Experimental Investigation of Mixed Convection on a Rotating Circular Cylinder in a Cavity Filled With Nanofluid and Porous Media

The present study, experimentally investigated the mixed convection in a square enclosure partitioned in two layers. The experiments were performed with Al 2 O 3 –water nanofluid (upper layer) and superposed porous medium (lower layer) with an adiabatic rotating cylinder at the center of the cavity. The boundary conditions of the experimental study were; the upper and lower walls were assumed adiabatic, the right wall was heated, and the left wall was cooled. Experimentally, 15 K-type thermocouples and thermal imaging camera were employed to measure the temperatures distribution inside the cavity when the concentration of nanoparticles (ɸ = 0.06), the temperature difference (∆T) between the cold and hot walls was (6, 8, and 10) °C, and angular rotational velocity (-50, -25, 0, 25, and 50) rpm. The results of experimental data showed that in general, the distribution of temperatures was very well along the upper half of the enclosure, while in the lower half the temperature distribution was confined near the hot wall region. When the circular cylinder rotates in counter-clockwise, it noted that the effect of speed is evident in the downside of the cylinder, while the temperature distribution in the left upper part of the enclosure decreasing. When the circular cylinder rotates in the clockwise direction, the results showed that the effect of cylinder rotation was around cylinder only. Moreover, the results demonstrated that the increasing temperature difference leads to a noticeable increment in the intensity of the flow.


Introduction
The studies of enhancement convection heat transfer inside the cavity, that has received the attention of researchers because of its multiple engineering applications such as solar collectors, nuclear reactors, electronic equipment cooling, and heat exchangers [1][2][3]. By adding solid nanoparticle to base fluid, the enclosure is developed by repair its geometry, adding fins or baffles and inserting objects inside the enclosure In fixed case or rotating case, etc. [4][5][6].
For controlling and enhancement of the convection heat transfer, a liddriven wall is used [7][8][9][10][11][12][13][14] and a rotating cylinder used inside the enclosures [15]. When using Rayleigh number 1-1400 show the influence of the rotating cylinder to increasing thermal transfer by Lewis [16]. Wu-Shung et al. [17] find the contribution of the counter-clockwise rotation cylinder in the heat transfer 60% when used Gr/Re 2 is 10 3 . It was compared with previous studies and the effect of the rotating cylinder on the heat transfer and flow patterns by Yang et al. [18]. Natural, mixed, and forced convection studied by Misirlioglu and Aydin [19] found when the cylinder rotates at a high speed, the maximum heat transfer is achieved. Costa et al. [20] appear the important effect of cylinder thermal properties inside enclosures on heat transfer. Hussain et al. [21] explain the effect of Richardson number and Reynolds number on average Nusselt number, heat transfer, and the field of fluid, as well as the effect of changing the cylinder location on the convection heat transfer. Roslan et al. [22] using several types of nanofluids with different concentrations and showed that rising the heat transfer when increasing the nanofluid concentrations, as well as the effect of heat transfer when the cylinder rotates in clockwise and counter-clockwise direction, The high heat transfer was found when using nanoparticles with the high concentration and good conductivity values when the rotating cylinder was in the centre of the cavity. By using a rotating cylinder inside horizontal annulus Matin et al. [23] explained the effect of Reynolds number on the average Nusselt number, as well as the effect of the location rotating cylinder. Mixed convection MHD was studied by Selimefendigil et al. [24] and show an increase in entropy and heat transfer with increasing concentration of nanoparticle. Numerical study of a hot rotating cylinder inside a cooled walls square enclosure by Liao et al. [25] and find average heat transfer at the rotating cylinder and the walls of the enclosure. Shih et al. [26] show that the triangular enclosure has been shown the greatest ability to dissipate thermal energy compared to the circular enclosure. By using two rotating cylinders inside a three-dimensional cavity, using nanofluid with different concentrations study by Selimefendigil et al. [27]. To enhance the heat transfer at two layers, a rotating cylinder has been added. Mixed convection that studied by Selimefendigil et al. [15] two layers nanofluid and porous medium inside the cavity and show the enhancement of Nusselt number when the cylinder rotates. Also two layers study by Hussein et al. [28]. The experimental study within nanofluid enclosure investigations by researchers [29,30]. Solomon et al. [31] experimental study on the influence of the aspect ratio of square cavity on natural convection heat transfer with Al2O3/Water nanofluids. Wen and Ding [32] showed a reduced mean size of accumulation by applying a highshear homogenizer. Joshi and Pattamatta [33] found that the lower value of Nusselt number when used (MWCNT/water), while get the higher value in (graphene/water) if compared with (alumina /Water). Khalili et al. [34] showed that the average size of the nanoparticles at the cold side wall was 3.10% higher than that along the hot wall.
Recently, many studies had been done with conviction/mixed heat transfer. A few studies have been done regarding using two layers (nanofluids and porous medium layers) with a rotating cylinder at different angular rotational velocity.
In this study, a mixed convection on the square cavity having two different horizontal layers with a rotating cylinder (clockwise and counterclockwise) is investigated experimentally. The upper and lower walls of the cavity are insulated thermally while the right wall is heated and the left wall is kept at low temperature.

The Experimental Work
This section describes the experimental rig design and its main components. It also shows the preparation of nanofluid and the type of porous medium used in the present work.

Experimental Rig
This section shows the description of the main units of the experimental setup. Fig. 1 and Fig. 2 illustrate the schematic diagram of the experimental setup and the schematic diagram of the experimental set-up respectively. This includes the enclosure, the stepper motor used to rotate the cylinder, the heating unit (consists of a heater, a controller, and a DC power supply), the cooling unit (water bath at a constant temperature) and the measurement system (thermal camera, data logger tool, thermocouples, and PC).   Fig. 3 shows the test rig which is a square enclosure. The top, bottom, front and back walls are made of Perspex glass having a thickness of 0.5 cm. The rotating circular cylinder made of PVC in 3 cm diameter at the center of the enclosure with thermal properties shown in Table 1. The right wall is made of the aluminum plate that heated by an electric heater to control the temperature. The left wall of the enclosure was designed from an aluminum heat exchanger to control the temperature by passing the water through it. Stepper motor was used to obtain rotate the internal cylinder of the enclosure at different rotational speeds. Three thermocouples were installed at each right and left walls to measure temperatures, and eight thermocouples were installed in the middle section of the cavity for measuring the temperature distribution inside the enclosure as shown in

Formulation and characterization of porous media
The porous medium is an object consisting of a fixed solid part called a solid matrix that contains empty spaces that can be filled with fluid. The solid matrix that used in present work made of glass bead with diameter 3 mm as shown in Fig. 5. The properties of the glass bead sphere take from Table 2. The porosity (ε) and permeability (K) of the porous medium are computed from Eqs. 1 and 2 respectively.
The mean value of the calculated porosity values for glass bead (3) mm diameter ε=0.37.  The average value of the permeability for glass bead (3) mm in diameter (K=7.65х10 -9 m 2 ).

Preparation of nanofluid
In the current work, Al2O3 nanoparticles are used having an average size between 40 nm and 80 nm and purity equal to 99.99%.
The preparation of the nanofluid requires several processes that are summarized in the following points: 1-To calculate the Nano-powder weight for all concentrations using Eq. 3 : 2-The sensitive balance that has been used to get the required quantity of mass as shown in Table 3. 3-To obtain an efficient and uniform distribution of nanoparticles as well as a steady mixture by using ultrasound waves (sonication bath for 4 hours). [37] with 0.25% of base fluid like surfactants for improving the stability of the nanofluid. 5-To ensure the homogeneity of the mixture and attain the required volume fraction of the nanoparticle using the mechanical stirrer for 1 hour.  6 shows how the prepared nanofluid, using the above points, is used in the test rig.

Measuring the thermo-physical properties of the nanofluid
In heat transfer, the thermophysical properties of the nanofluid play a significant role. The density, effective viscosity, and thermal conductivity of the nanofluid are computed from Eqs. 4 to 6. These parameters were introduced in the study achieved by Pak and Cho [ 33 ] and listed in Table4.
The dynamic viscosity is fined by using (Brinkman model): The thermal conductivity is fined by using (Maxwell correlation):

Experimental procedure and calculations
In this work, 45 experiments were performed. These experiments are divided into groups using different parameters; the concentration of nanoparticles, the difference in temperature between the hot and cold walls, and the insulated cylinder angular velocity. Firstly, obtaining the thermophysical properties of the nanofluid at nanoparticle concentrations which are specified as (0.02, 0.04, and 0.06). Secondly, the temperature differences between the cold and hot walls are taken equal to (6, 8, and10); which are not very large to avoid turbulent flow and also not very small to ensure a large flow in convective heat transfer. Third, after measuring all temperatures at steady state condition readings are taken inside the cavity for each insulated circular cylinder angular velocity of -50, -25, 0, 25, and 50 RPM. The temperature difference between the cavity and the cylinder wall was within the limits of the mixed convective heat transfer represented by Richardson's number (0.1 < Ri < 10). This next section introduces the device initialization and verifies the accuracy of readings to carry out experiments in a wide range of different operational conditions.

Device initialization
There are several processes, represented the initialization of the device, that should be taken into account before conducting the experiments.
1. The enclosure was initially washed by distilled water to remove impurities and sediments. 2. Half of the cavity is filled with a porous matrix, after that, the nanofluid is added completely to prevent the formation of gas bubbles so that accuracies of the test results are obtained.

3.
Heaters were placed at the hot wall including a controller and a suitable DC power supply; while a constant temperature water bath was used at the cold wall. 4. Temperature results are shown by displaying the data of thermocouples every 10 seconds and saved in the computer. 5. After all, temperatures are measured in a stable condition, the insulated circular cylinder is rotated at the required speed. 6. After stabilization is reached when the temperature difference is less than 0.25ºC in computer data for all points. Finally, a thermal image is taken by the thermal imaging camera (FLIR E30bx).
The experiments were conducted at the College of Engineering/ University of Al-Qadisiyah / Laboratories of the College of Mechanical Engineering.

Achieve device readings
To verify the device is working properly, the results of temperature distribution from a thermal camera were compared with eight thermocouples were installed in the middle section of the cavity. This includes two cases are shown in Fig. 7 for the temperature distribution at the horizontal axis (i.e. x-axis) at the enclosure middle line.
The comparison showed a very clear agreement for all cases and this is an important factor indicating the accuracy of the experimental part.

Results and Discussion
In this case, the results are shown form the test rig to study mixed convection heat transfer. The enclosure is filled with a porous media to half, after that, the nanofluid is added to full the enclosure. The experimental results were taken by a thermal imaging camera and thermocouples to show the distribution of the temperature inside the cavity. In addition, the experimental results were compared, as mentioned represented by the thermal imaging camera and thermocouples.
The experiments are divided into four cases based on specific parameters as follows: 1. The concentration of nanoparticles (0.02, 0.04, 0.06).

The angular rotational velocity of the insulated circular cylinder
(-50, -25, 0, 25, 50) RPM. 4. The inner circular cylinder radius (r = 1.5 cm). Fig.8 and Fig. 9 show the effect of angular rotational velocity on temperature distribution for Al2O3-water, ɸ=0.06, ∆T= 6. It can be seen from this figure the rotation of the cylinder is studied in five cases: two angular rotational velocities in counter-clockwise, two angular rotational velocities in a clockwise direction, and when the cylinder is stationary.

Effects of rotating circular cylinder with temperature difference (∆T= 6)
The figure shows that the distribution of temperatures is greater in the upper of the cavity, while it is confined near the hot wall in the lower half due to low permeability. The lower permeability impedes the fluid due to the use of low porosity of the porous matrix so that the conduction is dominant.
When Ꞷ=0 rpm the fluid moves by natural convection showing different temperatures between the right hot vertical wall and the cold left wall depending on the buoyancy force. Therefore, the fluid flow moves from the right hot vertical wall towards the upper horizontal wall that is thermally insulated. Then the flow completes its cycle down affected by the cold left vertical wall reaching the insulated lower horizontal wall. It can be noticed that the presence of a thermally insulated cylinder disturbs the movement of the fluid, resulting in a decrease the distribution of temperature in the upper half to the right of the cylinder (T=19 ºC), as well as in the lower half the influence of the hot wall and the buoyancy force are weak at the left of the cylinder and (T=17.3 ºC).
When the circular cylinder rotates in a counter-clockwise at (Ꞷ=-25 rpm). Note the effect of speed is evident in the down of the cylinder where (T =19 ºC), while the temperature distribution in the left upper part decreases. This is because the part of the fluid due to the rotation of the cylinder collides with the porous medium layer with little permeability to rise towards the isolated upper wall so it will prevent the buoyancy force. At (Ꞷ=-50 rpm) the influence of the rotational speed is more evident, where the same physical interpretation of (Ꞷ=-25 rpm), but the density of the heat distribution is greater. In this case, the temperature at the bottom of the cylinder increases from (19 ºC to 20.5 ºC). Also, the distribution of heat increases relatively to include the area near the cold wall.
At (Ꞷ=25 rpm) the cylinder rotates clockwise direction where the influence of a rotating cylinder is opposite to the influence of buoyancy force. Notice that the effect of the cylinder to rotate the fluid is limited to the area near it where (T=18.7 ºC), while the effect of buoyancy force appears in other areas, especially in the upper part of the cavity where (T=20.2 ºC).
At (Ꞷ= 50 rpm) the influence of the rotational speed of the cylinder is more influential than the buoyancy force inside most regions of the cavity, where the heat distribution density is greater. Therefore, the temperature at the bottom of the cylinder increases from (19 ºC to 20.5 ºC). The heat distribution also increases relative to the area near the cold wall while it decreases at the upper right of the cavity.  Fig. 11 shows the effect of angular rotational velocity on temperature distribution for Al2O3-water, ɸ=0.06, and ∆T= 8. The same five cases, related to the rotation of the cylinder, examined at ∆T= 6 are presented in this section for ∆T= 8. The physical interpretation of the cylinder rotation is the same as in the previous case, but the temperature difference here is greater. It is interesting to a reminder that the increase in temperature difference, results in increasing the intensity of the flow, which in turn leads to an increase in the buoyancy force.

Fig. 10 and
Considering mixed convection heat transfer inside an enclosure, there are two effects on the production of flow vortices, the first is the buoyancy force and the second is the sheer force of the rotational circular cylinder. Therefore, it can be seen that the influence of rotation of the cylinder is greater compared to the with the buoyant force if they are in the same direction while the effect of the cylinder is absent at lower speeds due to the high buoyancy force when there are different directions. Fig. 12 and Fig. 13 shows the effect of angular rotational velocity on temperature distribution for Al2O3-water, ɸ=0.06, ∆T= 01. The same five cases, related to the rotation of the cylinder, examined at ∆T= 8 are presented in this section for ∆T= 10. The physical interpretation of the cylinder rotation is the same as in the two previous cases, but the temperature difference here is greater. When (Ꞷ=0 rpm) the fluid moves by natural convection heat transfer at different temperatures between the right hot vertical wall and the cold left wall. It can be seen that due to the difference in high temperatures, the distribution of heat includes most regions of the cavity, and because of the accumulation of temperatures under the cylinder where it reaches T=22.3 ºC.

Effects of rotating circular cylinder with temperature difference (∆T= 10)
In general, when the cylinder rotates, the heat transfer inside the cavity increases, and the reason is the influence of buoyancy and cylinder rotational force. This shows the importance of mixed convection in increasing heat transfer.

Conclusions
From the experiment results, some important conclusions can be presented as follows: 1. In general, it is observed that the distribution of temperatures that is greater in the upper half of the considered enclosure, while it is confined near the hot wall in the lower half due to low permeability that impedes the fluid because of using low porosity of porous matrix so the conduction is dominant. 2. When (Ꞷ =0 rpm), the fluid moves by natural convection heat transfer at different temperatures, the flow field inside the cavity is controlled only by the buoyancy force, and the presence of a thermally insulated cylinder disturbs the movement of the fluid. 3. When the circular cylinder rotates counter-clockwise, the effect of the speed is evident in the down of the cylinder, while the temperature distribution in the left upper part decreases because a part of the fluid due to the rotation of the cylinder collides with the porous medium layer with little permeability to rise towards the isolated upper wall so it prevents the buoyancy force.
4. When the circular cylinder rotates clockwise direction, the influence of a rotating cylinder is opposite to the influence of the buoyancy force. Notice that the effect of the cylinder to rotate the fluid is limited to the area near it, while the effect of buoyancy force appears in other areas, especially in the upper part of the cavity. When temperature difference increases, a noticeable increase in the intensity of the flow is observed due to the increase in temperature, which leads to increased buoyancy force. Therefore, in mixed convection heat transfer, there are two effects on the production of flow vortices inside the cavity; the first is the buoyancy force and the second is because of the sheer force that resulted from the rotating cylinder.