Heat Transfer Enhancement Using Micro Porous Structured Surfaces

Abstract

The parabolic trough solar collector as a popular technique is widely used in solar concentrating technologies (SCTs). The solar absorber tube is the key position of the trough solar thermal power system. The internal modification of the absorber tube is one of the most interesting techniques for increasing the collector’s performance. At present, most of the methods to enhance heat transfer efficiency focus on designing alternative parabolic trough collectors (PTC) absorbers and improving the internal structure of absorption tubes. Due to the limitation of temperature range, most absorption tubes use oil as heat absorbing liquid, and very few heat absorbing tubes directly use water as working fluid. This is because water is limited by critical heat flux in high temperature environment, resulting in low heat transfer performance. In this work, we designed a new porous absorber tube with the function of allowing liquid resupply and vapor overflow from different paths, which can effectively improve the critical heat flux of the absorber tube when using distilled water as working fluid. In order to obtain better heat transfer performance of the absorber and verify the feasibility of vapor–liquid separation mechanism, a simplified model of the absorber was carried out in pool boiling. In this work, we fabricated an arterial porous structure with the function of regulating vapor–liquid flow path based on vacuum sintering technique, and the effect of different heating methods on boiling heat transfer performance are analyzed. The maximum heat flux of 450 W/cm² was achieved without any dry-out at the superheat of 42 °C, and the unique evaporation/boiling curve was obtained.

1. Introduction

The parabolic trough collector (PTC) seems to be the most mature solar heat collection technology in solar power generation after decades of rapid development. The solar receiver tube is one of the core light and heat transfer components of the system. It can absorb the solar radiation reflected by the parabolic trough [1]. In current, the annulus space between the absorber coating and the glass tube is typically designed to be evacuated to reduce heat loss at high operating temperatures, and to protect the absorber surface from oxidation.

Most studies focus on the design of alternative PTC absorbers due to the high cost of traditional vacuum tubes and their durability issues in high temperature conditions. Bortolato [2] studied the effect of flat absorbers with asymmetric reflectors for steam generation. The results show that this configuration has an optical efficiency of 82% and a thermal efficiency of 64% at a temperature level close to 100–120 °C. Halimi [3] studied the PTC with a U-tube absorber and found that the different positions of the internal U-tube had an impact on the thermal efficiency, and confirmed that the parallel structure obtains better heat transfer performance, and the thermal efficiency is 40%. Chen [4] proposed a PTC receiver with a V-cavity structure. The thermal efficiency of the structure with and without the addition of fins in the flow was investigated, separately. They found that the heat transfer rate could be increased by using inner fins. The results confirmed that at low temperature, the thermal efficiency of the system is about 55%, while at high temperature, the thermal efficiency is significantly reduced. Wang [5] proposed a PTC with a thermal siphon receiver. The system proved to perform well for operation in the temperature range of 200–400 °C, and due to its unique design, it does not have the traditional impediments compared to other natural phase circulation systems. It is worth mentioning that all thermal enhancement skills can improve the thermal efficiency of PTC, but they will lead to higher introjection costs.

Several techniques such as modifying the inner absorber geometry have been applied to enhance the thermal performance of PTC systems. Most of the studies [6] involved modifying the entire absorber surface or lower part due to the high concentration of solar heat flux in this area. Xiangtao [7] used D12 thermal oil to study the performance of the absorber with the fin pin arrays in the down part. The results show that the friction factor increases and the Nusselt number decreases. Bellos [8,9] studied the performance of absorber with internal longitudinal fins in all the periphery. The number and the location of the inner fins were optimized inside the absorber tube and finally found that this technique could increase thermal efficiency by 1.5%, but pressure drop would suffer. Munoz [10] presented an absorber with internal helical fins to enhance the thermal efficiency. Huang [11] developed a dimpled absorber tube and found that the Nusselt number increased by 20%. The current research shows that the use of internal modification can improve the performance and the pressure drop loss is relatively low. Most surface modification technologies are committed to increasing the convective heat transfer coefficient to improve the heat transfer efficiency. However, the improvement of single-phase convective heat transfer coefficient is limited. It is necessary to develop an efficient surface modification technology to significantly improve the heat transfer performance.

In the trough solar thermal power system, thermal oil such as Syltherm 800, Thermal oil D12 and Therminol VP-1 are usually used as working fluid to absorb solar energy in the absorber. The high-temperature hot oil turns water into high-temperature and high-pressure steam through the heat exchanger, and drives the steam turbine to generate electricity. In the current system, water is rarely used directly as a working fluid to absorb solar energy in the collector, because here exists a limit heat flux value when water operates at high temperatures. When the heat flux exceeds the so-called critical heat flux (CHF), a steam film will be formed between the heating surface and the liquid, covering the whole heating surface, seriously deteriorating the heat transfer performance. The heat transfer process of water in absorber is similar to that of pool boiling process, and the boiling enhanced heat transfer technology can also be applied to enhance the heat transfer process of trough solar thermal power system.

Extensive theoretical and experimental researchers have been proposed to strengthen the near surface effects, such as surface coating [12,13] and surface micro/nano structures fabrication [14,15,16]. The results show that micro/nano porous media can significantly improve the pool boiling heat transfer performance. Among them, the use of non-uniform surface engineering methods has been shown to significantly improve the boiling heat transfer performance as well as the boiling critical limit. This is because an artery is reasonably designed to reduce the resistance of vapor–liquid countercurrent, and the capillary effect of porous structure is used as the liquid supply path to supply the boiling surface. Li [17] prepared several uniform and modulated porous structures based on the sintered copper particle process, and experimentally studied the pool boiling CHF and heat transfer coefficient (HTC) of the porous structure. The CHF of the modulated porous structure is about 450 W/cm², which is three times that of the smooth surface. The highest heat transfer coefficient also reached a value of 20 W/cm²·K. Liter [18] made a periodically non-uniform thickness modulated porous layer coating surface to improve boiling heat transfer performance. Their results showed that pool boiling CHF increased to nearly three times that of smooth surface CHF. They attributed this enhancement to a modulated surface that separates the liquid and vapor phases, thereby reducing the liquid and vapor countercurrent resistance near the surface. The enhancement of boiling heat transfer can be attributed to the combination of several factors, such as increased nucleation point density, increased total surface area, liquid flow assistance of capillary effect and the formation of steam escape path [19,20,21]. As mentioned above, the realization of the boiling enhancement technology makes it possible to develop a new type of trough power generation technology that uses water directly as the working fluid. Although the current near wall enhancement technology improves the heat transfer performance to a certain extent, there is still a certain gap in the practical application of parabolic trough solar collector due to the lack of regulation of vapor–liquid flow path.

In this work, we attempted to develop a new type of absorber tube with an internal artery porous structure, which allows liquid replenishment and steam to escape from the heating surface from different paths. In this absorption tube, the porous structure separates the flow paths of the vapor and the liquid, the vapor flows in the pre-designed main channel, and the liquid uses the capillary characteristics of the porous structure to continuously supply from the middle liquid main channel to the outer wall of the absorption tube. In this process, the vapor generated in the vapor main channel can be directly collected and used to propel the steam turbine. The enhanced mechanism of vapor–liquid separation ensures that the heat transfer performance of the inner wall of the absorber will not be degraded due to high temperature operation. Using this technology, a large number of heat exchangers can be replaced, thereby achieving the purpose of simplifying the system, improving efficiency, and reducing costs. In order to obtain better heat transfer efficiency of trough solar thermal power system, the absorber tube with porous structure is simplified in this paper. The heat transfer performance of the simplified structure was investigated in pool boiling experiments and the gas-liquid separation mechanism was verified.

2. Experiments Details

2.1. Design

The absorber tube with artery porous structure proposed in this paper is shown in Figure 1. A porous layer of equal thickness is arranged on the inner wall of the heat collecting tube, and several vapor arteries are arranged inside the porous layer for the vapor rapid discharge. The top surface of the porous layer forms channels for liquid flow. Due to the great flow resistance of the steam in porous structure, the steam can only flow in pre-set arteries. As a result, the flow paths of vapor and liquid are separated by the porous structure.

Since the heat transfer performance of the absorber tube using water as the working fluid is limited by the critical heat flux under high temperature conditions, the maximum heat transfer limit should be considered when designing the absorber tube. In order to facilitate the research, this paper simplifies the structure of the absorber tube with artery porous structure to a flat artery porous structure which was shown in Figure 2. The boiling heat transfer performance of artery porous structure was studied in pool boiling system, which provides a research basis for better design of absorber tubes in the future.

In this porous structure, three rectangular arteries are located at the bottom of the copper porous structure, and there existed a thin, porous layer with uniform thickness of 0.25 mm between the arteries and the heating surface. The height of the artery is 2.0 mm, and the thickness of the top porous layer is 2.5 mm. The main parameters of porous artery structure are given in

Table 1. Main parameters of porous artery structure.

Parameters	Value
Porous artery structure diameter (D)/mm	8.0
Top microporous layer thickness (H1)/mm	2.5
Middle artery depth (H2)/mm	2.0
Bottom microporous layer thickness (H3)/mm	0.25
Artery width (W1)/mm	1.0
Porous fin width (W2, W3)/mm	1.2, 1.3

.

2.2. Sample Preparation

In this work, high purity (>99.99%) spherical copper particles with an average particle size of 100 microns were used to prepare porous structures. An electronic balance (OHAUS Explorer EX224, OHAUS, Parsippany, NJ, USA) with a measurement accuracy of 10⁻⁴ g were used to measure the weight of copper powders. The copper particles are put into the prepared graphite mold by loose sintering, and placed in the vacuum sintering furnace (BFG-12A) at 800 °C together with the graphite mold and copper powder for two hours to form a porous structure, and then naturally cooled to room temperature. The samples were washed with 20% citric acid, absolute ethanol, and distilled water in ultrasonic bath for 10 min. Clean the copper pillar made of c1020 oxygen-free copper with ultrasonic cleaning agent immersed in acetone, and then carefully rinse it with distilled water at ambient temperature. In the vacuum sintering furnace, these copper pillars are connected with the copper porous structure by direct sintering to form different samples.

2.3. Surface Characterization

The porosity Φ of the copper porous samples can be estimated as follows:

$$\Phi =1-\frac{m_1}{{\rho }_{1}v}$$

(1)

where m₁ is the mass of the porous samples, ρ₁ is the density of copper, v is the volume of the samples. In order to validate the above estimation, the infiltration method can also be used to calculated the porosity Φ of the test sample:

$$\Phi =\frac{m_2-m_1}{v{\rho }_2}$$

(2)

where m₂ is the mass of the porous samples filled with ethyl alcohol, ρ₂ is the density of ethyl alcohol. By calculation, the porosity of all test porous samples is between 38% and 41%. As shown in Figure 3, SEM photos of copper porous structure observed under scanning electron microscope (SUPRA 55, CARL ZEISS, Jena, Germany) are shown, in which the particle diameter is in the range of 96–106 μm. The average density of the copper porous samples is 5.39 g/cm³.

2.4. Experimental Apparatus

The schematic diagram of the boiling experiment system in this study is shown in Figure 4, which consists of a high-speed camera (EOS 5D Mark IV, CANON, Tokyo, Japan) to capture experimental phenomena, a serpentine reflux condenser to maintain the liquid level in the pool, a transparent boiling chamber to enable a visualization study, a copper pillar (C1020, Sambo, Osaka, Japan) to simulate heat source, a high-frequency electromagnetic induction heater (TD-25kW, TUODA, Suzhou, China) to provide high heat flux, an auxiliary heating thermostatic system to anti-heat leakage, and a temperature measurement and data acquisition system. The stainless steel plate was used as the support substrate, and the polyetheretherketone (PEEK) plate is installed in the middle to reduce the heat loss from the copper pillar to the stainless steel plate. The copper pillar passed through the hole in the center of the PEEK plate, and sealed the small gap between the copper pillar and the PEEK plate with an O-ring. Since only the central heated region where the boiling occurring was exposed, the heat losses through the supporting structures were negligible. The outer surface of the copper pillar was covered with aluminum silicate fiber insulation layer to prevent heat leakage of the copper pillar to the surrounding environment. The boiling chamber wall is customized by high temperature resistant quartz glass, with a thickness of 5 mm, an inner diameter of 150 mm, and a height of 140 mm. Maintain the water temperature at a constant value of 100 °C using an auxiliary heating thermostatic system (immersed in a water bath). The heating power of the auxiliary heater was controlled by an autotransformer. The focus of this paper is to explore the heat transfer limit of arterial porous structure and the heat transfer performance under different power. The heating method of solar concentrating is not enough to provide sufficient heat, so high-frequency electromagnetic induction heater is employed to simulate solar energy. The heating power of copper pillar is controlled by the electromagnetic induction heater, while the output power can be continuously controlled within the range of 0–15 kW. The top of the cylindrical copper pillar had the same size as the sample, which was 8 mm in diameter. Underneath the boiling surface, three K-type thermocouples (GG-K-30-SLE, 0.5 mm in diameter, Omega, Stamford, CT, USA) with a precision of about ±0.5 K were vertically embedded into the copper pillar at 4, 8, and 12 mm away from the boiling surface to measure the temperature distribution inside the copper pillar. An Agilent 34970A Data Acquisition System was used to monitor and analyze the measurement data.

In this work, the experiment was carried out at standard atmospheric pressure, and the working fluid was distilled water. During the experiments, the temperatures are recorded at a time interval of every 5 s and a steady state is judged, as the fluctuations of temperatures (T1, T2, T3) are smaller than 0.5 °C in 30 min. By changing the output power with a certain value, different heat flow test conditions can be realized.

2.5. Analysis of Experimental Errors

The axial temperature gradient of copper pillar was estimated using the three-point backward Taylor’s series approximation:

$$\frac{\mathrm{dT}}{\mathrm{ds}}=\frac{{3\mathrm{T}}_{\mathrm{T}1}-4{\mathrm{T}}_{\mathrm{T}2}+{\mathrm{T}}_{\mathrm{T}3}}{{2\mathrm{s}}_0}$$

(3)

where T_T1, T_T2, T_T3 are the temperature data of measuring points obtained from thermocouples, and s₀ is 4 mm indicating the distance between two adjacent temperature measuring points. The heat transfer on the boiling pillar can be expressed by the one-dimensional heat conduction equation:

$$\mathrm{q}=-\mathsf{\lambda }\frac{\mathrm{dT}}{\mathrm{ds}}$$

(4)

where λ is the thermal conductivity with value of 380 W/(m·K).

The convective heat transfer coefficient can be calculated by:

$$\mathrm{h}=\frac{\mathrm{q}}{\Delta \mathrm{T}}$$

(5)

where $\Delta \mathrm{T}$ is the wall superheat given by:

$$\Delta {\mathrm{T}=\mathrm{T}}_{\mathrm{w}}-{\mathrm{T}}_{\mathrm{s}}{=\mathrm{T}}_{\mathrm{TC}1}-\frac{{\mathrm{qs}}_0}{\mathsf{\lambda }}-{\mathrm{T}}_{\mathrm{s}}$$

(6)

Here, T_s is the water saturation temperature which is 100 °C. T_w is the surface temperature of the tested sample.

The parameters leading to uncertainty include temperature measurement, thermal conductivity of copper, and the distance between thermocouples in copper block. The uncertainty of temperature measurement (∆T) and saturation boiling temperature (T_s) are usually the same, all within ±0.5 °C. The uncertainty of thermal conductivity is very important in evaluating the overall uncertainty, and the accuracy is usually about 5%. The uncertainty of the thermocouple distance u(s) are usually determined by the uncertainty of the position of the thermocouple hole (u_hole) and the uncertainty of the position in the thermocouple hole (u_TC). The calculation method of the position uncertainty of thermocouple relative to surface u(s_surf) is similar. In this work, the uncertainties of u_hole and u_TC were within 0.2 mm and 0.05 mm, respectively.

According to the standard error analysis method based on the error transfer equation, the uncertainty of heat flux, surface superheat, and heat transfer coefficient were evaluated according to the following equation:

$$\mathrm{u}\left(\mathrm{s}\right)=\sqrt{{\mathrm{u}}_{\mathrm{hole}}^2+{\mathrm{u}}_{\mathrm{TC}}^2}$$

(7)

$$\mathrm{u}\left(\mathrm{q}\right)=\sqrt{2{\mathrm{u}}^2\left(\mathrm{T}\right){\left(\frac{\mathsf{\lambda }}{2\mathrm{s}}\right)}^2{+\mathrm{u}}^2\left(\mathrm{s}\right){\left[-\frac{\mathsf{\lambda }\left({\mathrm{T}}_{\mathrm{TC}3}-{\mathrm{T}}_{\mathrm{TC}1}\right)}{2{\mathrm{s}}^2}\right]}^2+{\mathrm{u}}^2\left(\mathsf{\lambda }\right){\left[\frac{{\mathrm{T}}_{\mathrm{TC}3}-{\mathrm{T}}_{\mathrm{TC}1}}{2\mathrm{s}}\right]}^2}$$

(8)

$$\mathrm{u}(\Delta \mathrm{T})=\sqrt{{\mathrm{u}}^2{\left(\mathrm{T}\right)+\mathrm{u}}^2\left(\mathrm{q}\right){\left(-\frac{{\mathrm{S}}_0}{\mathsf{\lambda }}\right)}^2{+\mathrm{u}}^2\left(\mathsf{\lambda }\right){\left(\frac{{\mathrm{qs}}_0}{{\mathsf{\lambda }}^2}\right)}^2{+\mathrm{u}}^2{(\mathrm{s}}_{\mathrm{surf}}){\left(-\frac{\mathrm{q}}{\mathsf{\lambda }}\right)}^2{+\mathrm{u}}^2{(\mathrm{T}}_{\mathrm{s}})}$$

(9)

$$\mathrm{u}\left(\mathrm{h}\right)=\sqrt{{\mathrm{u}}^2\left(\mathrm{q}\right)\frac{1}{{\left(\Delta \mathrm{T}\right)}^2}{+\mathrm{u}}^2(\Delta \mathrm{T})\frac{{\mathrm{q}}^2}{{(\Delta \mathrm{T})}^4}}$$

(10)

The calculated value of relative uncertainty is always less than 5.0% when the heat flux is greater than 100 W/cm². It is calculated that the maximum relative uncertainty of heat transfer coefficient is less than 6.0% in all cases. High temperature insulation material (aluminum silicate fiber) was used to reduce the heat loss of copper column surface to the environment. The thermal conductivity of aluminum silicate fiber is about 0.0252 W/(m K), and the total heat loss from copper pillar to environment is estimated to be less than 4.0%.

3. Results and Discussions

3.1. Comparison of Pool Boiling Heat Transfer Performance between Artery Porous Structure and Smooth Surface

To verify the boiling experimental setup, the smooth copper surface was repeatedly tested three times and the boiling curves for the smooth surface at 1 atm agreed well with those obtained by Calvin [17] and Cooke [22] with less than 6% discrepancy. In addition, the experimental CHFs on the smooth copper surface for water at 1 atm is 151 W/cm², which is also close to the previous results reported in the range of 128–150 W/cm² [15,23,24,25]. Figure 5 shows the boiling heat transfer performance curves of the smooth surface and the porous surface. The experimentally measured smooth surface data are in good agreement with those of other scholars, which proves the reliability of the experimental system and can provide high-quality and reproducible results. Small differences between the experimental results may be caused by the relative roughness of the smooth surface and the size of the heated surface.

It can be clearly seen from Figure 5 that the boiling heat transfer performance of the artery porous structure prepared in this experiment is significantly improved compared with the smooth surface. The maximum heat flux obtained was 450 W/cm² at superheat 42 °C. The initial superheat is reduced from 12 °C to within 5 °C. The boiling curve shows that the heat flux of the artery porous structure increases with the increase of superheat, especially in the initial stage of boiling, where the enhancement trend is very obvious, but with the increase of superheat, the enhancement trend gradually slows down.

Figure 6 shows the vapor–liquid distribution and movement for different heat fluxes. With the increase of heat flux, there are four vapor–liquid motion states of A, B, C, and D in turn. It can be seen from Figure 6 that the artery porous structure can plan the movement path of the vapor macroscopically. The vapor escaped from both sides of the artery, which confirms that the porous artery structure can effectively separate the vapor–liquid flow path. Figure 6A shows the initial state of boiling. Most of the liquid still exists in the artery. The tiny cavities provided by the porous structure act as the vaporization core to generate small bubbles first. Due to the small pressure change in the artery, the horizontal velocity of these small bubbles is small, and the small and dense vapor bubbles are discharged from both sides of the main channel and move up slowly against the outer wall of the porous structure. With the heat flux increases, large vapor bubbles first begin to appear on both sides of the artery. At this time, the liquid in the artery structure is gradually replaced by vapor, and a large amount of vapor accumulates in artery. The vapor pressure inside the artery keeps increasing, pushing the vapor out on both sides. Part of the liquid will enter the artery under the action of suction as the vapor is discharged. At this time, the discharge frequency of the bubbles is not high. Since the discharged bubbles have a certain initial horizontal velocity, they exhibit a parabolic trajectory motion under the influence of buoyancy, as shown in Figure 6B. Compared with the initial stage of boiling, the size of the bubbles becomes larger and the number thereof becomes smaller. Figure 6C shows the vapor–liquid motion state when the heat flow further increases. At this time, large bubbles are discharged from both sides of the three arteries, and the size, discharge frequency, and speed of the bubbles are both increased. As shown in Figure 6D, when the boiling state is at a high heat flux, the arteries are filled with vapor and the vapor temperature is in a superheated state. The bubbles on both sides of the three arteries converge into large bubbles, whose size even exceeds the outer diameter of the porous structure, and the movement frequency of the bubbles is faster. The small bubbles generated at the vapor–liquid interface merged during the movement of the artery, and converged to form large bubbles.

3.2. Temperature Profile with Different Heat Method

In this paper, the boiling surface is heated by continuous increasing heating method. The output power of the heater was first set at a proper min value, and when the measured temperature was stable for about 10 min, the output power was adjusted to a higher value. At this time, the temperature curve rises step by step with time. When the temperature measurement curve is stable, it can be considered that the boiling reaches a steady state. When calculating the heat flux, the steady-state experimental data points are selected according to Fourier’s law for calculation. However, in practical application, the increase of heating power is often not a step change. The instantaneous high-power heating condition puts forward higher requirements for the porous artery structure. In the face of the sudden increase of power, whether the porous structure can respond in time and still maintain good heat transfer performance is another important index to judge the performance. Therefore, it is necessary to study the transient heat transfer response ability and heat transfer performance. Most scholars use steady-state experimental data to describe the relationship between heat flux and superheat in boiling heat transfer experiments, and there is little analysis of transient experiments.

Figure 7 shows the temperature changes of three thermocouples placed on the upper part of the copper column under different heating power in the experimental test, and the heating methods are continuous incremental heating. With the increase of heater output power, the temperature difference of the three thermocouples between T1 and T2, and between T2 and T3 becomes larger. The two temperature differences are basically the same, indicating that the temperature is linearly transmitted along the copper pillar.

In order to better evaluate the heat dissipation capacity of the new structure, intermittent heating method was used as a comparison, which is shown in Figure 8. The output power of heat was also first set at a fixed value, and when the temperature had risen and tended to be stable for a while, the power output was shut down until the system temperature turned to ambient temperature. Then, we set a higher value of output power to repeat the previous experiment. When the output power is higher than 617 W, the temperature of three thermocouples continues to increase and cannot be stabilized. Due to considering the security of device, the experiment was forced to abort and the temperature’s sudden rise was not observed. These two heating methods are used to simulate the heat dissipation of electronic equipment under different working loads.

3.3. Comparison of Two Different Heating Methods

Figure 9 shows the superheat dependence of the heat flux under different heating conditions being as (a) continuous incremental heating, and (b) intermittent heating. Most scholars use steady-state experimental data to describe the relationship between heat flow and superheat, but in some cases, it is necessary to study transient experimental data. High-purity oxygen-free copper is used as the heat transfer medium. Since the thermal conductivity inside the copper pillar is very small and the thermal diffusivity is very large, the temperature data for transient acquisition is also processed by the steady-state Fourier’s law of heat conduction. Due to the thermocouple acquisition temperature interval of 5 s, it will reduce the error caused by transient temperature change to some extent. The difference between steady-state experiments and transient experiments were compared in Figure 9a, and it can be clearly observed that the transient experimental curve and the steady-state experimental curve are basically coincident. Since each successive output power gradient increases and a complete continuous steady-state point cannot be obtained, it is necessary to determine the trend between the two steady-state points. The coincidence of the steady-state curve and the transient curve further confirms the initial assumptions of the experiment and enriches the trend of the steady-state curve. The high agreement between the steady-state data and transient experimental curves proves the repeatability of the experiment on the one hand, and the transient experimental data curve can be used to better describe the whole boiling process on the other hand. For the working condition of instantaneous high power, the superheat increases when the superheat is about 20 degrees, that is, the boiling heat transfer performance deteriorates, and the turning point position is the same, which shows that the main channel porous structure can maintain stable boiling heat transfer performance for intermittent heating mode and instantaneous high power heating mode.

According to Figure 8, the difference between steady-state experiment and transient experiment in intermittent heating method was studied, which was shown in Figure 9b. Since the experiment included transient experiments that did not reach a steady state, the heater output power was used to distinguish the boiling characteristics under different operating conditions. In Figure 9, under the same structure, different output powers have a weaker influence on the boiling curve trend. Each curve has the same growth trend, but the final stability point is different. When the output power is less than 617 W, the heat flux will stabilize at a certain fixed value at a certain output power. The steady-state data curve and the multiple transient experiments are basically coincident. On the one hand, the reproducibility of the experiment is proved, and on the other hand, it is proved that the steady-state experimental data curve can be extended by transient experiments. An interesting phenomenon occurs when the output power is higher than 730 W; the heat flux did not continue to grow according to the previous trend, but showed a very slow growth trend when the degree of superheat was 25 K, and at this point, the heat flux is about 450 W/cm².

3.4. Effect of Different Heating Methods

The effect of heating conditions on pool boiling performance are compared. The difference between the two methods can be clearly seen in Figure 10. For the incremental heating method, a significant advantage is exhibited at small and moderate heat fluxes, with a smaller superheat at the same heat flux density. However, when the heat flux is higher than 250 W/cm², the temperature is difficult to stabilize, and keeps a certain slope continuously rising. For the intermittent heating method, first, the consistency between transient and steady-state experiments can be seen, and the maximum heat flux is higher than the continuous incremental heating method.

The cause of this phenomenon may be the hysteresis of the porous structure. In the process of stabilization, the final state is related to the initial state. The intermittent heating mode can ensure that the same initial state is the ambient temperature. In the continuous heating mode, the initial state of each stabilization process is different. In addition, when the vapor–liquid interface is inside the porous structure, liquid layer shrinkage and local overheating may occur inside, and the boiling curve may also be different. Due to the limitations of experimental visualization, the changes in the gas–liquid interface in the interior of the porous structure cannot be clearly observed. Therefore, the influence of the internal variation of the porous structure on the boiling performance remains to be further explored and studied.

4. Conclusions

In this paper, we propose a novel absorber tube with an internal artery porous structure, where the heat transfer performance of its simplified model is experimentally investigated in pool boiling. Based on the porous artery structure, the transient boiling characteristics have been widely studied experimentally. The positive experimental results provide experience for subsequent design and fabrication of efficient new absorber tubes. According to the preceding experimental results and discussion, some important concluding remarks can be drawn, as summarized below:

The vapor–liquid separation mechanism of the artery porous structure can significantly improve the heat transfer performance, and the visualized experimental results confirm the feasibility of the vapor–liquid separation mechanism. The maximum heat flux obtained in the experiment is 450 W/cm², which is three times that of smooth surface.

Experiments show that the transient experimental data are highly consistent with the steady-state experimental data. Due to the boiling hysteresis of porous materials, the artery porous structure will lead to the deterioration of heat transfer performance in the face of sudden power change.