Experimental Investigation of the Effects of Grooves in Fe₂O₄/Water Nanofluid Pool Boiling

Abstract

In this study, we systematically explored how changing groove surfaces of iron oxide/water nanofluid could affect the pool boiling heat transfer. We aimed to investigate the effect of three types of grooves, namely rectangular, circular, and triangular, on the boiling heat transfer. The goal was to improve heat transfer performance by consciously changing surface structure. Comparative analyses were conducted with deionized water to provide valuable insights. Notably, the heat transfer coefficient (HTC) exhibited a significant increase in the presence of grooves. For deionized water, the HTC rose by 91.7% and 48.7% on circular and rectangular grooved surfaces, respectively. Surprisingly, the triangular-grooved surface showed a decrease of 32.9% in HTC compared to the flat surface. On the other hand, the performance of the nanofluid displayed intriguing trends. The HTC for the nanofluid diminished by 89.2% and 22.3% on rectangular and triangular grooved surfaces, while the circular-grooved surface exhibited a notable 41.2% increase in HTC. These results underscore the complex interplay between groove geometry, fluid properties, and heat transfer enhancement in nanofluid-based boiling. Hence, we thoroughly examine the underlying mechanisms and elements influencing these observed patterns in this research. The results provide important insights for further developments in this area by shedding light on how surface changes and groove geometry may greatly affect heat transfer in nanofluid-based pool boiling systems.

1. Introduction

Recently, many studies have been conducted in pool boiling because of its many applications in the industry. The most common boiling applications are condensers, boilers, heat pumps, nuclear reactors, electronic chip cooling, and air conditioning systems. In order to develop the heat transfer capabilities of pool boiling, a number of alternatives could be utilized, namely adding nanoparticles or a surface activator to the base fluid, expanding the heating surface, applying an electric or magnetic field, and vibrating the thermal surface [1].

A colloidal solution of magnetic nanoparticles, namely iron, cobalt, iron oxide, gamma iron oxide, and cobalt oxide, which are completely stable in a base fluid, is called ferrofluid. The idea of using nanofluids in heating systems was first proposed by Choi [2]. Parameters such as nanoparticle concentration, nanoparticles size and type, surface roughness, the contact angle of bubbles, system pressure, etc., complicate the boiling mechanism.

Numerous studies have been devoted to nanofluids pool boiling [3,4,5,6,7,8,9]. Chandra et al. [10] studied copper/water at concentrations of $0.25\%$, $0.5\%$, and $1\%$ by weight on a 30 ${\mathrm{mm}}^2$ steel surface. They observed nanofluid had a higher critical heat flux than water, and the HTC decreased with increasing nanoparticles. Hia [11] studied water and aluminum oxide/water behaviour on a variety of surfaces. He observed a development in HTC of rough heater surface. Also, studying the nanoparticle concentration, a boiling test at two concentrations of 0.1% and 0.01% by volume was performed on a rough surface. According to the results, in high heat fluxes, with increasing concentration, the HTC decreases because of the tendency of the particles to clump. Shi et al. [12] studied aluminum oxide/water on copper heaters and found that the boiling HTC increased. In other research, methods such as tri-hybrid nanofluid [13] and using vortex generation for improving heat transfer [14] have been examined.

Norouzipour et al. [15] examined the effects of nanoparticle size on pool boiling. They found that by augmenting the nanoparticle diameter, the HTC increased. Generally, they reported a drop in HTC compared to pure water. Some researchers have also reported no considerable change in the HTC of the nanofluid boiling. For example, Vassallo et al. [16] investigated the 0.5% volumetric silicon oxide nanofluid on a nickel–chromium horizontal wire. In their results, little change in boiling HTC was observed.

Since surface finish could significantly affect pool boiling, a number of papers have been devoted to surface roughness, surface expansion, geometry change, and surface position. Hachem et al. [17] performed a comparative study of the HTC of water on a smooth and porous surface. They reported an increase in the HTC. This is because cavities increase nucleation site density and cause bubbles to form at lower surface temperatures. In another study, Das et al. [18] used surfaces with different structures to increase the HTC of water. These surfaces consisted of a number of parallel or intersecting grooves. They found that grooved surfaces improved the pool boiling of pure water due to their higher thermal surface. Also, Alimoradi et al. [19,20] studied the effects of surface modification on pool boiling. Pastuszko et al. [21] experimented with finned surfaces for water. A copper coating with micro-holes is attached to the fins. Their results show that at high fluxes, the finned surface with a fin of 1 mm height has the greatest effect on the HTC. Dadjoo et al. [22] carried out research on the effect of silicon oxide/water nanofluid surface on the copper heater and found that, as slope increases, the HTC of boiling decreases. In their study, bubbles form on the boiling surface and join with each other, which increases thermal resistance. In another study, researchers [23,24] proposed a boiling method for surface modification. Umesh et al. [25] studied pentane and pentane/copper oxide on a smooth and expanded surface at atmospheric pressure. The expanded surfaces were two levels with small and large fins. The HTC of pentane on the surface of the small fin increased by 10–15%, and on the surface of the large fin it decreased by 5–10%. They stated that because of the formation of more cracks during grooving and the capillary effect of the fluid, the HTC at the small fins increased, and on large fins, due to the lateral surface of the fins, more resistance to bubble separation from the surface was found. Further, in their results, it has been seen that the HTC of pentane/copper oxide on the surface of the fin was increased by about 15–25%, and for a concentration of 0.01%, it is almost equal to the boiling of pentane. This is because at higher concentrations a micro-layer forms under the growing bubble, blocking the nucleation sites. Narayan et al. [26] studied aluminum oxide/water nanofluid on rough surfaces. They used the nanoparticle surface interaction parameter (φ) as the criterion for their results. When $\phi \le 1$, the HTC decreases. Because the nanoparticles block the nucleation sites and create additional thermal resistance. When this parameter is greater than one, the HTC increases at low concentrations.

Research has also been done on utilizing ferroalloys in pool boiling. Abdollahi et al. [27] studied a magnetic field applied on water/iron oxide. They found an optimum volumetric concentration of 0.1% for ferrofluid, which increased the HTC by 43% [28]. Abdollahi et al. [29], in another study, studied time and surface modifications in the iron oxide/water pool boiling process and found that heat transfer on deposited surfaces decreased with low heat flux and increased with high heat flux. Lee et al. [30] compared the critical heat flux (CHF) of a ferrofluid with a water-based fluid, aluminum oxide/water, and titanium oxide/water. In their results, it was seen that nanoparticles increased the critical heat flux of boiling, and the highest CHF was related to ferrofluid and was 2.6 times higher than pure water. This is because of the accumulation of nanoparticles on the surface and increased surface wettability [31].

According to previous research, despite the fact that there is research on the use of nanofluids in pool boiling, few studies have been conducted on heated and grooved surfaces. In the present paper, the effect of rectangular, circular, and triangular grooved surfaces on the pool boiling of ionized water and iron oxide/water is investigated experimentally. Consequently, the pool boiling test of pure and ferrofluid was performed on a smooth surface and then on grooved surfaces. After which the depth of the grooves is increased, and its effect on heat transfer is investigated.

2. Materials and Methods

2.1. Preparation of Nanofluid

Nanofluids are prepared in two methods, one-stage and two-stage. In the first method, nanoparticles are produced directly in the fluid. However, in the later method, nanoparticles are dispersed in the form of dry powders in the fluid. The benefit of the first method is that the size and distribution of nanoparticles are controlled, and the stability of nanofluid in this method is higher than in the two-stage method.

To achieve stable nanofluids, methods such as ultrasonic rippling, acidification, and the addition of surfactants are used. The ultrasonic ripping method breaks the nanoparticle masses and makes the solution uniform in a short time. Adding surfactants to the solution causes the solution to be uniform, but affects the thermophysical properties of the fluid and does not work well in high-temperature applications. The method of changing the acidity is a suitable method for stabilizing the nanofluid.

In this paper, iron oxide/water nanofluids are fabricated by a one-step method developed by Berger [32] and a surfactant is added to stabilize them. First, 1 mL of $\mathrm{iron}\left(\mathrm{II}\right)\mathrm{chloride}$ is mixed with 4 mL of $\mathrm{iron}\left(\mathrm{III}\right)\mathrm{chloride}$. Then 2.64 mL of ammonia is added to 50 mL of deionized water and added to the solution with a syringe pump. Simultaneously, with the addition of ammonia, the solution is placed in a magnetic stirrer at a speed of 1000 rpm. After a few minutes, the iron oxide particles settle, and the water on top of the particles is removed. The resulting precipitate is then placed in a centrifuge for 5 min. After this step, a black deposit remains at the bottom of the container. The surfactant tetramethylammonium hydroxide is then added to the precipitate and placed on a magnetic stirrer at 500 rpm for half an hour. Finally, by removing excess ammonia, 1% by volume of iron oxide nanofluid is obtained. The prepared nanofluid is stable for at least one month, and no nanoparticles precipitate.

In order to get more insight of the nanofluid, dynamic light scattering (DLS) and transmission electron microscopy (TEM) tests are performed for the fabricated nanofluid. Figure 1 and Figure 2 show the results of DLS and TEM tests for nanofluid, respectively. It is clear that the average diameter of nanoparticles is about 15 $\mathrm{nm}$. To investigate the stability of our fluid, a potential test has also been performed. This test represents the magnitude of the electric charge at the nanoparticle surface. Many researchers [33,34,35,36,37,38,39] have reported that colloidal solutions are stable at zeta potentials greater than 30 $\mathrm{mV}$. This value for the nanofluid made at a concentration of 0.01% was −31 mV, which indicates the high stability of the nanofluid [40,41,42,43,44,45,46,47].

Before each boiling test, a diluted solution with a concentration of 0.01% is placed in an ultrasonic waveguide with 400 watts and 24 $\mathrm{kHz}$ for half an hour.

2.2. The Boiling System

The picture of the laboratory set made for the experiments is shown in Figure 3. The main components of the test apparatus are the copper and insulation heater set, boiling tank, condenser, and cooling system, voltage, current, and displays control system. The voltage regulator creates a constant voltage and current to generate a constant heat flux in the heater. The heater comprises two sections, namely heater cartridge and cooper block. The heater cartridge is inside the cooper block, and in the top section of the block, there are three thermocouples to record the temperature in the vicinity of the boiling surface. By using the thermocouples, the heat flux is calculated, the boiling surface temperature could be calculated as well. In the boiling vessel, there is a thermocouple to measure the bulk temperature of the fluid. By using these parameters, we are able to calculate the heat transfer coefficient. Additionally, in the top section of the boiling vessel, a condenser is installed to return the vapor to the vessel and keep the fluid volume constant.

The cylindrical heater is made of 99.99% pure copper, and it has a diameter of 45 $\mathrm{mm}$ and a length of 110 $\mathrm{mm}$. There are three holes in the heater, each of which has 10 mm diameter. These holes are 75 mm deep, and they are situated at relative $120°$. At the end of the heater, the bullet elements are created for inserting cartridge elements. The heating elements with a power of 650 $\mathrm{watts}$ are 10 $\mathrm{mm}$ in diameter and 80 $\mathrm{mm}$ long. Also, three holes with a diameter of 4 $\mathrm{mm}$ and a depth of 22.5 $\mathrm{mm}$ at a distance of 8, 18, and 30 $\mathrm{mm}$ from the surface of the heater with an angle of 120 $\mathrm{degrees}$ relative to each other are created by a honing machine to hold thermocouples.

The cylinder insulation is made of polytetrafluoroethylene (PTFE) whose inner and outer diameters are 45 $\mathrm{mm}$ and 100 $\mathrm{mm}$, respectively. The lower part of the insulation is fastened to the upper part by three socket head cap screws and holds the copper heater. Three springs with a length of 38 mm have been used to eliminate the effect of the expansion of socket head cap screws due to the heat and also to eliminate the threads of the screw holes. In addition to PTFE insulation, a 20 mm thick elastomeric insulation is wrapped around the heater and the insulation assembly. The thermal conductivity of the heater, PTFE insulation, and elastomer are measured using $K{D}_2$ and are 400 $\mathrm{W}/\mathrm{K}$, 0.25 $\mathrm{W}/\mathrm{K}$, and 400 $\mathrm{W}/\mathrm{K}$, respectively. Radial conductive heat transfer can be neglected, and the thermal conductivity transferred to the surface of the heater can be considered one-dimensional.

The boiling tank is a cylinder made of Pyrex whose diameter, height, and thickness are 60 $\mathrm{mm}$, 300 $\mathrm{mm}$, and 5 $\mathrm{mm}$, respectively. Moreover, it is glued to the insulation with Permatex glue. For sealing the tank and insulation, as well as the surface of the heater and insulation, two Whitney O-rings with a diameter of 59 and 40 mm, respectively, with a temperature tolerance of 250 °C, have been used.

A condenser has been used to condense the boiling vapors and keep the amount of boiling fluid constant. The condenser is made of Pyrex, which is 400 $\mathrm{mm}$ long, and its diameter is 50 $\mathrm{mm}$. Teflon tape is used to seal the end of the condenser. To cool the circulating condenser, water is pumped at a rate of 21 mL/s from the water and ice tank and then returned to the tank. The water and ice tank temperature are 20 °C and is controlled by a mercury thermometer. Usually at the end of each boiling test at high heat fluxes, more ice molds are used due to the increase in circulating water temperature.

A voltage controller (Auto-trans) with a maximum of 300 volts is used to adjust the voltage and achieve the desired power at each stage of the boiling test. An ammeter is used to measure current with an accuracy of 0.1 amps, and a voltmeter is used to measure voltage with an accuracy of 1 volt. The thermocouples are 80 mm long and 4 mm in diameter and are of PT100 type, and have been calibrated by the oil bath method in the calibration laboratory. Also, the Autonics temperature display model has an accuracy of 0.1 degrees.

2.3. Boiling Surface

The smooth boiling surface has a diameter of 40 mm; before each test it is sanded with sanding sheets No. 400, 600, and 800 using a lathe, and then its roughness is measured by a roughness tester. The average surface roughness of the boiling surface in all experiments is 213 nm (Figure 4). Rectangular, circular, and triangular grooved surfaces are made by wire cutting and have roughness similar to the smooth surface. The surface inside the grooves is sanded manually at a uniform speed before each test. Figure 5 shows the grooved surfaces of the boiling surface.

2.4. Relationships between Surface Temperature and Heat Flux

All experiments were started at low heat flux, 53 volts, and 2 amps. After boiling the fluid, the temperatures, voltage, and current are read, and the voltage of the device is added by the 5-volt voltage controller. Then the data is recorded after 5 min when the temperatures of the thermocouples are fixed. After the current reaches 6 amps, before the phenomenon of critical heat flux (CHF), the voltage is slowly reduced, and then the system is cut off. The amount of heat flux is calculated from Equations (1) and (2). The amount of heat loss is equal to the difference between the two values of heat flux obtained from Equations (1) and (2), which was less than 5% in all experiments.

$$q=K\frac{T_2-T_1}{\Delta x_2}$$

(1)

$$q=\frac{VI}{\pi D^2/4}$$

(2)

In Equations (1) and (2), $k$ is the thermal conductivity of the copper heater, and $T_1$ and $T_2$ are the temperature of thermocouples 1 and 2 at a distance of 8 and 18 mm from the boiling surface, respectively. $\Delta X_2$ is the distance between the two holes of the thermocouples. $V$, $I$, and $D$ are the voltage, the electric current, and the diameter of the heater.

The surface temperature is calculated by extrapolating the temperatures $T_1$ and $T_2$ due to the one-dimensional conductive heat transfer (Equation (3)). The HTC is obtained from Equation (4). $\Delta X_1$ is the $\mathrm{distance}\mathrm{between}\mathrm{thermocouple}$ number 1 and the boiling point, and $T_{sat}$ is the saturation temperature of the fluid.

$$T_w-T_{sat}=T_1-\frac{\Delta X_1}{\Delta X_2}\left(T_2-T_1\right)-T_{sat}$$

(3)

$$h=\frac{q}{T_w-T_{sat}}$$

(4)

2.5. Uncertainty Analysis

In this research, the Holman method [48] is utilized for uncertainty analyses. In this method, if the parameter F is a function of the variables $x_1, x_2,\dots , x_n$, then its uncertainty value is obtained from Equation (5). U is the amount of measurement error.

$$\frac{U_F}{F}={\left[{\displaystyle \sum }_{i=1}^n{\left(\frac{U_{x_i}}{x_i}\right)}^2\right]}^{0.5}$$

(5)

Therefore, the amount of heat flux uncertainty, surface and fluid temperature difference, and HTC are obtained from Equations (6)–(8), respectively. For all experiments, the highest heat flux uncertainty and HTC were 5.1 and 6.87%, respectively.

$$\frac{U_{q″}}{q″}=\sqrt{{\left(\frac{U_k}{k}\right)}^2+{\left(\frac{U_{\left(T2-T1\right)}}{T_2-T_1}\right)}^2+{\left(\frac{U_{\Delta x_2}}{\Delta x_2}\right)}^2}$$

(6)

$$\frac{U_{\left(T_w-T_{sat}\right)}}{\left(T_w-T_{sat}\right)}=\sqrt{{\left(\frac{U_{\Delta x_1}}{\Delta x_1}\right)}^2+{\left(\frac{U_{\left(T2-T1\right)}}{T_2-T_1}\right)}^2+{\left(\frac{U_{\Delta x_2}}{\Delta x_2}\right)}^2}$$

(7)

$$\frac{U_h}{h}=\sqrt{{\left(\frac{U_q}{q}\right)}^2+{\left(\frac{U_{\left(T_w-T_{sat}\right)}}{\left(T_w-T_{sat}\right)}\right)}^2}$$

(8)

3. Results

In each experiment, 150 mL of deionized water or nanofluid is added into the boiling chamber, and experiments are in the pre-CHF range. The laboratory temperature is 20 °C, and the ambient pressure is 623 mmHg. To ensure accuracy and repeatability, each experiment was performed three times in three days under the same conditions.

3.1. Accuracy of Results

Boiling of deionized water was performed on a flat surface, and its data was compared with the proposed equation of Rohsenow [49], Equation (9). In this equation, $h_{fg}$, $\sigma$, ${\rho }_l$, ${\rho }_v$, $Pr$, $C_{p,l}$, and ${\mu }_l$ $\mathrm{are}\mathrm{the}\mathrm{latent}$ boiling heat, surface tension, density of liquid, density of vapor, Prandtl number, the specific heat capacity of water, and the viscosity of liquid, respectively. $C_{sf}$ and $n$ are 0.013 and 1 for smooth cooper surface, respectively.

$$T_w-T_{sat}=\frac{h_{fg}{C}_{sf}}{c_{p,l}}{\left[\frac{q}{{\mu }_l{h}_{fg}}{\left(\frac{\sigma }{g\left({\rho }_l-{\rho }_v\right)}\right)}^{0.5}\right]}^{\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$3$}\right.}P{r}^n$$

(9)

Figure 6 shows the deionized water boiling curve and the Rohsenow’s curve, and there is a good correlation between the deionized water reference data [49]. Heat flux uncertainty is also shown in Figure 6.

3.2. Ferrofluid Boiling of Iron Oxide/Water on the Surface of the Smooth Heater

Boiling tests were performed for 0.01% volumetric iron oxide/water ferrofluid on a smooth surface. The diagram of heat flux in terms of excess temperature and HTC in terms of heat flux of iron oxide/water ferrofluid in comparison with deionized water is seen in Figure 7.

The saturation temperature of nanofluid is 96 °C. According to Figure 7, in all heat fluxes, the ferrofluid boiling diagram is shifted to the left, meaning that at the same heat flux, a smaller excess temperature difference is required. The HTC of nanofluid boiling relative to water increased by an average of 25.9%. This value is the average percentage increase of the boiling HTC results obtained from Equation (10).

$$\Delta h=\frac{h_{nf}-h_w}{h_w}\times 100$$

(10)

In Equation (10), $h_{nf}$ and $h_w$ are the nanofluid and water HTC, respectively. The reason for the improvement of heat transfer can be considered the high thermal conductivity of nanoparticles. Further, due to the average surface roughness (213 nm), and the average diameter of nanoparticles (15 nm), the value of the nanoparticle surface interaction parameter ($\phi$) is greater than one, which increases the HTC of the nanofluid.

3.3. Boiling Deionized Water on Grooved Surfaces

In this section, the deionized water boiling test was performed on three types of rectangular, circular, and triangular grooved surfaces, which can be seen in Figure 8. Boiling water on the surfaces with circular and triangular grooves had the highest and lowest HTC, respectively. The average HTC in boiling deionized water on the surfaces with circular and rectangular grooves increased by 91.5% and 48.7%, respectively; additionally, on the surface with triangular grooves, it decreased by 32.9%. It is assumed that this result is due to the unique geometry of these grooves, which can prevent proper bubble propagation and fluid flow. The sharp angles and corners of triangular grooves are likely to trap bubbles and cause local thermal resistance, which reduces heat transfer. In addition, nanoparticles have the potential to amass substantially in these corners, creating an insulating layer that impedes the passage of heat. Additional experimental investigations are planned to study this matter.

The areas of the central groove in rectangular, circular, and triangular surfaces are 1.6 ${\mathrm{cm}}^2$, 1.256 ${\mathrm{cm}}^2$, and 1.12 ${\mathrm{cm}}^2$, respectively. Although the area of a rectangular grooved surface is larger than a circle, due to the difference in the geometry of the groove, the boiling heat transfer on the circular grooved surface is improved. According to the results of Dadjoo et al. [22], by changing the position of the heater surface from horizontal to vertical, the intensity of boiling heat transfer decreases. In fact, the surface wettability on the vertical wall of the groove is less than the horizontal surface, because the contact angle of the bubble in the vertical surface is greater than the horizontal. In this paper, half of the surface of the rectangular groove is in the vertical position [50,51,52,53,54,55]. Also, in the rectangular groove, the presence of corners makes it more difficult to separate the bubbles from the surface. In a triangular groove, due to the slipping of bubbles and their accumulation in the center of the groove, the thermal resistance increases and the HTC decreases compared to deionized water.

3.4. Boiling of Iron Oxide/Water Nanofluid on Grooved Surfaces

A pool boiling test of 0.01% iron oxide/water nanofluid was also performed on grooved surfaces, the results of which can be seen in Figure 9. Due to the increase in surface area in grooved surfaces compared to the smooth surface, it is expected that the boiling heat transfer on the grooved surface will improve, but in rectangular and triangular grooved surfaces, the results have been contrary to expectations. HTC increased on average in ferrofluid boiling on the surface with the circular groove by 41.2%, but the percentage has decreased in rectangular and triangular surfaces by 22.3 and 89.2, respectively, compared to the smooth surface. The reason for this result can be considered to be increased accumulation of nanoparticles in the corners of the rectangular and triangular grooves and the formation of micro-layers under the bubbles, which clogs the nucleation sites and reduces the HTC. Easier separation of bubbles in circular grooves also improves heat transfer [56,57,58].

3.5. Effect of Groove Depth of Grooved Surfaces on Boiling of Water and Iron Oxide/Water Nanofluid

The depth of grooves in the grooved surfaces was increased from 1 mm to 2 mm by wire cutting. The roughness of the surfaces is the same as in the previous levels. Boiling experiments of deionized water and iron oxide/water nanofluid were performed on grooved surfaces with greater depth, which is shown in Figure 10 and Figure 11 of their boiling curves, respectively. According to Figure 10 and Figure 11, increasing the depth of the grooves has improved the boiling heat transfer of both water and nanofluid.

The average HTC of deionized boiling water on rectangular, circular, and triangular grooved surfaces with a depth of 2 mm has increased by 19.1%, 12.4%, and 42.9%, respectively, compared to a depth of 1 mm. In iron oxide/water nanofluid boiling, the average HTC of boiling in rectangular, circular, and triangular grooved surfaces has increased by 41.2%, 22.3%, and 41.4%, respectively, with increasing depth. As depth increases, the heat transfer surface area increases, the density of the nucleation sites increases, and the number of bubbles formed increases. In addition, when adding depth by wire cutting, the cracks in the grooves increase, which adds to the number of nucleation sites.

Groove geometry has a significant effect on the heat transfer efficiency of the fluid flow. In particular, the circular groove utilizes the symmetry and, therefore, demonstrates a better heat transfer and bubble release performance. However, in the case of triangular grooves, as there are sharp corners which could trap bubbles, a considerable thermal resistance is created [59,60,61]. Therefore, in that case, the heat transfer coefficient drops at a significant rate.

Our research demonstrates that using various groove geometry could substantially affect the heat transfer rate and critical heat flux. In particular, in the case of circular grooved surfaces, the heat transfer has increase significantly because of the better bubble behavior in these grooves. However, for triangular grooves, the mentioned problems would cause a drop in the HTC.

4. Conclusions

In this paper, boiling of deionized water and nanofluid of iron oxide/water with a volumetric concentration of 0.01% on smooth surfaces and grooved surfaces has been investigated experimentally. The use of nanofluid improves the HTC, provided that the surface-nanoparticle interaction parameter is greater than one, and the nanofluid is completely stable. The type and depth of grooves also affect the boiling results.

The most important results of the experiments are:

• In boiling deionized water on grooved surfaces, the surface with circular and rectangular grooves improves heat transfer, and the surface with triangular grooves reduces the heat transfer rate compared to the smooth surface. In a triangular groove, the thermal resistance increases due to the accumulation of bubbles in the center of the groove.
• The HTC in ferrofluid boiling increased on circular grooved surfaces and decreased on rectangular and triangular grooved surfaces due to the deposition of more nanoparticles in rectangular and triangular grooves than in circular ones due to the presence of groove corners.

By increasing the depth of the grooves, due to the increase of the heater surface, the boiling heat transfer rate of both water and ferrofluid has increased.