1. Introduction
In 1865, the following was proposed by Italian physicist Carlo Marangoni [
1]: when two liquids are in contact, the liquid with strong surface tension will pull the liquid with weak surface tension a phenomenon later academically named the Marangoni effect.
The working fluids often used as two-phase heat transfer devices, such as pure water, ethanol, acetone, etc., are generally pure material fluids.
Figure 1 shows that the surface tension of ordinary fluids (ordinary liquid) will gradually decrease with the increase in operating temperature. In 1973, scholars Vochten and Petre [
2], and in 2004, scholars Abe et al. [
3] measured the surface tension of different alcohol aqueous solutions, and found that alcohol aqueous solutions with high carbon chain (carbon chain number ≥ 4) have a different surface tension curve, such as the self-rewetting fluid of the red line in
Figure 1. In particular, when the temperature increases to the β point, the surface tension of the aqueous solution will increase with the temperature. With an increase to the γ point domain, it can be observed that the (β-γ) domain has a reverse gradient of surface tension that is reversed by the temperature rise. Fluids with this characteristic are different from ordinary fluids and are called self-rewetting fluids (SRF).
When the self-rewetting fluid enters the reverse surface tension gradient region (β-γ), due to the Marangoni effect, it means that the fluid at the vapor–liquid two-phase interface will flow from the place with low surface tension to the place with high surface tension, and the temperature will be lower. The liquid is brought to a place with a higher temperature and thus wets the surface with a high temperature, which has to delay the occurrence of drying out and enhance the heat transfer efficiency.
The working fluid in this research is intended to be a self-rewetting fluid, which has better wettability than the pure material working fluid. The relationship between concentration, temperature and surface tension is shown in Equation (1):
where
σ is the surface tension,
T is the temperature of the working fluid, and
C is the concentration of the working fluid.
Usually, the working fluids of the two-phase heat transfer device, such as pure water, ethanol, acetone, etc., are all pure substances, and their surface tension will gradually decrease with the increase in the operating temperature. In this paper, SRF is applied to the loop heat pipe (LHP), and the related literature is reviewed as follows.
In 2001, Zhang et al. [
4] pointed out that both traditional heat pipes and loop heat pipes are faced with a phenomenon. The surface tension of the working fluid decreases with the increase in temperature, which is unfavorable for the extension or rewetting of the heated surface liquid. Furthermore, they suggested that the best way to solve this problem is to use a working fluid whose surface tension increases with temperature.
In 2006, Abe [
5] sorted out the research over the years and pointed out that the self-rewetting fluid can push the cooler liquid to the hotter surface through the Marangoni effect because of the reverse surface tension gradient, delay the occurrence of drying, and then improve the heat transfer. The performance is helpful for the heat transfer performance of a non-capillary structure heat pipe, general heat pipe and boiling phenomenon.
In 2008, Francescantonio et al. [
6] applied the self-rewetting fluid of the 0.1% heptanol aqueous solution to the heat pipe. The experimental results showed that the maximum heat load was more than twice that of water, which not only reduced the thermal resistance, but also delayed the occurrence of dry-out.
In 2011, Morovati et al. [
7] used the butanol aqueous solution as the working fluid (the concentration range included 2%, 5%, and 7%), and carried out the pool boiling heat transfer experiment. The flux increased substantially, and the heat flux did not have a significant effect when the concentration increased to 7%.
In 2011, Launy et al. [
8] sorted out the literature on the theoretical prediction and parameter discussion of loop heat pipes over the years. The paper pointed out the selection and filling amount of the working fluid, the material of the capillary structure, the parameters and geometric dimensions, the temperature and the heat sink temperature. Ambient temperature, the design of evaporator and compensation chamber, the inclination of system operation, and the presence of non-condensable gas will all affect the performance of the loop heat pipe. Among them, the capillary structure parameters have the most important influence on the performance of the loop heat pipe.
In 2014, Hu et al. [
9] used the SRF fluid in micro oscillating heat pipes (MOHPs), and the SRF fluid of heptanol 0.1% aqueous solution and deionized water were used as working fluids to carry out the heat transfer experiments. The result shows that SRF has smaller surface tension than deionized water, which is beneficial to improve the efficiency and improve the heat transfer performance.
In 2015, Wu [
10] first reported that SRF fluid was applied to LHP with a Ni wick to measure the surface tension of SRF (including butanol, pentanol and hexanol aqueous solutions) with different concentrations of alcohol aqueous solutions. The Marangoni effect becomes more vigorous, but above the saturation concentrations, the effect does not increase any further. Therefore, the optimal concentration of SRF is the saturation concentration, and the surface tension measurement results show that the critical heat load is increased by 1.6 times, and the total thermal resistance is reduced by about 60% compared with the heat transfer performance of the water working fluid.
In 2017, Wu et al. [
11] first reported the application of SRF to LHP of the PTFE wick. The PTFE wick has a low thermal conductivity and can overcome heat leakage during LHP operation. However, PTFE is a hydrophobic material and cannot use pure water, the traditional working fluid. The butanol aqueous solution of SRF successfully passed the PTFE wick. After the LHP performance test, it was found that the butanol 6% aqueous solution PTFE system reduced the total thermal resistance by half, and the heat flux increased by more than 50% compared with the water–nickel system.
In 2017, Naresh et al. [
12] applied the SRF of aqueous solutions of different alcohols (butanol, pentanol, hexanol, and heptanol) to thermosyphon, and also pointed out that the optimal concentration of SRF for different alcohol aqueous solutions is its saturation concentration.
In 2018, Hu et al. [
13] used two SRFs, ethanol 5% and butanol 5% plus water, for the study of pool boiling. It was found that the critical heat flux (CHF) increase in the butanol 5% aqueous solution was 1.91 times higher than that of water due to the Marangoni effect caused by the surface tension gradient. It was pointed out that the Marangoni effect should be the key factor to enhance the heat transfer of pool boiling.
In 2019, Boubaker et al. [
14] soaked the wick itself with the SRF butanol 3% aqueous solution and pure water, respectively, and conducted evaporation visualization experiments. It was found that with the increase in applied power, the vapor film formed by the wick when the working medium was pure water also changed. However, the shape of the vapor film formed by SRF decreases with the increase in the applied power, which more effectively solves the obstacle of the vapor film.
In 2021, Zaaroura et al. [
15] applied the mixed fluid of the butanol 3% aqueous solution and gold nanofluid to the capillary heat pipe (CHP), pointing out that the butanol aqueous solution using SRF can reduce the evaporator temperature. Using visualization to observe the internal phase change, it was found that SRF can improve the vapor film barrier and effectively improve the heat transfer performance.
Summarizing the above literature, it is pointed out that the more obvious the Marangoni effect of SRF, the better its performance in the evaporative heat transfer operation. It is also found in the literature that the increase in the solute concentration of SRF [
10,
11] will increase the Marangoni effect and improve the LHP heat. Therefore, the purpose of this study is to develop an ideal SRF formulation for the improvement of heat transfer performance of LHP.
2. Materials and Methods
In this paper, the experimental method was firstly formulated and selected for the working fluid, then the aqueous solution of 2-butanol (10%, 15%, and 20%) was prepared, and the surface tension was measured with the aqueous solution of butanol 6% to confirm the SRF characteristics. The LHP system was firstly made with the Ni wick, and then the wick parameters were tested. Finally, the wick was placed into the LHP test system. The SRF, which was tested for surface tension and had an obvious Marangoni effect, was selected for the LHP performance test.
2.1. Working Fluid Choices
SRF formulations have the potential to enhance the heat transfer performance of LHP, so this paper focuses on finding the ideal SRF formulations to enhance the thermal transfer performance. In the literature [
10,
11,
12], butanol, pentanol, hexanol, and heptanol aqueous solutions are shown to have the potential to enhance the heat transfer performance, but it is also mentioned that the higher the concentration, the better the heat transfer performance, among which the butanol 6% aqueous solution has the highest concentration. This paper searched for SRF with higher solubility, and the results of this analysis showed that 2-butanol, an isomer of Butanol, has a higher saturation concentration. In this experiment, it was used 2-butanol (10%, 15%, and 20%) as the formulation of SRF and the butanol 6% aqueous solution, and a surface tension measurement with temperature was conducted. The surface tension measurements were carried out with the temperature variation to measure whether the SRF has the characteristics of LHP heat transfer performance enhancement as described in the literature.
2.2. Surface Tension Measurement
The surface tension measurement experiment is shown in
Figure 2. During the experiment, the DST-60 flat surface tension meter was used for measurement, and Ni was used as the substrate to observe the surface tension variation in different SRF. The substrate was dipped into the test working fluid (water, butanol 6% and 2-butanol (10%, 15%, and 20%)), and the specimen was subjected to surface tension. When the surface tension (
σ) is balanced with the other forces, the upward tension
F, the wetting length
L and the contact angle
θ between the specimen and the working fluid are recorded. Each sample was measured three times and the average was taken.
During the measurement process, the temperature of the working fluid is controlled by a constant temperature water bath, and the surface tension is measured at a fixed temperature (20, 40, 50, 60, and 70 °C) to observe the curve of the surface tension with the change in temperature and to judge the surface tension reverse phenomenon.
The surface tension value is calculated by substituting the measured data into Equation (2).
2.3. Wick Manufacture
The manufacturing process of the Ni wick is shown in
Figure 3. Type 255 nickel powder (2.2~2.8 μm) was selected as the wick material according to Tracey [
16]. This Ni powder has a three-dimensional dendritic structure, which provides a high capillary force and good permeability. In this experiment, the loose filling method [
17] was used to manufacture the Ni wick, in which the nickel powder is filled into the sintering mold without applying pressure to ensure the porosity of the capillary structure. The sintering temperature of Ni powder is set to increase to 625 °C within 60 min and hold the sintering temperature for 30 min. Hydrogen is passed through the sintering process for oxidation reduction. After the sintering is completed and the wick cools naturally, it can be removed from the mold to complete the wick production.
2.4. Wick Parameter Measurement
The parameters of the completed wick were measured using pure water as the working fluid and a self-designed test platform with reference to ASTM E128-99 [
18] (see
Figure 4). This platform measures important parameters, such as the effect radius (
), permeability (
), and porosity (
). The effective radius is related to the size of the wick capillary force, while the permeability is related to the resistance of the working fluid to flow in the wick, and the porosity represents the number of effective pores that can generate the capillary force.
When the air pressure exceeds the capillary pressure that can be generated by the wick itself, the air will penetrate through the wick and float out of the water in the form of air bubbles, as follows in Equation (3).
where
σ is the surface tension value of the working fluid, and
is the pressure difference corresponding to the first point of air bubble generation.
The measured parameters are brought into Equation (4) to calculate the permeability parameters.
is the wick length; is the liquid viscosity coefficient; is the mass flow rate; is the liquid density; ∆P is the pressure difference through both ends of the wick; and Tw is the wick thickness.
Porosity measurements are mainly performed using this system (e.g.,
Figure 4b), where the wick is first placed in a container with a working fluid, then vacuumed to ensure the complete wetting of the wick, and the porosity is calculated as Equation (5). Water is used as the test fluid and measured using the Archimedes method.
where ε is the porosity,
is the volume of the wick absorbing the test fluid, and
is the volume of the wick.
In addition, in order to verify the pore characteristics of the wick, SEM images were taken to observe the pore structure under microscopy.
2.5. LHP Heat Performance Test
The parameters of the LHP test system are shown in
Table 1. In the design of LHP, the basic requirements, such as heating area, space requirement, cooling target wattage, etc., as well as safety should be considered. Additionally, the materials of each component of the system, such as the working fluid, evaporator and other component materials, piping materials, wick materials, etc., should be selected to meet the requirements [
10,
19]. We should consider the space volume as a factor, then design the external dimensions of the components. In the heat transfer experiment, the Ni wick was used for the heat transfer test, and then aqueous solutions of water, butanol 6% and 2-butanol 20% were used as the working fluid.
Figure 5 is the schematic diagram of the heat transfer performance test experiment. During the performance test, the thermocouple (T-Type) was arranged to measure the temperature of each part of the LHP, and the measurement points of each part were arranged as shown in the figure. The total thermal resistance value (
) is obtained by bringing the condenser inlet (
), evaporator wall temperature (
), and power supply wattage (Q) into Equation (6). Meanwhile, Ref. [
10] pointed out that SRF has better heat transfer performance at higher operating temperatures with the need for high power density heat dissipation components, such as solar energy storage in the energy field according to Gasia et al. [
20]. Therefore, this experimental performance test continues to increase the heat load (Q) until LHP system performance failure, that is, the LHP system test stops.
In this paper, the performance of LHP is measured by the relative relationship between temperature–wattage (T − Q) and thermal-resistance–wattage (R − Q) to determine the performance.
Water, butanol 6% and 2-butanol 20% are tested as the working fluid. The ambient temperature is 25 °C, and the circulating water temperature of the condenser is 10 °C. The error of the thermocouple is ±0.2 °C. The error analysis is based on the relative inaccuracy analysis proposed by Moffat [
21], and the error of the total thermal resistance of the system is ±1.9–16.2%.