Case Studies in Thermal Engineering 52 (2023) 103788

Contents lists available at ScienceDirect

Case Studies in Thermal Engineering

journal homepage: www.elsevier.com/locate/csite

Heat transfer enhancement and temperature uniformity

improvement of microchannel heat sinks with twisted
blade-like fins
Shiyang Chen, Zhenwei Liu, Boyuan Wang, Ping Li *
MOE Key Laboratory of Thermo–Fluid Science and Engineering, Xi’an Jiaotong University, Xi’an, Shaanxi, 710049, China

A R T I C L E I N F O A B S T R A C T

Keywords: To enhance heat and fluid exchange between main flow and near-wall flow in microchannels, a
Twisted blade-like fin twisted blade-like fin with an advantage in stimulating both spanwise and normalwise secondary
Secondary flow flow is proposed. The cross sections of the fin are low-drag airfoils in different orientations. The
Heat transfer enhancement flow and heat transfer performances of microchannels with fins are numerically investigated at
Wall temperature uniformity Re = 50–700. The results show that the twisted blade-like fin improves heat transfer process
Microchannel heat sinks
significantly, especially wall temperature uniformity, and reduces the flow separation region
behind the fin, resulting in an obvious small pressure penalty. At small inflow (Re = 50), the best
heat transfer and lowest pressure penalty are obtained in the microchannel with a single fin.
When Re > 150, better heat transfer is obtained in the microchannel with three fins, and the
highest comprehensive thermal performance (TP) reaches 3.58. The twisted direction of fins has a
significant effect on heat transfer but less on flow drag. Due to the twisted fins, left and right walls
experience an overall decline in temperature, and the temperature uniformity of top and bottom
walls is improved. Compared with a smooth microchannel, the average and maximum temper­
ature of the investigated microchannels are reduced by 48.1 K and 49.0 K at most respectively.

1. Introduction
With the development of Very Large Scale Integration (VLSI) technology, Micro-Electro-Mechanical System (MEMS) and related
miniaturization technologies, the package density and heat flow density of these microsystem devices increase significantly. Micro­
channel Heat Sink (MCHS) is an effective method to solve the heat dissipation problem of microsystems with high heat flow density
[1]. In MCHS, the specific surface area is sharply increased and the thermal resistance is sharply reduced, so the cooling capability is
significantly strengthened. At the same time, the pressure drop increases rapidly as the flow rate increases, especially after crossing the
laminar flow [2], so the work fluid in the microchannel is usually kept in laminar to reduce the pressure drop. Sharp et al. [3]
investigated the transition in a circular glass tube channel with a hydraulic diameter between 50 μm and 247 μm, and the critical
Reynolds number for the transition is found to be between 1800 and 2000.
Applying passive control methods to improve the performance of MCHS is an important topic, with the aim of achieving higher heat
dissipation at a smaller cost. In other words, a higher heat transfer improvement with good temperature uniformity at the cost of a
smaller increase in flow drag.
The improvement of heat transfer in channels is a vast research field, and researchers typically approach it from two perspectives:

* Corresponding author.
E-mail address: [email protected] (P. Li).

https://doi.org/10.1016/j.csite.2023.103788
Received 23 September 2023; Received in revised form 13 November 2023; Accepted 18 November 2023
Available online 21 November 2023
2214-157X/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
S. Chen et al. Case Studies in Thermal Engineering 52 (2023) 103788

one involves optimizing the layout of channel flow passages [4,5], while the other focuses on arranging flow control structures within
the channels [6,7], and fins are the most classic and widely used flow control structures. Arranging fins on the heated walls signifi­
cantly improves the heat transfer capacity and temperature uniformity of channels. Tang et al. [8] improved the uniformity of the
heating surface temperature with micro pin-fin. Zohora et al. [9] investigated the hydrothermal performance of novel pin-fin heat sinks
with hyperbolic, wavy, and crinkle geometries and various perforations. Chang et al. [10] investigated oblique fins heat sink, and
compared to continuous fin heat sink, the maximum temperature is reduced by 3 K and the heat transfer area can be reduced by 31.3 %.
The large flow drag is an important factor limiting the comprehensive thermal performance of finned radiators, and many studies have
been devoted to reducing their flow drag. Mohit et al. [11] investigated the effect of fin height on heat transfer and flow characteristics,
and the results showed that the heat transfer capacity and flow drag coefficient is both decreased with the decrease of fin height.
Studies [12,13] showed that porous fins are effective in reducing pressure drop and increasing heat transfer compared to solid fins, and
it was also shown that the flow drag of forward triangular fin is the lowest among rectangular, elliptical, isosceles triangular, inverted
triangular, and forward triangular fins. Acharya [14] investigated the shape effect of square fin and elliptical fins on the heat transfer
from the heater surface and frictional loss, and varied parameters such as Reynolds number, fin height and number of fins.
However, it is inevitable to form local high temperature on the leeward region of fins. Studies by Meinders et al. [15] and Sara et al.
[16] showed that for rectangular blocks in the mainstream, convective heat transfer on leeward region is the lowest due to the presence
of wake vortex. Sahin et al. [17,18] introduced jets into the leeward region by setting holes in the fins, in which the heat transfer
coefficient is enhanced and the friction coefficient is reduced. By changing the cylindrical pin fin to conical shape, Ahmadian-Elm et al.
[19] reduced the wake and improved the wall temperature uniformity. Li et al. [20] proposed endwall fillet structure, and the local
high temperature region at the leeward endwall of the needle fins is effectively eliminated. Researchers show that the heat transfer on

Fig. 1. Single fin and schematic diagram of periodic unit: (a) Single twisted airfoil-section fin, (b) 3D model of TAMCHS with periodic fins and its periodic unit, (c)
TAMCHS periodic unit.

2
S. Chen et al. Case Studies in Thermal Engineering 52 (2023) 103788

leeward region can be manipulated by destroying or reducing the wake vortex. The boundary layer separation point of streamlined
structures occurs backward and the wake region is thus small. In convection heat transfer, this characteristic of streamlined structures
may result in a small drag coefficient, and a slight local high temperature region on the leeward side. Among all streamlined structures,
the airfoil is designed under a very strict requirement of delaying flow separation, so its profile suppress the flow separation and reduce
the flow drag well. NACA0012 is a typical airfoil with low flow drag and simple symmetrical geometry [21], which is easy to
manufacture in engineering application.
MCHS also suffers from high temperature on sidewalls, which is detrimental to its whole temperature uniformity. To reduce the
high temperature on sidewalls, researchers have done their research to improve the heat transfer capacity by diverting the flow to the
sidewalls or increasing secondary flows. Li et al. [22] proposed split protrusion based on the sharkskin bionic concept. The fluid is
directed the fluid to the sidewalls by split structures and the spanwise secondary flow is enhanced, thus the temperature on the corner
region of the microchannel sidewalls and the leeward of the protrusion is effectively reduced, and a high comprehensive thermal
performance (3.07) is obtained. Lu et al. [23] declared that the vortex generators transport the coolant in the core region of the
microchannel to the sidewalls, and significantly improve the cooling effect over the hotspot with a relatively lower pressure loss
penalty. Wang et al. [24] proposed a new vortex generator composed of two delta winglets and a backward inclination angle, the new
generator enhances the heat transfer performance by enhancing the mixing strength of the hotter air near the wall and the colder air in
the core flow region. Shen et al. [25] introduced deflectors into the double-layer microchannel, which can promote the mixing of the
upper and lower parts of the fluid, thereby improving the overall thermal performance and significantly reducing the substrate
temperature. To better facilitate fluid mixing, twisted tapes are good options. Ali et al. [26] first investigated the application of twisted

Fig. 2. Schematic diagram of TAMCHS with different fin numbers and deflection directions.

3
S. Chen et al. Case Studies in Thermal Engineering 52 (2023) 103788

tapes in MCHS, the average temperature of the heating surface is decreased by 10.5 K, the Nusselt number is increased by 40 %, and the
pressure loss penalty is doubled. Their subsequent work [27] systematically investigated four factors affecting thermal performance in
microchannel with twisted tapes, and the average temperature is decreased by 16 K compared to SMCHS at most. Song et al. [28]
arranged twisted tapes in the channel and got an increase in the average Nusselt number from 24 % to 179 %, an increase in the flow
drag from 50 % to 250 %, and a maximum comprehensive thermal performance of 1.55. Feng et al. [29] investigated the compound
effect of rectangular coils and twisted tapes on heat transfer, and showed that the helical structure effectively promotes flow mixing, so
lower and better wall temperature uniformity is achieved.
Based on above studies, a twisted blade-like fin with different oriented airfoil cross section is proposed in this work. Then, the
effects of numbers and deflection directions of the fin on flow and heat transfer characteristics are investigated in Twisted Airfoil-
Section fin Microchannel Heat Sink (TAMCHS).

2. Numerical method and validation

2.1. Physical models and boundary conditions
The twisted blade-like fin with airfoil cross section is shown in Fig. 1(a). The airfoil cross section is applied in fin to reduce the wake
on leeward region and decrease the flow drag. Moreover, the fin uniformly grows from the bottom to the top with an appropriate
twisted angle to stimulate secondary flow and guide the flow to the sidewalls. The design methods of the proposed fin are as follows：

1. The length (l) of the fin is 50 μm, the height (h) is 30 μm, and along the height direction, each cross-section is the low drag airfoil
NACA0012.
2. Twist a certain angle between each adjacent section, and the center point (the midpoint of the airfoil’s mean camber line) of all
sections is located in the twisting centerline. The twist centerline is a straight line through the midpoint of the airfoil’s mean camber
line and perpendicular to the airfoil plane.
3. The middle section of the fin is directly facing the direction of incoming flow. Taking the middle section as reference, the section
uniformly twists 30◦ (φ) in total from the middle to the top, and -30◦ (-φ) from the middle to the bottom, so the whole fin twist 60◦
(2φ) in total from the bottom to the top.
The proposed fins can be periodically arranged in microchannels. Taking three fins per row as an example, the three-dimensional
model of TAMCHS is shown in Fig. 1(b). The model can be further extended along streamwise direction (X), spanwise direction (Y) and
normal direction (Z), respectively, and flexibly adjusted according to the heat transfer requirements of equipment. For TAMCHS, fully
developed periodic velocity and temperature regions are formed when the fluid flow across through the periodic arranged fins, and the
periodic region are decisive for the whole performance of TAMCHS. Therefore, in order to reduce computational resource con­
sumption, a periodic unit of TAMCHS is studied in this work, and the inlet and outlet of the unit are given periodic boundary conditions
[30,31].
As shown in Fig. 1(c), the dimensions of TAMCHS periodic unit are 200 μm (W) × 50 μm (H) × 150 μm (L). Working fluid flows into
the fully developed domain along the positive X direction with a bulk temperature of 300.0 K. Both the inlet and the outlet are
subjected to periodic boundary condition, and the inlet Reynolds number ranges from 50 to 700. Four walls of the microchannel and
the wall of twisted blade-like fins are the heat sources and are no-slip boundaries, with a heat flux of q" = 5 × 105 W m− 2 [22].
In the periodic unit, the effects of numbers and deflection directions of the twisted blade-like fin are investigated, and the schematic
of different models are shown in Fig. 2. TAMCHS 1S and TAMCHS 2S represent the microchannels with one fin and two fins. TAMCHS
3S-A, TAMCHS 3S–B and TAMCHS 3S–C represent the microchannels with three fins sharing the same deflection direction, the top
cross sections of side fins deflecting to center, and the top cross sections of side fins deflecting to the sidewalls, respectively.
Considering the goal of placing the two side fins as close to the sidewalls as possible and maintaining uniform fin spacing, and
controlling variables when investigating the individual heat transfer enhancement of the middle and side fins, two distances of 70 μm
and 140 μm are chosen. For the arrangement of three fins, one of which is arranged in the center of the microchannel, and the side two
are symmetrically arranged on both sides with a 70 μm distance (w) from the center one. For two fins, the two are symmetrically
arranged on both sides with a 140 μm distance (2w) to each other. For one fin, the fin is arranged in the center of the microchannel.
The dimension data are summarized in Table 1.

2.2. Governing equations

Numerical simulations of MCHS are generally based on the Navier-Stokes equations under continuous medium assumption. The
applicability of the continuous medium assumption at microscale needs to be verified, and the Knudsen number is usually applied to
estimate whether the continuous medium assumption holds [32]. In this study, the Knudsen number is less than 0.001, so the
continuous medium assumption holds.
Similar with the previous studies [22], the following assumptions are used: (1) steady state, (2) three dimensional incompressible,
(3) laminar flow and (4) constant fluid properties, the governing equations are as follows:

Table 1
Dimension data of TAMCHS.

W/μm H/μm L/μm l/μm h/μm 2φ/◦ w/μm

200 50 150 50 30 60 70

4
S. Chen et al. Case Studies in Thermal Engineering 52 (2023) 103788

Continuity equation

∇·→
u=0 (1)
Momentum equation

ρ(→
u · ∇)→
u = − ∇p + μ∇2 →
u (2)
Energy equation

ρCp (→
u · ∇T) = k∇2 T (3)

The pressure and velocity are coupled with SIMPLEC method. The pressure is discretized using standard scheme. The momentum
and energy equations are solved with second-order upwind scheme [22]. The solver is Fluent and the convergence of simulation is
judged by monitoring the residual of continuity, energy, velocities, and the temperature on special points, in which convergence
criterion is set as 1 × 10− 6.

2.3. Data reduction

In this study, Reynolds number (Re) is defined by
ρUm,in Dh
Re = (4)
μ

where Um,in is the average velocity of inlet, Dh is hydraulic diameter and defined by
2WH
Dh = (5)
W +H

where W is the width of the microchannel, H is the height of the microchannel.

The Nusselt number (Nu) is given by

hDh q″ Dh
Nu = = · (6)
k ΔT k

where k is the thermal conductivity of working fluid, h is the heat transfer coefficient, q" represents the heat flux, ΔT is the difference of
average wall temperature Tw,ave and average fluids temperature Tf,ave
ΔT = Tw,ave − Tf ,ave (7)

The Fanning friction factor (f) is defined as

(Δp/L)Dh
f= − 2
(8)
2ρUm,in

where ΔP is the pressure drop between inlet and outlet.

The comprehensive thermal performance (TP) is described as
( ) ( )− 1/3
Nu f
TP = · (9)
Nu0 f0
TP has been widely used as a comprehensive thermal performance evaluation index in similar studies [22,33,34]. This parameter
considers both heat transfer increase and pressure penalty, and depends on the ratios of Nu to Nu0 and f to f0, where Nu0, f0 are the
experimental results in smooth channels [35].
The wall temperature uniformity is evaluated by ΔTu
ΔTu = Tmax − Tmin (10)
where Tmax and Tmin are the maximum and minimum temperature on the walls [8,34,36].

Table 2
Grid independence validation at the conditions of TAMCHS 3S–C, q" = 5 × 105 W m− 2, Re = 700.

Elements Nu Difference% f⋅Re Difference%

Mesh 1 871,611 16.880 21.11 45.271 5.00

Mesh 2 2,009,638 16.332 17.17 44.634 3.51
Mesh 3 2,881,862 14.001 0.45 43.099 − 0.04
Mesh 4 4,766,764 13.938 – 43.117 –

5
S. Chen et al. Case Studies in Thermal Engineering 52 (2023) 103788

2.4. Model validation

Tetrahedral and hexahedral mixed grids are applied in this study. The simulation accuracy is ensured by fine grids near fins and
boundary layer grids near walls, and the computational resource is saved by a bit coarse hexahedral grids in other fluid domains. In
order to balance the simulation accuracy and efficiency, the grid independence validation is carried out at the case of TAMCHS 3S–C,
q" = 5 × 105 W m− 2, Re = 700. In the validation procedure, Nu and f are selected as evaluation criteria, starting from a coarse mesh and
refining it until these parameters are independent on the mesh size. As shown in Table 2, the relative differences of Nu and f in Mesh3
are 0.45 % and − 0.04 %, respectively. It is considered that Mesh3 has reached the reasonable standard of grid independence vali­
dation, so it is selected for the following studies.
To validate the simulation method, the flow and heat transfer of rectangular water-cooled SMCHS is simulated and compared with
referenced experimental results [35]. The width, height and length of SMCHS are 200 μm, 50 μm and 150 μm, respectively, and the
wall heat flux (q") is 5 × 105 W m− 2. The verification results are shown in Table 3. It can be seen that the largest difference is 1.43 %, so
the numerical simulation results show good agreement with the experimental results. Thus, the proposed simulation method in this
work is expected accurate enough for the analysis of flow and heat transfer in TAMCHS.

3. Results and discussions

3.1. Analysis of performance parameters
In this study, the relative Nusselt number (Nu/Nu0), relative Fanning friction coefficient (f/f0), and comprehensive thermal per­
formance (TP) of five models are investigated.
As can be seen from Fig. 3, the Nu/Nu0 of TAMCHS ranges from 0.77 to 4.76, and the minimum value of 0.77 is obtained in
TAMCHS 2S, Re = 50, and its heat transfer is the only deteriorating case among all. The reason is that the two fins arranged near the
sidewalls result in higher flow drag near the sidewalls, and the fluid tends to flow toward the center of the channel. While the vortices
generated near the fins are not so strong at small flow rates, and the additional heat brought by fins surface make the heat removal
capacity of TAMCHS 2S insufficient. As Re increases, the vortex intensity increases and then Nu/Nu0 increases, with TAMCHS 2S and
TAMCHS 1S rising approximately linearly, while TAMCHS 3S-A/B/C rising fast and then slow. When Re = 50, the best heat transfer
enhancement is achieved in TAMCHS 1S, but thereafter the Nu/Nu0 increases slightly from 1.33 to 1.74 as Re increases from 50 to 700.
By contrast, for TAMCHS 2S, the rise rate of Nu/Nu0 is greater than that of TAMCHS 1S and the Nu/Nu0 exceeds TAMCHS 1S when Re
= 250. For TAMCHS 3S-A/B/C, the heat transfer enhancement is higher than that of TAMCHS 1S and TAMCHS 2S (Re = 50 excluded).
There is also an obvious difference among them, it is highest in TAMCHS 3S–C, second in TAMCHS 3S-A, and lowest in TAMCHS 3S–B,
indicating that when arranging three fins, the arrangement in TAMCHS 3S–C is the most favorable for heat transfer enhancement.
Compared with other similar studies [22,26,33,37], there is an obvious advantage for heat transfer enhancement in TAMCHS. In
addition, in heat transfer process, the microchannel with fins not only has an enhanced convective heat transfer coefficient but also has
an increased heat transfer area compared to SMCHS, and in this study, the heat transfer area of TAMCHS 3S-A/B/C increases by 1/8.
For SMCHS, the flow drag is dominated by frictional drag, while in the microchannel with fins, the development of flow boundary
layer is disrupted, so the frictional drag decreases, but at the same time, the pressure drag increase significantly due to the flow
separation near the flow control structures [38]. In order to compare the flow drags, this study employs relative Fanning friction
coefficient f/f0 to show the flow drag of TAMCHS, in which f0 is the flow drag of SMCHS.
As shown in Fig. 4, the f/f0 of all five TAMCHS is greater than 1. This is because the introduction of fins inevitably increases the flow
drag. The f/f0 of TAMCHS in this work ranges from 1.14 to 2.39, and the drag coefficient increases with the increase of Re and fin
numbers. The f/f0 value and increase rate of TAMCHS 1S and TAMCHS 2S are significantly smaller than that of TAMCHS 3S-A/B/C. For
TAMCHS 3S-A/B/C, the drag is very close to each other, and the difference among them only begins to manifest when Re increase from
500 to 700. Compared to the similar studies [22,26,37], the drag coefficient of TAMCHS is lower, which attributes to the decrease in
flow separation region and windward region.
In this study, the heat transfer enhancement ratio is greater than the drag increase ratio at most cases, except TAMCHS at Re = 50.
In order to evaluate the heat transfer improvement and drag penalty comprehensively, the comprehensive thermal performance
evaluation index TP, which is widely used in similar studies, is adopted. As shown in Fig. 5, the variation trend of TP is generally similar
to Nu/Nu0, and the range of TP is 0.71–3.58, with the minimum value of 0.71 obtained in TAMCHS 2S (Re = 50), and the maximum
value of 3.58 obtained in TAMCHS 3S–C (Re = 700). The TP of TAMCHS 1S and TAMCHS 2S increases approximately linearly as the
increase of Re, but for TAMCHS 2S it increases faster and exceeds TAMCHS 1S when Re = 250. The TP rise rate of TAMCHS 3S-A/B/C is

Table 3
Comparison of numerical simulation results with reference values.

Re Nu Difference % f⋅Re Difference %

Referenced result [35] Simulation results Referenced result [35] Simulation results

50 2.940 2.982 1.43 18.233 18.057 0.96

150 2.955 0.51 18.075 0.87
250 2.949 0.31 18.094 0.76
350 2.947 0.24 18.113 0.66
500 2.945 0.17 18.142 0.50
700 2.944 0.14 18.180 0.29

6
S. Chen et al. Case Studies in Thermal Engineering 52 (2023) 103788

Fig. 3. Variation of Nu/Nu0 with Re in different TAMCHS.

Fig. 4. Variation of f/f0 with Re in different TAMCHS.

Fig. 5. Variation of TP with Re in different TAMCHS.

7
S. Chen et al. Case Studies in Thermal Engineering 52 (2023) 103788

large at small Re, but becomes smaller as Re increases, and when Re > 500, it is smaller than that of TAMCHS 2S. Compared with other
similar studies [22,33,34,37], the comprehensive thermal performance of TAMCHS is more significantly improved due to its higher
Nu/Nu0 and lower f/f0. It is important to emphasize that although the increase in the number of fins leads to an increase in heated area,
the outstanding heat transfer capacity of TAMCHS 3S-A/B/C in this study is not merely due to the increase of fin number. It benefits
more from the geometry features of the proposed fin and resulting secondary flow, and it is advantageous even compared to the
channels with vortex generator arrays [39] or pin arrays [40].

3.2. Temperature characteristic analysis

For chip cooling, the maximum temperature and average temperature are very important parameters because they are directly
related to the reliability and performance of chips [34]. Therefore, wall temperature uniformity is an important evaluation criterion of
MCHS, and researchers [8,34,36] generally apply the temperature gradient between maximum temperature and minimum temper­
ature on heating surface to estimate wall temperature uniformity.
The maximum temperature, average temperature and wall temperature uniformity of heating walls for SMCHS and TAMCHS under
different Re numbers are compared in this study. The temperature of SMCHS is shown as a red horizontal line in Fig. 6, as can be seen,
the average temperature of SMCHS is almost constant under different Re. In contrast, after the arrangement of twisted blade-like fins,
the average temperature of all walls in TAMCHS decreases, with TAMCHS 2S and TAMCHS 3S-A/B/C decreasing faster and then
slower, and TAMCHS 1S decreasing slowly. The sharpest decrease for TAMCHS 2S and TAMCHS 3S-A/B/C is observed when Re in­
creases from 50 to 150, with a greatest decrease of 28.6 K. By comparison, the total decrease is 49.0 K when Re increases from 50 to
700. In all TAMCHS, the average temperature of four walls showed a clear difference, with low temperature on the top and bottom
walls and high temperature on the left and right walls. Particularly, this difference is more obvious at lower Re numbers, but with the
increase of Re, the temperature difference becomes small. However, such a difference does not vary with the increase of Re and remains
constant at a high level in SMCHS. Moreover, for TAMCHS, the temperature of left and right walls decreases more than that of top and

Fig. 6. Comparison of average temperature on different walls of SMCHS and TAMCHS.

8
S. Chen et al. Case Studies in Thermal Engineering 52 (2023) 103788

bottom walls, and this difference reduces the temperature difference among four walls. Compared with SMCHS, the average tem­
perature of left and right walls in TAMCHS 2S and TAMCHS 3S-A/B/C decreases by 41.3 K, 45.7 K, 38.0 K, 48.1 K, and 41.0 K, 36.5 K,
37.7 K, 39.9 K at most, respectively, and remained below 313.2 K. At the same time, the temperature of top and bottom walls is
between 303.2 and 313.2 K when Re > 350, in which the average temperature on all four walls is close to each other, so the wall
temperature uniformity is significantly improved compared to SMCHS. However, for TAMCHS, the average temperature deteriorates
at small Re numbers, in which, on the top and bottom walls, the average temperature for some TAMCHS is deteriorated compared to
SMCHS when Re < 150, especially for TAMCHS 2S. On the left and right walls, the average temperature in some TAMCHS is dete­
riorated when Re = 50, especially on the right wall. Among all TAMCHS, the average temperature characteristic for TAMCHS 2S is the
worst at small Re number, with average temperature deteriorating on all walls when Re = 50. By contrast, the average temperature
characteristic for TAMCHS 1S is the best when Re = 50, with average temperature on all walls decreasing.
Different from average temperature, the maximum temperature on four walls shows a consistency. As can be seen from Fig. 7, for
SMCHS, the maximum temperature of four walls is equal and maintains at a high level, and for TAMCHS, the maximum temperature
varies in a similar value range and trend. Specifically, when Re = 50, the maximum temperature characteristic is the best for TAMCHS
1S, with a significant decrease on four walls compared to SMCHS, but thereafter it decreases slowly. By contrast, the maximum
temperature for TAMCHS 2S and TAMCHS 3S-A/B/C does not decrease significantly or even shows an increase at Re = 50, but as Re
increases from 50 to 150, the temperature rapidly decreases to be lower than that of the SMCHS temperature, and the decrease rate
become smaller thereafter. In the process of Re = 50–150, the temperature drop is nearly half of the total drop in the process of Re =
50–700. When Re > 500, on the top and left walls, the temperature for TAMCHS 3S-A/B/C decreases slowly and for TAMCHS 3S–B it
even rises, but for TAMCHS 1S and TAMCHS 2S it still maintains a high decreasing trend.
The temperature characteristics can be explained as follows. When the Reynolds number is small, the disturbance of the fluid by the
twisted fins is weak, while the stagnation effect is more pronounced. Therefore, the heat transfer performance is poor, and the heat
generated on the wall cannot be removed in a timely manner, resulting in the formation of hot zones and deterioration of heat transfer

Fig. 7. Comparison of maximum temperature on different walls of SMCHS and TAMCHS.

9
S. Chen et al. Case Studies in Thermal Engineering 52 (2023) 103788

compared to smooth microchannel. As the flow velocity increases, the disturbance effect becomes much greater than the stagnation
effect, leading to enhanced heat transfer. At the same time, at higher flow rates, the average temperature of the fluid decreases,
resulting in lower wall temperature. For TAMCHS 1S, at low Reynolds number, the stagnation effect is small, resulting in better heat
transfer compared to smooth microchannel, and lower wall temperature. However, as the Reynolds number increases, the disad­
vantage of the weak disturbance effect of TAMCHS 1S becomes apparent. Compared to other TAMCHS, its heat transfer enhancement is
smaller, and the temperature decreases slightly.
Fig. 8 shows the calculated wall temperature uniformity, and smaller ΔTu represents better wall temperature uniformity. In
addition to the wall temperature uniformity, overall temperature of the wall is also evaluated. In general, a wall with low overall
temperature and fine wall temperature uniformity is in its ideal situation. In this study, the overall temperature is measured by
minimum temperature (Tmin). The right half of the figure is ΔTu, and the left half is Tmin. By analyzing the length of left and right bars of
four walls, it is obvious that the top and bottom walls are in high ΔTu but in low Tmin. In contrast, the left and right walls are in low ΔTu
but in high Tmin. It can be seen that as Re increases, for SMCHS, the top and bottom walls always maintain a high ΔTu of 66.0 K, and the
left and right walls always remain a high Tmin of 346.4 K, which is far from ideal situation. It is necessary to make necessary ad­
justments to decrease the ΔTu of top and bottom walls and the Tmin of right and left walls. Fortunately, for TAMCHS, this problem is

Fig. 8. Comparison of ΔTu and minimum temperature on four walls in SMCHS and TAMCHS.

10
S. Chen et al. Case Studies in Thermal Engineering 52 (2023) 103788

Fig. 8. (continued).

well solved, for example, the ΔTu of the top and bottom walls, which is originally very high, decrease significantly with the increase of
Re, especially for TAMCHS 3S–C, where the ΔTu of top wall and bottom wall is decreased to 14.9 K and 21.4 K respectively. At the same
time, the Tmin of left and right walls, which is originally at high level, significantly decreases with the increase of Re, especially for
TAMCHS 3S–C, in which Tmin of left wall and right wall decreases to 300.0 K and 308.0 K, respectively. In comparison with other study
[41], the wall temperature uniformity of TAMCHS 3S–C in this work is further enhanced. On top and bottom walls, the Tmin in each
TAMCHS shows different degrees of rise, which attributes to the mixing of hot and cold fluids stimulated by twisted fins.
The temperature contours of SMCHS and TAMCHS are shown in Fig. 9, taking Re = 150 and Re = 700 as examples. The temperature
contours of the bottom wall, left wall, right wall and part of the top wall near the upper corner region are displayed, and different walls
are separated by black dashed lines. The wall temperature distributions directly reflect the cooling effect by the liquid coolant. As can
be seen from Fig. 9, the temperature distribution of SMCHS does not vary with Re, and as shown in the dashed box, the sidewalls,
especially the upper corner region and lower corner region, are at high temperature. Compared with SMCHS, the overall temperature
for TAMCHS decreases, especially the sidewalls and the corner regions between two neighboring walls, and it is more obvious as Re
increases from 150 to 700.
The differences in fin numbers and deflection directions lead to temperature differences in different TAMCHS. The single fin in

11
S. Chen et al. Case Studies in Thermal Engineering 52 (2023) 103788

Fig. 9. Temperature contours of SMCHS and TAMCHS (Re = 150 and Re = 700).

TAMCHS 1S enhances the heat transfer in the core region significantly, with a wide low temperature region appears, and the region
width increases with the increase of Re. However, the sidewalls especially right wall and its corner regions are still at high temperature.
In contrast, the two fins near the sidewalls of TAMCHS 2S strengthen the heat transfer on the sidewalls and corner region, but their
effect is relatively weak in core region. For TAMCHS 3S-A/B/C, the low temperature regions are wider compared to TAMCHS 1S and
TAMCHS 2S, but there is a left-right asymmetry. For example, the temperature of left wall in TAMCHS 3S-A and TAMCHS 3S–C is
significantly lower than that of their right walls. The temperature of left wall and right wall in TAMCHS 3S–B is close to each other, but
it is higher than that of left walls in TAMCHS 3S-A and TAMCHS 3S–C. For TAMCHS 3S-A and TAMCHS 3S–B, the bottom wall
temperature is more uniform than that of TAMCHS 3S–C, but the temperature on upper corner region is relatively high. Based on these
differences, it can be concluded that heat transfer is significantly enhanced in the region where the top trailing edge of the twisted fin is
deflected. It is important to note that the twist direction depicted in the figure is for the bottom section of the twisted fin, whereas the

12
S. Chen et al. Case Studies in Thermal Engineering 52 (2023) 103788

Fig. 9. (continued).

top section is deflected in the opposite direction. The effect of the deflection is not only reflected in the temperature asymmetry be­
tween left and right walls, but also the asymmetry between upper and lower corner region in TAMCHS 3S-A and TAMCHS 3S–B. As the
top trailing edge of the twisted fin deflects away from the sidewalls, the temperature near the upper corner region is higher than the
lower corner region. In the case of TAMCHS 3S–C, the top trailing edges of the two twisted fins near the sidewalls are deflected towards
the wall, so the same temperature is obtained on upper and lower corner regions. Moreover, the previously observed high temperature
near the upper corner region in previous studies [22,42] is significantly decreased, which can be attributed to the normalwise sec­
ondary flow generated by the twisted blade-like fins. Last but not least, the inevitable local hot region on the leeward side in traditional
ribbed radiators down to a slight level in TAMCHS.
Furthermore, a typical section, shown in Fig. 10(a), is employed to analyze the flow characteristic. For SMCHS, the fluid near the
wall and the fluid in the core of the microchannel cannot be mixed well. Therefore, the fluid temperature near the wall is high, and the
mainstream temperature is low, forming an obvious temperature stratification, and the temperature field does not change in different
flow velocities. After the introduction of twisted fins, shown in Fig. 10(b), the twisted blade-like fin we propose, characterized by the

13
S. Chen et al. Case Studies in Thermal Engineering 52 (2023) 103788

Fig. 9. (continued).

spatial features, generates both spanwise and normalwise secondary flows as the fluid flows over the curved surfaces. Simultaneously,
the primary flow serves as the base flow, and the superposition of the three components forms a longitudinal vortex, which spreads
outwards from the structure, forming a wider range of vortices. A single structure can produce two vortices in opposite directions.
Furthermore, due to the smaller windward area between the structure and the fluid, and the fact that the curved surfaces of the
structure guide the fluid rather than obstruct it, the increase in fluid drag is small. Based on these two characteristics of twisted blade-
like fins, achieving higher heat transfer and improved temperature uniformity with lower pressure loss becomes feasible.
By analyzing Fig. 10(b), the temperature characteristics reflected in Fig. 9 can be explained. The vortices generated by the single fin
in TAMCHS 1S enhance disturbance in the core region but show limited impact on sidewalls, so the temperature is low in the core
region, but high on the sidewalls. By contrast, for TAMCHS 2S, the vortices generated by the two fins near the sidewalls strengthen the
disturbance nearby, so the temperature on sidewalls decreases significantly compared to TAMCHS 1S, but at the same time the vortices
near the sidewalls have limited influence on the core region, where the temperature is higher than that of TAMCHS 1S. When the
twisted fins increase to three, the influence of vortices fills the whole space, so the low temperature region of TAMCHS 3S-A/B/C is
wider than that of TAMCHS 1S and TAMCHS 2S. Furthermore, due to different deflection directions of three fins, the vortices also show
differences. For example, for TAMCHS 3S-A, its vortex on the right side of the upper part is far away from the right wall, so the
temperature near the upper right corner region is higher than that on other regions. For TAMCHS 3S–C, though the two vortices on
upper two sides are close to the sidewalls, there is a difference in temperature between the left part and right part: low on the left and
high on the right. This is because the vortex between the middle fin and the right fin is weak, and the fluid exchange between these two
parts is weak, so the right part forms two relatively independent vortices, and the heat and mass exchange mainly carry out in these
independent vortices. Therefore, fails to mix in more core cold fluid, the fluid temperature in the right region is high, which leads to a
relatively high wall temperature. By contrast, the left part exchanges a lot with the core fluid, so the fluid temperature on the left region
is low and thus the wall temperature is relatively low. Therefore, to obtain low sidewall temperature, on the one hand, it is necessary to
generate vortices as close as possible to the sidewalls, and on the other hand, these vortices also need to exchange sufficiently with the
core cold fluid to obtain lower temperature.

4. Conclusions
In this work, a twisted blade-like fin with airfoil cross section is proposed. The heat transfer and temperature uniformity of
TAMCHS with different fins and deflection directions are numerically studied. The following conclusions are obtained.

14
S. Chen et al. Case Studies in Thermal Engineering 52 (2023) 103788

15 (caption on next page)

S. Chen et al. Case Studies in Thermal Engineering 52 (2023) 103788

Fig. 10. Flow vectors in characteristic section: (a) location of characteristic section, (b) fluid flow vectors (Re = 700).

(1) The Nu/Nu0, f/f0, and TP of TAMCHS range from 0.77 to 4.76, 1.14 to 2.39, and 0.71 to 3.58, respectively. When Re = 50, the
highest comprehensive thermal performance is observed in TAMCHS 1S. For TAMCHS 3S-A/B/C, the heat transfer and flow
drag are higher than that of others at Re > 50. The flow drag of TAMCHS 3S-A/B/C is close, but the heat transfer performance
differs significantly.
(2) The average and maximum temperature of TAMCHS are reduced by 48.1 K and 49.0 K at most respectively compared to SMCHS.
When Re = 50, the best temperature characteristic is observed in TAMCHS 1S, but other TAMCHSs play advantage at higher Re,
and the most significant temperature reduction occurs between Re = 50–250.
(3) With the introduction of twisted fins, the high overall temperature on the left and right walls is significantly decreased and the
wall temperature uniformity on the top and bottom walls is greatly improved. The overall temperature and wall temperature
uniformity of four walls are maintained at a better level.
(4) The fin number has a great influence on heat transfer and flow drag, and the deflection direction has a slight influence on flow
drag, but a significant effect on heat transfer. The region where the top trailing edge of the twisted fins deflecting produces
better heat transfer.
(5) The fin not only stimulate spanwise secondary flow but also stimulate normalwise secondary flow, so the heat transfer in normal
direction is also strengthened. The high temperature near upper corner region significantly decreases. Moreover, the inevitable
local hot region on the leeward side in traditional ribbed radiators down to a slight level in TAMCHS.
(6) To obtain low sidewall temperature, on the one hand, it is necessary to generate vortices as close as possible to the sidewalls, and
on the other hand, these vortices also need to exchange sufficiently with the core cold fluid to obtain lower temperature.

Author statement
Shiyang Chen: Numerical simulation、Formal analysis、Original Draft.
Zhenwei Liu: Data Curation、Visualization.
Boyuan Wang: Data Curation、Visualization.
Ping Li: Research design、Review & Editing、Funding acquisition、Supervision.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

Data availability

No data was used for the research described in the article.

Acknowledgments
The authors acknowledge the financial support from the National Natural Science Foundation of China (Grant No. 51976152).

References
[1] D.B. Tuckerman, R.F.W. Pease, High-performance heat sinking for VLSI, IEEE Electron. Device Lett. 2 (5) (1981) 126–129.
[2] H.Y. Wu, P. Cheng, Friction factors in smooth trapezoidal silicon microchannels with different aspect ratios, Int. J. Heat Mass Tran. 46 (14) (2003) 2519–2525.
[3] K.V. Sharp, R.J. Adrian, Transition from laminar to turbulent flow in liquid filled microtubes, Exp. Fluid 36 (5) (2004) 741–747.
[4] H. Shen, H.L. Liu, X.D. Shao, G.N. Xie, C.C. Wang, Thermofluids performances on innovative design with multi-circuit nested loop applicable for double-layer
microchannel heat sinks, Appl. Therm. Eng. 219 (2023), 119699.
[5] H. Shen, G.N. Xie, C.C. Wang, Thermal performance and entropy generation of novel X-structured double layered microchannel heat sinks, J. Taiwan Inst. Chem.
Eng. 111 (2020) 90–104.
[6] Y.L. Zhang, Y.F. Liu, M. Qu, Q. Yao, C.C. Fu, S.L. Yin, Optimization analysis of heat dissipation performance of microchannel heat sinks with the addition of
columnar inserts of spiral distribution, Int. J. Therm. Sci. 195 (2024), 108633.
[7] Z.F. Feng, Q.Y. Zhang, Z.Z. Li, Y.R. Bian, Z.J. Hu, et al., Hydrothermal and energy-saving performances in a mini-channel heat sink under different wire coil
inserts induced swirling flow, Int. J. Heat Mass Tran. 216 (2023), 124515.
[8] J.G. Tang, X. Li, R. Hu, Z.Y. Mo, M. Du, A novel designed manifold ultrathin micro pin-fin channel for thermal management of high-concentrator photovoltaic
system, Int. J. Heat Mass Tran. 183 (2022), 122094.
[9] F.T. Zohora, M.R. Haque, M.M. Haque, Numerical investigation of the hydrothermal performance of novel pin-fin heat sinks with hyperbolic, wavy, and crinkle
geometries and various perforations, Int. J. Therm. Sci. 194 (2023), 108578.
[10] S.W. Chang, A. Sadeghianjahromi, W.J. Sheu, C.C. Wang, Numerical study of oblique fins under natural convection with experimental validation, Int. J. Therm.
Sci. 179 (2022), 107668.
[11] M.K. Mohit, R. Gupta, Numerical investigation of the performance of rectangular micro-channel equipped with micro-pin-fin, Case Stud. Therm. Eng. 32 (2022),
101884.
[12] M.S. Lori, K. Vafai, Heat transfer and fluid flow analysis of microchannel heat sinks with periodic vertical porous ribs, Appl. Therm. Eng. 205 (2022), 118059.
[13] F. Li, Q.M. Ma, G.M. Xin, J.C. Zhang, X.Y. Wang, Heat transfer and flow characteristics of microchannels with solid and porous ribs, Appl. Therm. Eng. 178
(2020), 115639.
[14] S. Acharya, Thermo-fluidic analysis of microchannel heat sink with inline/staggered square/elliptical fins, Int. Commun. Heat Mass Tran. 147 (2023), 106961.
[15] E.R. Meinders, K. Hanjalic, Vortex structure and heat transfer in turbulent flow over a wall-mounted matrix of cubes, Int. J. Heat Fluid Flow 20 (3) (1999)
255–267.

16
S. Chen et al. Case Studies in Thermal Engineering 52 (2023) 103788

[16] O.N. Sara, T. Pekdemir, S. Yaplcl, M. Yilmaz, Enhancement of heat transfer from a flat surface in a channel flow by attachment of rectangular blocks, Int. J.
Energy Res. 25 (2001) 563–576.
[17] B. Sahin, A. Demir, Performance analysis of a heat exchanger having perforated square fins, Appl. Therm. Eng. 28 (2008) 621–632.
[18] B. Sahin, A. Demir, Thermal performance analysis and optimum design parameters of heat exchanger having perforated pin fins, Energy Convers. Manag. 49
(2008) 1684–1695.
[19] M. Ahmadian-Elmi, A. Mashayekhi, S.S. Nourazar, K. Vafai, A comprehensive study on parametric optimization of the pin-fin heat sink to improve its thermal
and hydraulic characteristics, Int. J. Heat Mass Tran. 180 (2021), 121797.
[20] J.X. Li, Z.L. Pan, P. Li, Improvement of temperature uniformity in microchannel with pin-fin based on endwall fillet structure, Chin. Sci. Bull. 63 (1) (2018)
108–116.
[21] Y.L. Mustafa, K. Hasan, C.K. Erkan, C. Ziya, A comparative CFD analysis of NACA0012 and NACA4412 airfoils, J. Energy Syst 2 (4) (2018) 145–159.
[22] P. Li, D.Z. Guo, X.Y. Huang, Heat transfer enhancement, entropy generation and temperature uniformity analyses of shark-skin bionic modified microchannel
heat sink, Int. J. Heat Mass Tran. 146 (2020), 118846.
[23] G.F. Lu, J.R. Yang, X.Q. Zhai, X.L. Wang, Hotspot thermal management in microchannel heat sinks with vortex generators, Int. J. Therm. Sci. 161 (2021),
106727.
[24] J.B. Wang, T. Fu, L.C. Zeng, G. Chen, F. Lien, Numerical and experimental investigations of micro thermal performance in a tube with delta winglet pairs,
Micromachines 12 (2021) 786.
[25] H. Shen, G.N. Xie, C.C. Wang, H.L. Liu, Experimental and numerical examinations of thermofluids characteristics of double-layer microchannel heat sinks with
deflectors, Int. J. Heat Mass Tran. 182 (2022), 121961.
[26] A.M. Ali, A. Rona, H.T. Kadhim, M. Angelino, S. Gao, Thermo-hydraulic performance of a circular microchannel heat sink using swirl flow and nanofluid, Appl.
Therm. Eng. 191 (2021), 116817.
[27] A.M. Ali, A. Rona, H.T. Kadhim, Numerical investigation of various twisted tapes enhancing a circular microchannel heat sink performance, Int. J. Heat Fluid
Flow 98 (2022), 109065.
[28] Y.Q. Song, N. Izadpanahi, M.A. Fazilati, Y. Lv, D. Toghraie, Numerical analysis of flow and heat transfer in an elliptical duct fitted with two rotating twisted
tapes, Int. Commun. Heat Mass Tran. 125 (2021), 105328.
[29] Z.F. Feng, Z.Z. Li, Z.J. Hu, Y.Q. Lan, S.Y. Zheng, Q.Y. Zhang, J.Y. Zhou, J.X. Zhang, Combined influence of rectangular wire coil and twisted tape on flow and
heat transfer characteristics in square mini-channels, Int. J. Heat Mass Tran. 205 (2023), 123866.
[30] H.S. Rad, S.M. Mousavi, A. Sarmadian, Comparative study on the thermal–hydraulic performance of tubes enhanced with three different types of teardrop
protrusions, Appl. Therm. Eng. 263 (2024), 121682.
[31] J.Y. Zheng, L. Zhang, C.G. Liu, Z. Liang, X. Deng, A numerical study on heat transfer enhancement and fluid flow of enhanced tube with ellipsoidal dimples and
protrusions, J. Therm. Sci. Eng. Appl. 14 (9) (2022), 091007.
[32] A.N. Ilikan, R. Aydin, Analysis of the slip flow in the hydrodynamic entrance region of a 2D microchannel, J. Therm. Eng. 9 (3) (2023) 733–745.
[33] Z.F. Feng, Z.J. Hu, Y.Q. Lan, Z.Q. Huang, J.X. Zhang, Effects of geometric parameters of circular pin-fins on fluid flow and heat transfer in an interrupted
microchannel heat sink, Int. J. Therm. Sci. 165 (2021), 106956.
[34] G.D. Xia, D.D. Ma, Y.L. Zhai, Y.F. Li, R. Liu, M. Du, Experimental and numerical study of fluid flow and heat transfer characteristics in microchannel heat sink
with complex structure, Energy Convers. Manag. 105 (2015) 848–857.
[35] R.K. Shah, A.L. London, Laminar Flow Forced Convection in Ducts, Academic Press, New York, 1978.
[36] G.D. Xia, J. Jiang, J. Wang, Y.L. Zhai, D.D. Ma, Effects of different geometric structures on fluid flow and heat transfer performance in microchannel heat sinks,
Int. J. Heat Mass Tran. 80 (2015) 439–447.
[37] Y.T. Jia, J.W. Huang, J.T. Wang, Heat transfer and fluid flow characteristics of microchannel with oval-shaped micro pin fins, Entropy 23 (11) (2021) 1482.
[38] F.M. White, Fluid Mechanics, McGraw-Hill Education, New York, 2016.
[39] J.F. Zhang, Y.K. Joshi, W.Q. Tao, Single phase laminar flow and heat transfer characteristics of microgaps with longitudinal vortex generator array, Int. J. Heat
Mass Tran. 111 (2017) 484–494.
[40] R. Kumar, R. Abiev, G. Ribatski, S. Abdullah, M. Vasilev, New approach of triumphing temperature nonuniformity and heat transfer performance augmentation
in micro pin fin heat sinks, J. Heat Transf.-Trans. ASME 142 (6) (2020), 062501.
[41] Z.Q. He, Y.F. Yan, S. Feng, X.Q. Li, Z.Q. Yang, Numerical study of thermal enhancement in a micro-heat sink with ribbed pin-fin arrays, J. Therm. Anal. Calorim.
143 (2021) 2163–2177.
[42] P. Li, D.Z. Guo, X.Y. Huang, Heat transfer enhancement in microchannel heat sinks with dual split-cylinder and its intelligent algorithm based fast optimization,
Appl. Therm. Eng. 171 (2020), 115060.

17