Experimental study on spray cooling heat transfer enhancement using MWCNT and TiO 2 hybrid nanofluid

Abstract

In the present study, two nanofluids namely Multi-walled carbon nanotubes and Titanium dioxide dispersed in distilled water, have been used with three different concentrations; 0.1%, 0.2%, and 0.3% by volume, to investigate heat transfer enhancement in spray cooling applications. A square copper specimen of 10 mm² has been spray cooled with a volumetric flux of 2.5 cm³/cm²s using a 0.4 mm diameter pressure nozzle. The results revealed that the average heat transfer coefficient and corresponding critical heat flux enhanced significantly at 0.2 vol% for both nanofluids in contrast to distilled water. The results showed an improvement of 21.9% and 26.3% in heat transfer coefficient for Multi-walled carbon nanotubes and Titanium dioxide nanofluid, respectively. The effect of hybrid nanofluid in various concentrations has also been investigated on heat transfer performance. The results obtained showed a maximum value of 425 W/cm² for critical heat flux and 4.42 W/cm²K for heat transfer coefficient, with a heat transfer coefficient improvement of nearly 40.31% in comparison to distilled water. This results in an enhancement ratio of 1.6 for critical heat flux and 1.4 for heat transfer coefficient. In contrast to single nanoparticle nanofluid, the hybrid nanofluid indicated a moderate improvement in heat transfer coefficient of 11.7% and 7.8% for Multi-walled carbon nanotube and Titanium dioxide nanofluids, respectively. It has been observed that heat transfer performance was first enhanced and then declined as the volume fraction increased from 0.1% to 0.3% for all the nanofluids.

1. Introduction

The increased demand for high-performance electronic devices has emphasized the use of efficient cooling methods. The conventional methods such as natural convection, forced convection, and heat sinks are proving to be insufficient to meet the high demand for cooling. As a result, aggressive cooling techniques such as jet impingement, microchannel, and spray cooling are becoming popular in high heat flux applications. Spray cooling is a low-cost method of dissipating heat from high heat flux electronic devices, datacenters, ¹ nuclear fuel rods, ² metallurgical processes, ³ and combustion technology. ⁴ Further, it has additional benefits of high cooling power, low surface superheat, least contact thermal resistance, and low working fluid inventory. ⁵

Many experimental and theoretical studies have been conducted to optimize spray characteristics such as flow rate, droplet diameter, droplet velocity in order to improve heat transfer. Other aspects such as enhanced surfaces and modification of the thermo-physical properties of the base fluid by incorporating additives (surfactants and nanofluids), have also been studied extensively. The primary goal of introducing nano-particles is to increase the rate of evaporation and improve the spread of spray droplets across the heater surface. ⁶

Nanofluids provide high thermal conductivity, internal micro-convection, large specific area, and less corrosion. ⁷ The word nanofluid is a new class of engineered fluids which have suspended nanoparticles in the base fluid that was first used by American scientist Choi. ⁸ Base fluid can be: water, acetone, refrigerants, oil or mixtures whereas metals, metal-oxides, carbon, and other materials can be used as nanoparticles.

Bansal et al. ⁹ prepared alumina-water nanofluid by a two-step method using ultrasonic bath to examine the effect of various concentration and flow rates on spray cooling heat transfer. The results revealed that heat flux values for nanofluid obtained were lower in the temperature range between 45 °C to 140 °C. However, unlike for water, in that range, there was no indicated CHF for nanofluid. The lower values were attributed to nanoparticle deposition on the surface, which influenced heat transfer. Wang et al. ¹⁰ investigated the effect of nanoparticles (Cu, CuO, and Al₂O₃) and surfactant on spray cooling heat transfer enhancement. The results showed that nanofluid containing Cu nanoparticles performed better than other nanofluids, with heat flux increasing from 3.36 MW/m² to 3.48 MW/m² as the particle volume fraction increased from 0.1% to 0.5%. Furthermore, the addition of Tween-20 to nanofluid improved stability, reduced agglomeration and decreased the contact angle, which improved wettability. Duursma et al. ¹¹ investigated the effect of alumina particles in dimethyl sulfoxide and ethanol at 0.1% and 3.23% by weight, respectively. The results did not reveal any significant enhancement in heat flux associated to nanofluids when compared to water. On contrary, Hsieh et al. ¹² performed experimental analysis with seven distinct types of nanofluids: Ag, Al, Al₂O₃, Fe₃O₄, SiO₂, TiO₂ and MWCNTs, which were dispersed in deionized water at three different concentrations, namely 0.04%, 0.07%, and 0.1% by volume. The heat transfer coefficient (HTC) and critical heat flux improved by 1.7 and 1.84 times, respectively, as the volume percentage increased from 0.04% to 0.1%, when compared to deionized water. Chakraborty et al. ¹³ investigated the effect of Cu-Zn-Al LDH nanofluid thermo-physical properties on cooling of high temperature steel plate. When the nanofluid concentration was 160 ppm, the cooling rate increased by 18.5% compared to water. Results also indicated that, in addition to agglomeration and poor suspension, heat transfer dissipation deteriorated above optimal concentration due to formation of a layer of nanoparticles over the surface that inhibit direct contact of coolant and surface. Kaya et al. ¹⁴ numerically calculated the effect of TiO₂/water nanofluid in semi-circular cross-sectioned micro-channels at concentrations ranging from 1% to 4%. In comparison to the base fluid, increasing the nanoparticle volume fraction improved the averaged Nusselt number by 10%.

It is evident from the literature that in jet/spray impingement heat transfer, most of the studies are related to the performance of the nanofluid above the superheat temperature. The shifting of the peaks in the boiling curve denotes the coolant efficiency to extract the high heat flux after the nucleate boiling region. But this efficiency restricted due to the poor stability of the nanofluids and also the lower thermal conductivity of the fluid. Another concern with using only one type of nanoparticle-based nanofluid is its monopoly in physical and chemical characteristics. ¹⁵ Metallic nanoparticles (Ag, Cu, and so on) have strong thermal characteristics, but they are less stable in base fluids. On the other hand, metal oxide nanoparticles (Al₂O₃, ZnO, etc.) are more stable in base fluids. These nanoparticles do, however, have an extremely low thermal conductivity value. As a result, researchers are currently focusing on the production and use of hybrid nanofluids. Hybrid nanofluids are composed of two or more types of nanoparticles disseminated in a base fluid by mixing. Many authors have studied and found enhanced thermo-physical properties suitable for heat transfer applications.^16,17

Researchers recently attempted to investigate nanofluid capabilities by dispersing more than one nanoparticle into the base liquid and referred to it as a hybrid nanofluid. This hybrid nanofluid has improved characteristics compared to single nanofluid due to synergistic effects. Siddiqui et al. ¹⁸ studied the impact of Silver (Ag)-Graphene (GNP) hybrid nanofluid droplet and residue size on evaporation at different mixing ratios. The results demonstrated that droplet evaporation increased with increasing residual size and decreased mixing ratio. At a mixing ratio of (0.1:0.9) of Ag:GNP, the highest evaporation rate of 370% was obtained. However, the evaporation rate diminished as the mixing ratio for Silver was increased. Barewar et al. ¹⁹ carried out an experimental investigation to investigate the effect of a hybrid nanofluid of silver and zinc oxide (Ag/ZnO). In comparison to water, the HTC and boiling performance of hybrid nanofluid improved by 88.53% and 87.17%, respectively. The study of hybrid nanofluids although complex, presents multiple opportunities for flow-boiling heat transfer applications. The area of nanofluids in spray cooling needs to be further explored.

In the above mentioned research studies, it can be found that the various authors have tried different nanoparticle (metals, oxides etc.) based nanofluids to better understand their effect on phase change heat transfer but most of these studies have been conducted in pool boiling. The results of these studies revealed contradictions, as nanoparticles both increased and decreased the efficiency of heat transfer in the base fluid. It might be owing to the varying proportions of nanoparticles. Despite the fact that spray cooling has numerous benefits over traditional cooling methods, it appears that the experimental data is scarce for spray cooling methods using a low volume concentration of (0.1%) for MWCNT and TiO₂ nanoparticles. Further, the use of hybrid nanofluid comprising of MWCNT and TiO₂ has not been reported with the spray cooling method. In addition, due to very high thermal conductivity of MWCNTs, researchers have shifted their focus to explore its benefits in various heat transfer industrial applications. Also, Titanium dioxide based nanofluid have strong heat transfer capability with low viscosity and pressure drop while retaining very good dispersion stability for long time. Moreover, it is inexpensive and non-toxic, and employing it as a nanofluid will be cost effective and safer for many industries. Furthermore, much of the research study described above focused on the cooling phase of steel fabrication at high temperatures. Experimental studies need to be carried out with an emphasis on cooling of electronic devices. With this motivation, an attempt has been made to find optimal concentration of MWCNT and TiO₂ nanofluid in addition to hybrid nanofluid of MWCNT and TiO₂ using spray cooling method. Experiments have been conducted in the present study at various concentrations (0.1%–0.3%) of single nanoparticle-based nanofluid and hybrid nanofluid to understand the mechanism of spray cooling enhancement.

2. Experimental setup

2.1. Test facility

The details of the experimental test rig are given in the author's earlier work. ²⁰ Figure 1(a) and (b) displays the schematic view and pictorial view of the experimental test rig built to conduct the experiments at various concentrations of nanofluid for a wide range of volumetric flux.

Figure 1.

(a) and (b) Schematic and pictorial representation of the spray cooling test facility.

Figure 2 shows the pictorial view of the copper specimen, cartridge heaters and test heater assembly. Four holes were drilled at the bottom of the copper specimen for fixing cartridge heaters as heat source. A single cartridge heater capable of providing heat input of q_in = 125 W can be controlled using a voltage regulator by adjusting the AC voltage. A square of 10 mm² in size was prepared over the copper specimen as the impact surface. A uniformly distributed heat flux is assumed across the copper block. The copper block is housed inside a stainless-steel cover and the gap between the cover and copper block is filled with ceramic wool to minimize the heat loss. Teflon insulation has been provided at the top of the test specimen, and sealants to check leakage into the gap between the heater and teflon. Sizing and dimensions of the copper heater block and positioning of various thermocouples in the heater block have been provided in the author's earlier work. ²⁰

Figure 2.

Pictorial view of the (a) copper specimen (b) cartridge heaters (c) test heater assembly.

2.2. Test fluid

In the present study, two distinct nanofluid systems consisting of single nanoparticles of Multi-walled carbon nanotubes (MWCNT) and Titanium dioxide (TiO₂), as well as one hybrid nanofluid (MWCNT and TiO₂), have been prepared for experimental investigations. Table 1 shows the nanoparticle characteristics used in this study. Table 2 shows the thermo-physical characteristics of the distilled water (DI water) at 25°C and 1 atm.

Table 1.

Thermo-physical properties of nanoparticles parameters used at 25 °C and 1 atm.

	TiO2 nanoparticles	MWCNT nanoparticles
Characteristics length	10∼30 (nm)	10∼15 (nm) in diameter
		10∼15 (nm) in length
Surface ratio(m2/g)	≥150	≥200
Density (g/cm3)	10.51	2.7
Melting point (°C)	1843	3550
Thermal conductivity(W/m K)	13	2250
Specific heat capacity (Kj/kg K)	0.7	0.45

Table 2.

Thermo-physical properties of DI water at 25 °C and 1 atm.

Average molecular weight (KJ/kg mole)	18.16
Critical temperature (°C)	374.2
Saturation temperature (°C)	99.9
Density (Kg/m3)	997
Heat of vaporization (Kj/kg)	2256.7
Thermal conductivity (W/m K)	0.606
Specific heat (Kj/kg K)	4.22
Surface tension (N/ m)	0.07275
Thermal diffusivity (m2/s)	1.440 × 10-7

Table 3 enlists the experimental conditions for pressure nozzle having a 0.4 mm orifice diameter. The pressure nozzle is henceforth denoted as N1. The optimum concentration for single nanoparticle of MWCNT and TiO₂ nanofluid have been obtained by varying the nanoparticle concentration from 0.1–0.3% by volume in DI water. For each of the hybrid nanofluid systems, three distinct mixtures have been prepared by varying concentration from 0.1% to 0.3% by volume in proportions of 25%, 50% and 75% for each nanofluid as shown in Table 4.

Table 3.

Operational experimental conditions.

Coolant	Spray height (H) mm	Inlet temperature (Tin) °C	Inlet pressure (P) MPa	Chamber Pressure (P) kPa	Volumetric flux (Q″) cm3/cm2s
DI water	10.25	25	0.39	101	0.83, 1.637, 2.50
Nanofluids	10.25	25	0.39	101	2.50

Table 4.

Details of concentration of nanoparticles employed.

Sr. No.	Nanofluids	Vol%	Proportion %
1	MWCNT	0.1,0.2,0.3	100
2	Titanium dioxide (TiO2)	0.1,0.2,0.3	100
3	MWCNT:Titanium dioxide (TiO2)	0.1,0.2,0.3	25:75, 50:50,75:25

The hybrid nanofluid is assumed to give high heat transfer performance in contrast to nanofluid consisting of single nanoparticle. ¹⁶ For estimating the average size of the nanoparticles, a scanning electron microscopic (SEM) image of the MWCNT and TiO₂ nanoparticles are shown in Figure 3. The measured length and outer diameter of MWCNT are between 10–30 µm and 10–15 nm, respectively. The average size of TiO₂ nanoparticles measured is in the range 15–30 nm. The hybrid nanofluid has been prepared by adding commercial grade of nano-particles of MWCNT and TiO₂ in three different volume percentage i.e. 0.1%, 0.2%, and 0.3% in DI water. The aqueous solution is then subjected to high shear stirring to produce a stable nanofluid. The nanofluid is then ultra-sonicated for nearly 1 h at a frequency of 40 Hz. After the sonication process is completed, the solution is used for spray cooling within 1 h.

Figure 3.

SEM pictures of (a) TiO₂ (b) MWCNT.

2.3. Data acquisition system

The parameters measured in this study are the liquid flow rate (Q″), fluid inlet pressure (P), the inlet and outlet temperatures of the liquid (T_l), and heater surface temperature (T_w). To avoid interfering with the formation of a thin fluid layer on the heated surface, the desired surface temperature has been obtained indirectly using Fourier's 1D heat transfer equation. The indirect method ensures no interference with the film formation over the heater surface and also ensures continuous spray droplet impact without any blockage. Four thermocouples (type-K) have been embedded in the holes drilled in the protruded cube at the top of copper heater block as shown in Figure 2(a). This form of indirect measurement, as previously indicated, minimizes spray interruption and preserves ideal convection boundary conditions near the heater surface. The thermocouple signals are transferred to the data acquisition system for display and recording of temperature data. All thermocouples have been pre-calibrated before using in the present study.

The heat flow on the heater surface can be estimated using Fourier's equation of heat conduction, which is as follows:

$${\mathrm{q}}^{\mathrm{″}}=\mathrm{k}{\displaystyle \frac{({\mathrm{T}}_{2,\mathrm{a}}-{\mathrm{T}}_{1,\mathrm{a}})}{{\mathrm{x}}_2}}$$

(1)

The equation below can be used to calculate the average temperature of the heater surface: ²⁰

$${\mathrm{T}}_{\mathrm{w}}={\displaystyle \frac{{\mathrm{T}}_{1,\mathrm{a}}-{\mathrm{q}}^{\mathrm{″}}{\mathrm{x}}_1}{\mathrm{k}}}$$

(2)

HTC has been estimated using the equation given below:

$$\mathrm{h}={\displaystyle \frac{{\mathrm{q}}^{\mathrm{″}}}{({\mathrm{T}}_{\mathrm{w}}-{\mathrm{T}}_{\mathrm{sat}})}}$$

(3)

2.4. Experimental procedure

Before carrying out experiments, it was ensured that there was no leakage in the pipeline. The working fluid was first heated to its saturation temperature T_sat using band heaters for 15–20 min to remove any non-condensable gas (NCG) present in the fluid. It is done to avoid the interference of NCG with the heat transfer mechanism during the experiments. It was ensured that copper specimen was wiped off using acetone before each experiment to remove any impurities and layer of oxidation over the surface. After that, the nozzle-to-surface distance (H), volumetric flux (Q″) and the inlet temperature (T_in) of the working fluid was adjusted to the desired value. The working fluid is allowed to flow through the pressure nozzle which breaks it into droplets that sprays on the surface. Once the spray is stabilized, AC voltage is turned ON to avoid overheating of the specimen. The secondary fluid is allowed to flow through the secondary loop. Measurements of various parameters of interest are taken such as Q″, P, T_in and temperature from every thermocouple after the steady-state was achieved. The entire system is a closed loop to avoid the wastage of working fluid.

3. Uncertainty analysis

All measurements taken during the experiments have error tolerance. The data acquisition system (make: Agilent) with multiplexer have been used to measure temperature and to calculate the uncertainties in Q″. Precision mercury thermometer with ± 0.05 °C rated accuracy is used to calibrate all the thermocouples. Throughout the experiments, the accuracy of this system configuration is found to be within 0.25 °C. As a result, when the temperature fluctuation is less than 0.25 °C, it is considered that a steady state has been established. The maximum uncertainty of fixing and measuring type-K (Chromel-Alumel) sheathed thermocouple is about ± 0.01 mm and ± 0.25 °C. The volumetric flux and the fluid inlet temperature of the DI water are obtained by coriolis mass flow meter. Uncertainties associated to direct measure quantities are listed in Table 5.

Table 5.

Uncertainties of the direct measured quantities.

Measured parameter	Uncertainty
Pressure (P)	± 1.2%
Flow rate (Q″)	± 1.1%
Temperature (°C)	± 0.25 °C
Distance between the heater surface and plane of thermocouple location (mm)	± 0.6%
Distance between planes of thermocouple locations (mm)	± 0.2%

Uncertainty can be estimated using the equation below:

$${\sigma }_{\mathrm{y}}=\sqrt{{{\sum }_{\mathrm{i}=1}^{\mathrm{n}}\left({\displaystyle \frac{\mathrm{\partial }\mathrm{f}}{\mathrm{\partial }{\mathrm{x}}_{\mathrm{i}}}}\right)}^2{\sigma }_{{\mathrm{x}}_{\mathrm{i}}}^2}$$

(4)

As a result, the uncertainties in surface temperature, heat flow, and HTC are 1.64%, 5.35%, and 6.29%, respectively.

Measurements were taken four times to ensure the reproducibility of results before experimentations. It is estimated that the deviation in the heat flux curve is due to the fluctuation of spray characteristics, such as volumetric flux and spray droplet size, and not by the input power.

4. Results and discussion

4.1. Physical properties measurement

Each coolant has been characterized before experiments by measuring its surface tension, contact angle (CA), viscosity, and thermal conductivity. To achieve accuracy in the results, all measurements have been performed four times and then averaged. The resultant thermo-physical properties of MWCNT and TiO₂ nanofluid have been listed in Table 6. The surface tension which determines capacity of a liquid to wet a solid surface has been measured using a surface tensiometer. The test samples’ contact angles have been determined with apex contact angle measuring equipment. Every sample surface has been washed with acetone and dried before measuring the CA to reduce the error. A viscometer has been used to measure the viscosities of all of the working fluids. The thermal conductivity of the hybrid nanofluid has been measured using the KD2 pro thermal property analyzer.

Table 6.

Thermo-physical properties of the nanofluid at 25 °C and 1 atm.

Coolant	Vol%	Density (kg/m3)	Surface tension (N/m)	Viscosity (N-s/m2)	Thermal conductivity (W/m K)
MWCNT	0.1	1000.5	.0655	8.883 × 10−4	0.63542
	0.2	1000.8	.0645	8.903 × 10−4	0.64496
	0.3	1001.5	.0637	8.943 × 10−4	0.65348
TiO2	0.1	1000.5	.0832	8.885 × 10−4	0.62542
	0.2	1000.8	.0825	8.913 × 10−4	0.62696
	0.3	1001.4	.0817	8.957 × 10−4	0.62748

Figure 4(a)–(d) shows the variation of surface tension, CA and viscosity for hybrid nanofluid at various proportions of MWCNT and TiO₂ at three different concentrations by volume i.e. 0.1%, 0.2% and 0.3%. Concentration of primary nanofluid is shown on primary x-axis, and concentration of secondary nanofluid is shown on secondary x-axis.

Figure 4.

Effect of concentration of hybrid nanofluid on surface tension, contact angle (CA), and viscosity.

Effect of concentration of 0.1–0.3 by vol% of hybrid nanofluid on surface tension and CA is shown in Figure 4(a)–(c). Higher percentage of TiO₂ nanofluid concentration reduced the surface tension and CA for hybrid nanofluid monotonically when compared to single nanoparticle nanofluid. Further, at 25:75 proportion of MWCNT:TiO₂ at 0.1 vol%, the drop in surface tension and CA is more appreciable. The reason for this change can be attributed to the synergism, phenomena that results in properties of hybrid nanofluid, which is better than that of single nanoparticle nanofluid.

Density and heat capacity of the single nanoparticle based nanofluid can be estimated using the equation below. ²¹

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

(5)

$${\mathrm{C}}_{\mathrm{p},\mathrm{nf}}=(1-\phi ){\mathrm{C}}_{\mathrm{p},\mathrm{bf}}+\phi {\mathrm{C}}_{\mathrm{p},\mathrm{np}}$$

(6)

The viscosity for the nanofluid (single nanoparticle based) can be estimated using the equation given by Bachelor ²² as:

$${\mu }_{\mathrm{nf}}={\mu }_{\mathrm{bf}}(1+1.25\phi )\mathrm{for}\mathrm{MCNT}$$

(7)

Thermal conductivity of single nanoparticle based nanofluid can be estimated using equation given by Maxwell ²³ as:

$${\mathrm{k}}_{\mathrm{nf}}={\mathrm{k}}_{\mathrm{nf}}{\displaystyle \frac{{\mathrm{k}}_{\mathrm{np}}/{\mathrm{k}}_{\mathrm{bf}}+\mathrm{k}-\mathrm{k}\phi (1-{\mathrm{k}}_{\mathrm{np}}/{\mathrm{k}}_{\mathrm{bf}})}{{\mathrm{k}}_{\mathrm{np}}/{\mathrm{k}}_{\mathrm{bf}}+\mathrm{k}+\mathrm{k}\phi (1-{\mathrm{k}}_{\mathrm{np}}/{\mathrm{k}}_{\mathrm{bf}})}}$$

(8)

Effect of three different concentrations of hybrid nanofluid on viscosity is shown in Figure 4(d). It is clearly observed that increase in the concentration of hybrid nanofluid from 0.1% to 0.3% by volume in the base fluid increased the viscosity. The increase in viscosity can be attributed to the agglomeration and clustering of the nanoparticles in the base fluid. Clustering has been shown to increase heat transfer under certain conditions, but it must be used with caution since it causes nanoparticle agglomeration. ²⁴ The hybrid nanofluid with low viscosity have better spreadability of the liquid film on the heated surface, which is crucial for achieving high extraction of heat. Furthermore, the improved droplet spreadability increased surface wetting, resulting in improved liquid film boiling over the heater surface. As a result, high heat transfer rates can be achieved. Although there has been an increase in the viscosity of hybrid nanofluid, but the increase is negligible in comparison to the single nanoparticle type nanofluid.

4.2. Effect of surface superheat on heat flux and heat transfer coefficient

In the following subsections, the effect of distilled water (DI water), single nanoparticle based nanofluids and hybrid nanofluid have been investigated at various concentrations using spray cooling method.

4.2.1. Distilled water

Figure 5 shows heat flux and HTC curves drawn for DI water at three different volumetric flux levels (Q″) of 0.83, 1.67, and 2.5 cm³/cm²s, respectively at 25°C. It can be observed that the heat flux increased with the increasing trend of Q″. Two main regions of heat transfer have been observed from the heat transfer curves as reported by Hsieh et al. ²⁵ In the first region, the main modes of heat transfer were single phase forced convection and evaporation of thin film. With the further increase in heat flux, the slope of the curve becomes steep with the onset of nucleate boiling. As the heat flux increased further, CHF was observed with the appearance of dry outs on the heater surface. It can observed from Figure 5 that with the increase in volumetric flux, boiling curves shifts towards left indicating the decrease in surface superheat. The CHF achieved at Q″ = 0.83 is 231 W/cm², Q″ = 1.67 is 255 W/cm², and Q″ = 2.5 is 271 W/cm², respectively. A maximum HTC of 3.15 w/cm²K have been obtained at Q″ = 2.50 cm³/cm²s.

Figure 5.

(a) Boiling curve and (b) heat transfer coefficient vs. surface superheat for DI water.

Figure 6 depicts the comparison of HTC and the pressure drop curve plotted with the change in volumetric flux (Q″). The increase in volumetric flux results in steep increase in the pressure drop curve. The pressure drop increased from 35–299 kPa as the volumetric flux has increased from 0.83–2.5 cm³/cm²s. Because of the increase in volumetric flux, the heat trafer coefficient has also increased from 2.64–3.15 W/cm²K. It can be noticed from Figure 6 that the increase in HTC is comparitively less than the increase in pressure drop for the same range of volumetric flux. This can be attributed to the fact that more spray droplets goes without actually participating in the heat transfer process at high volumetric flux levels, thereby reducing the HTC values with increased pressure drop. Hence, reducing the spray cooling efficiency.

Figure 6.

Comparison of pressure drop and heat transfer coefficient (HTC) with the change of volumetric flux (Q″).

4.2.2. MWCNTs/Di water nanofluid

Figure 7 illustrates the influence of three different concentrations of MWCNT in the base fluid (distilled water) on the boiling curve at a volumetric flux of 2.50 cm³/cm²s. The addition of varying amounts of nanoparticles in DI water caused all of the curves to change quantitatively. The maximum heat flux and HTC achieved using N1 pressure nozzle is 410 W/cm² and 3.84 W/cm²K at 0.2 vol%. In comparison to the results obtained with DI water, MWCNT nanofluid showed an enhancement of about 22% and 51.30% in HTC and heat flux, respectively. It is interesting to see that the heat transfer performance increased as the concentration of the MWCNT is increased from 0.1% to 0.2%. However, with the further increase in the concentration from 0.2% to 0.3%, heat transfer performance deteriorated. Although, the thermal conductivity of the MWCNT nanofluid improved significantly as the concentration of the nanoparticles increased, heat transfer diminished at high concentrations. Therefore, it seems that heat transfer is not dependent on thermal conductivity alone. Enhancement in HTC from 0.1% to 0.2% concentration can be attributed to the thermal boundary layer over the heated surface that developed highly favourable conditions for heat transfer. As a result, heat transfer with nanofluids appears to be highly dependent on film thickness rather than only on thermal conductivity.

Figure 7.

(a) Boiling curve and (b) heat transfer coefficient vs. surface superheat for MWCNT nanofluid.

4.2.1.1. Titanium dioxide (TiO₂/Di water) nanofluid

Figure 8 shows the effect of different concentrations of TiO₂ nanoparticles on the boiling curve at a volumetric flux of 2.50 cm³/cm²s. All boiling curves quantitatively changed with the addition of different amounts of nanoparticle concentrations in the base fluid. The maximum surface heat flux of 393 W/cm² have been achieved at 0.2 vol% concentration. In contrast to DI water, there has been an increase of about 20.9% in HTC and 45% in heat flux for TiO₂ nanofluid. The boiling curves in the present study also followed traditional boiling regimes i.e. single phase convection, and CHF. There has been no shift to the left in the boiling curves as the volume concentration increased from 0.1 vol% to 0.3 vol% as previously seen with DI water. This can be attributed to the fact that the presence of nanoparticles in the fluid perturbs and deteriorates the vapour film on the heated surface and alters the wettability of the surface. Figures 7 and  8 shows that the addition of nanoparticles to the fluid induced a shift in the CHF towards higher superheats. This might be due to the nanofluid's boundary layer interacting with vapour bubbles leaving the heated surface at low superheats. The results obtained for MWCNT and TiO₂ nanofluid in terms of average heat flux and HTC are listed in Table 7.

Figure 8.

(a) Boiling curve and (b) heat transfer coefficient vs. surface superheat for TiO₂ nanofluid.

Table 7.

Heat flux and heat transfer coefficient (HTC) results for DI water and nanofluid systems at T_in = 25 °C.

Parameter	DI water	MWCNT			TiO2
Vol %	0%	0.1%	0.2%	0.3%	0.1%	0.2%	0.3%
q	271	393	410	384	379	393	373
HTC	3.15	3.78	3.84	3.75	3.83	3.98	3.79
Tw (°C)	133	155	158	160	145	147	143

4.2.3. Hybrid nanofluid (MWCNT and TiO₂)

The final coolant system prepared in this study is a hybrid nanofluid between MWCNT and TiO₂ nanoparticles in DI water at various concentrations. Figure 9 shows that all the curves vary quantitatively at different concentrations of nanoparticle mixtures in the base fluid. The surface heat flux and HTC achieved at a concentration of 0.1 vol% for (25:75 proportion of MWCNT:TiO₂) is highest, and heat flux diminishes as the volume concentration of hybrid nanofluid is increased from 0.1 vol% to 0.3 vol%. The maximum heat flux and HTC achieved at this concentration is 425 W/cm² and 4.42 W/cm²K, respectively. In contrast to DI water, an enhancement of about 40.31% and 56.8% in HTC and heat flux, respectively is achieved. At this proportion, superior heat transfer performance has also been reported in the literature. ²⁶ The results obtained for hybrid nanofluid (MWCNT:TiO₂) in terms of average heat flux and HTC are listed in Table 8. In comparison to the results of single nanoparticle nanofluid, hybrid nanofluid systems has shown considerable improvement in heat transfer performance. For instance, MWCNT:TiO₂ hybrid nanofluid solution at 0.1 vol% provided nearly 15% enhancement in HTC over single MWCNT nanofluid, respectively. The enhancement in HTC in contrast to TiO₂ is nearly about 11% in HTC. The reason for this superior performance can be attributed to the synergistic effect between MWCNT and TiO₂ hybrid nanofluid system, which has reduced droplet size upon exit from the pressure nozzle. Besides that, low surface tension produced a small CA, resulting in a large droplet contact area with the heated surface. Droplets with a small CA and a large contact area can evaporate quickly from the surface. Because the thickness of the liquid film is proportional to the diameter of the droplet, a smaller droplet diameter has resulted in the formation of a thin liquid film that can be easily pierced by nanoparticles. As a result, spray droplets make intermittent contact with the heated surface. Furthermore, bubbles near the surface are more likely to nucleate and accumulate, which can eventually increase the rate of evaporation as well as the performance of heat transfer. The effect of nanoparticle deposition on the heater surface has a significant impact on its surface properties. Several research studies (Wang et al.; ¹⁰ Siddiqui et al. ¹⁸ ) have concluded that surface wettability and shape are the most important characteristics in terms of CHF. If the surface is extremely wet, the surface dry out may be delayed, leading in high CHF values. The influence of nanoparticles on CHF is clearly shown in the current study for both MWCNTs and TiO₂ nanofluids. It is likely that the nanoparticle layer that forms over the heater surface during spray impingement acts as a conductivity layer, influencing the CHF for various nanofluids. Despite MWCNTs’ strong thermal conductivity, it is possible that the comparable CHF values are obtained for TiO₂ due to its low contact angle (very wettable) in comparison to MWCNTs. The improvement in HTC is consistent with a recent work (Hseih et al. ¹² ) which found an improvement in HTC of 1.5–1.9 times for nanofluid in nucleate boiling and at CHF, respectively.

Figure 9.

Variation in heat flux and heat trnasfer coefficient (HTC) with surface superheat at various concentrations of MWCNT and Titanium dioxide (TiO₂).

Table 8.

Heat flux and HTC results for hybrid nanofluid systems at T_in = 25 °C.

	Hybrid nanofluid – MWCNT:TiO2
Vol %	0.1:0.1	0.1:0.2	0.1:0.3	0.2:0.1	0.2:0.2	0.2:0.3	0.3:0.1	0.3:0.2	0.3:0.3
q	393	404	425	395	411	383	395	417	387
HTC	4.21	4.29	4.42	4.27	4.29	4.19	4.17	4.18	4.13
Tw (°C)	164	167	171	163	167	162	161	163	160

5. Conclusions

In this study, the heat transfer characteristics of MWCNT and TiO₂ single nanoparticle nanofluid have been analyzed using spray cooling with pressure nozzle of an orifice diameter of 0.4 mm. The concentration of single nanoparticle type nanofluid ranges between 0.1%–0.3% by volume. The spray cooling results revealed that the surface heat flux and HTC improved significantly using single nanoparticle nanofluids in contrast to DI water. In addition, the effect of hybrid nanofluid of MWCNT and TiO₂ in varying proportions have been analyzed using spray cooling for enhanced heat transfer performance. Hybrid nanofluid was prepared in three different concentrations i.e. 0.1%, 0.2%, and 0.3% by volume in DI water. At each concentration, three different proportions of hybrid nanofluid were prepared i.e. 25:75, 50:50, and 75:25 of MWCNT:TiO₂. All of above prepared coolants were analyzed and comparison of results have been made with DI water to analyze the enhancement in heat transfer performance. For spray cooling experiments, a copper heater block with a surface area of 10 mm² was considered. The present work is quite novel in the sense that it has employed MWCNT and TiO₂ nanofluids as well as hybrid nanofluid to investigate the effects of surface tension and CA in heat transfer applications in pressure atomized spray systems. Results showed that the surface cooling rate achieved for MWCNT:TiO₂ hybrid nanofluid is better in comparison to the other single nanoparticle type nanofluid. A 0.1 vol% hybrid nanofluid outperformed all due to the low surface tension of the coolant, which increased surface activity. Higher surface activity aids in generation of fine droplets during atomization. Low surface tension has reduced the CA that improved wettability of the heated surface by the spray droplets. Fine droplets improved evaporation over the heated surface because of higher contact area, resulting in film thickness of small size, which improved overall boiling heat transfer.

The following conclusions have been drawn from the data of the present study:

The surface tension of DI water is reduced by 15.4% when nanoparticles (MWCNT and TiO₂) have been added to the base fluid, aiding in easy wettability by the droplets. Also, the CA improved with the addition of nanoparticle concentration. The HTC tends to increase with the increase in volumetric flux.

In case of single nanoparticle type nanofluid, the maximum heat transfer performance is obtained with MWCNT nanofluid at 0.2% volume concentration. The maximum value of heat flux and HTC obtained is 410 W/cm² and 3.84 W/cm²K at T_in = 25 °C. Therefore, HTC and heat flux increased by about 22% and 51.30%, respectively, compared to DI water. The critical heat flux achieved for TiO₂ nanofluid is 393 W/cm² and HTC obtained is 3.98 W/cm²K at the same fluid inlet temperature.

The effect of hybrid nanofluid on CHF values is quite significant. Heat transfer enhancement upto 1.6 times that of the base fluid (DI water) have been obtained. Hybrid nanofluid in proportion of 25:75 (MWCNT:TiO₂) at 0.1% by volume has provided the best overall performance and showed an enhancement of 40.31% and 56.8% in HTC and heat flux, respectively in contrast to DI water.

TiO₂ nanofluids performed similarly well in a two-phase nucleate boiling regime when compared to MWCNT nanofluids, despite the fact that the latter has a greater effective thermal conductivity. This has been due to the TiO₂ nanofluid solution, having an equally distributed phase with no agglomeration, as opposed to MWCNT nanofluids. The greater mixing, rather than the nanofluid's usual better thermal conductivity, is the primary heat transfer augmentation mechanism.

It is also observed that heat transfer performance is not only dependent on thermal conductivity but film thickness has also played a dominant role in heat dissipation. Nanoparticle size can also affect the thermo-physical properties of the nanofluid and may contribute in heat transfer enhancement. Therefore, the study on nanoparticle size and film thickness needs to be carried out as a future work in order to substantiate the above statement.

6. Social significance of present study

In recent years, thermal management of engineering devices with high heat flux has grown increasingly difficult. The quantity of heat generated per unit area in the electronics devices has increased significantly as semiconductor device sizes have reduced. As a result, an advanced thermal management system, such as spray cooling, can provide good system cooling performance and reliability from heat generated by electronic devices, particularly when the electronic components are subjected to heat fluxes more than 500 W/cm². Spray cooling is a high heat flux, consistent, and efficient cooling approach, which has proven helpful in a variety of applications such as industrial operations, data centres, and combustion technologies. Further, spray cooling capacity and efficiency must be increased continuously to satisfy the needs of next-generation ultrahigh-power applications. Surfactants and nanofluids can have a considerable impact on liquid–wall interactions in spray cooling, making them the most viable option for increasing spray cooling efficiency. The present study's findings can assist designers of high heat duty applications/devices in designing these devices in more effective way, resulting in a lower rate of failure, higher reliability and efficiency, and lower material costs.

Supplemental Material

sj-docx-1-pie-10.1177_09544089221094991 - Supplemental material for Experimental study on spray cooling heat transfer enhancement using MWCNT and TiO₂ hybrid nanofluid

Supplemental material, sj-docx-1-pie-10.1177_09544089221094991 for Experimental study on spray cooling heat transfer enhancement using MWCNT and TiO₂ hybrid nanofluid by Sukhdeep Singh and Rajeev Kukreja in Proceedings of the Institution of Mechanical Engineers, Part E: Journal of Process Mechanical Engineering

Supplemental Material

sj-docx-2-pie-10.1177_09544089221094991 - Supplemental material for Experimental study on spray cooling heat transfer enhancement using MWCNT and TiO₂ hybrid nanofluid

Supplemental material, sj-docx-2-pie-10.1177_09544089221094991 for Experimental study on spray cooling heat transfer enhancement using MWCNT and TiO₂ hybrid nanofluid by Sukhdeep Singh and Rajeev Kukreja in Proceedings of the Institution of Mechanical Engineers, Part E: Journal of Process Mechanical Engineering