Effect of Magnetic Field on the Forced Convective Heat Transfer of Water–Ethylene Glycol-Based Fe₃O₄ and Fe₃O₄–MWCNT Nanofluids

Abstract

This paper discusses the forced convective heat transfer characteristics of water–ethylene glycol (EG)-based Fe₃O₄ nanofluid and Fe₃O₄–MWCNT hybrid nanofluid under the effect of a magnetic field. The results indicated that the convective heat transfer coefficient of magnetic nanofluids increased with an increase in the strength of the magnetic field. When the magnetic field strength was varied from 0 to 750 G, the maximum convective heat transfer coefficients were observed for the 0.2 wt% Fe₃O₄ and 0.1 wt% Fe₃O₄–MWNCT nanofluids, and the improvements were approximately 2.78% and 3.23%, respectively. The average pressure drops for 0.2 wt% Fe₃O₄ and 0.2 wt% Fe₃O₄–MWNCT nanofluids increased by about 4.73% and 5.23%, respectively. Owing to the extensive aggregation of nanoparticles by the external magnetic field, the heat transfer coefficient of the 0.1 wt% Fe₃O₄–MWNCT hybrid nanofluid was 5% higher than that of the 0.2 wt% Fe₃O₄ nanofluid. Therefore, the convective heat transfer can be enhanced by the dispersion stability of the nanoparticles and optimization of the magnetic field strength.

1. Introduction

Heat exchange is an essential process in various industrial applications, and forced convective heat transfer occurs during heat exchange processes. Different properties of heat transfer fluids are significant for effective heat transfer. Increasing the heat transfer area of a heat exchanger, increasing the turbulence effect by changing the shape at the heat transfer channel entrance, and improving the heat transfer by reducing the channel size are some of the methods of increasing the heat transfer efficiency. However, these methods are limited due to various constraints in modifying the heat exchanger to enhance the heat transfer efficiency. Therefore, to improve the thermal properties of working fluids, researchers have synthesized heat transfer fluids using different nanoscale particles with high thermal conductivities with existing fluids and have performed various studies on them. These fluids are generally called nanofluids. Nanofluids are prepared by uniformly dispersing metal or non-metallic nanoparticles with high thermal conductivities in conventional working fluids such as water, oil, and ethylene glycol. The thermal conductivity of nanofluids is much higher than that of conventional working fluids. Choi et al. [1] were the first to publish nanofluids, and various studies on nanofluids actively followed.

Several studies have been conducted on the characteristics of the forced convective heat transfer for a single nanofluid. Jarahnejad et al. [2] experimentally studied viscosity characteristics according to the concentration and size of Al₂O₃ and TiO₂ nanofluids and the effect of surfactant addition. The results indicated that the viscosity decreased with increasing temperature, and the addition of a surfactant secured the dispersion stability of the nanofluid and increased the viscosity. The thermal conductivity and viscosity of Al₂O₃ nanofluid were studied by H. W. Chiam et al. [3]. Different ratio of water–ethylene glycol (EG) was used as base fluid. The result showed that the thermal conductivity of the nanofluid was increased, and viscosity was decreased with the increase in EG percentage. Ashrafmansouri et al. [4] confirmed an improvement of micro-convection of less than 10% by measuring the tracer diffusion coefficient of tert-butanol in a water-based silica nanofluid according to temperature and various concentrations. In addition, Yu et al. [5] studied the flow and heat transfer characteristics of Al₂O₃ nanofluids in microchannels using the lattice Boltzmann method. The experimental results confirmed that the average Nusselt number (Nu) increased as the Reynolds number (Re) and nanofluid concentration increased. Yu et al. [6,7] examined the heat transfer characteristics of various nanofluids in turbulent flow and confirmed their effect on improving the heat transfer performance of nanofluids. They also reported that the heat transfer was enhanced by approximately 18% in nanofluids with low concentrations (<2.00 vol%) at high temperatures. Hu et al. [8] studied the forced convective heat transfer of a solar salt-based Al₂O₃ nanofluid. They demonstrated that the specific heat and heat transfer coefficient increased by 12.3% and 7.26%, respectively, for a 2 wt% Al₂O₃ nanofluid. Rostami et al. [9] made a review on the experimental and numerical studies of the control parameters that influence the natural convection in different shaped cavities with and without nanofluids. The forced convection can be increased by the addition of nanoparticles, magnetic fields, fins, and porous structures, but they reduce the free convection. Mamand [10] used the Prasher analytical model for calculating the thermal conductivity of nanofluids with Al₂O₃, CuO, ZnO, and SiO₂ nanoparticles and water/ethylene glycol as base fluid. Goncalves et al. [11] performed a numerical study on the optimization of microchannel geometry for nanofluid flow using COMSOL Multiphysics software. The result showed that the rectangular shaped collector shows greater uniformity in the distribution of the hat transfer in microchannel.

Mikkola et al. [12,13] investigated the convective heat transfer performance of nanofluids containing solid–liquid phase change materials. They investigated properties such as nanoparticle size and thermal conductivity, which affected the convective heat transfer, and reported that nanofluids with small nanoparticles were desirable for convective heat transfer applications. Sheikholeslami et al. [14] studied the characteristics of the forced convective heat transfer of Fe₃O₄–water nanofluids using the finite element method. Saarinen et al. [15] experimentally studied the convective heat transfer characteristics of decane and micelle nanofluids in the turbulent region. The results indicated that the Nu number increased up to 15% at a high Re number. Sundar et al. [16] demonstrated that when a 0.6 vol% Fe₃O₄ nanofluid was used, the heat transfer coefficient and friction factor increased by 31% and 10%, respectively, compared with the existing working fluid. Ibrahim et al. [17] comparatively studied the thermal conductivity of graphene nanoparticles/ethylene glycol nanofluid and ethylene glycol nanofluid in the temperature range of 25–79 °C. The result showed that the incorporation of nanoparticles increases the thermal conductivity. More addition of nanoparticles reduces the positive effect on thermal conductivity, and the increase in temperature strengthens the positive effect.

Recently, studies on the forced convective heat transfer characteristics of hybrid nanofluids in which two or more materials are dispersed in a fluid to complement the shortcomings and develop more effective nanofluids have been actively conducted. Better convective heat transfer characteristics can be achieved using a hybrid nanofluid. Han et al. [18] prepared a nanofluid by adding three types of nanoparticles, namely Al₂O₃, Fe₂O₃, and carbon nanotubes (CNTs), to the existing fluid based on polyalphaolefin. The results confirmed that the thermal conductivity increased by 21% compared with that of the existing fluid. Khosravifard et al. [19] studied the thermal properties of TiO₂ and CNTs by dispersing them in a mixture of water and propylene glycol (50:50), and the results confirmed that the thermal conductivity of the nanofluid increased compared with that of the conventional fluid. Chen et al. [20] prepared Ag–multiwall carbon nanotubes (MWCNT) nanofluids, which exhibited a higher thermal conductivity than single MWCNT nanofluids. Sundar et al. [21] studied the heat transfer and friction coefficient of an MWCNT–Fe₃O₄ hybrid nanofluid. When the concentration was 0.3 vol%, the Nu number increased by approximately 32.7%. Huang et al. [22] investigated the pressure drop and heat transfer characteristics of Al₂O₃ and MWNCT hybrid nanofluids. The experimental results indicated that the heat transfer coefficient was higher than that of the Al₂O₃ single nanofluid, with a lower pressure drop, confirming that it is suitable for heat transfer applications. Arevalo-Torres et al. [23] executed an experimental study of forced convective heat transfer in a coiled flow inverter using TiO₂-water nanofluid. The authors presented the determination of flow ranges for coil flow inverter and limitations on nanoparticle concentration with the aim of heat transfer enhancement.

Among the various types of nanofluids, magnetic nanofluids can improve heat transfer performance through the effect of a magnetic field and have been studied for different applications. Generally, magnetic nanofluids can enhance the efficiency of heat exchange devices such as heat pipes, thermal siphons, and heat exchangers by applying an external magnetic field. Recently, several studies have been conducted on the effect of the magnetic field on Fe₃O₄ nanofluids. Sha et al. [24,25] experimentally studied the effect of a magnetic field on the convective heat transfer characteristics of Fe₃O₄ nanofluids. At 40 °C with a 3 vol% and a magnetic field of 0 G, the convective heat transfer coefficient increased by 1.2–2.3% in the laminar flow region and 4.7–5.6% in the turbulent flow region. The experimental results confirmed that Fe₃O₄ nanofluids increased the heat transfer coefficients by 4.2% and 8.1% for a constant magnetic field strength at 800 G and a gradient magnetic field, respectively. In addition, Sheikholeslami et al. [26,27] analyzed the flow of Fe₃O₄ nanofluids in an elliptical cylinder and inside a moving wall in the presence of an external magnetic field using a simulation. The results indicated that the heat transfer improved as the Re increased, but the heat transfer decreased as the Hartmann number increased. Esmaeili et al. [28] prepared Fe₃O₄ nanofluids using the solvothermal method and observed that the heat transfer characteristics of magnetic nanofluids significantly improved when an external magnetic field was applied. Bahiraei et al. [29] modeled the convective heat transfer characteristics of Fe₃O₄ magnetic nanofluids in a rectangular channel using four magnets. When a magnetic field was applied, the average convective heat transfer coefficient increased by 40.8% and 58.2% compared with the Fe₃O₄ magnetic nanofluid without a magnetic field and pure water, respectively. Dogonchi et al. [30] studied the convective heat transfer characteristics of nanofluids in a new test section consisting of a wavy circular cylinder and a rhombus. The results indicated that the shape of the nanoparticles also affected heat transfer improvement. S. Ahangar Zonouzi et al. [31] experimentally investigated the heat transfer and flow performance of a magnetic nanofluid in a vertical tube under the magnetic quadrupole effect. The result showed that the maximum enhancement of 23.4%, 37.9%, and 48.9% in the local heat transfer coefficient for the magnetic nanofluid with 2 vol% of Fe₃O₄ in the presence of quadrupole magnets at three different positions for Re of 580 and the enhancement in heat transfer coefficient was decreased with increase in Re. Faouzi et al. [32] numerically investigated the magnetohydrodynamic natural convection of Fe₃O₄-water nanofluid in a round diagonal corner square cavity. The effect of various parameters such as Rayleigh number, Hartmann number on the nanofluid flow, and heat transfer was evaluated and discussed. The magnetic field effect on micro cross jet injection of water/Al₂O₃ nanofluids in a microchannel was investigated by Bagherzadeh et al. [33]. The nanofluid jet was used for the heat transfer enhancement under homogeneous magnetic field strength with Ha of 0, 20, and 40. Goshayeshi et al. [34] conducted an experimental study on the effect of inclinational angle on the heat transfer enhancement of a ferrofluid under the influence of a magnetic field. The heat transfer coefficient of the heat pipe was improved by increasing the input heat flux. The inclination angle of the heat pipe plays a significant role in the performance of the heat pipe. The same author has studied the effect of magnetic field on the heat transfer rate of kerosene/Fe₂O₃ nanofluid in a heat pipe [35]. The experimental results showed that the temperature difference between the surface and the vapour core of the evaporator section was 3.1 and 2 °C for with and without the influence of the magnetic field. Hajatzadeh et al. [36] numerically investigated the effect of radiation on the convective heat transfer rate and entropy generation of the nanofluid in a diagonal rectangular chamber. The result showed that the Nusselt number and entropy generation increase, and the Bejan number decreases with increasing Raleigh number. Moreover, the heat transfer rate and entropy generated were reduced, and the Bejan number increases by increasing the angle of the magnetic field.

Sun et al. [37] conducted an experimental study on the convective heat transfer characteristics of Fe₃O₄ magnetic nanofluids in a circular tube subjected to a magnetic field. When the 0.5 vol% Fe₃O₄ magnetic nanofluid had a Re of 1080 and magnetic field strengths of 415 and 700 G, the Nu number increased by 4.36% and 7.19%, respectively. Mehrali et al. [38] analyzed convective heat transfer and entropy generation of graphene–Fe₃O₄ hybrid nanofluid. The results indicated that the local convective heat transfer increased by up to 82%, and the total entropy production rate was reduced by 41% compared with water. Shahsavar et al. [39] experimented on forced convective heat transfer in laminar flow using a Fe₃O₄–MWCNT hybrid nanofluid in a constant magnetic field. Under the same Re of 548, the Nu number of a 0.5 vol% Fe₃O₄–1.35 vol% MWCNT hybrid nanofluid exhibited a maximum increase of 20.5% compared with the scenario without a magnetic field. Similarly, Alsarraf et al. [40] conducted an experiment on forced convective heat transfer in laminar flow for a Fe₃O₄–MWCNT hybrid nanofluid. When the Re was 500, the Nu number and pressure drop of 0.9 vol% Fe₃O₄–1.35 vol% MWNCT hybrid nanofluids increased by 109.31% and 25.02%, respectively, when compared to the scenario without a magnetic field. N. Abu-Libdeh et al. [41] investigated the effect of a constant magnetic field on Ag/MgO/H₂O nanofluid in a cavity form. The result showed that heat transfer was limited by the increase in Ha number, and therefore, the magnetic field can be used as a heat transfer controller. Shahravar et al. [42] conducted an experimental study on the preparation of the liquid paraffin-Fe₃O₄ mixture and further developed models using an artificial neural network for the prediction of thermal conductivity and viscosity. Bagherzadeh et al. [43] developed a new approach using an enhanced artificial neural network on functionalised-MWCNTs-Fe₃O₄/EG hybrid nanofluid. The proposed method provides more precise results in less time. In addition, Gao et al. [44] investigated the thermal conductivity and stability of graphene oxide-Al₂O₃ hybrid nanofluid. The result showed that the pH, dispersant, ultrasonication power, and duration affect the stability of the hybrid nanofluid.

Previous studies have demonstrated that nanofluids can increase the efficiency of heat exchange systems through forced convective heat transfer. In addition, studies on forced convective heat transfer characteristics using hybrid nanofluids consisting of two or more nanoparticles have been conducted. The magnetic field effect on the forced convective heat transfer characteristics using Fe₃O₄ magnetic nanofluids has also been studied. However, the research on hybrid magnetic nanofluids is insufficient, and it can hardly be found in the literature. Although hybrid magnetic nanofluids may improve the forced convective heat transfer characteristics through the synergistic effect of two or more nanoparticles, most studies on the effect of magnetic fields focused on single magnetic nanofluids. To date, no study has been conducted to confirm the synergistic effect of hybrid magnetic nanofluids by comparing the improvement rate of convective heat transfer of hybrid magnetic nanofluids with that of single magnetic nanofluids.

Generally, producing single and hybrid magnetic nanofluids for each concentration and experimentally confirming the characteristics of forced convective heat transfer according to changes in the magnetic field are time-consuming. Hence, experimental studies on the forced convection heat transfer performance of hybrid magnetic nanofluids are lacking. Therefore, this study investigated the thermal properties of the Fe₃O₄ nanofluid and Fe₃O₄–MWCNT hybrid nanofluid, such as thermal conductivity and viscosity, and the magnetic field effect on the forced convective heat transfer characteristics in the laminar flow region was experimentally analyzed. Experiments were conducted using the Fe₃O₄ nanofluid and Fe₃O₄–MWCNT hybrid nanofluid at concentrations of 0.025, 0.05, 0.1, and 0.2 wt%. A water and ethylene glycol mixture at a ratio of 80:20 was used as the base fluid. The forced convective heat transfer experiment was conducted in the laminar flow region with a Re range of 1000–1600. In addition, the magnetic field strength was changed from 0 to 750 G to investigate the effect of the magnetic field on the forced convective heat transfer characteristics of magnetic nanofluids.

2. Materials and Methods

2.1. Nanoparticles and Nanofluids

The Fe₃O₄ nanoparticles used in this study were prepared using a co-precipitation method. Commercially available MWCNTs (US Research Nanomaterials Inc., Houston, TX, USA) were used for hybrid nanofluid preparation.

Table 1. Physical properties of MWCNT and Fe₃O₄ nanoparticles.

Item	MWCNT	Fe3O4
Purity	>95%	99.9%
Thermal conductivity (W/m·K)	3000	80
True density (kg/m3)	1350	4950
Specific heat (J/kg·K)	650	670

shows the thermal properties of the Fe₃O₄ and MWCNT nanoparticles used in the experiments. Measured thermal properties are provided by nanoparticle manufacturers (AVENTION, Incheon, Korea). MWCNT nanoparticles have a high thermal conductivity of 3000 W/m·K, and Fe₃O₄ nanoparticles have a high density of 4950 kg/m³. Both nanoparticles have a similar specific heat of 650–670 J/kg·K. Therefore, for preparing Fe₃O₄–MWCNT hybrid nanofluid in which the two nanoparticles were dispersed in the base fluid at an appropriate ratio, synergy is expected to be generated in the direction with excellent thermal properties through thermal conductivity and the magnetic effect.

The nanofluid used in this study was prepared using a two-step process. Initially, the base fluid was prepared by mixing water and EG at a ratio of 80:20. EG is known for its low freezing temperature and is widely used as a working fluid in heat transfer systems, particularly in cold winters. An 80:20 ratio water–EG mixture is commonly used in industries to prevent freezing and minimize viscosity while increasing conductivity [45]. Fe₃O₄ nanoparticles, which were prepared using the co-precipitation method, were added to the base fluid, stirred for 1 h with a stirrer, and sonicated for 3 h using an ultrasonic disperser. To ensure reliable dispersion stability of the Fe₃O₄ nanofluid, a small amount of NaOH was added to adjust the pH to 11. This is because when the pH value of the Fe₃O₄ nanofluid is 11, magnetite is formed, and the dispersion stability can be ensured [46]. To prepare the Fe₃O₄–MWCNT hybrid nanofluid, Fe₃O_4, and MWCNT nanoparticles were added to the base fluid at a 1:1 ratio, and the mixture was stirred for 1 h. After stirring, the mixture was sonicated for 2 h using an ultrasonic disperser. In addition, a small amount of Arabic gum was added as a surfactant to disperse the MWCNT nanoparticles in the base fluid evenly [47].

In this study, Fe₃O₄ nanofluids were prepared at different concentrations of 0.025, 0.05, 0.1, and 0.2 wt%, and Fe₃O₄–MWCNT hybrid nanofluids were prepared by adding Fe₃O₄ and MWCNT nanoparticles at a 1:1 ratio. The prepared Fe₃O₄ nanofluids and Fe₃O₄–MWCNT hybrid nanofluids are shown in Figure 1. The Fe₃O₄ nanofluid was dark brown, and as the concentration increased, the nanofluid became darker. The Fe₃O₄–MWCNT hybrid nanofluid was black compared with the base fluid owing to the effect of MWCNT nanoparticles. To investigate the dispersion stability of the manufactured nanofluids, observation and measurement of zeta potential methods were conducted simultaneously. A one-month visual inspection of the nanofluids manufactured in this study revealed no significant precipitation. In addition, the zeta positional values of the Fe₃O₄ nanofluid and Fe₃O₄–MWCNT hybrid nanofluid were measured and observed to be stable at −34 to −47 mV and −32 to −56 mV, respectively. Generally, when the measured Zeta potential is more than the ±30 mV, it is judged that the dispersion stability of nanofluids is secured [48]. The dispersion stability of the prepared Fe₃O₄ nanofluid and Fe₃O₄–MWCNT hybrid nanofluid was confirmed through visual inspection and zeta potential tests.

2.2. Experimental Methods

2.2.1. Thermal Conductivity Measurement

Thermal conductivity is a significant parameter for evaluating forced convective heat transfer characteristics. In this study, the thermal conductivity was measured using the transient hot-wire method. The KD2 Pro (Decagon Devices, Inc., Pullman, DC, USA) device used for the thermal conductivity measurements is shown in Figure 2. The thermal conductivity was measured by placing the nanofluid in a constant temperature atmosphere by using a double water bath method.

2.2.2. Viscosity Measurement

The viscosity of the nanofluid prepared with different concentrations of nanoparticles was measured using a viscosity measuring device (SV-10) that operates based on the tuning fork vibration method. A photograph of the instrument is shown in Figure 3. The error rate of the viscosity measuring device is ±1%. The viscosity of the nanofluid was measured at a constant temperature maintained in the viscosity measurement device using a thermostat. Moreover, in order to increase the accuracy of the viscosity, the average viscosity was used through more than 5 repeated measurements.

2.2.3. Measurement of Forced Convective Heat Transfer

The forced convective heat transfer of Fe₃O₄ nanofluid and Fe₃O₄–MWCNT hybrid nanofluid was experimentally measured. A schematic of the experimental setup for measuring forced convective heat transfer is shown in Figure 4. The nanofluid from the storage tank was circulated through the tubular test section using a pump. The nanofluid was cooled before returning to the storage tank using the heat exchanger. The test section tube is made of stainless steel with an outer diameter of 6.3 mm, an inner diameter of 3.7 mm, and a length of 1500 mm. A heating wire was used to provide a constant heat flux to the test section tube. To stabilize the fluid in the test section, straight section of 1 m was provided at the inlet and outlet of the test section. In addition, in order to minimize heat loss to the surrounding, an insulation material was installed with a thickness of 10 cm or more using two layers of insulation material, and a silicone tube was used before and after the test section to reduce heat conduction loss through the test section tube. Figure 5 shows the magnetic field device used to investigate the characteristics of forced convection heat transfer according to the magnetic field effect of the magnetic nanofluid. Magnetic field devices were installed on both sides of the test section tube (Figure 4). Magnetic field strengths of 250, 500, and 750 G were applied in this study.

2.3. Data Reduction

The density and specific heat of the Fe₃O₄ nanofluid used in this study can be expressed by Equations (1) and (2) [49].

$${\rho }_{nf}=(1-\varphi ){\rho }_{bf}+\varphi {\rho }_{np}$$

(1)

$$c_{p,nf}=\frac{(1-\varphi ){\rho }_{bf}{c}_{p,bf}+\varphi {\rho }_{np}{c}_{p,np}}{(1-\varphi ){\rho }_{bf}+\varphi {\rho }_{np}}$$

(2)

where $\varphi$ is the volume fraction of the nanofluid, ${\rho }_{nf}$ and ${\rho }_{np}$ are the densities of the base fluid and nanoparticles, respectively, and $c_{p,bf}$ is the specific heat of the nanofluid.

The density and specific heat of the Fe₃O₄–MWCNT hybrid nanofluid were calculated using Equations (3) and (4) [50].

$${\rho }_{np}=\frac{{\varphi }_1{\rho }_{np1}+{\varphi }_2{\rho }_{np2}}{\varphi }$$

(3)

$$c_{p,nf}=\frac{{\varphi }_1{\rho }_{np1}{c}_{p,np1}+{\varphi }_2{\rho }_{np2}{c}_{p,np2}}{{\rho }_{np}\varphi }$$

(4)

The local convective heat transfer coefficient was calculated using Equation (5).

$$h=\frac{Q}{A(t_w-t_b)}$$

(5)

The bulk temperature of the fluid at each point where the thermocouple was attached was calculated using the measured inlet and outlet temperatures, specified specific heat, and heat flux measured using a flow meter. The local convective heat transfer coefficient at that point was calculated using the difference between the bulk temperature of the fluid and the temperature measured by the thermocouple attached to the surface of the tube. By applying the convective heat transfer coefficient calculated using Equation (5) and the measured thermal conductivity of the nanofluid, the Nu number of the nanofluid was calculated using the following equation:

$$Nu=\frac{hD}{k}$$

(6)

In this study, before experimenting on the characteristics of the convective heat transfer coefficient of the nanofluid, the reliability of the fabricated experimental device was verified by comparison with the widely known Shah’s equation in the laminar flow region, as presented in Equation (7), with deionized water as the working fluid [25].

$$Nu(x)=\{\begin{array}{cc}1.953{\left(\mathrm{Re}.\mathrm{Pr}.\frac{D}{x}\right)}^{0.33333}& (\mathrm{Re}.\mathrm{Pr}.\frac{D}{x})\ge 33.3\\ \mathrm{4.3640.0722}\mathrm{Re}.\mathrm{Pr}.\frac{D}{x}& (\mathrm{Re}.\mathrm{Pr}.\frac{D}{x})<33.3\end{array}$$

(7)

Figure 6 shows the comparison between the measurement results obtained in this study and the Nu number of water suggested by Shah. As Figure 6 shows, the maximum error between the experimental results using water in the laminar flow region with a Re of 1000–1600 and the result of Shah’s equation was approximately 4.75%, and the average error was approximately 3.25%. These results confirmed the reliability of the fabricated experimental device.

In this study, the factors affecting the uncertainty of the obtained experimental results included a T-type thermocouple, differential pressure transmitter, mass flow meter, power meter, and magnetic strength meter. The uncertainties of the experimental devices used in this study are listed in

Table 2. Uncertainties of measuring devices.

Parameter	Uncertainty (%)
T-Type thermocouple (K)	0.75
Power meter (V)	3.0
Pressure transmitter (PSID)	0.2
Mass flow meter (kg/s)	1.0
Magnetic strength meter (G)	5.0
Convective heat transfer coefficient (W/m2 K)	0.3
Nusselt number	0.3

.

3. Results and Discussions

3.1. Characteristics of Nanoparticles

Figure 7 shows the transmission electron microscopy (TEM) images of the Fe₃O₄ nanoparticles and Fe₃O₄–MWCNT hybrid nanoparticles used in this experiment. The Fe₃O₄ nanoparticles were as spherical as most conventional nanoparticles and had an average size of 20 nm. In addition, as shown in Figure 7b, the Fe₃O₄ nanoparticles were attached to the surface of the MWCNT by cohesion, thus forming Fe₃O₄–MWCNT hybrid nanoparticles. The thickness of the MWCNT nanoparticles with Fe₃O₄ nanoparticles was 10–30 nm.

The structures of the Fe₃O₄ and Fe₃O₄–MWCNT nanoparticles were investigated using X-ray diffraction (XRD). The XRD results for Fe₃O₄ and Fe₃O₄–MWCNT nanoparticles are shown in Figure 8. The XRD measurement results of the Fe₃O₄ nanoparticles indicated the characteristics of a cubic spinel structure at (220), (311), (400), (422), (511), and (440). The strongest reflection was observed at (311) [51]. The XRD measurements of the Fe₃O₄–MWCNT hybrid nanoparticles indicated the characteristic of the cubic spinel structures of both the Fe₃O₄ and MWCNT nanoparticles at (002) and (100). This indicated that the structure of the MWCNTs was not modified during the synthesis of composite nanoparticles with Fe₃O₄ nanoparticles [52].

3.2. Thermal Conductivity and Viscosity of Fe₃O₄ and Fe₃O₄–MWCNT Hybrid Nanofluids

Figure 9 shows the thermal conductivity of the MWCNT and Fe₃O₄ single nanofluids and Fe₃O₄–MWNCT hybrid nanofluids of various concentrations at a temperature of 25 °C. The thermal conductivity of MWCNT nanofluids is used here for comparison from a previous study [45]. For Fe₃O₄ nanofluids, the thermal conductivity was 0.515, 0.518, 0.521, and 0.521 W/m·K at concentrations of 0.025, 0.05, 0.1, and 0.2 wt%, respectively. We observed that the thermal conductivity of Fe₃O₄ nanofluids was lower than that of MWCNT nanofluids at the corresponding concentrations. Adding a small amount of MWCNT nanoparticles to the Fe₃O₄ nanofluid can synergistically enhance its convective heat transfer due to the high thermal conductivity of the MWCNT nanoparticles. The thermal conductivity of the Fe₃O₄–MWCNT hybrid nanofluid was 0.516, 0.519, 0.524, and 0.531 W/m·K at concentrations of 0.025, 0.05, 0.1, and 0.2 wt%, respectively. Compared with the Fe₃O₄ single nanofluid, the thermal conductivity of the Fe₃O₄–MWCNT hybrid nanofluid increased by approximately 2%, which confirmed the low thermal conductivity of the Fe₃O₄ nanofluid through the addition of MWCNT nanoparticles.

Figure 10 shows the change in viscosity according to the various concentrations of the MWCNT and Fe₃O₄ single nanofluids and Fe₃O₄–MWNCT hybrid nanofluid at 25 °C. The viscosity values of MWCNT nanofluids were used for comparison from a previous study [45]. Since Fe₃O₄ nanoparticles had the highest density among the single nanoparticles, the viscosities of Fe₃O₄ nanofluids at 0.25, 0.05, 0.1, and 0.2 wt% were 1.81, 1.82, 1.85, and 1.87 mPa·s, respectively. As the concentration increased, the viscosity of the nanofluids increased slightly. By adding MWCNT nanoparticles to Fe₃O₄ nanofluids, the viscosities of the Fe₃O₄–MWCNT hybrid nanofluids were 1.75, 1.77, 1.81, and 1.83 mPa·s at concentrations of 0.25, 0.05, 0.1, and 0.2 wt%, respectively. According to the increase in the Fe₃O₄–MWCNT hybrid nanofluids concentration, the increase in viscosity exhibited a similar tendency to that of Fe₃O₄ nanofluid but had a lower viscosity than the Fe₃O₄ nanofluid. The results confirmed that the viscosity of the Fe₃O₄–MWCNT hybrid nanofluid was 2.6% lower than that of the Fe₃O₄ nanofluid on average.

3.3. Convective Heat Transfer Performance of Fe₃O₄ and Fe₃O₄–MWCNT Nanofluids under a Magnetic Effect

Figure 11 shows the enhancement in the convective heat transfer coefficient of the Fe₃O₄ nanofluid compared with that of the base fluid. Generally, the convective heat transfer coefficient of Fe₃O₄ nanofluids is higher than that of the base fluid. Under a constant Re, the convective heat transfer coefficient improvement was affected by the concentration and magnetic field intensity. Figure 11a shows the enhancement in the convective heat transfer coefficient of the Fe₃O₄ nanofluid according to the concentration compared with the base fluid at Re = 1000. At Re = 1000, the convective heat transfer coefficient of the base fluid was 1232 W/m² K, and the convective heat transfer coefficient of Fe₃O₄ nanofluid at a 0.2 wt%, the concentration increased by 21% to 1491 W/m² K. A similar trend was observed for other Re values other than 1000. As shown in Figure 11b–d, the increase in the convective heat transfer coefficient of the Fe₃O₄ nanofluid was compared with that of the base fluid at different Re values of 1200, 1400, and 1600. When Re numbers were 1200, 1400, and 1600, the convective heat transfer coefficients of the base fluid were 1268, 1323, and 1347 W/m² K, respectively. In addition, when the magnetic field was not applied, the convective heat transfer coefficients of the 0.2 wt%, Fe₃O₄ nanofluid were 1535, 1597, and 1616 W/m² K, respectively. Figure 11a shows that the average increase in the convective heat transfer coefficient was 20.6% for the 0.2 wt% Fe₃O₄ nanofluid and Re = 1000. As the concentration of the nanofluid increased, Brownian motion intensified, causing the nanoparticles in the tube to move irregularly and actively, thereby improving the convective heat transfer of the Fe₃O₄ nanofluid.

The convective heat transfer coefficient was increased by the increase in both the concentration of nanofluids and magnetic field strength. Under the magnetic field effect, the Fe₃O₄ nanoparticles formed a chain-like structure, thereby creating fine turbulence in the flow, resulting in improved convective heat transfer. Overall, as shown in Figure 11a–d, a significantly enhanced convective heat transfer coefficient at 750 G was observed, particularly the maximum convective heat transfer coefficient at 0.2 wt%. When the strength of the magnetic field was 750 G, the convective heat transfer coefficients of the 0.2 wt% Fe₃O₄ nanofluid were 1534, 1578, 1654, and 1676 W/m² K at Re = 1000, 1200, and 1400, respectively, which indicated an improvement of approximately 24.6% on average compared with the base fluid. As the concentration increased, the convective heat transfer coefficient was increased by forming more chain-shaped nanoparticles according to the effect of the magnetic field.

Figure 12 shows the enhancement in the convective heat transfer coefficient of the Fe₃O₄–MWCNT hybrid nanofluid compared with the base fluid as the concentrations of nanofluids and magnetic field strength varied. The maximum convective heat transfer coefficient of the Fe₃O₄–MWCNT hybrid nanofluid occurred at 0.1 wt%. In the absence of a magnetic field, the convective heat transfer coefficients of the 0.1 wt% Fe₃O₄–MWCNT hybrid nanofluids were 1514, 1555, 1620, and 1647 W/m² K for Re values of 1000, 1200, 1400, and 1600, respectively. It was confirmed that the convective heat transfer coefficient increased by 22.6% on average compared with the base fluid. When a magnetic field was applied to the Fe₃O₄–MWCNT hybrid nanofluid at various concentrations, the addition of Fe₃O₄ nanoparticles increased the convective heat transfer coefficient. Under the given operating conditions, the maximum convective heat transfer coefficient of the 0.1 wt% Fe₃O₄–MWCNT hybrid nanofluid was observed at 750 G. For Re values of 1000, 1200, 1400, and 1600, the convective heat transfer coefficients were 1545, 1597, 1679, and 1688 W/m² K, respectively, indicating an improvement of 25.9% on average compared with the base fluid.

Based on the thermal conductivity measurement results of magnetic nanofluids presented in Figure 9, the maximum convective heat transfer coefficient was expected in the 0.2 wt% Fe₃O₄–MWCNT hybrid nanofluid. However, in this experiment, the convective heat transfer coefficient decreased slightly compared with that at 0.1 wt%. This indicated that the thermal conductivity increases as the concentration of the nanofluid increases, but for hybrid nanofluids with two or more different nanoparticles, the dispersion stability can decrease compared with that of a single nanofluid. The agglomeration of nanoparticles occurs because the magnetic field forms fine turbulence, which increases convective heat transfer. However, convective heat transfer is considered to be decreased by blocking heat transfer between the inner wall and the fluid when more nanoparticles aggregate beyond the limit of dispersion stability.

3.4. Pressure Drop of Fe₃O₄ and Fe₃O₄–MWCNT Hybrid Nanofluids under a Magnetic Field

Figure 13 shows the pressure drop of the Fe₃O₄ nanofluid as the magnetic field intensity varied in the tube. As Re increased, the convective heat transfer coefficient of the working fluid and pressure drop increased. In addition, the pressure drop increased owing to the effect of the magnetic field. Figure 13a shows the pressure drop of the 0.025 wt% Fe₃O₄ nanofluid according to the change in the magnetic field. When the magnetic field strength was 0 G at Re = 1600, the pressure drop of 0.025 wt% Fe₃O₄ nanofluid was 2.55 kPa, and when the magnetic field strength was 250, 500, and 750 G under the same Re number, the pressure drops were 2.84, 2.85, 2.86 kPa, respectively. The pressure drop of the 0.025 wt% Fe₃O₄ nanofluid exhibited an average increment of 11.5% under a magnetic field. Figure 13b shows the pressure drop of the 0.05 wt% Fe₃O₄ nanofluid according to the change in the magnetic field. For Re = 1000, the pressure drops at 0, 250, 500, and 750 G were 1.7, 1.87, 1.88, and 1.9 kPa. At a maximum Re of 1600, the pressure drops were 2.68, 2.9, 2.9, and 2.95 kPa for magnetic fields of 0, 250, 500, and 750 G, respectively. As Figure 13c shows, when the magnetic field of the 0.1 wt% Fe₃O₄ nanofluid was varied as 0, 250, 500, and 750 G, the pressure drops at Re = 1000 were 1.78, 1.85, 1.88, and 1.9 kPa, respectively, and at a maximum Re of 1600, the pressure drops were 2.8, 2.9, 2.92, and 2.95 kPa, respectively. Similar to Figure 13a, the pressure drop increased by 10% according to the change in the magnetic field.

Figure 13d shows the pressure drop of the 0.2 wt% Fe₃O₄ nanofluid as the magnetic field varied. Even in the absence of a magnetic field, the pressure drop was high due to the addition of a large amount of Fe₃O₄ nanoparticles. The pressure drop of the 0.2 wt% Fe₃O₄ nanofluid without a magnetic field at Re = 1600 was 3.22 kPa. The pressure drops of the 0.2 wt% Fe₃O₄ nanofluid at 250, 500, and 750 G under the same Re were 3.35, 3.37, and 3.39 kPa, respectively, which increased by 4.73% on average compared with the pressure drop without the magnetic field. Based on the experimental results of this study, we confirmed that the pressure drop increased as the concentration of the Fe₃O₄ nanofluid increased in the laminar flow region. When the magnetic field strength was increased from 0 to 250 G, the pressure drop of the Fe₃O₄ nanofluid increased significantly, but the increase in the pressure drop decreased as the strength of the magnetic field increased. This is because the magnetic nanofluid flow increased the microturbulence in the tube through the aggregation and arrangement of Fe₃O₄ nanoparticles on the wall due to the magnetic field, thereby increasing the pressure drop and convective heat transfer coefficient.

Figure 14 shows the pressure drop of the Fe₃O₄–MWCNT hybrid nanofluid in the tube as the nanofluid concentration and magnetic intensity varied. Similar to the variation in the pressure drop of the Fe₃O₄ nanofluid, the pressure drop of the Fe₃O₄–MWCNT hybrid nanofluid increased as Re increased. When Re increased from 1000 to 1600, the overall pressure drop increased, with an average increment of approximately 52.7%. In addition, the pressure drop of the Fe₃O₄–MWCNT hybrid nanofluid increased as the magnetic intensity increased. When the magnetic field strength was increased from 0 to 750 G, an average pressure drop increase of 6.7% was observed for Re = 1000. In this study, the maximum heat transfer coefficient and Nu of the Fe₃O₄–MWCNT hybrid nanofluid were observed at 0.1 wt%, but the maximum pressure drop was observed at 0.2 wt%. This is due to the increase in thermal conductivity as the concentration of the nanofluid increased. However, as shown in Figure 10, the viscosity increased proportionally with the concentration; thus, the pressure drop increased. Although MWCNT nanoparticles were added to increase the thermal conductivity of Fe₃O₄ nanofluids, the MWCNT nanoparticles are considered as a significant factor that causes aggregation of particles, resulting in a high-pressure drop [20]. In addition, because of the effect on the external magnetic field, the dispersion stability of the two nanoparticles in the hybrid nanofluid reduced compared with that of the single nanofluid, which caused an increase in the pressure drop. Although the nanofluid concentration increased over a certain value, the decrease in the dispersion stability of the nanofluid resulted in a decrease in the heat transfer performance of the nanofluid as a result of the reduction in the convective heat transfer coefficient. Moreover, it was a major cause of the increase in the pressure drop of the fluid in a tube.

3.5. Comparison of Convective Heat Transfer Performance between Fe₃O₄ and Fe₃O₄–MWCNT Nanofluids

This study compared the convective heat transfer characteristics of a single nanofluid and a hybrid nanofluid. The convective heat transfer coefficient and Nu of the 0.2 wt% Fe₃O₄ and 0.1 wt% Fe₃O₄–MWCNT nanofluids presented the maximum convective heat transfer coefficient for each nanofluid, are compared and illustrated in Figure 15. When comparing the Fe₃O₄ and Fe₃O₄–MWCNT nanofluids, we observed that the convective heat transfer coefficient of the latter was higher than that of the former. This indirectly indicated that the effect of increasing the convective heat transfer coefficient of the Fe₃O₄ nanofluid was achieved through the addition of MWCNT nanoparticles with high thermal conductivities. The convective heat transfer coefficients of the 0.2 wt% Fe₃O₄ and 0.1 wt% Fe₃O₄–MWCNT nanofluids at Re = 1400 were compared. When the magnetic field strength was 0, 250, 500, and 750 G, the convective heat transfer coefficients of the 0.1 wt% Fe₃O₄–MWCNT hybrid nanofluid were 1620, 1650, 1662, and 1679 W/m² K, respectively, and those of the 0.2 wt% Fe₃O₄ nanofluid were 1597, 1618, 1631, and 1654 W/m² K, respectively. We confirmed that the convective heat transfer coefficients of the 0.1 wt% Fe₃O₄–MWCNT hybrid nanofluid increased by 1.73% on average compared with that of the 0.2 wt% Fe₃O₄ nanofluid, although the total concentration of the Fe₃O₄–MWCNT hybrid nanofluid was lower than that of a single Fe₃O₄ nanofluid.

Most previous studies reported that convective heat transfer characteristics are increased by the effect of an external magnetic field [21,22,23,24,25,26]. However, some authors have reported a tendency to decrease convective heat transfer characteristics due to an external magnetic field [39]. This is caused by the aggregation of magnetic nanoparticles on the inner wall of the tube due to the effect of an external magnetic field. Moreover, under the magnetic field effect, magnetic nanoparticles form a chain-like structure, thereby creating fine turbulence and micro-vortices in the flow of the fluid. However, studies indicated that the convective heat transfer could be reduced by blocking heat transfer if a large amount of magnetic nanoparticles aggregate in the tube. In this study, Fe₃O₄ nanoparticles were aggregated under the effect of an external magnetic field to form micro-vortices, thereby improving convective heat transfer. However, when the MWCNT nanoparticles were added to the Fe₃O₄ nanofluid, the convective heat transfer performance was lower than the expected value for heat transfer improvement considering the increase in thermal conductivity because more nanoparticles were aggregated under the effect of an external magnetic field. Therefore, the increase in convective heat transfer performance degraded as more nanoparticles clumped on the inner wall of the tube. Even if MWCNT nanoparticles were added to the Fe₃O₄ nanofluid, it is observed that the convective heat transfer coefficient did not significantly increase.

Figure 16 shows the comparison of Nu for the 0.2 wt% Fe₃O₄ and 0.1 wt% Fe₃O₄–MWCNT nanofluids. Similar to the convective heat transfer coefficient, the Nu was higher in the latter than in the former. As Re increased, Nu was also increased. The increase in Nu for the Fe₃O₄–MWCNT hybrid nanofluid was due to the increase in the Brownian motion of MWCNT with high thermal conductivities, thermal and physical properties, and large surface area of the nanoparticles. In addition, it was confirmed that the heat transferability of the Fe₃O₄–MWCNT hybrid nanofluid increased due to the effect of the magnetic field. In particular, when the magnetic field strengths were 0, 250, 500, and 750 G at Re = 1400, the Nu values of the 0.1 wt% Fe₃O₄–MWCNT hybrid nanofluid were 12.4, 12.6, 12.7, and 12.8, respectively, and the Nu values of the 0.2 wt% Fe₃O₄ nanofluid were 12.3, 12.4, 12.5, and 12.7, respectively. The Nu of the 0.1 wt% Fe₃O₄–MWCNT hybrid nanofluid increased by 1.20% on average compared with that of the 0.2 wt% Fe₃O₄ nanofluid. Similar to the convective heat transfer coefficient changes, the Nu is not increased significantly because the convective heat transfer coefficient increased due to the increase in Re at the same thermal conductivity. Few studies have compared the convective heat transfer characteristics of Fe₃O₄ nanofluids and Fe₃O₄–MWCNT hybrid nanofluids. However, similar trends have been reported in previous studies [31,32]. Due to the slight improvement in the convective heat transfer coefficient, Nu also exhibited a slight improvement rate.

4. Conclusions

An experimental study was conducted to investigate the magnetic field effect on the forced convective heat transfer in a tube according to the magnetic field strength of water–ethylene glycol-based Fe₃O₄ nanofluid and Fe₃O₄–MWCNT hybrid nanofluid. It is confirmed that the convective heat transfer coefficient increased as the concentration of the nanofluid and the magnetic field strength increased. Moreover, the convective heat transfer coefficient and pressure drop increased due to the effect of the external magnetic field. When the magnetic field strength was varied from 0 to 750 G, the maximum convective heat transfer coefficients were observed for the 0.2 wt% Fe₃O₄ and 0.1 wt% Fe₃O₄–MWNCT nanofluids, and the improvements were 2.78% and 3.23%, respectively. In addition, when the strength of the magnetic field was 750 G, the convective heat transfer coefficient of the 0.2 wt% Fe₃O₄ and 0.1 wt% Fe₃O₄–MWCNT nanofluids exhibited 24.6% and 25.9% improvements on average, respectively, when compared with the base fluid. The average pressure drops for the 0.2 wt% Fe₃O₄ and 0.2 wt% Fe₃O₄–MWNCT nanofluids were 4.73% and 5.23%, respectively. Furthermore, it was observed that the convective heat transfer coefficient of 0.1 wt% Fe₃O₄–MWCNT hybrid nanofluid was 5% higher than that of 0.2 wt% Fe₃O₄ nanofluid, although it had a lower concentration. This was because of the complex aspects of blocking heat transfer owing to the aggregation of Fe₃O₄ nanoparticles generated by the external magnetic field and the large amount of aggregates generated on the inner wall of the tube. Therefore, to increase the convective heat transfer of the Fe₃O₄–MWCNT hybrid nanofluid using an external magnetic field, the dispersion stability, concentration of the nanofluid, and the magnetic field strength should be optimized.