Performance enhancement of straight and wavy miniature heat sinks using pin-fin interruptions and nanofluids

Accepted Manuscript Title: Performance enhancement of straight and wavy miniature heat sinks using pin-fin interruptions and nanofluids Authors: M. Khoshvaght-Aliabadi, S.M. Hassani, S.H. Mazloumi PII: DOI: Reference:

S0255-2701(17)30506-8 https://doi.org/10.1016/j.cep.2017.10.002 CEP 7086

To appear in:

Chemical Engineering and Processing

Received date: Revised date: Accepted date:

23-5-2017 20-9-2017 1-10-2017

Please cite this article as: M.Khoshvaght-Aliabadi, S.M.Hassani, S.H.Mazloumi, Performance enhancement of straight and wavy miniature heat sinks using pin-fin interruptions and nanofluids, Chemical Engineering and Processing https://doi.org/10.1016/j.cep.2017.10.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Performance enhancement of straight and wavy miniature heat sinks using pin-fin interruptions and nanofluids M. Khoshvaght-Aliabadi a,*, S.M. Hassani b, S.H. Mazloumi b a

b

Department of Chemical Engineering, Shahrood Branch, Islamic Azad University, Shahrood, Iran.

Chemical Engineering Department, Faculty of Engineering, Ferdowsi University of Mashhad, Mashhad, Iran.

*(Corresponding author): E-mail 1: [email protected] E-mail 2: [email protected] Phone: +98 9151811311, Fax: +98 5147244818, Postal address: 36199-43189.

Graphical Abstract

Highlights    

► Effects of pin-fin interruptions in straight and wavy MHSs are studied. ► Water and Al2O3/water nanofluid are considered as coolants. ► Interrupted MHSs have better hydrothermal performance. ► Al2O3/water nanofluid improves overall hydrothermal performance of MHSs.

Abstract The current study reports both experimental and numerical studies on flow and heat transfer characteristics of straight and wavy miniature heat sinks (MHSs). Effects of different interruptions

1

of pin-fin and concentrations of Al2O3/water nanofluids (0, 0.1, and 0.4 wt.%) are investigated. For the experimental part, three pin-fin interruptions and nanoparticle concentrations are considered and tested for Reynolds number ranged from 100 to 900. For the numerical part, a conjugate simulation is carried out using the finite volume approach and SIMPLEC algorithm. A comparison is conceded between the integral and the interrupted MHSs. The results are presented in terms of base temperature, heat transfer coefficient, Nusselt number, pressure drop, pumping power, and performance factor. It is found that the interrupted MHSs have better hydrothermal performance and more uniform temperature distribution. Also, the application of Al2O3/water nanofluids improves the overall hydrothermal performance of straight and the wavy MHSs. The maximum performance factors of 2.65 and 2.46 are obtained, respectively, for the 0.4 wt.% nanofluid flow in the straight and wavy MHSs with interrupted-staggered arrangement of pin-fins. Based on the current evaluation, it can be concluded that the pin-fin interruption and nanofluid flow have considerable effects on the performance of straight and wavy MHSs. Keywords: Straight and wavy MHSs; Pin-fin interruption; Al2O3/water nanofluid; Experimental study; Numerical simulation.

1. Introduction Micro/mini channel heat sinks that have been proposed by Tuckerman and Peace [1] in 1981 are a kind of thermal conductive metal devices designed to absorb and dissipate excess heat from a high temperature object such as a micro electromechanical system. Generally, two passive techniques are employed in order to enhance the thermal performance of water-cooled MHSs. The first technique is to modify the surface geometry of MHSs, while the second approach involves the modification of thermal properties of water. For instance, the effect of fin spacing on the performance of a flat-plate integral MHS was examined by Jajja et al. [2]. It was found that the thermal performance of MHS enhanced by decreasing the fin spacing. In the other work [3], they 2

concluded that using multiwalled carbon nanotube nanofluid can decrease the base temperature of MHS. An investigation was done on TiO2/water nanofluid in a microchannel with a focus on the effects of uncertainties in thermal conductivity and viscosity models on the prediction of Nusselt number and friction factor [4]. It is found that the difference among the Nusselt numbers predicted by the various models is about 2% (at ϕ = 2%), which is not considerable. However, for the friction factor, it is observed that the differences between the friction factors predicted by the three models at ϕ = 2% are great, so that using the viscosity obtained based on the experimental data is necessary to estimate the friction factor. Arshad and Ali [5,6] applied TiO2 nanoparticle and Graphene nanoplatelet based nanofluids in a straight integral MHS at the laminar flow regime. It was reported that the thermal performance of nanofluids in the MHS decreased as the heat flux increased. Tafarroj et al. [7] used an artificial neural network approach to model the heat transfer coefficient and Nusselt number of TiO2/water nanofluid flow in a MHS. It was elucidated that an appropriately trained network can act as an appropriate approach for costly and time-consuming experiments on the nanofluid flow in the MHSs. For example, in the considered case, the average relative errors in the prediction of heat transfer coefficient and Nusselt number were 0.2% and 0.3%, respectively. In addition to the MHSs, the application of nanofluids was expanded in different heat exchange devices [8–11]. On the other hand, using indirect channels or interrupted pins to increase the heat transfer surface area corresponds for large number of experimental and numerical analysis. In recent years, different water based nanofluids have been tested inside the MHSs by many researchers. However, it is clear from the reviewed literature that studies on the hydrothermal performance of MHSs with the modified geometries working with water based nanofluids are very scarce.

3

A numerical study on flow and heat transfer behaviors of CuO/water and Al2O3/water nanofluids inside the pin-fin MHSs was carried out by Seyf and Feizbakhshi [12]. It was reported that the performance of MHSs enhanced as the nanofluids were replaced with the base fluid. Also, decreasing the diameter of Al2O3 nanoparticles in the base fluid increased Nusselt number values while the trend was reverse for CuO nanoparticles. Tokit et al. [13] analyzed numerically the performance of Al2O3, CuO, and SiO2 nanoparticles as additives in water flowing in an interrupted MHS. The highest Nusselt number enhancement was predicted for Al2O3, followed by CuO and SiO2. In another numerical investigation, Kuppusamy et al. [14] studied the temperature and flow fields of different nanofluids in a trapezoidal grooved MHS. The effects of different geometrical parameters, nanoparticle diameters and concentrations, and Reynolds numbers were explored. The results clarified that the trapezoidal grooved MHS with higher maximum-width and lower minimum-width displayed the maximum thermal performance. It was also reported that Al2O3/water had the highest thermal performance with 4% volume fraction and 25 nm nanoparticle diameter. The heat transfer and pressure drop specifications in a heat sink fitted with dynamic mixers using Al2O3/water nanofluid were studied by Selvakumar et al. [15]. A significant enhancement in heat transfer for the simultaneous application of dynamic mixer and nanofluid was reported. The performance of a pin-fin MHS with different configurations (square, triangular, and circular) was numerically investigated in the presence of diamond/water and Al2O3/water nanofluids by Hasan [16]. The results showed that using of nanofluids instead of the base fluid enhanced the thermal performance with a certain penalty in the pressure drop for all configurations. Zhai et al. [17] examined complex structure and Al2O3/water nanofluid in a MHS under constant heat flux. It was detected that the applied compound heat transfer enhancement technique can obviously enhance the heat transfer. The Nusselt number displayed an increase of 1.4 times

4

compared to that of base line. The simultaneous application of wavy channel and nanofluid as a compound technique was numerically examined by Sakanova et al. [18]. The effects of different geometrical parameters (wavy-amplitude and wave-length) and nanofluid types (Diamond/water, SiO2/water and CuO/water) were studied. It was reported that nanofluid had higher effect on the performance of MHS with straight channel. The performance of a MHS with inline arrangement of circular pin-fins in presence of ZnO/water and SiO2/water nanofluids was experimentally investigated by Duangthongsuk and Wongwises [19,20]. The acquired data explained that the thermal performance of nanofluid-cooled MHS was better than that of water-cooled MHS. Comparison between ZnO/water and SiO2/water nanofluids, higher heat transfer performance for ZnO/water nanofluid was observed by about 3–9%. The effects of ribs turbulator with different design parameters on the hydrothermal performance of a nanofluid-cooled MHS were numerically investigated by Ghale et al. [21]. The results showed that both the heat transfer coefficient and the pressure drop of nanofluid in the ribbed MHS were higher than those of the simple MHS, and this enhancement intensified with increasing the width of ribs. Ali and Arshad [22] performed an experimental study on a nanofluid-cooled MHS with inline and staggered arrangements of square pin-fins. TiO2(Anatase)/water and TiO2(Rutile)/water nanofluids were tested and their results were compared with water. The minimum base temperature of 29.4 °C was reported for the TiO2(Rutile)/water nanofluid flow in the MHS with staggered arrangement of pin-fins. A new innovative design of MHS with rectangular and triangular double-layered channels working with Al2O3/water and SiO2/water nanofluids was tested by Ahmed et al. [23]. The results showed that the MHS with triangular double-layered channels provided a 27.4% reduction in the wall temperature comparing with the MHS with rectangular double-layered channels. Experimental tests were conducted for Al2O3/water nanofluid through a MHS with corrugated channels by

5

Khoshvaght-Aliabadi and Sahamiyan [24]. The effects of geometrical parameters, nanoparticles mass fraction, and flow rate were investigated. The base temperature and convective thermal resistance were found to drop by decreasing the wave-length and by increasing the waveamplitude. Different corrugated pin-fins in the MHS were also tested by the same team [25,26]. Ali and Arshad [27] performed an experimental study to examine the angle effect of pin-fins on the performance of MHS. It was concluded that the MHS with 22.5 degree channel angle had the lowest convective thermal resistance. Finally, Khoshvaght-Aliabadi et al. [28] conducted a parametric study on MHS with offset-strip pin-fins and Al2O3/water nanofluid. Effects of different geometrical parameters, including strip thickness, strip length, strip transversal pitch, and strip longitudinal pitch, were examined on the performance of MHS. It was detected that the strip length had a stronger influence than the other studied parameters, and the cooling performance of the nanofluid-cooled MHS was greater than the base fluid-cooled MHS. In the current work, both the modifications in the surface area of MHS and the thermal properties of water are also considered. The idea of interrupting the pin-fins in the straight and wavy MHSs and adding the α-Al2O3 nanoparticles in water is investigated experimentally. To show their effects on hydrothermal behavior of the water-cooled MHSs, three pin-fin interruptions and nanoparticle concentrations are considered. It is expected that heat transfer and pressure drop characteristics of the Al2O3/water nanofluid through the straight and wavy MHSs will be considerably affected with pin-fin interruptions. Furthermore, numerical simulations are carried out to provide a more comprehensive qualitative assessment which is difficultly accessible in the experimental part. The first objective is to validate the numerical results with the obtained experimental data. The second one is to evaluate the influence of pin-fin interruptions on

6

temperature counters and velocity vectors inside the straight and wavy MHSs, as two useful configurations of MHS. 2. Experimental part 2.1. Test setup and MHSs configuration Generally, the MHSs are fabricated in two types: active and passive. A water-cooled MHS system is an active type and consists of four main parts, I. II.

A reservoir for water, A pump to drive water,

III.

A pin-fin MHS as the heat transfer media,

IV.

A water-to-air heat exchanger (radiator + fan) to restore the temperature of water.

In the current work, the cooling performance of water-cooled MHSs is evaluated in a closed flow loop (Fig. 1). It comprises all the mentioned parts for a water-cooled MHS system as well as measurement instruments and control facilities. In hydrothermal studies for the MHS, the measurement parameters are the flow rate and inlet/outlet bulk temperatures of water, the temperature of MHS base surface, and the pressure drop of water through the MHS. Also, the rate of heat inducted to the base area of MHS, i.e. heat flux, is the control parameter. Details about the applied test setup are tabulated in Table 1. More information about applied apparatus and procedure are included in the previous works by Khoshvaght-Aliabadi et al. [29,30] which, for the purpose of brevity, are not repeated here. Please insert Fig. 1 here Please insert Table 1 here Although the water-cooled MHS with straight integral pin-fins is simple and low cost, the inherent defect is that straight pin-fins make water path smoother, which is undesirable form the

7

heat transfer point of view. To enhance the heat transfer, the corrugated pin-fins are proposed, which are constructed in triangular, trapezoidal, and sinusoidal shapes [30]. Certainly, the pressure drop and pumping power tend to increase as the corrugated pin-fins are applied in the water-cooled MHSs. However, up to now, the straight and wavy pin-fins in the water-cooled MHSs are usually in integral shapes due to the restriction of fabrication technology. In the current analysis, different interruptions of the straight and wavy pin-fins in the water-cooled MHSs are investigated for the first time. Fig. 2(a–b) shows the considered configurations for the pin-fin interruptions in both the straight and the wavy MHSs. To make a logical comparison condition, all geometrical parameters in the straight and wavy MHSs are set equal. Details of geometrical parameters of the fabricated MHSs are presented in Table 2. The test module is constructed with an interchangeable test section. It consists of a base housing, a MHS, eight cartridge electrical heaters, nine K-type thermocouples, two T-type thermocouples, two digital pressure transmitters. The base housing and MHSs are manufactured by CNC milling from blocks of aluminum alloy 6061 with the thermal conductivity of 170 W/m.K. Please insert Fig. 2 here Please insert Table 2 here 2.2. Nanofluid preparation and properties A two-step procedure is adopted for the preparation of nanofluids in which alpha aluminum oxide (α-Al2O3) nanomaterial and deionized-water are selected as solid phase (or nanoparticles) and fluid phase (or base fluid), respectively. The preparation procedure is extensively explained in Ref. [24] and only the essential details are given here. Firstly, desire weight of nanoparticles are separated by a precise digital-electronic balance (CPA1003S, Sartorius). Then, they are gently

8

dispersed in to the weighted deionized water using an electro-magnetic stirrer (MR Hei-End, Heidolph) for 1 hour at 25 °C. It must be noted that no surfactant is added into the base liquid. A proper nanofluid, which is not a simple suspension of solid in liquid, should have good homogeneity, high stability, and low sedimentation. Therefore, the mixture is continuously sonicated by an ultrasonic processor (UP400S, Hielscher GmbH) for 1 hour at 400 W and 24 kHz. The specifications of initial phases are presented in Table 3. Please insert Table 3 here We tested three concentration samples of 0, 0.1, and 0.4 wt.% of Al2O3/water nanofluid because it was shown that the weight concentration less than 0.5% shows good stability and thermal conductivity [31]. The stability of prepared nanofluids are investigated by performing photo capturing tests on close containers for 10 days. It can be pointed out that no visually observable sedimentation or stratification is found even after three days while in the present work, the time taken to complete an experiment is less than 5 hours. Repeatability is an important issue for experiments associated with nanofluids. The experiments of nanofluids are repeated after three days of the preparation. The obtained results show good repeatability values for hydrothermal results within ±5%. It should be noted that the repeatability of the base fluid is about ±2%. Also, after each experiment associated to the nanofluids, the control experiment (i.e., the water flow inside the MHSs) was performed to ensure that the system performance is returned to the base-line condition. The recorded data for the base fluid confirmed the repeatability of the experimental results and returning to the base-line condition. This illustrates that precipitation of nanoparticles cause the formation of nanofins (enhanced surface area resulting in enhanced heat transfer even when the experiment is repeated with pure water) or excessive precipitation leads to fouling (causing degradation in heat flux when

9

the experiment using pure water) is negligible. Likewise, it should be mentioned that the possible agglomeration of nanoparticles may poorly occur in indirect and complex geometries in comparison with direct and simple ones, due to boundary layers regeneration and secondary and swirl flows, which break down the possible agglomerated particles. Fig. 3(a–c) demonstrates the X-ray diffraction (XRD), scanning electron microscope (SEM), and transmission electron microscopic (TEM) images of α-Al2O3 nanoparticles. Fig. 3(a) signifies a single-phase that there is no impurity in the structure of α-Al2O3 nanoparticles. As can be seen in Fig. 3(b), α-Al2O3 nanoparticles are identical in morphology. Based on Fig. 3(c), αAl2O3 nanoparticles have spherical shape with a mean dimeter size of 20 to 40 nm. As depicted in the figure, no visible aggregation was detected for the nanoparticles. On the other hand, the possible agglomeration of nanoparticles may poorly occur in complex geometries, due to boundary layers regeneration and swirl flows, which break down the possible agglomerated particles. Although, it was shown that the aggregation of nanoparticles has both negative and positive effects on the thermal performance of nanofluids [32]. Please insert Fig. 3 here In order to measure the thermal and rheological characteristics of nanofluids, a sample is prepared in volume of 100 milliliter for each mass fraction [33,34]. The thermal conductivity of nanofluids is evaluated using KD2 Pro thermal property analyzer of Decagon Devices, Inc., USA which is based on the transient line heat source method. The accuracy of the sensor is ±5% over the thermal conductivity range of 0.2–2 W/m.K and the temperature span of 0 to 50 °C. The dynamic viscosity of nanofluids is measured using Physica MCR 301 rheometer of Anton Paar, Ashland, USA. Also, the pH value of nanofluids is around 6.5, which is measured using Metrohm 691 pH meter. To make sure of the equipment accuracies, each measurement is repeated five times

10

and the average value is obtained. The variations of measured experimental values of thermal conductivity and dynamic viscosity are compared with those of the base fluid and presented in Fig. 4. Please insert Fig. 4 here Furthermore, the following equations are used to compute the density and specific heat of nanofluids,

nf  1    f   p

 c 

p nf

(1)

 1     c p     c p  f

(2)

p

2.3. Data processing and uncertainty The heat produced by the electrical heaters is absorbed with the flowing fluid in the MHSs. The rate of heat transfer is expressed by the conservation of energy principle, Eq. (3),

Qconv  mc p Tout T in 

(3)

where, m, cp, Tin, and Tout are the mass flow rate, specific heat, inlet and outlet temperature of the fluid, respectively. Eq. (4) is used for the estimation of the experimental heat transfer coefficient,

h

Qconv At Tw T m 

(4)

where, At is the total surface area in contact with the fluid, Tw is the average wall temperature, and Tm is the mean temperature between the inlet and the outlet. The values of total surface area of the fabricated MHSs are presented in Table 4. As depicted in Eq. (4), this parameter affects directly the heat transfer coefficient values. Also, the effects of this parameter on the Reynolds number, Nusselt number, and friction factor depend on the hydraulic diameter, see Eqs. (8) to (11).

11

Fourier’s law, Eq. (5), is also used to evaluate the local wall temperature on the base surface of MHSs,

Twi  T i 

sqcond s

(5)

where, Ti is the temperature indicated by the base thermocouples (i = 1 to 9), s is the distance between the base thermocouples location and the surface of MHSs, qcond is the heat flux, and κs is the thermal conductively of base housing. To ensure that the MHSs can absorb as much heat as possible, a suitable thermal paste is used to create a seal between the MHSs and the base housing. The average wall temperature is calculated by Eq. (6),

1 9 Tw  Twi 9 i 1

(6)

The pumping power can be evaluated by Eq. (7),

PP 

m p f

(7)

where, ∆p is the pressure drop across the MHSs and ρ is the density of fluid. Finally, Reynolds number is considered based on the inlet parameters and calculated by Eq. (8),

Re 

GD h

(8)

f

where, G is the mass velocity and μ is the dynamic viscosity of the fluid. in which,

Dh 

4 Ac L At

(9)

12

where, Ac is the minimum free flow area and L is the length of MHSs. The hydraulic diameter of MHSs is calculated based on the total surface area and free flow area for each MHS and the obtained values are tabulated in Table 4. Also, the average Nusselt number is defined as, Nu 

hD h

(10)

f

where, κf is the thermal conductively of fluid. Finally, the friction factor is determined from the pressure drop measurements,

f 

2 f pD h LG 2

(11)

Please insert Table 4 here In the current work, all the instruments are calibrated for estimating the accuracy of measured data and the uncertainty of calculated parameters. The accuracy of applied measurement instruments are presented in Table 1. Furthermore, the uncertainties for the calculated parameters are evaluated according to the propagation analysis, Eq. (12) [35]. 12

WR

2 2 2   R   R   R     w  w  ...   w      X 1 1    X 2 2  X n n     

(12)

where, R is a function of independent variables X1, X2, …, Xn and w1, w2, …, wn are uncertainties in the independent variables. Therefore, WR is the uncertainty of dependent variable. For instance, as presented in Eq. (13), the uncertainty in calculating the Nusselt number is found to be less than ±1.6%,

13

12

W Nu

2 2 2   Nu    Nu      Nu   w h   w Dh    w f      Dh   h    f   2 12

 D   h   hD     h w h    w D h    2h w f     f   f   f   2

2

(13)

Also, the uncertainty in the pumping power is estimated to be less than ±2.2% and ±1.8% at the lowest and the highest Reynolds numbers, respectively. Note that a repeated measures technique is performed for every case, and the average value of centralized data is applied for the presented results in the current study. 3. Numerical part 3.1. Problem statement and physical models In order to provide a more comprehensive qualitative assessment on the studied straight and wavy MHSs, a separate numerical simulation is also carried out in this report. The main scope of the current simulation is to illustrate the effects of the pin-fin interruption on temperature counters and velocity vectors of the coolant in the straight and wavy MHSs, so only water flow is simulated. As depicted in Fig. 5, the physical models considered in the numerical part have the same configurations as the ones used in the experimental part. However, due to the symmetric and periodic arrangement of pin-fins in the spanwise direction of the MHSs, a certain branch of those and water domain are modeled as the computational zone to reduce the grid number and computational time (Fig. 6) [36]. Note that as illustrated in Fig. 6, the coordinates x, y, and z represent the streamwise, normal, and spanwise directions, respectively. At the inlet of original portion, the ratio of entrance portion length to the original length is 2 in order to ensure the inlet uniformity. The outlet of original portion is also extended by the same length in order to overcome the convergence issues stemming from the reversed flow at the outlet of computational zone. The

14

bottom of computational zone is heated at a constant heat transfer rate of 50 W, that is, the heat flux at the base of MHSs is 25000 W/m2. It is assumed that water flow enters to the MHSs with a uniform velocity and a constant temperature of 27 °C. Based on the experimental test module (Fig. 1), it exchanges the heat with the pin-fins of MHSs, while the upper plexiglass layer is considered thermally insulated. Other assumptions made in the numerical model include: the conductive thermal contact resistance between the MHSs and the base housing is negligible and the heat loss by natural convection from the test module to the ambient is neglected. Please insert Fig. 5 here Please insert Fig. 6 here 3.2. Governing equations and boundary conditions The applied governing equations for the laminar flow and heat transfer of water-cooled straight and wavy MHSs are the Navier-Stokes and energy equations [37] with the following assumptions, (1) water flow is steady-state and incompressible; (2) the thermo-physical properties of both the coolant (water) and the MHS (aluminum alloy 6061) are independent of temperature; (3) the gravity and other forms of body forces are neglected; (4) the buoyancy and radiation heat transfer are ignored; (5) the viscous dissipation is neglected. V  0

f V .V   p  .  f V

(14)



(15)

f c p V .T    f 2T

(16)

For the solid region, the energy equation is expressed as,

 s 2T  0

(17)

The boundary conditions applied for the computational zone are as follow, (1) a constant heat flux condition at the bottom of original portion (red section in Fig. 6); 15

u x  u y  u z  0,

 s

T q y

(18)

(2) no-slip and adiabatic conditions at the walls of entrance and exit portions (colorless sections in Fig. 6); u x  u y  u z  0,

T T T   0 x y z

(19)

(3) no-slip and adiabatic conditions at the top of original portion (gray section in Fig. 6); u x  u y  u z  0,

T T T   0 x y z

(20)

(4) a velocity inlet condition at the inlet of computational zone (blue section in Fig. 6);

u x  u in  const ,

u y  u z  0, T  T in  const

(21)

(5) a symmetry condition at all the side surfaces of original portion (yellow section in Fig. 6); u x u y   u z  0, z z

 s

T s 0 z

(22)

(6) the outlet condition at the exit of computational zone (orange section in Fig. 6); u x u y u z    0, x x x

T 0 x

(23)

(7) the conjugate boundary condition at the internal surfaces (liquid/solid interface);

u x  u y  u z  0, T f  T s ,

 f

T T f  s s n n

(24)

3.3. Numerical method and grid independency A finite volume method (FVM) is used to perform the conjugate numerical simulation and solve the governing equations along with the considered boundary conditions. The pressure-based solver with an implicit solution formulation is applied. It was shown that for the steady-state

16

conditions, the semi implicit method for pressure linked equation (SIMPLE) procedure gives a shortest computational run time [38], so it is introduced for the velocity-pressure coupling. The standard scheme is used for the pressure discretization following the second order upwind for the momentum and energy as the discretization scheme. The diffusion terms in the momentum and energy equations are also approximated by a second order central difference which gives a stable solution. The numerical simulations are considered to be converged when the absolute convergence criterion is achieved for all variables. It is set to be 10-4 for the continuity and velocities and 10-6 for the energy. The numerical method is based on the commercial CFD package (Fluent V6). For checking the independency of numerical simulation results from the grid number, five sets of grid with different numbers are tested for the straight integral MHS. The heat transfer coefficient and pressure drop are considered as check variables and their results at the lowest, middle, and highest Reynolds numbers in this study are shown in Fig. 7. It is clear from the results that the effect of grid number on both the parameters increases as Reynolds number goes up. It is also detected that the values of heat transfer coefficient and pressure drop for the last two grid sets, i.e. fourth and fifth, are very close to each other and the average differences are less than 1%. Therefore, in order to save the numerical simulation time and computer memory, the fourth grid set is employed in the main numerical simulations. Please insert Fig. 7 here 4. Results and discussion Firstly, experiments are conducted to examine the effects of pin-fin interruption on the hydrothermal performance of the straight and wavy water-cooled MHSs for a range of Reynolds number between 100 and 900 corresponding to the laminar regime with a heating power of 50 W.

17

Then, the Al2O3/water nanofluid at 0.1% and 0.4% mass fractions is tested in the MHSs as coolant and the obtained results are compared with water data. Finally, a numerical simulation is carried out to capture temperature counters and velocity vectors in the straight and wavy water-cooled MHSs. The acquired qualitative results in the numerical part are utilized to explain the quantitative data in the experimental part. 4.1. Validation In order to validate the current experimental and numerical results, we use the experimental data obtained by Ho and Chen [39]. Fig. 8 compares Nusselt values obtained experimentally and numerically in the current work and those of Ho and Chen [39]. The comparison shows that the tendency of the current results is consistent with that of Ho and Chen [39]. The mean absolute deviation between the current experimental results and Ho and Chen [39] data is about 9.7%. Furthermore, the maximum deviation between the experimental and the numerical data in the current work is not more than 5%. It may be due to the considered simplifying assumptions in the numerical part, such as neglecting the variation of water properties with temperature and the heat transfer between the MHS and the ambience, and also the uncertainties in the experimental part, such as in the fabrication of MHS and the measurement of flow rate and temperature. It should be noted that the trends of experimental pressure drop values are successfully captured in the numerical simulations and the mean absolute deviation is less than 9%. It clarifies that the numerical simulation predicts reasonably well the current experiments. Hence, the numerical simulation will be employed to conduct a detailed analysis to describe the hydrothermal characteristics of the studied MHSs. Please insert Fig. 8 here 4.2. Effect of pin-fin interruption

18

For micro electromechanical cooling, minimizing the base temperature is the main target because it is directly related to the reliability and performance of system. Fig. 9 shows variations of the average temperature on the top surface of base housing equipped with different straight and wavy water-cooled MHSs. The main results obtained from the figure can be explained as follows, o The base temperature decreases with increasing of Reynolds number for all MHSs. For instance, based on Celsius degree, as Reynolds number enhances from 100 to 900, the base temperature of straight integral MHS decreases about 1.6%. At the same enhancement in Reynolds number, a maximum decrement of 4.4% is observed for the wavy interrupted-inline MHS. It clarifies the effect of corrugated shape and interrupted arrangement of pin-fins on the base temperature of MHSs particularly at higher Reynolds numbers. o Generally, the base temperature values of wavy MHSs are lower than those of straight MHSs. The base temperature of wavy integral MHS is averagely 1.6% lower than that of straight integral MHS, it of wavy integral-interrupted MHS is averagely 2.3% lower than that of straight integral-interrupted MHS, it of wavy interrupted-inline MHS is averagely 3.4% lower than that of straight interruptedinline MHS, and it of wavy interrupted-staggered MHS is averagely 4.1% lower than that of straight interrupted-staggered MHS. o The interrupted MHSs propose lower values of base temperature compared with the integral MHSs. The integral-interrupted, interrupted-inline, and interruptedstaggered arrangements cause averagely 1.2%-1.9%-2.3% and 1.9%-2.6%-3.2% decrements in the base temperature of straight and wavy MHSs compared with the integral arrangement.

19

o The staggered arrangement of pin-fins results to a considerable improvement in the base temperature of both the straight and the wavy MHSs. The base temperature of interrupted-staggered straight MHS is averagely 5.3% lower than that of interrupted-inline straight MHS and the base temperature of interrupted-staggered wavy MHS is averagely 7.6% lower than that of interrupted-inline wavy MHS. Please insert Fig. 9 here For a better insight, Fig. 10 compares the temperature contours on the base surface of straight and wavy MHSs (in plane of y = 4 mm) at Reynolds number of 500. Due to the periodical nature of hydrothermal behavior in the MHSs, only the temperature contours corresponding to the simulated part are displayed in the figure. It is clear that there is a significant difference between the temperature contours of straight and wavy MHSs. The highest temperature zones are located around the pin-fins region, and utilizing of the pin-fin interruption leads to a substantial decrease in the temperature of base surface. In other words, the temperature of fluid (water) in the vicinity of integral pin-fins is considerably higher than that of fluid at the center zone. Therefore, the temperature of integral MHSs increases significantly to sustain heat transfer from the pin-fins to the high temperature water flow within the thermal boundary layer. On the other hand, in the interrupted MHSs the thermal and hydraulic boundary layers are repeatedly broken over the interrupted pin-fins leading to a considerable heat transfer. The values of outlet temperature for the studied Reynolds number are also presented in Fig. 10; note that the temperature at the outlet of contours is not uniform due to the elimination of exit portion in the presented contours. These values are in excellent agreement with the experimental data. It can be seen that the outlet temperature values of wavy MHSs are higher than those of straight MHSs and the temperature distribution at the outlet is more uniform for the wavy MHSs.

20

Obviously, as depicted in Fig. 12, swirl flows generated by corrugations enhance the fluid mixing in the wavy MHSs compared to the straight MHSs (see Fig. 12). It causes a heavy exchange between the cold fluid in the core part and the hot fluid near the pin-fins thereby higher heat transfer coefficient values (see Fig. 11). This means that the corrugation on pin-fins have a significant influence on the temperature field of MHS, especially for the interrupted arrangements. Please insert Fig. 10 here The effects of pin-fin interruption on the heat transfer coefficient of different MHSs are illustrated in Fig. 11. As Reynolds number is increased, the heat transfer coefficient of MHSs begins to improve. As depicted in Fig. 12, the mean fluid temperature in the MHSs decreases and the thermal boundary layer becomes thinner with increasing of Reynolds number, which enhance the heat transfer. Indeed, for lower Reynolds number, the heat transfer performance of MHSs is poor. It can deteriorate the micro electromechanical system due to the formation of hot spots. It is clear from Fig. 12 that at higher Reynolds number, the presence of inline and staggered interrupted pin-fins can induce recirculation flows in both the straight and the wavy MHSs. Although at lower Reynolds number a small recirculation flow is generated behind the interrupted pin-fins, the main flow velocity is not enough to wash away the vortex flow. On the other hand, from Fig. 11 it can be seen that for a given Reynolds number the pin-fin interruption intensifies the heat transfer coefficient in the MHSs, particularly for the wavy MHS. The interrupted pin-fins disturb and redevelop periodically thermal boundary layer and enlarge the local heat transfer area across the MHSs, while the fluid proceeds through the integral MHSs without any disturbance, slowly developing a velocity profile. Please insert Fig. 11 here Please insert Fig. 12 here

21

In order to have a better understand from the existing fluid flow and heat transfer phenomena in the interrupted pin-fins, the pressure, velocity, and temperature contours at a zoom view in the wavy interrupted-staggered MHS are demonstrated in Fig. 13 as a typical case. Note that in the convection heat transfer, the temperature and velocity fields are coupled. As the mainstream reaches to an interrupted pin-fin, it experiences a restriction due to the decreasing channel cross sectional area and this causes a local increase in the pressure at the frontal of interrupted pin-fin (Region 1). The pressure in the side channels of the interrupted pin-fin shows a local decrease due to the increase in the channel cross sectional area (Region 2). Being driven by this pressure difference, some recirculation flows are generated in the channel (Region 3). It can also be seen that the fluid changes its direction after impinges to the interrupted pin-fin. A main part of impinging fluid turns 90° in the y-z plane and then is pressed into the two side gaps between the two other interrupted pin-fins (Region 4). Due to the small width of the gaps, the thermal boundary layer close to the right and left wall is reduced (Region 5). When the fluid enters the back region of interrupted pin-fin, flow velocity dramatically decreases due to abrupt expansion of flow area and a transverse velocity component is produced simultaneously (Region 6). Subsequently, a recirculation flow is produced behind the pin-fins (Region 7). As depicted in the temperature contour, it mixes the fluid and interrupts the thermal boundary layer, which can produce a higher energy dissipation rate in the MHS (Region 8). Please insert Fig. 13 here Fig. 14 shows the enhancement in Nusselt number for the interrupted MHSs compared to the integral MHSs. It clarifies the significant influence of pin-fin interruption on the thermal performance of both the straight and the wavy MHSs. The results show that the inline interruption of pin-fins has the highest enhancement values in the wavy MHS at lower Reynolds numbers,

22

whereas at higher Reynolds numbers, the staggered interruption of pin-fins proposes the highest enhancement values. A similar discussion can be made for the straight MHS. Please insert Fig. 14 here In addition to the thermal performance, the hydraulic characteristic is another important factor for the application of MHSs. Hence, efforts should be made to reduce the pumping power and improve the efficiency of energy utilization in the MHSs. We expect that the pin-fin interruption diminishes the pressure drop or pumping power in the MHSs. Fig. 15 displays the comparison of pressure drop values in different MHSs obtained from the experiments. It is seen that the pressure drop increases rapidly with growing of Reynolds number, and that of the wavy MHSs is larger than the straight MHSs. It is worth to state that the pin-fin interruption reduces considerably the pressure drop in both the MHS types; the integral arrangement of pin-fins shows the largest values of pressure drop, while the interrupted arrangements present lower values. At the studied range, the pressure drop is the lowest for the straight interrupted-inline MHS. As it can be seen in Figs. 2 and 4, this configuration of MHS has the minimum blocking effects thereby the coolant proceeds easily along the pin-fins gaps. It can be concluded that always, the large pressure drop in the MHSs cannot be balanced by the heat transfer enhancement because the current results show that the integral pin-fins in the MHSs result in a small thermal enhancement compared with the interrupted pin-fins, which have lower pressure drop values (see Fig. 14). Please insert Fig. 15 here As shown in Fig. 16, utilizing the interrupted pin-fins instead of the integral pin-fins reduces drastically the pumping power for both the MHS types. For instance, at the same operating condition, the use of integral-interrupted pin-fins in the wavy MHS causes approximately 50% less power consumption of pumping compared with the integral pin-fins.

23

Please insert Fig. 16 here The hydrothermal performance factor, defined as Eq. (25), is considered to appraise the effectiveness of interrupted MHSs compared with that of integral MHSs.



 Nu f

Interrupted

Interrupted

Nu Integral 

(25)

f Integral 

13

Fig. 17 plots the hydrothermal performance curves of different interrupted MHSs compared to the straight and wavy integral MHSs as a function of Reynolds number. It is shown that, at a given Reynolds number, the performance factor of the interrupted MHSs is 1.2–2.3 times of that of the integral MHSs and mostly, the enhancement increases as Reynolds number is increased. It means an efficiency improvement of energy utilization in the MHSs equipped with the interrupted pin-fins. Among the interrupted MHSs, the straight integral-interrupted MHS shows the lowest values of performance factor, and for Reynolds number ranging from 300 to 700, the highest values are obtained for the wavy interrupted-staggered MHS. In the studied range of Reynolds number, there are averagely 22.1%, 67.7%, and 85.8% enhancements in the performance factor, respectively, for the straight integral-interrupted, straight interrupted-inline, and straight interrupted-staggered MHSs comparing with the straight integral MHS. These enhancements are averagely 53.7%, 76.8%, and 96.8% for the wavy integral-interrupted, wavy interrupted-inline, and wavy interrupted-staggered MHSs comparing with the wavy integral MHS. Based on the current evaluation, it can be concluded that the pin-fin interruption has more effects on the hydrothermal performance of wavy MHSs. Please insert Fig. 17 here

24

Recently, different modifications were proposed to improve the hydrothermal performance of MHSs. A comparison of the Nusselt number and friction factor between the obtained results in the present work and those from the previous works [40–43] is shown in Fig. 18(a–b). Please insert Fig. 18 here 4.3. Effect of Al2O3/water nanofluid Fig. 19(a–b) displays variations of the heat transfer coefficient and pressure drop for the nanofluid-cooled straight integral MHS in different Al2O3 nanoparticle mass fractions and Reynolds numbers. In order to save the space, the plots corresponding to the other straight and wavy MHSs are not presented and the average variations compared with the base fluid are tabulated in Table 5. The measurement results show that the heat transfer coefficient and pressure drop increase with an increase in nanoparticle mass fraction and Reynolds number. The enhancement in the heat transfer coefficient with the nanoparticle concentration is due to higher nanoparticle participation in the nanofluid and effective thermal conductivity modification. However, it seems that in such low concentrations, the enhancement of the heat transfer coefficient could not be only attributed to the increase of the thermal conductivity. As an example, at the highest Reynolds number in the straight integral MHS, the enhancement in the heat transfer coefficient of the 0.4 wt.% nanofluid is approximately 21%, whereas the measurement of thermophysical properties data shows about 11% enhancement in the thermal conductivity (please see Fig. 4). Also, the augmentation of pressure drop is because of the fact that suspending solid particles in a fluid generally increases the dynamic viscosity relative to the base fluid. Since, the dynamic viscosity is in direct relation with the pressure drop, the higher value of viscosity for nanofluids leads to increase the amount of pressure drop. Please insert Fig. 19 here

25

Please insert Table 5 here Fig. 20 shows the effect of particle loading (or nanoparticle mass fraction) on the trend of heat transfer coefficient enhancement for the straight integral MHS. Obviously, as the particle loading in the base fluid goes up, the enhancement ratio increases with almost linear slope. Assuming variation of the temperature as linear in the thermal boundary layer, the local heat transfer coefficient can be approximately given as κf/δt, that κf is the thermal conductivity of fluid and δt is the thickness of thermal boundary layer [44]. This ratio indicates that two parameters, i.e. κf and δt, affect the convective heat transfer coefficient. Both the increase of κf and/or the decrease of δt enhance the heat transfer coefficient. As depicted in the Fig. 4, the nanofluids have higher thermal conductivity compared to the base fluid. Please insert Fig. 20 here Our experimental measurements illustrate that the specific heat of nanofluid is lower than that of its base fluid [45–47]. Hence, at a constant heat flux boundary condition, nanofluid will decrease the temperature difference between the wall of MHSs and the coolant because of higher thermal conductivity. At the same time, it will increase the local mean temperature relative to that associated with use of the base fluid, because of lower specific heat. Because of these competing effects, the net benefit in terms of reduced base temperature associated with the use of nanofluid is unknown [48]. However, our measured results demonstrate that the presence of Al2O3 nanoparticles in water slightly decreases the base temperature of MHSs. For example, the 0.1 and 0.4 % wt. nanofluids averagely reduce the base temperature of straight integral MHS about 2.8% and 3.3% compared to that of water. Al2O3 nanoparticles collide with the channel wall of MHS and absorb the energy, then mix back to the core of fluid flow. Probably, this phenomenon is the reason which reduces the base temperature of MHS.

26

Also, to have the Al2O3/water nanofluid as a successful working fluid in practical applications of the MHSs, the variations of considered hydrothermal performance factor against to Reynolds number for different straight and wavy nanofluid-cooled MHS are presented in Fig. 21(a–b). Note that water flow in the integral MHSs is considered as the base line. It is detected that at the studied ranges, the nanofluids have greater values of performance factor compared to those associated with water as base fluid. The maximum performance factors of 2.65 and 2.46 are found, respectively, for the 0.4% nanofluid in the straight and wavy MHSs with the interruptedstaggered arrangement of pin-fins. Generally, it can be concluded that the effect of nanofluid flow on the hydrothermal performance of wavy MHSs is higher than that of straight MHSs. Please insert Fig. 21 here 5. Conclusion Heat transfer and pressure drop characteristics of Al2O3/water nanofluid flow inside the straight and wavy miniature heat sinks (MHSs) with integral and interrupted pin-fins are studied both numerically and experimentally. The numerical simulation is shown to predict the current experimental results well; up to 5% for the heat transfer and 9% for the pressure drop measurements. The results indicate that the influences of pin-fin interruption on the hydrothermal performance of MHSs are significant, and the interrupted MHSs give better thermal performance compared to the integral MHSs. For the Reynolds number ranging from 100 to 900, the interrupted straight MHSs show the hydrothermal performance factor 1.19–2.32 times larger than the integral straight MHSs, while those with the wavy MHSs show 1.23–2.28 times higher. Based on the numerical simulation, the temperature contours and velocity vectors under the same Reynolds number are also investigated. It is shown that the interrupted pin-fins can create recirculation flows, which leads to flow mixing and thus more uniform temperature distribution along the

27

MHSs. The experimental results show that Al2O3 nanoparticles dispersed into the based fluid, i.e. water, enhance both the heat transfer coefficient and the pressure drop of MHSs. Amount of augmentations increase with increasing nanoparticles mass fraction. The results show that the heat transfer coefficient for nanofluids is 2.2–28.3% more than the base fluid, although the pressure drop is 1.3–12.7% more, depending on the Reynolds number within its range. It can be concluded that applying the interrupted pin-fins instead of the integral pin-fins is a more effective way to enhance the heat transfer compared to the second method, which is using the nanofluid instead of the base liquid. The maximum enhancement in the heat transfer coefficient for the first technique is evaluated about 1.77. However, the corresponding value for the second technique is computed 1.28. Finally, the current study can provide the great values to select the optimum pin-fin configuration for use in the MHSs. Acknowledgements Acknowledgment is given to the Islamic Azad University (IAU) of Shahrood Branch for the supports through the setup fabrication and research implementation. Nomenclature Ac

minimum free flow area (m2)

At

active heat transfer area (m2)

cp

specific heat (J kg-1 K-1)

Dh

hydraulic diameter (m)

f

friction factor

G

mass velocity (kg m-2 s-1)

h

heat transfer coefficient (W m-2 K-1)

L

length of MHS (m)

28

m

mass flow rate (kg s-1)

Nu

Nusselt number

Q

heat transfer rate (W)

q

heat flux (W m-2)

Re

Reynolds number

∆p

pressure drop (Pa)

PP

pumping power (W)

s

distance between thermocouples location and MHS (m)

T

temperature (K)

V or u

velocity (m s-1)

Greek symbols ρ

density (kg m-3)

μ

dynamic viscosity (Pa s)

κ

thermal conductivity (W m-1 K-1)

φ

nanoparticle fraction

η

hydrothermal performance factor

Subscripts cond

conduction

conv

convection

f

fluid

in

inlet

m

mean

nf

nanofluid

29

out

outlet

p

particles

w

wall

s

solid

Acronyms MHS

miniature heat sink

SEM

scanning electron microscope

TEM

transmission electron microscopic

XRD

X-ray diffraction

References [1]

D.B. Tuckerman, R.F. Pease, High performance heat sinking for VLSI, IEEE Electron Device Letters 15 (5) (1981) 126–130.

[2]

S.A. Jajja, W. Ali, H.M. Ali, A.M. Ali, Water cooled minichannel heat sinks for microprocessor cooling: Effect of fin spacing, Applied Thermal Engineering 64 (1–2) (2014) 76–82.

[3]

S.A. Jajja, W. Ali, H.M. Ali, Multiwalled carbon nanotube nanofluid for thermal management of high heat generating computer processor, Heat Transfer–Asian Research 43 (7) (2014) 653–666.

[4]

P. Nitiapiruk, O. Mahian, A.S. Dalkilic, S. Wongwises, Performance characteristics of a microchannel heat sink using TiO2/water nanofluid and different thermophysical models, International Communications in Heat and Mass Transfer 47 (2013) 98–104.

30

[5]

W. Arshad, H.M. Ali, Graphene nanoplatelets nanofluids thermal and hydrodynamic performance on integral fin heat sink, International Journal of Heat and Mass Transfer 107 (2017) 995–1001.

[6]

W. Arshad, H.M. Ali, Experimental investigation of heat transfer and pressure drop in a straight minichannel heat sink using TiO2 nanofluid, International Journal of Heat and Mass Transfer 110 (2017) 248–256.

[7]

M.M. Tafarroj, O. Mahian, A. Kasaeian, K. Sakamatapan, A.S. Dalkilic, S. Wongwises, Artificial neural network modeling of nanofluid flow in a microchannel heat sink using experimental data, International Communications in Heat and Mass Transfer 86 (2017) 25– 31.

[8]

O. Mahian, A. Kianifar, S.A. Kalogirou, I. Pop, S. Wongwises, A review of the applications of nanofluids in solar energy, International Journal of Heat and Mass Transfer 57 (2) (2013) 582–594.

[9]

A. Kasaeian, R. Daneshazarian, O. Mahian, L. Kolsi, A.J. Chamkha, S. Wongwises, I. Pop, Nanofluid flow and heat transfer in porous media: A review of the latest developments, International Journal of Heat and Mass Transfer 107 (2017) 778–791.

[10]

M. Gupta, V. Singh, R. Kumar, Z. Said, A review on thermophysical properties of nanofluids and heat transfer applications, Renewable and Sustainable Energy Reviews 74 (2017) 638–670.

[11]

N.A.C. Sidik, I.M. Adamu, M.M. Jamil, G.H.R. Kefayati, R. Mamat, G. Najafi, Recent progress on hybrid nanofluids in heat transfer applications: A comprehensive review, International Communications in Heat and Mass Transfer 78 (2016) 68–79.

31

[12]

H.R. Seyf, M. Feizbakhshi, Computational analysis of nanofluid effects on convective heat transfer enhancement of micro-pin-fin heat sinks, International Journal of Thermal Sciences 58 (2012) 168–179.

[13]

E.M. Tokit, H.A. Mohammed, M.Z. Yusoff, Thermal performance of optimized interrupted

microchannel

heat

sink

(IMCHS)

using

nanofluids,

International

Communications in Heat and Mass Transfer 39 (10) (2012) 1595–1604. [14]

N.R. Kuppusamy, H.A. Mohammed, C.W. Lim, Numerical investigation of trapezoidal grooved microchannel heat sink using nanofluids, Thermochimica Acta 573 (2013) 39–56.

[15]

P. Selvakumar, S. Suresh, S. Salyan, Investigations of effect of radial flow impeller type swirl generator fitted in an electronic heat sink and Al2O3/water nanofluid on heat transfer enhancement, Chemical Engineering and Processing: Process Intensification 72 (2013) 103–112.

[16]

M.I. Hasan, Investigation of flow and heat transfer characteristics in micro pin fin heat sink with nanofluid, Applied Thermal Engineering 63 (2) (2014) 598–607.

[17]

Y.L. Zhai, G.D. Xia, , X.F. Liu, Y.F. Li, Heat transfer enhancement of Al2O3-H2O nanofluids flowing through a micro heat sink with complex structure, International Communications in Heat and Mass Transfer 66 (2015) 158–166.

[18]

A. Sakanova, C.C. Keian, J. Zhao, Performance improvements of microchannel heat sink using wavy channel and nanofluids, International Journal of Heat and Mass Transfer 89 (2015) 59–74.

[19]

W. Duangthongsuk, S. Wongwises, An experimental study on the thermal and hydraulic performances of nanofluids flow in a miniature circular pin fin heat sink, Experimental Thermal and Fluid Science 66 (2015) 28–35.

32

[20]

W. Duangthongsuk, S. Wongwises, A comparison of the heat transfer performance and pressure drop of nanofluid-cooled heat sinks with different miniature pin fin configurations, Experimental Thermal and Fluid Science 69 (2015) 111–118.

[21]

Z.Y. Ghale, M. Haghshenasfard, M.N. Esfahany, Investigation of nanofluids heat transfer in a ribbed microchannel heat sink using single-phase and multiphase CFD models, International Communications in Heat and Mass Transfer 68 (2015) 122–129.

[22]

H.M. Ali, W. Arshad, Thermal performance investigation of staggered and inline pin fin heat sinks using water based rutile and anatase TiO2 nanofluids, Energy Conversion and Management 106 (2015) 793–803.

[23]

H.E. Ahmed, M.I. Ahmed, I.M.F. Seder, B.H. Salman, Experimental investigation for sequential triangular double-layered microchannel heat sink with nanofluids, International Communications in Heat and Mass Transfer 77 (2016) 104–115.

[24]

M. Khoshvaght-Aliabadi, M. Sahamiyan, Performance of nanofluid flow in corrugated minichannels heat sink (CMCHS), Energy Conversion and Management 108 (2016) 297– 308.

[25]

M. Khoshvaght-Aliabadi, E. Ahmadian, O. Sartipzadeh, Effects of different pin-fin interruptions on performance of a nanofluid-cooled zigzag miniature heat sink (MHS), International Communications in Heat and Mass Transfer 81 (2017) 19–27.

[26]

M. Khoshvaght-Aliabadi, S.M. Hassani, S.H. Mazloumi, Comparison of hydrothermal performance between plate fins and plate-pin fins subject to nanofluid-cooled corrugated miniature heat sinks, Microelectronics Reliability 70 (2017) 84–96.

33

[27]

H.M. Ali, W. Arshad, Effect of channel angle of pin-fin heat sink on heat transfer performance using water based graphene nanoplatelets nanofluids, International Journal of Heat and Mass Transfer 106 (2017) 465–472.

[28]

M. Khoshvaght-Aliabadi, O. Sartipzadeh, S. Pazdar, M. Sahamiyan, Experimental and parametric studies on a miniature heat sink with offset-strip pins and Al2O3/water nanofluids, Applied Thermal Engineering 111 (2017) 1342–1352.

[29]

M. Khoshvaght-Aliabadi, M. Sahamiyan, M. Hesampour, O. Sartipzadeh, Experimental study on cooling performance of sinusoidal–wavy minichannel heat sink, Applied Thermal Engineering 92 (2016) 50–61.

[30]

M. Khoshvaght-Aliabadi, F. Nozan, Water cooled corrugated minichannel heat sink for electronic devices: Effect of corrugation shape, International Communications in Heat and Mass Transfer 76 (2016) 188–196.

[31]

D. Zhu, X. Li, N. Wang, X. Wang, J. Gao, H. Li, Dispersion behavior and thermal conductivity characteristics of Al2O3–H2O nanofluids, Current Applied Physics 9 (2009) 131–139.

[32]

O. Mahian, A. Kianifar, S. Zeinali Heris, D. Wen, A.Z. Sahin, S. Wongwises, Nanofluids effects on the evaporation rate in a solar still equipped with a heat exchanger, Nano Energy 36 (2017) 134–155.

[33]

M. Khoshvaght-Aliabadi, M. Eskandari, Influence of twist length variations on thermal– hydraulic specifications of twisted-tape inserts in presence of Cu–water nanofluid, Experimental Thermal and Fluid Science 61 (2015) 230–240.

34

[34]

M. Khoshvaght-Aliabadi, A. Alizadeh, An experimental study of Cu–water nanofluid flow inside serpentine tubes with variable straight-section lengths, Experimental Thermal and Fluid Science 61 (2015) 1–11.

[35]

H.W. Coleman, W.G. Steele, Experimentation, validation, and uncertainty analysis for engineers, John Wiley & Sons, New York, 2009.

[36]

L. Chai, G.D. Xia, J. Qi. Experimental and numerical study of flow and heat transfer in trapezoidal microchannels, Heat Transfer Engineering 33 (11) (2012) 972–981.

[37]

R.J. Phillips, L.R. Glicksman, R. Larson, Forced-convection, liquid-cooled, minichannels heat sinks. US patent 4894709[P], 1990.

[38]

J.P. Van Doormaal, G.D. Raithby, Enhancements of the SIMPLE method for predicting incompressible fluid flows, Numerical Heat Transfer 7 (2) (1984) 147–163.

[39]

C.J. Ho, W.C. Chen, An experimental study on thermal performance of Al2O3/water nanofluid in a minichannel heat sink, Applied Thermal Engineering 50 (2013) 516–522.

[40]

L. Chai., G. Xia, M. Zhou, J. Li, J. Qi, Optimum thermal design of interrupted microchannel heat sink with rectangular ribs in the transverse microchambers, Applied Thermal Engineering 51 (2013) 880–889.

[41]

G. Xia., Y. Zhai, Z. Cui, Numerical investigation of thermal enhancement in a micro heat sink with fan-shaped reentrant cavities and internal ribs, Applied Thermal Engineering 58 (2013) 52–60.

[42]

Y.F. Li, G.D. Xia, D.D. Ma, Y.T. Jia, J. Wang, Characteristics of laminar flow and heat transfer in microchannel heat sink with triangular cavities and rectangular ribs, International Journal of Heat and Mass Transfer 98 (2016) 17–28.

35

[43]

I.A. Ghani, N. Kamaruzaman, N.A.C. Sidik, Heat transfer augmentation in a microchannel heat sink with sinusoidal cavities and rectangular ribs, International Journal of Heat and Mass Transfer 108 (2017) 1969–1981.

[44]

Holman JP (1995) Heat and Mass Transfer, fifth ed., Mcgraw Hill, New York.

[45]

M. Khoshvaght-Aliabadi, F. Hormozi, A. Zamzamian, Experimental analysis of thermal– hydraulic performance of copper–water nanofluid flow in different plate fin channels, Experimental Thermal and Fluid Science 52 (2014) 248–258.

[46]

M. Khoshvaght-Aliabadi, F. Hormozi, Heat transfer of Cu–water nanofluid in parallel, corrugated, and strip channels, Journal of Thermophysics and Heat Transfer 29 (4) (2015) 747–756.

[47]

M. Khoshvaght-Aliabadi, S. Pazdar, O. Sartipzadeh, Experimental investigation of water based nanofluid containing copper nanoparticles across helical microtubes, International Communications in Heat and Mass Transfer 70 (2016) 84–92.

[48]

T.L. Bergman, Effect of reduced specific heats of nanofluids on single phase, laminar internal forced convection, International Journal of Heat and Mass Transfer 52 (2009) 1240–1244.

Table 1. Specification, range, and accuracy of test setup components. Table 2. Details of geometrical parameters. Table 3. Specifications of initial phases. Table 4. Total surface area and hydraulic diameter of MHSs. Table 5. Variations of heat transfer coefficient and pressure drop with nanofluid concentration for different MHSs (%).

36

Caption of figures Fig. 1. Applied test setup. Fig. 2. Considered configurations for water-cooled straight and wavy MHSs (a) graphics (b) schematic. Fig. 3. (a) XRD (b) SEM (c) TEM images of α-Al2O3 nanoparticles. Fig. 4. Variations of thermal conductivity and dynamic viscosity with nanoparticle mass fraction. Fig. 5. Top view of studied physical models in numerical simulation (red dashed lines show the considered parts). Fig. 6. Schematic of the computational zone. Fig. 7. Variations of heat transfer coefficient and pressure drop versus grid number. Fig. 8. Comparison between current results and Ho and Chen [39] data. Fig. 9. Effect of pin-fin interruption on base temperature of straight and wavy water-cooled MHSs. Fig. 10. Effect of pin-fin interruption on temperature distribution of straight and wavy water-cooled MHSs.

37

Fig. 11. Effect of pin-fin interruption on heat transfer coefficient of straight and wavy water-cooled MHSs. Fig. 12. Temperature contours and velocity vectors around interrupted pin-fins in straight and wavy MHSs at minimum and maximum Reynolds numbers. Fig. 13. Pressure, velocity, and temperature contours on a zoom view in wavy interrupted-staggered MHS at Reynolds number of 900. Fig. 14. Nusselt number enhancement for interrupted MHSs compared with integral MHSs. Fig. 15. Effect of pin-fin interruption on pressure drop of straight and wavy water-cooled MHSs. Fig. 16. Pumping power reduction for interrupted MHSs compared with integral MHSs. Fig. 17. Effect of pin-fin interruption on hydrothermal performance factor of straight and wavy water-cooled MHSs. Fig. 18. Comparison of present results and previous data (a) Nusselt number (b) friction factor. Fig. 19. Effect of nanofluid flow on heat transfer coefficient and pressure drop of straight integral MHS. Fig. 20. Effect of particle loading on trend of heat transfer enhancement for straight integral MHS. Fig. 21. Effect of nanofluid flow on hydrothermal performance factor of (a) straight MHSs (b) wavy MHSs.

38

‘ Bulk Thermocouple Base Thermocouple Pressure transmitter

P

Pressure transmitter indicator

T

Bulk/Surface Thermocouple indicator

P-1

T-3

P-2

Selector switch Radiator + Fan Three way

T1 T2 T3 T4 T5 T6 T7 T8 T9 Ball valve

Reservoir

Cartridge heaters

Needle valve

Digital timer

T-1

T-2

Pump Digital multifunction instrument

Variac variable transformer

Electrical valve

Flow line Input signal Output signal Digital balance

Fig. 1. Applied test setup.

39

(a)

aig Str

v Wa

y

ht

eg int

ral

M

HS

h aig Str

eg int

ral

t in

te

l-in gra

Wa

teg

MH

S

aig Str

S MH in vy

ru ter

d pte

ra

ru ter l- i n

dM pte

ht

pte rru e t in

n d-i

Wa

ru ter

HS

a Str

HS

in vy

M line

d- s pte

tag

g

d ere

MH

t igh

u err int

d pte

a -st

gg

d ere

S

vy Wa

u err int

d pte

ag - st

g

d ere

MH

MH

S

S

(b) Fig. 2. Considered configurations for water-cooled straight and wavy MHSs (a) graphical (b) schematic.

40

XRD

(a)

SEM (b)

41

TEM (c) Fig. 3. (a) XRD (b) SEM (c) TEM images of α-Al2O3 nanoparticles.

42

Fig. 4. Variations of thermal conductivity and dynamic viscosity with nanoparticle mass fraction.

43

100 mm

4 mm

20 mm

2.5 mm

Straight integral MHS

Straight integral-interrupted MHS

Straight interrupted-inline MHS

Straight interrupted-staggered MHS

Wavy integral MHS

Wavy integral-interrupted MHS

Wavy interrupted-inline MHS

Wavy interrupted-staggered MHS

Fig. 5. Top view of studied physical models in numerical simulation (red dashed lines show considered parts).

44

X = 500 mm

Wa te

X = 300 mm

it Ex

on r ti po

y=

l na igi Or

on r ti po

ri Pe

od

ic

X = 200 mm

r in

tic

He

at

Normal, y

,z

Ad

a iab

i se

Wa te

t En

on r ti po

ic

nw

X=0

ce ran

od

Sp a

ri Pe

ea Str

mw

i se

,x

flu x

let

Z = 4 mm Z=0

Fig. 6. Schematic of the computational zone.

45

0

ia Ad

ti ba

c

ro

y=

utl

et

5m

m

Fig. 7. Variations of heat transfer coefficient and pressure drop versus grid number.

46

Fig. 8. Comparison between current results and Ho and Chen [39] data.

47

Straight integral MHS

Wavy integral MHS

Straight integral-interrupted MHS

Wavy integral-interrupted MHS

Straight interrupted-inline MHS

Wavy interrupted-inline MHS

Straight interrupted-staggered MHS

Wavy interrupted-staggered MHS

Base temperature (K)

310 309 308 307 306 305 304 0

200

400

600

800

1000

Reynolds number Fig. 9. Effect of pin-fin interruption on base temperature of straight and wavy water-cooled MHSs.

48

Fig. 10. Effect of pin-fin interruption on temperature distribution of straight and wavy water-cooled MHSs.

49

Heat transfer coefficient (W m-2 K-1)

Straight integral MHS

Wavy integral MHS

Straight integral-interrupted MHS

Wavy integral-interrupted MHS

Straight interrupted-inline MHS

Wavy interrupted-inline MHS

Straight interrupted-staggered MHS

Wavy interrupted-staggered MHS

6000 5000 4000 3000 2000 1000 0 0

200

400

600

800

1000

Reynolds number Fig. 11. Effect of pin-fin interruption on heat transfer coefficient of straight and wavy water-cooled MHSs.

50

Fig. 12. Temperature contours and velocity vectors around interrupted pin-fins in straight and wavy MHSs at minimum and maximum Reynolds numbers.

51

Flow direction

Pressure contour

Region 1

Region 2

Velocity contour

Region 6

Region 3

Velocity vector

Region 7

Region 4

Temperature contour

Region 8

Temperature (K)

Velocity (m/s)

Pressure (Pa)

Region 5

Fig. 13. Pressure, velocity, and temperature contours on a zoom view in wavy interrupted-staggered MHS at Reynolds number of 900.

52

Straight integral-interrupted MHS

Wavy integral-interrupted MHS

Straight interrupted-inline MHS

Wavy interrupted-inline MHS

Straight interrupted-staggered MHS

Wavy interrupted-staggered MHS

Nusselt number enhancement

3

2.5

2

1.5

1 0

200

400

600

800

1000

Reynolds number Fig. 14. Nusselt number enhancement for interrupted MHSs compared with integral MHSs.

53

Straight integral MHS

Wavy integral MHS

Straight integral-interrupted MHS

Wavy integral-interrupted MHS

Straight interrupted-inline MHS

Wavy interrupted-inline MHS

Straight interrupted-staggered MHS

Wavy interrupted-staggered MHS

10000

Pressure drop (Pa)

8000 6000 4000 2000 0 0

200

400

600

800

1000

Reynolds number Fig. 15. Effect of pin-fin interruption on pressure drop of straight and wavy water-cooled MHSs.

54

Straight integral-interrupted MHS

Wavy integral-interrupted MHS

Straight interrupted-inline MHS

Wavy interrupted-inline MHS

Straight interrupted-staggered MHS

Wavy interrupted-staggered MHS

Pumping power reduction

0.6 0.5 0.4 0.3 0.2 0.1 0 0

200

400

600

800

1000

Reynolds number Fig. 16. Pumping power reduction for interrupted MHSs compared with integral MHSs.

55

Hydrothermal performance factor

Straight integral-interrupted MHS

Wavy integral-interrupted MHS

Straight interrupted-inline MHS

Wavy interrupted-inline MHS

Straight interrupted-staggered MHS

Wavy interrupted-staggered MHS

2.5

2

1.5

1 0

200

400

600

800

1000

Reynolds number Fig. 17. Effect of pin-fin interruption on hydrothermal performance factor of straight and wavy water-cooled MHSs.

56

Nusselt number

25

Straight interruptedstaggered MHS

20

Wavy interruptedstaggered MHS

15

Chai et al. Xia et al.

10

Li et al. 5 Ghani et al. 0 0

200

400

600

800

1000

Reynolds number

(a) 2.5

Straight interruptedstaggered MHS

Friction factor

2

Wavy interruptedstaggered MHS Chai et al.

1.5

Xia et al.

1

Li et al. 0.5 Ghani et al. 0 0

200

400

600

800

1000

Reynolds number

(b) Fig. 18. Comparison of present results and previous data (a) Nusselt number (b) friction factor.

57

Water-cooled straight integral MHS, h 0.1% nanofluid-cooled straight integral MHS, h 0.4% nanofluid-cooled straight integral MHS, h 0.1% nanofluid-cooled straight integral MHS, ∆p

3000

3000

0.4% nanofluid-cooled straight integral MHS, ∆p

2500

2500

2000

2000

1500

1500

1000

1000

500

500

0 0

200

400

600

800

Pressure drop (Pa)

Heat transfer coefficient (W m-2 K-1)

Water-cooled straight integral MHS, ∆p

0 1000

Reynolds number Fig. 19. Effect of nanofluid flow on heat transfer coefficient and pressure drop of straight integral MHS.

58

Fig. 20. Effect of particle loading on trend of heat transfer enhancement for straight integral MHS.

59

Hydrothermal performance factor

3

2.5

Integral-interrupted MHS, water Integral-interrupted MHS, 0.1% nanofluid Integral-interrupted MHS, 0.4% nanofluid Interrupted-inline MHS, water Interrupted-inline MHS, 0.1% nanofluid Interrupted-inline MHS, 0.4% nanofluid Interrupted-staggered MHS, water Interrupted-staggered MHS, 0.1% nanofluid Interrupted-staggered MHS, 0.4% nanofluid

2

1.5

1 100

300

500

700

900

700

900

Reynolds number

Hydrothermal performance factor

(a)

3

2.5

Integral-interrupted MHS, water Integral-interrupted MHS, 0.1% nanofluid Integral-interrupted MHS, 0.4% nanofluid Interrupted-inline MHS, water Interrupted-inline MHS, 0.1% nanofluid Interrupted-inline MHS, 0.4% nanofluid Interrupted-staggered MHS, water Interrupted-staggered MHS, 0.1% nanofluid Interrupted-staggered MHS, 0.4% nanofluid

2

1.5

1 100

300

500

Reynolds number (b) Fig. 21. Effect of nanofluid flow on hydrothermal performance factor of (a) straight MHSs (b) wavy MHSs.

60

61

Table 1. Specification, range, and accuracy of test setup components. Component Specification 1. Reservoir Stainless steel 2. Pump Gear 3. Bulk thermocouples PT-100 4. Surface thermocouples K-type 5. Pressure transmitters PTCHC060BCIA, Sensys 6. Electrical heater Cartridge 7. Digital multifunction instrument mfm 3430, Ziegler 8. Variac variable transformer 3PN1010B, DAM 9. Radiator + Fan Kooshesh 10. Electrical valve Festo 11. Digital timer ATE-10S, Autonics 12. Digital balance TE1502S, Sartorius 13. Bulk thermocouples indicators SU-105PRR, Samwon 14. Base thermocouples indicators SU-105KRR, Samwon 15. Pressure transmitters indicators MT4W, Autonics a Based on calibration process b Based on manufacturer claim

62

Range – – –50 to 200 °C –73 to 260 °C 0 to 60 kPa 0 to 125 W – – – – – – – – –

Accuracy – – 0.1 °Ca 0.1 °Ca 10 Pab – – – – – 0.1 sa 0.001 kgb – – –

Table 2. Details of geometrical parameters. Parameter Length of MHSs Width of MHSs Thickness of MHSs Length of integral pin-fins Length of interrupted pin-fins Width of integral/or/interrupted pin-fins Height of integral/or/interrupted pin-fins Width of miniature channel Deep of miniature channel

Value 100 mm 20 mm 5 mm 100 mm for straight MHS and 110 mm for wavy MHS 5 mm for straight MHS and 5.5 mm for wavy MHS 1 mm 1 mm 1 mm 1 mm

63

Table 3. Specifications of initial phases. Physical specifications of nanoparticles Purity Specific Average surface particle area size +99% +19 m2/g 20–40 nm Thermal-hydraulic specifications of nanoparticles and base fluid Density Specific heat (kg/m3) (J/kg.°C) Nanoparticles Base fluid

3950 998.2

773 4182

64

Color

Morphology

White

Spherical

Thermal conductivity (W/m.°C) 36 0.597

Dynamic viscosity (kg/m.s) – 9.98×10-4

Table 4. Total surface area and hydraulic diameter of MHSs. MHS Total surface area, At (mm2) Straight integral 2400 Straight integral-interrupted 1280 Straight interrupted-inline 1040 Straight interrupted-staggered 1040 Wavy integral 2640 Wavy integral-interrupted 1400 Wavy interrupted-inline 1130 Wavy interrupted-staggered 1130

65

Hydraulic diameter, Dh (mm) 1.33 2.50 3.08 3.08 1.21 2.28 2.83 2.83

Table 5. Variations of heat transfer coefficient and pressure drop with nanofluid concentration for different MHSs (%). MHS Parameter 0.1% nanofluid 0.4% nanofluid Straight 1. Integral 2. Integral-interrupted 3. Interrupted-inline 4. Interrupted-staggered

h ∆p h ∆p h ∆p h ∆p

4.9 2.3 6.1 2.9 3.3 1.6 7.2 1.7

19.2 11.4 23.5 12.7 18.1 10.1 28.3 9.3

h ∆p h ∆p h ∆p h ∆p

2.2 1.3 4.9 1.9 2.8 2.2 5.6 1.4

18.2 8.6 17.1 9.5 19.5 9.4 22.3 9.7

Wavy 1. Integral 2. Integral-interrupted 3. Interrupted-inline 4. Interrupted-staggered

66