HEAT TRANSFER PERFORMANCE OF SILVER / WATER NANOFLUID IN A SOLAR FLAT-PLATE COLLECTOR

An experimental study is carried out to investigate the heat transfer characteristics of silver/water nanofluid in a solar flatplate collector. The solar radiation heat flux varies between 800 W/mand 1000W/m, and the particle concentration varies between 0.01%, 0.03%, and 0.04%. The fluid Reynolds number varies from 5000 to 25000. The influence of radiation heat flux, mass flow rate of nanofluid, inlet temperature into the solar collector, and volume concentration of the particle on the convective heat transfer coefficient and the collector efficiency are studied. Both parameters increase with increase in the particle volume concentration and flow rate. The maximum percentage increase obtained in the convective heat transfer coefficient is 18.4% for the 0.04% volume concentration at a Reynolds number of 25000. An increase in the performance of nanofluid is also witnessed when compared to the base fluid, which has a strong dependency on volume concentration and mass flow rate.


INTRODUCTION
Suspension of nano-particles in a conventional heat transfer fluids [1][2][3][4] results in notable enhanced thermal properties.These properties of thermal conductivity, thermal diffusivity, viscosity, and design parameter for convective heat transfer are enhanced in comparison to base fluid properties [5][6][7][8], and these results would be beneficial in saving equipment costs and increasing performance.Wong and De Leon [9] carried out a review paper detailing the current and future applications of nanofluids.The need for economical, energy-efficient, and technologically sustainable green technologies is being met by nanofluids in many key applications to control the flow of heat [10], cool nuclear reactors [11], and cool microchips [12], etc. Experimental investigations on the application of nanofluids in solar energy has been done as well [13].
Li et al. [14] investigated the forced convective heat transfer of nanofluids in solar collectors during the day and night, with distilled water and nanoparticles of Al2O3, ZnO, and MgO.The nanofluid achieved a 3°C temperature difference during the daytime peak solar radiation compared with the base fluids.With a concentration of 0.2% ZnO, a temperature difference of 2.55°C for daytime and 1°C for nighttime was reached, and this was determined to be the most attractive option for solar energy utilization.Yousefi et al. [15] witnessed a 28% performance improvement in a flat-plate collector when it was operated with water-Al2O3 nanofluids.Tyagi [16] theoretically compared the conventional flat-plate collector with a direct absorption solar collector (DAC) and observed the former to be 10% more efficient.Otanicar [17] studied the conomic and environmental influences of using nanofluids to enhance solar collector efficiency with conventional solar collectors.Dongxiao et al. [18] presented excellent photothermal properties of carbon-black aqueous nanofluids at highvolume fractions.Further work on nanofluids' application to direct solar absorption has been carried out by Lijuan Mu [19] using a custom-made direct solar absorber.The radiative properties of several nanofluids are tested for the highest temperature difference across the heat exchangers.
Based on the above-mentioned review of the literature, it has been clearly observed that most of the previous studies on solar flat-plate collectors were conducted using metal oxide nanoparticles in relatively high concentrations.These high concentrations of metal oxide nanoparticles cause a higher pressure drop that then requires a higher pumping power.Since a limited number of studies exists in the literature with respect to pure metal nanoparticles, it is recommended to study the heat transfer characteristics of pure metal nanoparticles with relatively low concentrations (<1%) by volume and high thermal conductivity compared with metal oxides.Therefore, in the present study, the efficiency of a solar flat-plate collector is studied with a low particle volume concentration of less than 0.04% silver-water nanofluid.These experiments are conducted for a solar radiation flux ranging from 800 W/m 2 to 1000 W/m 2 , and the Reynolds number varying from 5000 to 25000.The effect of radiative heat flux, mass flow rate, inlet temperature, and volume concentration on the convective heat transfer coefficient and the collector efficiency are studied.The tailormade setup for a collector area is 2.4m 2 and the collector plate is made of nine parallel copper strips.

EXPERIMENTS
2.1 Nanofluid preparation Silver nanofluids with particle volume concentrations of 0.01%, 0.03%, and 0.04% are prepared by suspending the required amount of silver nanoparticles in water with polyvinyl pyrolidine (PVP) used as the surfactant.The mixture is composed of Ag (silver) nanoparticles with an average diameter of less than 100 nm and PVP dispersed in water.In order to produce the required particle volume fractions, a dilution with water followed by a stirring action was used.Moreover, an ultrasonic vibrator with a frequency range from 0 Hz to 100 Hz is used to sonicate the solution continuously for approximately 30 minutes in order to break down the agglomeration of nanoparticles.The Scanning Electron Microscope (SEM) analysis has been done to observe the size and morphology of the suspended silver nanoparticles.[21], thermal conductivity by Yu and Choi [22], viscosity by Einstein's equation [23].The specific heat of the nanofluid is calculated by Xuan and Roetzel [24].
( 1) In these equations, ρnf is the density of the nanofluid, ρp the density of nanoparticle, and ρw the density of water.μ represents viscosity, k thermal conductivity, Cp specific heat, and the subscripts w and p are the properties of water and nanoparticles.Here Ø is the volume concentration of the nanofluid.The uncertainty analysis is done using the method proposed by Robert J. Moffat [25] and the maximum error in the observed readings and calculated parameters are found to be less than 5% ± .

FIGURE 2a . SCHEMATIC DIAGRAM OF THE EXPERIMENTAL LAYOUT FIGURE 2 b. PHOTOGRAPHIC VIEW OF THE TEST SECTION
The schematic diagram of the test facility (figure 2a) is a closed loop consisting of a flat-plate collector, liquid pump, heat exchanger, and storage tank.A bypass valve is provided around the pump so that the mass flow rate can be adjusted to the prescribed value.The heat exchanger is used to eliminate heat from the outlet water.The combination of a heat exchanger and a storage tank enables the user to adjust the fluid inlet temperature.A set of 27 K-type thermocouples (figure 2b) with 0.1% ± accuracy are attached at several pre-determined positions of the riser tubes to measure the wall temperature of the collector.Two K-type thermocouples are immersed in the fluid to measure the bulk fluid temperatures between the inlet and outlet of the test section.A pyranometer with a digital micro-voltmeter is used to determine the incident radiation on the flat-plate collector, and two pressure sensors with 0.1% ± accuracy are immersed in the fluid between the inlet and outlet of the collector to measure the pressure drop.For the detailed specifications of the collector, see table 1.

Occupied area 200mm x 120mm x 15mm
Absorber area 190mm x 115mm Header pipe (Cu) Φ 25.4mm Connector riser pipe Φ 12.5mm Absorber sheet (Cu) ---- The experiments are conducted according to the ASHRAE 93-86 [20] testing of solar flat-plate collectors.The standard specifies that the collector will be tested under clearsky conditions to determine its efficiency characteristics.On any given day, data is recorded under steady-state conditions for fixed values of mass flow rate and fluid inlet temperature.The collector is considered to be operating under steady-state conditions if the deviation of the experimental readings observed for a given mass flow rate is less than the following specified limits over a 15-minute time period: Global radiation incident on collector plane ± 50 W/m 2 Ambient temperature Temperature rise across the collector ± 0.1°C Other specification are that the value of total radiation IT is greater than 600W/m 2 , the wind speed between 3 m/s and 6 m/s, and the fluid flow rate at approximately 0.02 kg/s per square meter of collector gross area.With respect to the present study, the maximum variation in total/global radiation observed is 15 W/m 2 , which is still far less than the value given in the ASHRAE standards.While the maximum variations between the inlet and outlet temperatures are found to be less than 1°C.The wind velocity measured with the anemometer vary between 3m/s and 5 m/s.The solar collector is tested for various mass flow rates varying from 1 L/min to 10 L/min with corresponding Reynolds numbers ranging from 5000 to 25000; however, the efficiency calculations are limited to a flow rate of 6 L/min and each reading is taken at 15-minute time intervals.During the experimental test runs, the mass flow rate, incident radiation, wall temperatures, inlet and outlet fluid temperatures, and wind velocity are observed and stored in the data log for further processing and data reduction.

Data reduction and efficiency calculation
The incident radiation or total radiation, IT, is determined by a pyranometer.The useful qu heat gain is then calculated as ( ) ( ) where Ap is the collector plate area, S is the absorbed radiation, ql is the heat loss by convection and re-radiation, Ul is the overall loss coefficient, Tpm is the mean plate temperature of the absorber plate, Ta is the ambient temperature, and τα stands for the absorbance transmittance product.
The convective heat transfer coefficient of the fluid is determined by the following equations.From the useful heat gain equation, the heat flux on the riser tube is calculated as: where t A is the lateral surface area of the riser tube.
The fluid temperature corresponding to the measured wall temperature is calculated as: The bulk fluid temperature is calculated as: The Reynolds number is calculated by the equation: Re The convective heat transfer coefficient is then calculated by the equation: The Nusselt number for the nanofluid is given as: The efficiency of the flat-plate collector can be defined by: ( ) Equation number 14 can be further modified in terms of the heat removal factor FR: The calculated efficiencies are then plotted with the parameter ((Tfi -Ta) / IT) × 1000.The experimental values of efficiency are plotted against the parameter ((Tfi -Ta) / IT) × 1000 and generally yield straight lines; however, the scatter of the experimental data is found to be less than 10% ± deviation.

RESULTS AND DISCUSSION
The experimental test facility is validated for its accuracy by conducting some preliminary experiments with water.The experimental Nusselt number obtained from the present study are compared with the Dittus-Boelter correlation [26].A deviation of ± 9 % is observed between the experimental results and those obtained by the correlation, thus validating the experimental setup's accuracy.The variations of temperature over time with incident solar radiation of 900W/m 2 are shown in figure 3.

FIGURE 3. EVOLUTION OF FLUID TEMPERATURE PROFILE WITH RESPECT TO TIME IN SOLAR COLLECTOR
The temperature rise of the nanofluid across the collector over time is higher when compared to that of the base fluid.For the particle concentration of 0.04%, the nanofluid has the highest temperature rise of 4°C.Nanofluids are known to have higher heat absorption capabilities compared to that of water as the dispersed nanoparticles tend to absorb more incident energy due to the pure metallic nature and higher thermal conductivity [16].
The wall and bulk fluid temperatures for various particle concentrations are shown in figure 4.
The particle concentration is observed to have an indirect relation with the wall temperature; for instance, a decrease of 3°C is observed for the Ø = 0.04% concentration.This effect on the decreased wall temperature can be attributed to the fact that on suspending silver nanoparticles in the base fluid, the thermo-physical properties of the resulting mixture are improved.The bulk fluid temperature recorded while operating with a nanofluid is also found to be lower compared to that of water.

FIGURE 4. VARIATION OF WALL AND BULK FLUID TEMPERATURES AGAINST DIMENSION LESS LENGTH
Figures 5 and 6 show the variations of the convective heat transfer coefficient and Nusselt number against the Reynolds number.It is observed that the heat transfer coefficient increases with the increases in particle concentration and flow rate.The enhancement in convective heat transfer coefficient of the nanofluid over the base fluid is significant at higher Reynolds numbers.For example at a Reynolds number of 25000 and volume concentration of 0.04%, an enhancement of 18.14% (Nu = 170) has been observed compared to that of 8.41% (Nu = 141) when the Reynolds number is 5000 at the same volume concentration.The percentage enhancements in the heat transfer coefficient are 12.2%, 16.17%, and 18.4% at 0.01%, 0.03%, and 0.04% volume concentrations, respectively.The enhancement in the heat transfer is due to the quick development of the hydraulic and thermal fields compared to that in laminar flow.The intensification of turbulence (eddy formation) also plays a vital role in high heat transfer rates, along with the addition of nanoparticles to the base fluid.The enhancement of the heat transfer coefficient directly depends on the thermal conductivity and indirectly on the thermal boundary layer thickness.Thermal conductivity of the nanofluid increases with increase in volume concentration.Decreasing of the thermal boundary layer thickness may be due to the mobility of particles near the wall, their migration to the center of tube, and reduction of viscosity at the wall region.Figure 7 shows a typical Hottel-Whittlier-Bliss performance characterization of the collector, which expresses the collector's efficiency as the fraction of incident radiation that is collected by the working fluid as per the ASHRAE 93-86 standards [20].In the present study, it is observed that, at a flow rate of 6 L/min, the efficiency operated with nanofluid with a 0.04% volume concentration and the water efficiency is found to be 68.7% and 60.7%, respectively.The main reason for this increase in efficiency can be explained by considering the two parameters, namely, the energy absorbed parameter FR(τα) and the removed energy parameter, FRUL.Table 2 shows the FR(τα) and FRUL values of water and other concentrations of nanofluids.In the case of high circulation rates of the fluid, the smaller the temperature change from inlet to outlet and the closer the inlet fluid temperature to the average collector plate temperature, then the higher the value of the parameter FR, for nanofluids.This value of FR at a flow rate of 6 L/min is found 0.94, 0.91, and 0.89 for Ø =0.04%, 0.03%, and 0.01%, respectively, and compared to water, where the FR value is 0.87.The energy absorbed parameter FR(τα) is dominant in lower temperature rises corresponding to ((Tfi -Ta) / IT) × 1000 < 20 and the removed energy parameter FRUL is dominant in higher temperature differences.The FR(τα) values for lower concentrations (0.01% and 0.03%) nanofluids would be greater than those of a higher concentration nanofluid (0.04%) at lower and higher temperature differences, as for ((Tfi -Ta) / IT) × 1000 < 20, the FRUL value for 0.04% nanofluid would be higher than 0.03% and 0.01% volume concentration nanofluids.Therefore, the efficiency of these ranges would be higher for 0.04% volume concentration nanofluid.

FIGURE 6. VARIATION OF AVERAGE NUSSELT NUMBER WITH RESPECT TO REYNOLDS NUMBER
This similar trend is observed by Yousefi et al. [15].The thermal conductivity of nanofluid increases with increases in temperature and particle concentrations, and for higher temperatures, the thermal conductivity enhancement of the 0.04% nanofluid would be higher compared to two other concentrations, Ø = 0.01% and 0.03%.Moreover, the chaotic movement of the nano-particles suspended in the base fluid promotes thermal conductivity, and such chaotic movement is more dominant in the 0.04% volume concentration silver/water nanofluid.By increasing the particle concentration, the effect in temperature rise becomes important.Mass flow rate = 0.02 kg/s, IT = 800 W/m 2 , water Table 3 shows that the efficiency of the collector decreases sharply with the increasing values of the fluid inlet temperature Tfi, as the value falls from 60% to 29% as Tfi increases from 40°C to 90°C.This decrease is cause by the higher temperature level at which the collector as a whole operates when the fluid inlet temperature increases.Because of this the top-loss coefficient (UT) and the surroundings' temperature increased difference, the heat lost increases and the heat gain decreases.4, it is seen that the useful heat gain is higher for nanofluids and so the efficiency of nanofluids over water is higher.This efficiency increases with increasing incidents of solar flux.The fluid inlet temperature and ambient temperature essentially determine the losses from a collector.Similarly, when the incident radiation flux increases, the useful heat gain and efficiency increase (table 5).Mass flow rate = 0.02 kg/s, IT = 800 W/m 2 , 0.04% silver/water nanofluid Figure 8 shows the variations of pressure drops against the Reynolds number for the different particle concentrations.Pressure drop is found to increase as the particle volume concentration increases, which is primarily due to the increase in viscosity.The increase is more dominant for higher Reynolds numbers; for example, at Reynolds number of 25000 and 0.04% volume concentration, the pressure drop is 67.24 kPa, whereas in the case of water the pressure drop is 62.48kPa, which is a 7.5% increase in pressure drop.For a Reynolds number between 10000 and 15000, the pressure drop increases are 6.45%, 5.25%, and 3.17% for 0.04%, 0.03%, and 0.01% volume concentrations, respectively.It is observed from the present study that the Reynolds number is to be maintained between 10000 and 15000 for low pumping power and best efficiencies.Mass flow rate = 0.02 kg/s, IT = 1000 W/m 2 , 0.04% silver/water nanofluid

CONCLUSION
The performance characteristics of a solar flat-plate collector operated with silver/water nanofluid is experimentally investigated and the effect of the nanofluid on the efficiency of the collector is compared with that of water.The particle volume concentration increases the convective heat transfer rate, and the maximum percentage increase obtained is 18.4% under the tested condition.Similarly, the efficiency of the flatplate collector is found to increase with increasing particle concentration and flow rate.The maximum efficiency of the system is found to be near 70% for 0.04% particle volume concentration at 6 L/min.The primary reason for the enhancement in the heat transfer coefficient is due to the enhanced thermal conductivity of the suspended nanoparticles.

FIGURE 1 .
FIGURE 1. SEM IMAGE OF 0.04% VOLUME CONCENTRATION OF SILVER/WATER NANOFLUID Fig 1 shows the 0.04 % volume concentration of silver/water nanofluid taken at 10X magnification.It is clearly observed from Fig 1 that the silver nanoparticles are uniformly dispersed and have size of less than 100 nm.The thermo physical properties of the nanofluid are calculated as follows: density by Pak and Cho[21], thermal conductivity by Yu and Choi[22], viscosity by Einstein's equation[23].The specific heat of the nanofluid is calculated by Xuan and Roetzel[24].

FIGURE 7 .
FIGURE 7. THE EFFICIENCY OF THE SOLAR FLAT PLATE COLLECTOR AT THE FLOW RATE OF 6 LPM

FIGURE 8 .
FIGURE 8. VARIATION IN PRESSURE DROP WITHRESPECT TO REYNOLDS NUMBER.