Influence of water based binary composite nanofluids on thermal performance of solar thermal technologies: sustainability assessments

Recent technological advances have made it possible to produce particles with nanometer dimensions that are uniformly and steadily suspended in traditional solar liquids and have enhanced the impact of thermo-physical parameters. In this research, a three-dimensional flat plate solar collector was built using a thin flat plate and a single working fluid pipe. The physical model was solved computationally under conditions of conjugated laminar forced convection in the range 500 ≤ Re ≤ 1900 and a heat flux of 1000 W/m2. Distilled water (DW) and different types of hybrid nanofluids (namely, 0.1%-Al2O3@Cu/DW, 0.1%-MWCNTs@Fe3O4/DW, 0.3%-MWCNTs@Fe3O4/DW, 0.5%-Ag@MgO/DW, 1%-Ag@MgO/DW, 1%-S1 and 1%-S2, where MWCNTs are multi-wall carbon nanotubes, S1 means 2CuO–1Cu and S2 means 1CuO–2Cu nanocomposites) were evaluated via a set of parameters. The numerical results revealed that, by increasing the working fluid velocity (the Reynolds number), the average heat transfer coefficient, pressure loss, heat gain and solar collector efficiency were increased. Meanwhile, outlet fluid temperature and flat plate surface temperature were decreased. At Re = 1900, 1%-S2 and 1%-S1 presented higher thermal performance enhancement by 44.28% and 36.72% relative to DW. Moreover, low thermal performance enhancement of 7.59% and 7.44% were reported by 0.1%-Al2O3@Cu/DW and 0.3%-MWCNTs@Fe3O4/DW, respectively.


Adopted literature review on nanofluids-based FPSCs
According to several earlier research works (Shah & Ali, 2019), hybrid nanofluids are more efficient than single nanofluids in terms of thermo-physical properties.Two single nanofluids (0.1 wt.%-Al 2 O 3 -H 2 O with 20 nm and 0.1 wt.%-TiO 2 -H 2 O with 15 nm) and a hybrid nanofluid (Al 2 O 3 @TiO 2 -H 2 O) were evaluated in experiments and numerical investigations as HTFs inside a flat plate solar collector by Farajzadeh et al. (2018).The surfactant cetyltrimethylammonium bromide (CTAB) was added to the mixture to accelerate the chemical reaction, and three volume flow rates, 1.5, 2.0 and 2.5 LPM, were trialled.Their results showed that the overall FPSC thermal efficiency was enhanced by 19%, 21% and 26% when using 0.1 wt.%-Al 2 O 3 /H 2 O, 0.1 wt.%-TiO 2 /H 2 O and Al 2 O 3 @TiO 2 /H 2 O, respectively.In another study (Ranga Babu et al., 2018), the entropy generation of Cu@CuO/H 2 O, Cu/H 2 O and CuO/H 2 O nanofluids was reduced by 3.31%, 1.35% and 2.96%, respectively.Furthermore, using the hybrid nanofluid (Cu@CuO/H 2 O) increased the exergy efficiency by 2.59%.Two hybrid nanofluids, 80%-MgO@20%-MWCNTs and 80%-CuO@20%-MWCNTs, were compared in terms of energy performance by Verma et al. (2018).The nanoparticle concentration was in the range 0.25-2.0%with volume flow rates of 0.5-2.0LPM.The application of MgO@MWCNTs/H 2 O and CuO@MWCNTs/H 2 O inside the FPSC enhanced the thermal performance by 18.05% and 20.52%, respectively, relative to H 2 O. Hybrid nanofluids containing three nanocomposites, such as CF-MWCNTs@CF-GNPs with h-BN, were suspended in H 2 O in the presence of Tween-80 by Hussein et al. (2020).In their study, the solar collector efficiency reached up to a value of 85% at 4 LPM.The application of 0.1 wt.%-MWCNTs@GNPs/h-BN increased the heat gain parameter and heat loss parameter up to 21.9% and 78.3%, respectively.Nano-diamond-cobalt oxide (ND@Co 3 O 4 ) hybrid nanofluids in different concentrations of 0.05-0.15wt.% were studied at various volume flow rates in the range 0.56-1.35LPM by Sundar et al. (2021).The thermal conductivity and dynamic viscosity of 0.15 wt.%-ND@Co 3 O 4 -H 2 O were increased by 15.71% and 45.83% at an inlet temperature of 60°C.Moreover, the employment of 0.15 wt.%-ND@Co 3 O 4 -H 2 O increased the average Nusselt number (Nu avg ) by 21.23% with a maximum friction factor ( f ) penalty of 1.13 times that of the base fluid.
The solar collector efficiency of 0.15 wt.%-ND@Co 3 O 4 -H 2 O was found to be 59%, while it was 48% for H 2 O. Okonkwo et al. (2020) proved that a single nanofluid (0.1%-Al 2 O 3 -H 2 O) was more efficient than a hybrid nanofluid (0.1%-Al 2 O 3 @Fe-H 2 O) with an enhancement in the thermal performance by 2.16% and 1.79%, respectively, relative to the base fluid.A considerable increase in the dynamic viscosity of the hybrid nanofluid compared with the single nanofluid caused this phenomenon.
On the other hand, 0.1%-Al 2 O 3 @Fe-H 2 O enhanced the exergy efficiency by 6.9% as against 5.7% using 0.1%-Al 2 O 3 -H 2 O.The combination of using a porous medium and Ag/Al 2 O 3 -H 2 O hybrid nanofluids was solved computationally in the presence of thermal radiation by Xiong et al. (2021).According to their study, at a given Reynold number the heat transfer was found to be increased substantially by increasing the porosity coefficient and decreasing the thermal radiation parameter, the Darcy number and the thermal conductivity ratio.Different concentrations of hybrid CuO-Cu/water nanofluid were prepared to study its influence on a solar thermal energy storage system by Alrowaili et al. (2022).The collection area was lowered by up to 38% by using a hybrid nanofluid.The calculated heat removal factor was 0.894.Hybrid CuO 2.5 g + Cu 1.5 g nanofluid outperformed water and mono CuO in terms of thermal-optical efficiency by 61.7% and 14.9%, respectively.The results of Khetib et al. (2022) showed that Nu avg was increased by augmenting the Re and concentration of DWCNTs-TiO 2 /water nanofluid.In addition, at Re = 28,000 and ϕ = 3%, the installation of turbulators with innovative geometry (TIG) with PR (Pitch Ratio) = 4 within the solar collector increased the Nu avg by 63.46%.In the case of ϕ = 3% and by augmenting the Re from 7000 to 28,000, the energy and exergy performance was augmented by 22.19% and 23.26% for PR = 4 and PR = 1, respectively.Recently, novel machine-learningbased experimental data was implemented by the use of Bayesian optimisation coupled with boosted regression trees to predict the thermal performance of MWCNT-Fe 3 O 4 /water hybrid nanofluids inside FPSC systems (Said et al., 2022).The boosted regression tree model was optimised by 99.9% prognostic efficiency, and a peak thermal efficiency of 63.84% was attained at a Reynolds number of 1413.

Research objectives
A considerable amount of literature has been published on experimental and numerical studies of hybrid nanofluids inside FPSCs.However, there has been relatively little technological literature published on the effects of different nanocomposite types in terms of different nanoparticle shape, different nanoparticle size and different nanomaterial mixing ratios inside solar collectors.Thus, this study aims to discuss the computational study of the influences of using seven different types of hybrid nanofluid, namely 0.1%-Al 2 O 3 @Cu/DW, 0.1%-MWCNTs@Fe 3 O 4 /DW, 0.3%-MWCNTs@Fe 3 O 4 /DW, 0.5%-Ag@MgO/DW, 1%-Ag@MgO/DW, 1%-S1 and 1%-S2.The FPSC thermal model was solved under the conditions 293K inlet temperature, 1000 W/m 2 heat flux and 500 ≤ Re ≤ 1900.Six different parameters, i.e. surface heat transfer coefficient, pressure drop, outlet temperature, surface temperature, heat gain and thermal efficiency, were analysed to examine the best alternatives for base fluids.Previous experimental and computational investigations of the use of nanofluids as working fluids in FPSCs are summarised in Table 1.

Problem description
A conjugated forced convection model is solved numerically via a three-dimensional physical problem using a thin flat plate made of aluminium attached to a working fluid pipe made of copper (see Figure 1).In this study, the radiation model is not activated and is replaced by constant wall heat flux for heating the solar collector surface (Liu et al., 2020).The physical dimensions of the solar collector model are as follows: length of thin flat plate, L = 914.4mm; width of thin flat plate, w = 128 mm; thickness of thin flat plate, t = 2 mm; inner hydraulic diameter (ID h ), 12.5 mm; outside hydraulic diameter (OD h ), 14.8 mm; and wall thickness, 1 mm.The computational domain is divided into a large number of elements (cells) in order to improve grid size control and meshing efficiency.The feature of inflation is selected near the working fluid pipe wall, as shown in Figure 1.

Mathematical model
A physical model of a 3D FPSC with tilt angle 30°w as tested numerically using seven different types of hybrid nanofluid with different nanomaterials and different nanoparticle concentrations.As mentioned above, the thermo-physical properties of the hybrid nanofluids and H 2 O were collected from the literature review at an inlet temperature of 293 K (Sarsam et al., 2022;Vatani & Mohammed, 2013).The following assumptions are made in the current model: the flow is a steady state, the flow is Newtonian, the nanofluid is undergoing single-phase flow and the flow is laminar.The wall heat flux was constant at 1000 W/m 2 , being the intensity of solar radiation.Meanwhile, non-slip and adiabatic boundary conditions were applied in the current simulation cases.Moreover, gravity was activated with a magnitude of −9.81 m/s 2 in the normal direction of the y-axis.The conservation equations of continuity, momentum and energy are written in the mathematical expressions as follows (Akbarinia & Behzadmehr, 2007;Cerón et al., 2015;Edalatpour & Solano, 2017): Here, the symbol "i" equals 1,2,3, and the velocity vector "u i " equals (u, v, w).

Thermo-Physical properties
A wide range of hybrid nanofluids are compared in this numerical evaluation in terms different nanomaterials, different mixing ratios and different concentrations.Furthermore, the thermo-physical properties and essential data were collected from previously published articles in order to test for the most efficient hybrid nanofluids as alternative working fluids for solar collector applications.Table 2 shows the thermo-physical properties of the base fluid (H 2 O) and different types of hybrid nanofluid that were involved in the current assessments.Firstly, the nanocomposites used in the experiments of Suresh et al. (2011) were chosen.Also, the MWCNT@Fe 3 O 4 hybrid nanofluid in the mixing ratio 74%@Fe 3 O 4 and 26%@MWCNTs (Lomascolo et al., 2015) were used.

Validation and verification approaches
In order to conduct the grid independence test, five different grids were tested in the current numerical simulations.The grid independence test is carried out firstly using H 2 O as a working fluid to ensure the validity and accuracy of the tested computational grids.The working fluid was under the conditions Re = 500 and inlet temperature = 293 K. Six different output parameters were used in the evaluations, i.e. surface heat transfer coefficient, pressure drop, outlet temperature, heat gain, solar collector efficiency and surface temperature.Table 3 indicates that, as the number of elements rises, so does the accuracy of the data obtained.Owing to its correctness and accuracy, grid number 5 with 872,000 elements was acceptable for processing the remaining tests in the current research.Moreover, the current numerical results were compared with the experimental data reported by Verma et al. (2018) as shown in Figure 2 in order to validate the current physical problem.In Figure 2   and 0.025 kg/s with different intensities.The current data showed an average error of 6.33%, 6.66% and 6.54%, respectively, when compared with the experimental results.Another validation was carried out between the current model and the previous (experimental and numerical) report by Farajzadeh et al. (2018).Figure 3(a) shows a comparison with DW and 0.1 wt.%-Al 2 O 3 @TiO 2 /DW at 1.5 LPM.Meanwhile, Figure 3(b) illustrates the comparison for 0.1 wt.%-Al 2 O 3 @TiO 2 /DW at 1.5, 2 and 2.5 LPM.The present results show an average error of 4.7%, 5.2%, 4.8% and 5.6% for DW, and 0.1 wt.%-Al 2 O 3 @TiO 2 /DW at different volume flow rates, respectively.

Outlet and flat plate surface temperatures
The outlet fluid temperature and flat plate surface temperature are discussed in this section as a function of 500 ≤ Re ≤ 1900 for the base fluid (DW) and seven water based hybrid nanofluids.As shown in Figures 5  and 6, outlet and flat plate surface temperatures decrease as the Reynolds number increases.When the Reynolds number is increased from 500 (lower value) to 1900 (higher value), the outlet and flat plate surface temperatures are decreased by 4.37% and 3.47% for DW, 3.87% and 2.93% for 0.1%-Al 2 O 3 -Cu/DW, 3.86% and 3.01% for 0.1%-MWCNTs-Fe 3 O 4 /DW, 3.55% and 2.69% for 0.3%-MWCNTs-Fe 3 O 4 /DW, 4.18% and 3.27% for 0.5% Ag-MgO/DW, 3.96% and 3.08% for 1% Ag-MgO/DW, 2.66% and 1.86% for 1%-S1, and 2.49% and 1.71% for 1%-S2.The base fluid shows the higher values for outlet fluid temperature and flat plate surface temperature followed by 0.5%-Ag-MgO/DW, 1%-Ag-MgO/DW, 0.1%-Al 2 O 3 -Cu/DW, 0.1%-MWCNTs-Fe 3 O 4 /DW and 0.3%-MWCNTs-Fe 3 O 4 /DW.Meanwhile, 1%-S1 and 1%-S2 present the lower values.This decline in outlet/surface temperatures can be credited to the improved thermal conductivity in the presence of water-based hybrid nanofluids as a result of the increased convective heat transfer coefficient (Sarsam et al., 2015(Sarsam et al., , 2020)).As per Table 2, 1%-S1 and 1%-S2 nanocomposites showed better improvements in thermal conductivity over the base fluid with 32.06% and 40.03%, respectively.Previous studies on laminar and turbulent nanofluids flows proved that the heat transfer fluids (HTFs) with better thermal conductivities than their base fluids (DW) will also have greater convective heat transfer coefficients (Akram et al., 2021;Kumar et al., 2021).Moreover, Appendix compares the temperature profiles of different hybrid nanofluids at Re = 700.As can be seen from the temperature labels, in all cases the temperature is increased from the centre of the tube toward the radial direction.The reason for this is justified by the formation of boundary layers and their growth.

Heat transfer coefficient and pressure drop
The heat transfer coefficient and pressure drop are discussed in this section as a function of 500 ≤ Re ≤ 1900 for the base fluid (DW) and seven water based hybrid nanofluids.As shown in Figures 7 and 8, the heat transfer coefficient and pressure drop are increased as Reynolds number increases.When the Reynolds number is increased from 500 (lower value) to 1900 (higher value), the heat transfer coefficient and pressure drop are increased by 34.59% and 81.57% for DW,    heat transfer coefficients than base fluids.The thermal conductivity of nanocomposite fluids is important for increasing the heat transfer parameters.In previous work, as well as in laminar and turbulent regimes, nanofluids with a higher thermal conductivity than their base fluids frequently have higher convective heat transfer coefficients (Sadri et al., 2018).

Heat gain and collector efficiency
The   5) resulting in lower heat losses from the collector, i.e. increased collector efficiency/heat gain (Elcioglu et al., 2020;Tong et al., 2019).This behaviour is confirmed through Figures 9 and 10, which display the variation of energy efficiency/heat gain versus Reynolds number (fluid velocity)for different types of hybrid nanofluid (Moravej et al., 2020).Furthermore, according to Equation ( 14), the values of energy gain and efficiency are related to mass flow rate, specific heat capacity, and the temperature differential between the outlet and input.

Conclusions
This paper discusses recent breakthroughs in computational analyses of hybrid nanofluid applications in FPSCs.The use of nanocomposites as new solar fluids can be considered a powerful approach to improving FPSCs' performance.Base fluids and water-based hybrid nanofluids were tested and evaluated inside solar collectors in terms of heat transfer, fluid flow and energy  effectiveness.The thermo-physical parameters of DW and seven hybrid nanofluids at 293 K were collected from published works and used as heat transfer fluids in numerical simulations.The working fluids were simulated under the condition of laminar flow in the range 500 ≤ Re ≤ 1900 and the heating rate was 1000 W/m 2 .The followings conclusions can be drawn.

Figure 1 .
Figure 1.A drawing of the FPSC numerical model and the grid domain.
(a), the base fluid and CuO@MWCNTs and MgO@MWCNTs hybrid nanofluids under conditions T a = 300 K, I = 800 W/m 2 and 0.025 kg/s were validated with different volume fractions.Meanwhile, as shown in Figure 2(b), the base fluid and CuO@MWCNTs and MgO@MWCNTs nanocomposites were compared under the settings T a = 298

Figure 2 .
Figure 2. Comparison between the current data and the previous results of Verma et al. (2018).

Figure 3 .
Figure 3.Comparison between the present work and the experimental results of Farajzadeh et al. (2018).

Figure 4 .
Figure 4. Thermo-physical properties of different nanocomposites and DW at 293 K.

Figure 5 .
Figure 5. Outlet temperature versus Reynolds number for DW and different hybrid nanofluids.

Figure 6 .
Figure 6.Surface temperature versus Reynolds number for DW and different hybrid nanofluids.

Figure 7 .
Figure 7. Surface heat transfer coefficient versus Reynolds number for DW and different hybrid nanofluids.

Figure 8 .
Figure 8. Pressure drop versus Reynolds number for DW and different hybrid nanofluids.

Figure 9 .
Figure 9. Heat gain versus Reynolds number for DW and different hybrid nanofluids.

Figure 10 .
Figure 10.Thermal efficiency versus Reynolds number for DW and different hybrid nanofluids.

Table 1 .
Previous experimental and computational studies on the use of hybrid nanofluids inside FPSCs.

Table 3 .
Grid independence test using different grid domains under the conditions 293 K and Reynolds number 500.