabatic and isothermal boundaries;

(4)

Soil thermophysical property tests suffer from low accuracy. The sampling process can introduce unavoidable errors as soil thermophysical properties are influenced by soil compactness.

By addressing these challenges and improving the experimental techniques, a more accurate understanding of the heat transfer characteristics and performance of GHEs in practical projects can be achieved.

4.3. Perspectives

The GHE sandbox experiment can not only study the heat transfer mechanism but also verify the heat transfer model. This is significant for the design and application of different types of GHEs under different conditions. Combined with new technologies, there will be more and more studies using GHE sandbox in the future, with some possible aspects as follows:

(1)

Complex environmental conditions, such as three-dimensional seepage and multi-layered soil, pose challenges to current seepage experiments. Most current seepage experiments investigate the horizontal seepage process, while there is a lack of seepage experiments for shallow boreholes, including both horizontal seepage and vertical seepage caused by rainfall. Current layered soil experiments are mainly performed with two to three layers of soil, while the multi-layered soil and inclined-layered soil experiments in practical projects have not been considered;

(2)

Novel composite energy geostructures: Emerging energy geostructures, such as energy piles, energy walls, and energy tunnels, incorporate heat exchange tubes within building infrastructures, offering the advantage of reducing initial investment costs and occupying less space compared to traditional vertical boreholes. While some types of energy geostructures have been widely adopted globally, current research primarily focuses on vertical boreholes, with a limited investigation into energy geostructures and composite energy geostructures combined with boreholes. Future studies should aim to bridge this research gap and explore the potential of these innovative composite energy geostructures;

(3)

The integration of new materials and new technologies presents opportunities for further enhancing the performance of energy geostructures. For example, the use of nanofluids can enhance heat transfer within the heat exchange tubes, while incorporating phase change materials between the heat exchange tube and the soil can mitigate soil temperature variations and improve heat exchange efficiency. To ensure the practical application of these new materials and technologies, relevant sandbox experiments focusing on nanofluids and phase change materials are necessary to verify the feasibility and characterize the performance. These experiments will provide valuable insights into the potential benefits and challenges associated with these advancements.

GSHP technology plays a vital role in improving indoor comfort while striving for carbon neutrality and sustainable development. GHE sandbox experiments will contribute to the wider adoption and popularization of this green technology. This review serves as a valuable reference for the design, construction, and characterization of various sandbox test setups in future research endeavors.