Numerical assessment of the influence of helical baffle on the hydrothermal aspects of nanofluid turbulent forced convection inside a heat exchanger.

This study is devoted to the numerical assessment of the influence of helical baffle on the hydrothermal aspects and irreversibility behavior of the turbulent forced convection flow of water-CuO nanofluid (NF) inside a hairpin heat exchanger with 100 mm length, 10 mm inner tube internal diameter, and 15 mm outer diameter internal diameter. The variations of the first-law and second-law performance metrics are investigated in terms of Reynolds number (Re = 5000-10,000), volume concentration of NF ([Formula: see text]) and baffle pitch (B = 25-100 mm). The results show that the NF Nusselt number grows with the rise of both the Re and [Formula: see text] whereas it declines with the rise of B. In addition, the outcomes depicted that the rise of both Re and [Formula: see text] results in the rise of pressure drop, while it declines with the increase of B. Moreover, it was found that the best thermal performance of NF is equal to 1.067, which belongs to the case B = 33.3 mm, [Formula: see text]=2%, and Re = 10,000. Furthermore, it was shown that irreversibilities due to fluid friction and heat transfer augment with the rise of Re while the rise of B results in the decrease of frictional irreversibilities. Finally, the outcomes revealed that with the rise of B, the heat transfer irreversibilities first intensify and then diminish.

Introduction

It is well known that the turbulent flow has a higher heat exchange rate and pumping power than the laminar flow, the former being desirable and the latter undesirable[1-3]. So, this idea came into the researchers' minds to put equipment in the path of laminar flow and create local turbulence[4-9]. The idea was very successful and was widely used in the industry today[10-12]. This equipment is called turbulator, and so far, various types of turbulators were introduced and their performance was studied experimentally and numerically. Also, the use of nanostructures was developed to increase the efficiency of engineering systems[13-16].

Although the use of turbulators has led to an improvement in the performance of heat transfer systems, this has not prevented researchers from looking for ways to further improve the performance of these systems. One of these amazing techniques, which are originated from the low thermal conductivity of heat transfer fluids, is the use of nanofluids (NFs) instead of common heat transfer fluids[17-19]. Choi[20] first made these modern fluids and called them NFs. After the introduction of NFs and their amazing thermal properties, much research was done on their performance in diverse applications such as thermal management of electronic components[21,22] and photovoltaic panels[23,24], nuclear applications[25,26], and performance improvement of solar thermal collectors[27-29], heat exchangers[30-32] and car radiators[33-35]. The literature inspection shows that the performance of thermal systems with NF coolants equipped with turbulators was investigated by various researchers. Bellos et al.[36] analyzed the efficacy of oil-CuO NF in a parabolic trough collector equipped with turbulators. They found that using the combination of NF and turbulator causes a 1.54% thermal efficiency improvement. Nakhchi and Esfahani[37] inspected the efficacy of aqueous Cu NF inside a heated tube equipped with the perforated conical rings in a turbulent regime. It was reported that using compound NF and turbulator results in considerable heat transfer intensification. Akyurek et al.[38] experimentally evaluated the forced convection flow of water-Al2O3 NF inside a horizontal tube equipped with a wire coil turbulator. They utilized two turbulators with different pitches and found that the performance metrics of a tube filled with NF without any turbulator are superior to that of the cases with turbulator. Xiong et al.[39] simulated the efficacy of aqueous CuO NF flowing inside a tube having a compound turbulator. The outcomes portrayed that using turbulator elevates the rate of heat exchange between tube wall and NF. Xiong et al.[40] numerically investigated the forced convection of NF through a pipe equipped with a complex-shaped turbulator. They inspected the consequence of elevating width ratio, flow rate, and pitch ratio on the performance features. It was found that the Nusselt number intensifies with boosting pitch ratio of turbulator. In a numerical investigation, Ahmed et al.[41] explored the forced convection of water-Al2O3 and water-CuO NFs inside a triangular duct with a delta-winglet pair of turbulator under a turbulent flow regime. They reported the significant effect of using both NF and turbulator on the performance aspects. The design of a heat transfer system can be done both according to the first law and the second law of thermodynamics. If the first law applies, the system must have the highest overall hydrothermal performance, and if the second law applies, the system performance must have the least irreversibility. The efficacy of NF flow in turbulator-equipped thermal units has rarely been investigated from a second-law perspective[42-45]. Sheikholeslami et al.[42] inspected the irreversibility features of turbulent flow of aqueous CuO NF inside a pipe equipped with complex turbulators. Li et al.[43] simulated the NF irreversibility in a tube with helically twisted tapes. The thermal irreversibility was found to be declined with elevating the height ratio of turbulator, while the opposite was true for the frictional irreversibility. Farshad and Sheikholeslami[44] analyzed the irreversibility aspects for aqueous Al2O3 NF flow in a solar collector having a twisted tape. The outcomes revealed that the irreversibility diminishes with the increase of diameter ratio. Al-Rashed et al.[45] examined the influence of nanomaterial type on the irreversibility production of an NF in a heat exchanger. It was reported that the maximum irreversibility belongs to the platelet shape nanomaterials.

This numerical work aims to evaluate the features of turbulent flow of water-CuO NF through a hairpin heat exchanger equipped with helical baffles in the annulus side from both the first and second-law perspectives. The impacts of , , and on the NF efficacy are assessed. This investigation is the first work on the consequences of using a helical baffle on the irreversibility production inside the annulus of a hairpin heat exchanger filled with NF.

Problem statement

Figure 1 gives a demonstrative sketch of the geometry under investigation. It is a hairpin heat exchanger with 100 mm length, 10 mm inner tube internal diameter, and 15 mm outer diameter internal diameter. Additionally, the wall thickness for both the inner and outer tubes is 1 mm. Moreover, the considered values are 25 mm, 33 mm, 50 mm, and 100 mm. The purpose of using this device is to cool the water-CuO NF passing through the annulus with the help of water passing through the inner tube. In both water and NF streams, time is of the essence and both streams are in a turbulent regime. Both streams enter the device at a uniform velocity and temperature and are discharged into the atmosphere and as a result, the relative pressure at the outlets is zero. Also, the exterior wall of the heat exchanger is considered insulated and the no-slip condition is utilized on the walls.

Figure 1

Schematic sketch of the problem.

Governing equations and problem parameters

To simulate the steady, incompressible, and turbulent flow of nanofluid and water in the annulus and inner tube of the hairpin heat exchanger studied in the present study, the conservation equations of mass, momentum, and energy must be solved. These equations are as follows[46-48]:wherewhere is the total heat flux that the fluid receives over the entire computational domain and is calculated as follows:

Continuity equation

Momentum equation

Energy equationwhere and are defined as follows;

The - turbulence scheme is employed to establish the turbulence. To this end, two additional equations for turbulent kinetic energy () and the turbulent dissipation rate () are added which are as follows[42]:where , ; ; ; ; ; .

The irreversible generation in forced convection flow of NF flow is due to two sources, namely fluid friction and heat exchange. Therefore, the total irreversibility is computed as[47]:

By integrating the whole computational domain, the total irreversibility can be obtained as [47]:

Bejan number[49]:

The characteristic of an NF includes its thermophysical properties that establish the relationship between the base fluid and the nanoparticles. These basic relationships for the water-CuO NF are defined as follows[50-53]:

The heat transfer rates for hot and cold fluids are calculated as follows:

Average Nusselt number[10]:

The heat transfer coefficient[54-56]:

The pressure drop is defined as follows:

Thermal performance can be computed as:

Numerical scheme, validation, and grid independency

In the current numerical investigation, the simulations were conducted using the ANSYS Fluent 18.1 software. The governing equations were discretized using the second-order upwind technique. Besides, the SIMPLE scheme was utilized to perform the pressure–velocity coupling. The convergence metric was set to 10–6[57-63]. Four different types of grids were used to make the outputs of the problem independent of the mesh. The tetrahedral mesh was used. The walls had a boundary layer mesh with a factor of 5%. The Nusselt number was used to perform the mesh study. Finally, it was found that the most appropriate mesh is the one with a 650,538 element number. To ensure the accuracy and validity of the results of the present work, the experimental findings of Wen and Ding[64] and the numerical outcomes of Goktepe et al.[65] for the local convection coefficient of NF flowing through a uniformly heated tube were employed. The outcomes are reported in Fig. 2. As can be observed, there is a good consistency between the outcomes, and the largest discrepancy between our results and the data from the other works (i.e. Refs.[64,65]) is 10%. To further validate the simulations, the average Nusselt number of NF flow inside a channel with rectangular rib (in-line) reported by Vanaki and Mohammed[66] was compared with our results. As can be seen in Fig. 3, at low velocities, a good consistency is observed, and as the flow velocity elevates, the discrepancy of the results is elevated and reaches 7%.

Figure 2

Validation with experimental[64] and numerical[65] works.

Figure 3

Validation with numerical work[66].

Results and discussion

The main focus of this numerical investigation is to analyze the turbulent flow of water-CuO NF inside a hairpin heat exchanger with the helical baffle on the annulus side. The effects of (0–4%), (5000–10,000), and (25–100 mm) on the performance metrics are investigated. Figure 4 depicts the contour plots of NF velocity and temperature for different values of at and . It is seen that as augments, the NF velocity diminishes.

Figure 4

Contours of NF velocity (left) and temperature (right) in terms of at and .

Figure 5 gives the consequences of elevating on the velocity and temperature contours of NF for =25 mm. As is seen, intensifying the entails an elevation in the NF velocity and a decrement in the NF temperature.

Figure 5

Contours of NF velocity (left) and temperature (right) in terms of for .

Figure 6 gives the Nusselt number in terms of for various . The Nusselt number improves by elevating both the and , while it declines by augmenting the . For instance, at and , elevating the from 25 to 100 mm results in a 19.05% decrease in the Nusselt number of NF, while this amount for 4% is 15.62%. In addition, at =25 mm and , intensification of from 5000 to 10,000 causes a 61.83% decline in the Nusselt number of NF. Elevating the entails a rise in the thickness of the velocity and thermal boundary layer, which elevates the temperature gradient and, consequently, elevates the convective heat transfer coefficient and Nusselt number. Moreover, intensification of causes augmentation in the and, thereby, an augmentation in the convective heat transfer and Nusselt number. Furthermore, intensifying the baffle pitch entails a decrease in the NF velocity and, therefore, a decrease in the convective heat transfer coefficient and Nusselt number.

Figure 6

Variations of Nusselt number versus in terms of for (a) , (a) and (b) .

One of the important issues when choosing a working fluid is that the fluid performance should be examined from both the heat transfer and pressure drop (pumping power) aspects. Figure 7 gives the variations of the pressure drop of NF versus and for different . It is observed that the pressure drop intensifies by boosting both the and , while it reduces by boosting the . For example, at and , boosting the from 25 to 100 mm results in a 73.59% reduction in the pressure drop of NF, while this amount for 4% is 73.55%. Additionally, at =25 mm and , augmentation of from 5000 to 10,000 results in a 232.82% decrease in the pressure drop of NF. Boosting both the and causes an elevation in the NF velocity and, therefore, according to Eq. (24), the pressure drop of NF elevates. On the other hand, elevating the causes an increment in the annulus fluid path and, as a result, an elevation in the pressure drop, while the NF velocity declines by boosting the which results in a decrease in the pressure drop of NF. According to the obtained results, it can be concluded the effect of elevated annulus flow path on the pressure drop outweighs the effect of decreased NF velocity and, therefore, the NF pressure drop reduces with an elevation in the .

Figure 7

Variations of pressure drop versus in terms of for (a) , (a) and (b) .

The results presented so far have shown that intensifying the entails an elevated Nusselt number and pressure drop, which the first is a favorable effect and the latter is undesirable. For the final decision on the usefulness of using NF in the heat exchanger under investigation, the performance index should be examined. Figure 8 demonstrates the variations of the performance index versus and for different values. The outcomes reveal that the hydrodynamic performance of the water-CuO NF in the considered heat exchanger is superior to that of the pure water in just the following four cases:

Figure 8

Variations of performance index versus in terms of for (a) , (b) and (c) .

mm, and where .

mm, and where .

mm, and where .

mm, and where .

In the remainder of this section, the flow of water-CuO NF with in the considered heat exchanger is examined from the perspective of irreversibility production. Figure 9 displays the influences of and on the frictional irreversibility of NF. As can be seen, the frictional irreversibility declines and rises with increasing and , respectively, and the effect on the frictional irreversibility at lower is greater. For example, at , increasing the from 25 to 100 mm results in a 67.84% decrease in frictional irreversibility. In addition, at mm, the augmentation of from 5000 to 10,000 causes a 209.45% increment in the frictional irreversibility. By boosting the at a constant (i.e. constant Prandtl number) and , the thickness of the velocity boundary layer decreases, which results in an elevated velocity gradient and thus elevated frictional irreversibility. On the other hand, increasing the at a constant and , results in an elevated NF velocity and hence a decrease in the average NF temperature which results in elevated frictional irreversibility. Elevating the results in a decrease in the flow mixing, which results in a decrease in the velocity gradient and a decrease in the NF temperature, which in turn decreases and elevates the frictional irreversibility. It can be concluded that the decreasing effect of velocity gradient on the frictional irreversibility is overcome by the increasing effect of NF temperature and, therefore, the frictional irreversibility decreases with increasing .

Figure 9

Variations of frictional irreversibility of NF with versus in terms of .

Figure 10 gives the changes of thermal irreversibility of NF with versus in terms of . It is seen that the thermal irreversibility declines with the elevation of . For example, at mm, the intensification of from 5000 to 10,000 causes a 52.57% decrease in thermal irreversibility. As mentioned before, the NF temperature decreases with increasing at a constant and , which results in elevated thermal irreversibility. Also, the augmentation of results in improved mixing of the flow and consequently, a decrease in the temperature gradient which ultimately results in a decrease in the thermal irreversibility. It can be said that the impact of temperature gradient on the thermal irreversibility outweighs the impact of NF temperature and therefore, the thermal irreversibility declines with the rise of . Moreover, Fig. 10 shows that with the rise of at a constant and , the thermal irreversibility first elevates and then reduces. The highest thermal irreversibility occurs at =50 mm. Boosting of at a constant and reduces the flow mixing which results in the decrease of both the NF temperature and temperature gradient which respectively elevates and diminishes the rate of thermal irreversibility. Figure 10 reveals that for the lower than 50 mm, the increasing impact of temperature outweighs the decreasing of the temperature gradient, and ultimately, the thermal irreversibility declines while the opposite is true for the higher than 50 mm.

Figure 10

Variations of thermal irreversibility of NF with versus in terms of .

The results presented in Figs. 9 and 10 show that increasing the baffle pitch results in a decrease in both the thermal and frictional irreversibilities, and therefore, it can be easily deduced that increasing the entails a declined total irreversibility. However, the influence of on the total irreversibility cannot be predicted because the thermal irreversibility elevates with decreasing and then decreases.

Conclusion

In this study, the first law and the second law of thermodynamics are employed to investigate the turbulent flow of aqueous CuO NF through a hairpin heat exchanger equipped with the helical baffle on the annulus side. The influence of (5000–10,000), (0–4%), and (25–100 mm) on the performance metrics are assessed. The following results can be deduced from this simulation:

With the rise of at a constant and , the thermal irreversibility first elevates and then reduces.

Intensifying the entails an elevated NF velocity and a declined NF temperature.

Pressure drop intensifies by boosting both the and , while it reduces by boosting the .

Frictional irreversibility declines and rises with increasing and , respectively.

effect on the frictional irreversibility at lower is greater.

In future studies, the effect of using perforated and porous turbulators on the results presented in the present study will be investigated.