International Journal of Thermo uids 24 (2024) 100827

Contents lists available at ScienceDirect

International Journal of Thermofluids

journal homepage: www.sciencedirect.com/journal/international-journal-of-thermofluids

Heat transfer enhancement and pressure drop performance of Al2O3

nanofluid in a laminar flow tube with deep dimples under constant heat
flux: An experimental approach
Alireza Khashaei , Mohammad Ameri *, Shahram Azizifar
Energy Conversion Department, Faculty of Mechanical and Energy Engineering, Shahid Beheshti University, Tehran, Iran

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

Keywords: This study examines the heat transfer enhancement and pressure drop of Al2O3 nanofluid in deep dimpled tubes
Heat transfer enhancement in both longitudinal and circumferential directions. It explores mechanisms that improve the thermal perfor­
Dimpled tube mance of this novel tube geometry. Experiments were conducted using plain and deep dimpled tubes under
Al2O3/water nanofluid
laminar flow with Reynolds numbers from 500 to 2250, a constant heat flux of 10,000 W/m2, and nanofluid
Experimental setup
Forced convection
concentrations from 0.1 wt% to 1 wt%. The findings indicate that local velocity enhancement, vortex generation,
Performance evaluation criteria and flow rotation and mixing are the three main mechanisms that improve the thermal performance of deep
dimpled tubes. The results demonstrate that a deep dimpled tube with 1 wt% nanofluid can increase the
convective heat transfer coefficient by up to 3.42 times compared to a smooth tube at Re = 2250. At this
Reynolds number, the Nusselt number reaches a maximum of 41.80, and the friction factor ratio increases by
only 1.82. Additionally, circumferential analysis reveals how dimple-induced vortices enhance heat transfer. The
results also indicate that the tube geometry modification changes the flow regime zones, allowing turbulent flow
at lower Reynolds numbers near Re = 2000, as identified by Nusselt number and friction factor plots. The deep
dimpled tube has a low improvement penalty, with the highest friction factor of 0.38 at Re = 500 and high
thermal enhancement, resulting in a performance evaluation criterion (PEC) of up to 2.80 in the studied region.
However, the deep dimpled tube is unsuitable for Reynolds numbers below 1000. For higher velocities, replacing
simple tubes with deep dimpled tubes in traditional heat exchangers is highly recommended.

CuO [10], TiO2 [11], or a combination of two different nanoparticles

[12] are used to improve the fluid physical properties.
1. Introduction
The features of dimpled tubes have attracted the attention of many
researchers due to their lower pressure drop compared to other heat
Technologic and economic growth have led to a sharp rise in energy
transfer enhancement methods [13], leading to a higher Performance
demand. This has also increased the emission of pollutants from fossil
Evaluation Criterion (PEC), which is desirable. Nowadays, Researchers
fuels, which are still the principal source of energy consumption. Despite
are exploring various dimple geometries to enhance heat transfer and
the efforts made in the field of renewable energies, a large amount of
lower pressure drop. Different dimple geometries (conical, spherical,
energy consumption in all areas is still supplied by fossil fuels [1]. Heat
ellipsoidal) in dimpled tubes were numerically investigated by Sabir
transfer in various devices, especially heat exchangers, accounts for
et al. [14] to determine the optimal geometry. They performed this
much energy consuming. By improving heat transfer and enhancing the
study under constant heat flux condition in a laminar regime. Their
efficiency of heat exchangers, energy consumption and CO2 emissions
results showed that an ellipsoidal dimple with a 45-degree angle has the
can be lowered. Heat transfer enhancement methods are classified into
highest performance, about 58 % higher than a simple tube. In addition,
active and passive, based on energy consumption. Passive methods in­
Xie et al. [15] conducted a numerical comparison between the spherical,
crease the thermal performance by altering the physical characteristics
elliptical, and teardrop dimples. They found that the smooth flow pas­
of the tube or the fluid [2]. The tube surface characteristics are enhanced
sage over the teardrop dimple resulted in a lower pressure drop than the
by using a helical tube [3], twisted tape [4], dimpled fin [5], fins [6], or
other dimples. The maximum performance evaluation criteria (PEC)
dimpled tube [7]. Furthermore, nanofluids such as Fe3O4 [8], Al2O3 [9],

* Corresponding author.
E-mail address: [email protected] (M. Ameri).

https://doi.org/10.1016/j.ijft.2024.100827

Available online 25 August 2024

2666-2027/© 2024 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-
nc/4.0/).
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

Nomenclature Pw wetted perimeter

Di inner diameter [cm] Subscripts

D diameter [cm] in inlet
d depth [mm] out outlet
Do outer diameter [cm] nf nanofluid
q′′ heat flux [W/m2] bf base fluid
f friction factor [-] np nanoparticle
cp fluid specific heat capacity [J/kg.K] s simple tube
Ac cross-section area [m2] h hydraulic
h convective heat transfer coefficient [W/(m2.K)] b bulk
T temperature [◦ C] i index
Ti.w inner wall temperature [◦ C] R independent variable
Tout.w outer wall temperature [◦ C] Greek letters
U average axial velocity [m/s] p pitch ratio [mm]
Q volumetric flow rate [lit/min] ρ density [kg/m3]
ΔP pressure drop [pa]
Φ̇˙ heat of unit volume [W/m3]
Re Reynolds number [-]
μ dynamic viscosity [kg/(m.s)]
Nu Nusselt number [-]
φ nanoparticle volume fraction
PEC performance evaluation criteria
ω uncertainty [-]
HTC convection heat transfer coefficient
ṁ˙ Mass flow rate [kg/s]

value of 2.06 was obtained at Re = 5000, which was the highest Rey­ among the studied particles, respectively. Bretado de los Rios et al. [26]
nolds number examined. Cheraghi et al. [16] performed a numerical examined the thermal effect of Al2O3 nanofluid (1 % and 3 % volume
investigation on the deep dimple, which is a novel dimple shape. In the concentration) in a parabolic trough linear collector. They reported that
laminar regime and under constant heat flux condition, they assessed nanofluid enhanced the efficiency by up to 57.7 %. In an experimental
the impact of the dimple diameter, depth, and pitch ratio on the heat study, Briclot et al. [27] examined the behavior of Al2O3 nanofluid in a
transfer, pressure drop, and overall performance of the tube. They found simple tube under laminar and transient conditions. The research
that the best overall performance is obtained at Re = 2000. Zhang et al. focused on Reynolds numbers ranging from 500 to 4500 and utilized
[17] numerically studied the heat transfer of a tube with cross-combined nanofluid with a weight concentration varying from 0.75 % to 5 %. The
ellipsoidal dimples in the turbulent regime (Reynolds numbers 6880 to findings revealed that the nanofluid demonstrated inferior thermal
37,500). They found that this tube geometry increases the Nusselt performance compared to distilled water in the transient region.
number by up to 2.1 times and the friction factor by up to 1.9 times, with Furthermore, the nanofluid concentration was found to have no impact
a maximum PEC of 1.9. In studying dimples’ different shapes, Xie et al. on the critical Reynolds number. Zhang et al. [28] investigated the
[18] investigated a tube with helical dimples by varying the number, hybrid CuO-Al2O3 nanofluid (1 %–3 % mass concentration) in the tur­
size, spiral pitch, and starting points of the dimples. They observed that bulent regime numerically and experimentally. They compared different
the PEC reaches its maximum value of 1.93 at Re = 5,000. They numerical methods with experimental results and showed that PEC
attributed this improvement to the better mixing and uniform temper­ increased with nanofluid concentration and decreased with Reynolds
ature distribution caused by the vortices generated by the dimples. number, ranging from 1.01 to 1.29.
Wang et al. [19] combined a three-start twisted tube and oval dimples Regarding the mentioned studies, nanofluids are typically combined
and studied the Reynolds numbers range of 2000 to 20,000. They with other active or passive heat transfer enhancement methods because
concluded that the vortices which are induced by dimples, enhances the of their negligible heat transfer improvement when used solely. For
heat transfer. This tube has the PEC of 1.38 compared to a simple tube instance, Gürsoy et al. [24] conducted an experimental study investi­
and 1.19 compared to a simple twisted tube. The mentioned researches gating the impact of magnetic field location on heat transfer enhance­
provide evidence that using tubes with different dimple configurations ment when a ferrofluid Fe3O4 passes a tube with a novel dimpled design.
can significantly improve heat transfer efficiency. This study, which also evaluated the tube’s thermal and hydrodynamic
Working fluids like water, ethylene glycol, and oil, are frequently performance, considered a Reynolds number range of 1131 to 2102 and
used in heat transfer due to their physical properties. However, they low positioned the magnetic field at varying distances from the tube’s start
thermal conductivity limits heat transfer efficiency [20]. This challenge at a constant intensity. Their findings indicate that the PEC value rea­
led Maxwell [21] to propose the addition of solid particles to these fluids ches approximately 1.5 at the highest Reynolds number. Subsequently,
to enhance their thermal conductivity in 1873. As a result of techno­ Gürdal et al. [29], in a numerical study, investigated the effect of using a
logical progress and the emergence of nanoparticles, Choi [22] intro­ magnetic field when a nanofluid passes through a tube with dimples.
duced the concept of nanofluids, which are suspensions of fine particles This study was conducted in the Reynolds numbers range of 10,000 to
in a liquid. These nanofluids have garnered significant interest from 50,000 under constant heat flux and utilized a magnetic nanofluid Fe3O4
researchers due to their characteristics, including low pressure drop, in three volume concentrations of 0.5 %, 1 %, and 2.5 %. The results
high thermal conductivity, and extensive contact surface [23]. They show that increasing the Reynolds number increases the Nusselt number
have found wide application in heat sinks, solar collectors, heat ex­ while increasing the dimple pitch decreases it. Also, at the highest
changers, and microchannels [24]. In an empirical investigation, Loong concentration and field intensity, the Nusselt number increases by
et al. [25] evaluated the thermal performance of nanofluids composed of approximately 72.48 %, and the PEC reaches a maximum of 1.126.
various metal oxides. The study focused on metal oxide particles with a Altunay et al. [30] simulated a similar setup with an alternating mag­
mean diameter ranging from 20 to 30 nanometers. The findings revealed netic field and investigated pressure drop and heat transfer simulta­
that MgO, CuO, and Al2O3 exhibited the highest thermal performance neously. This study was conducted in the laminar regime with a

2
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

Reynolds number of 1131 < Re < 2102 and investigated a knowledge, only Cruz et al. [33] have conducted similar work just on
ferro-nanofluid Fe3O4 with a volume concentration of 1 %. The magnetic helical corrugated tubes, and it has not been done on dimpled tubes.
field also oscillated with frequencies of 1, 2, and 5 Hz. Their results Hence, this study contributes to the existing literature by offering the
showed that the highest PEC value is nearly 1.30, which is obtained in following novel insights: (i) Combination of novel tube geometry with
the absence of a magnetic field. Ahmed et al. [31] conducted a numer­ nanofluid, (ii) Experimental investigation of low-concentration nano­
ical study on the performance of a tube with dimples when a Al2O3 fluid enhancement mechanisms, (iii) Studying different flow regime’s
nanofluid passes through it. This study varied the nanofluid concen­ zones when nanofluid passes through the tube, (iv) Circumferential heat
tration from 0 to 4 % and investigated different tube geometries. They transfer characteristics of the tube.
concluded that using a tube with dimples increases heat transfer by The deep dimple tube, with its tube size and consequential pressure
approximately 27 %, and utilizing a nanofluid with a volume concen­ drop, has the potential to replace conventional tubes in existing heat
tration of 4 % increases this improvement by about 32 %. Lastly, Gürdal exchangers. By using these tubes, the size of heat exchangers can be
et al. [32] numerically and experimentally studied the effect of placing a significantly reduced, thus leading to a decrease in energy consumption
tube with a dimple under a magnetic field when a hybrid nanofluid and environmental pollution. The study’s findings indicate that the deep
passes through it. This study was conducted in the laminar regime and dimple tube could be a promising alternative to conventional tubes in
used a combination of nanoparticles Fe3O4 and CuO to prepare the heat exchangers. It is worth mentioning that the pre-heating section of a
nanofluid. Their results showed that when using a hybrid nanofluid and heat exchanger for recovering waste heat from an automobile operates
a magnetic field, the PEC reaches approximately 2.4. In numerical within or close to the transitional flow regime [34], similar to the cur­
simulations, the Nusselt number and friction factor increase by 11.87 % rent study consideration. Therefore, the use of deep dimple tubes would
and 6.19 %. However, in experiments, this increase is about 174.65 % be beneficial for this objective.
and 169.4 %, respectively.
The literature review indicates that researchers have explored 2. Experimental apparatus system and methods
various dimple tube geometries. In this trend, they desire to enhance
thermal performance and minimize pressure drop in order to achieve the Fig. 1 concisely overviews the procedure of the current study,
highest PEC. In addition, nanofluids are often combined with other heat including the order of different steps and some details about them. An
transfer enhancement methods due to their minimal effectiveness when experimental apparatus was prepared and validated by testing a simple
used alone. The current study follows the trends of previous research tube and comparing the results with empirical correlations. Next, a
[14–19] in exploring new dimple tube shapes to enhance thermal per­ novel deep dimple tube shape was investigated. After thorough analysis,
formance. It introduces a novel hybrid method combining dimple tube nanofluids with varying concentrations were prepared, and their
geometry, called the deep dimple, with a nanofluid and fulfills the thermal-hydraulic performances were studied.
mentioned requirements. It should be considered that the geometrical
characteristics of current tube is totally different from those have been
investigated yet. In addition, this study investigates the use of Al2O3 2.1. Experimental details
nanofluid in low concentrations, which are both easily accessible and
affordable to produce. Examination of heat transfer in both longitudinal Cheraghi et al. [16] studied various geometrical configurations of
and circumferential directions is another feature of the current study. deep dimple tubes and ultimately selected an arrangement with superior
Previous researches have largely neglected this topic. Finally, this performance. The Current investigation evaluated the distinct mass
research experimentally explores the mechanisms that enhance heat conservation of Al2O3 − γ nanofluid in the novel geometry that was
transfer and the different flow regimes present in the new geometry chosen based on their results. This section introduces the essential pa­
tube. Among the numerous studies reviewed, to the best of the authors’ rameters for assessing data reliability and discusses the validation of
data obtained from the setup.

Fig. 1. Outline sketch of the experimental procedure to investigate.

3
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

2.1.1. Dimpled tube geometry a variac and transformer device (11) with a constant heat flux boundary
For assessments in this experimental study, a tube with an inner condition, thermocouples (9) that measure the wall temperature along
diameter (Di) of 1.8 cm, a length (l) of 100 cm, and a thickness (t) of 0.5 the test section, and a flowmeter (12) that measures the volumetric flow
cm was utilized. An extra 10 cm is added to each side of the test tube to rate of the working fluid.
mitigate any undesired flow fluctuations at connection points. The The cooling system (13) regulates the temperature of the working
longitude of the 306 stainless steel tube margins is approximately equal fluid before returning to the reservoir tank. It consists of a two-pipe heat
to the diameter of the dimples. This distance is insignificant relative to exchanger with chilled water as the cooling medium.
the length of the dimple distribution. In order to improve the heat This type of connection enables the test section to be replaced by
transfer efficiency, the tube surface was modified with longitudinal deep other tubes with different surface features. The setup is insulated with
dimples. These dimples have a 2 mm depth (d), 4D mm pitch ratio (p) (i. polyethylene foam (5 mm thick) to minimize heat loss. The Reynolds
e., 72 mm), and 18 mm diameter (D). Fig. 2 depicts these features in (a) number of the working fluid ranges from 500 to 2250, which covers the
front cross-sectional view and (b) longitude of the tube. Also, in Fig. 2 laminar flow regime.
(a), the angle between dimples in a defined tube length is demonstrated. To calculate the heat transfer coefficient (HTC), previous studies on
The modified tube’s dimensions allow it to operate in conventional heat enhanced tubes measured the outer wall temperature along the tube [7,
exchangers without significant structural changes, providing user- 37,38]. However, this study examined the dimple effect experimentally
friendliness. by measuring around three dimple-sets and an inlet/outlet circumfer­
ence of another dimple-set using K-type thermocouples. Figs. 4 and 5
2.1.2. Setup characteristics illustrate the longitudinal and circumferential views of the dimple tube
An experimental apparatus was adjusted to evaluate the deep dimple in detail.
tube configuration in a laminar regime. The adjustment followed con­ Fig. 4 shows 15 thermocouples at an equal distance and 2.5 cm (x1)
ventional enhanced tube setups, as in Altunay et al. [30], Gürdal et al. offset from the start of the test region (i.e., the central 100 cm length that
[32], Dagdevir et al. [35,36], and Sun et al. [37], with minor modifi­ was heated). A previous study [16] indicated that the HTC varied along
cations due to the current study conditions. For this purpose, the setup the tube due to the periodic presence of dimples. Therefore, points be­
consists of a DC centrifugal pump with a speed controller module that tween two adjacent dimple-sets were selected to eliminate the effect of
can regulate the flow up to 4 lit/m, and a needle valve that can adjust the dimple-induced turbulence in measurement and obtain average HTC for
flow precisely. As a result, these pump features eliminate the need for reliable comparison and conclusion.
bypass piping in the setup. Moreover, a ventilation bypass is another The dimple tube has 14 dimples, as seen from the top view. Three
modification in this setup. The deep dimple tube diameter and Reynolds dimple-sets at the beginning, middle, and end of this novel geometry, as
number of investigations cause air to emerge and mix with the working depicted in Fig. 5(a) (dimple-set number 2,7,13), were investigated to
fluid, which may affect the results of experiments. So, increasing the examine the circumferential condition of the dimples. Six thermocou­
flow rate and eliminating air bubbles from the system is necessary to ples were placed around each dimple-set to measure the temperature at
prevent this problem. Nevertheless, the flowmeter’s precision is affected selected points, as shown in Fig. 5(b). The area just before and after
by the high flow value. Thus, the bypass protects the flowmeter from any dimple-set number three was also investigated to verify the effectiveness
damage. of the designed enhancement mechanisms. Fig. 5(c) shows the location
The experimental setup operates in a closed loop because of the of four equally distributed points around the tube on both sides of the
constraints of nanofluid preparation and stability. As shown in Fig. 3, the dimple-set. Finally, brief information about the deep dimple tube’s
apparatus consists of four main parts: a reservoir tank, a pumping sys­ characteristics and the test conditions are presented in Table 1.
tem, a testing region, and a cooling system.
The reservoir tank (1) stores the working fluid, which is either water
2.2. Operating conditions and calculations
or water-based nanofluid with different concentrations of alumina
nanoparticles. The pumping system (2–4) circulates the working fluid
This section describes the experimental procedure and data collec­
through the loop. It comprises a DC centrifugal pump (2), a speed
tion methods used in this study. Before installation, all data-collecting
controller module (3), and a needle valve (4) that controls the flow rate
instruments, such as thermocouples, differential pressure transmitter,
of the working fluid.
and flowmeter, undergo calibration in a laboratory. This standardization
The testing region (5–12) measures the heat transfer performance of
ensures accuracy and prevents errors from miscalibration. The computer
different tubes. It comprises two bulk temperature thermocouples (5,10)
stores these data from each data logger for future data processing. To
that measure the inlet and outlet temperatures of the working fluid, a
achieve a hydraulic fully developed flow, the pump fixes the Reynolds
developing tube (6) that ensures fully developed flow before entering
number, and the fluid recirculates for at least 15 min. The test section
the test section, a differential pressure transmitter (7) that measures the
receives a constant heat flux in the next step. However, the experimental
pressure drop across the test section, a dimpled tube (8) that is heated by
condition remains stable for a minimum of 45 min. Finally, data is

Fig. 2. 3D geometric representation of the deep dimple tube: (a) front cross-sectional view and (b) top view.

4
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

Fig. 3. Simplified illustration of the experimental setup.

Fig. 4. Installing thermocouples lengthwise.

collected and averaged over 15 min without significant parameter ˙

the heat loss (Q) from the environment from the electrical power con­
fluctuations to ensure accuracy. Each experiment is repeated at least l
three times, with confirmation based on consistent trends in behavior to sumption by the power supply. Alternatively, it can be computed by the
ensure repeatability. The data loggers measure the parameters Q, ΔP, thermal energy transfer rate to the working fluid between the inlet and
and T directly. MATLAB software processes their mean value to obtain outlet of the tube:
Re, f, HTC, and Nu. Moreover, an ampere meter and a voltmeter on a
˙
transformer measure the current (A) and voltage (V) of the variac Pnet = VI − Q = ṁ˙cp (Tout − Tin ) (1)
output, determining the applied heat flux. Finally, a crucial procedure loss

after each experiment is to evacuate the setup to protect the tube and The following symbols represent the variables used in the equation:
nanofluid from damage. V is the transformer voltage, I is the transformer current, ṁ˙ is the mass
As noted before, to remove the air bubbles that may form in the flow rate, cp is the fluid-specific heat capacity, Tout and Tin are the outlet
working fluid due to the laminar regime of the Reynolds numbers, a and inlet bulk temperatures, respectively.
ventilation path with a high flow volume capacity is used. The flow rate The heat flux (q′′) applied to the tube is calculated as:
is increased to about 3 liters per minute at the start of the test, which
facilitates the ventilation of the working fluid and expels the air. The Pnet
qʹʹ = (2)
flow rate is kept constant until all the air bubbles are eliminated, and πDi L
then the pump runs for five min more for assurance. Then, the flow rate Di and L represent the inner diameter and length of the test section
is slowly reduced and switched to the cooling system path within the tube, respectively. In order to evaluate the thermal performance, the
allowed range. Finally, the flow rate further decreases until it reaches following equation is used to calculate the local convective heat transfer
the desired value. coefficient along the length of the tube:
The main objective of this study is to experimentally examine the
heat transfer and pressure drop of nanofluid in the dimpled tube by qʹʹ
h(x) = (3)
evaluating the Nusselt number (Nu), friction factor (f), and performance Tin.w (x) − Tb (x)
evaluation criteria (PEC) at various Reynolds numbers (Re). The The local inner wall temperature of the dimple tube Ti.w(x) and the
following steps describe the data processing method for this aim: local bulk temperature of the working fluid Tb(x) are used in this
The net power (Pnet) applied to the tube is determined by subtracting equation. By applying the energy balance principle, Tb(x) is expressed

5
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

Fig. 5. Installing thermocouples around the circumference.

axial velocity, which is computed by:

Table 1
Experimental condition and characteristics of the deep dimple tube. Q
U= (5)
Parameter Value Unit Ac

Outlet diameter of the 20 mm Q is the volumetric flow rate of the nanofluid and can be directly
tube, Do
Inlet diameter of the tube, 18 mm
measured through a flow meter. Since the thermocouples measure the
Di outer wall temperature of the tube without affecting the flow, the inner
Depth of a dimple, d 2 mm temperature can be determined by solving the radial heat conduction
Pitch ratio of dimples, p 72 mm equation below, which provides more accurate data [39]:
Diameter of a dimple, D 18 mm
( )
Fully developed region 2 m 1 d dT Φ̇˙
length r + =0 (6)
r dr dr ks
Length of the test section 1 m
Inlet temperature, Tin 25 ◦
C By considering boundary conditions:
Heat flux, q′′ 10000 W/
m2 dT
Reynolds number, Re 500–2250 - − ks = qʹʹ when r = rin
dr (7)
Working fluid Distilled water - 0.1 %, 0.5 % and 1 wt% -
Al2O3/water T = Tout.w when r = rout
Eq. (4) can be solved to obtain the following expression for the inner
as: wall temperature:
( ) ( )
4qʹʹ 2rin qʹʹ + r2 in Φ̇˙ rout Φ̇˙ ( 2 )
Tb (x) = X + Tin (4) Tin.w = Tout.w − ln − r in − r2 out (8)
ρ Di U Cp 2ks rin 4ks

In this equation, ρ represents the nanofluid density at the mean The thermal conductivity of stainless steel 304 is denoted by ks. The
temperature between the inlet and outlet. U is the nanofluid’s average heat flux, Φ̇˙, is obtained by:

6
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

Pnet Table 3
Φ̇˙ = (9)
π(r2 out − r2 in )L Quantifying the uncertainty of dependent variables.
Dependent parameter Uncertainty
The Reynolds number and the local Nusselt number are the thermal
and hydraulic dimensionless parameters that can be expressed as Re (Reynolds number) 0.8%
follows: Nu (Nusselt number) 3.89%
F (Friction factor) 2.76%
U Di ρ h (Heat transfer coefficient) 2.1%
Re = (10)
μ

h(x) Di Shah’s equations [42] for thermally developing flow under uniform heat
Nu(x) = (11) flux and the Hagen–Poiseuille equation [43] for conventional laminar
kf
flow friction factor of circular tubes:
μ is the dynamic viscosity at the average bulk temperature of the inlet ⎧
and outlet. The overall Nu number of the tube can be calculated by ⎨ 3.302x∗ − 1/3 − 1.00 x∗ ≤ 0.00005
Nux = 1.302x∗ − 1/3 − ( 0.50 0.000005 < x ≤ 0.0015 (16)
substituting havg with h(x), where havg is: ⎩ )− 0.506 − 41x∗ ∗
4.364 + 8.68 103 x∗ e x∗ > 0.001
∫L 14
1 ∑ h(xi ) + h(xi+1 ) [xi+1 − xi ]
havg = h(x)⋅dx ≈ (12) Re
L 2 [x15 − x1 ] f= (17)
0
i=1 64

The Fanning friction factor (f) is a dimensionless parameter that The local Nusselt number on the simple tube is calculated from ten
quantifies the pressure loss along the test section tube due to fluid equally spaced thermocouples and compared with Shah’s equations.
friction. It is obtained from the differential pressure transmitter and Fig. 6 shows the thermal and hydraulic performance of the experimental
given by: setup and the empirical correlations. The local Nusselt numbers at Re =
2000 are consistent with Shah’s equation, except for point one, where
f =(
ΔP
) (13) welding defects caused a 19 % deviation. Despite this, errors are rela­
ρU2
× DLh tively low, ranging from 0.04 % to 10 % for the different distances from
2
the inlet, as shown in Fig. 6(a). There is also the same trend for other
Heat transfer enhancement usually requires more energy consump­ investigated Reynolds numbers. As demonstrated in Fig. 6(b), the fric­
tion, such as additional friction in the flow. Therefore, the enhanced tion factor is within an acceptable margin of error, with a minimal de­
tube should be evaluated in terms of both heat transfer and energy ef­ viation of 0.31 % at Re = 2000. Although there is a deviation at Re =
ficiency. A prevalent method for this is applying the performance eval­ 2250, it is justified by considering that the flow is in the vicinity of the
uation criteria (PEC) proposed by Webb [40]: transition region. In this region, the mentioned correlation loses its ac­
curacy. Furthermore, it is noteworthy that errors do not exceed 11 % for
( )( )− 1/3
other velocities. Finally, it should be considered that there are some
Nunf fnf
PEC = (14) minor challenges, such as adjusting the precise flow rate, local defects in
Nubf fbf
insulation, and disruption in flow pattern at the joints. These are un­
The subscripts nf and bf indicate the nanofluid in the deep dimple avoidable in experimental studies and contribute to these errors.
and distilled water in the smooth tube, respectively.
This study applies the methods of Kline and McClintock [41] to ac­ 3. Nanofluid preparation and properties
count for the error of measurement devices. Table 2 shows the measured
parameters, their units, and the maximum error of each instrument. 3.1. Preparation and stability
This method assumes that R is a dependent variable and u1,2, … are
independent variables. So, the uncertainty is calculated as: The nanofluid samples were prepared by dispersing Al2O3 nano­
√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅ particles of 10 <dp< 30 nm (99.5 % purity) from US Research Nano­
√ n (
√∑ ∂ R )2 materials in distilled water as the base fluid. Three weight
ωR = √ ω (15)
∂ui i concentrations of nanoparticles (0.1, 0.5, and 1 wt%) were applied to
i=1
investigate their effects on the thermal performance of the nanofluid in
Table 3 shows the results of the uncertainty analysis for the depen­ the deep dimple tube. The nanofluids were purchased from a nanofluid
dent experimental data in this study, namely Re, Nu, f, and h. research institute in Iran, and their stability was guaranteed for at least
one week. Therefore, each weight concentration at different Reynolds
numbers was tested within one week (maximum four days) to avoid any
2.3. Experimental system validation possible instability issues.
This study chose an Al2O3 nanoparticle based on the findings of
In order to ensure the accuracy of experimental results, it is essential Loong et al. [25], who investigated the performance factors of different
to compare it with the empirical correlation. Therefore, the simple nanofluids. According to their results, MgO, CuO, and Al2O3 had the
tube’s local Nusselt number and friction factor are verified by applying highest performance factors compared to other metal oxide nanofluids.
Because of budget constraints in this study, Al2O3 was used at the given
Table 2 weight concentration. Moreover, many studies on heat transfer
Evaluating the uncertainty of independent variables. enhancement have employed Al2O3 nanofluid [27,29,30,44–46].
Independent parameter Uncertainty Images of the three nanofluid samples obtained in this study are
presented in Fig. 7, demonstrating the appropriate stability and homo­
Wall temperature (◦ C) ± 0.1◦ C
Fluid temperature (◦ C) ± 0.1◦ C geneity of the purchased nanofluids. It also reveals that the transparency
Test section’s length (m) ± 0.01% of the samples decreases with increasing the weight concentration (the
Test section’s diameter (cm) ± 0.1% white color intensifies with higher particle concentration). Furthermore,
Pressure drop (kPa) ± 0.25%
the SEM view and XRD analysis of the Al2O3 are presented in Fig. 8. XRD
Volumetric flow rate (lit/min) ± 0.2%
Heat power (W) ± 0.05%
analysis is a technique that can determine the chemical structure and

7
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

Fig. 6. Comparison of (a) local Nu number and (b) friction factor with empirical correlation.

Fig. 7. Visualizing the effects of concentration on nanofluid properties.

Fig. 8. (a) SEM and (b) XRD Studies of nanoparticle features.

8
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

percent purity of nanoparticles. elevating the fluid’s overall thermal conductivity and potentially
through other mechanisms such as Brownian motion and thermopho­
3.2. Thermophysical properties resis [28]. It is important to note that in a deep dimple tube, the vortex
and swirling flow enable solid nanoparticles, which have absorbed more
Nanofluid, a fluid containing nanoparticles, exhibits distinct ther­ heat due to their superior thermal conductivity, to mix more with the
mophysical properties such as thermal conductivity, viscosity, density, flow. This results in a more homogeneous fluid temperature distribution
and specific heat. These properties diverge from those of distilled water, and a warmer core flow than a conventional tube [52]. Additionally, the
whose values are available from thermodynamic tables. The concen­ turbulent flow of solid particles disrupts the thermal boundary layer,
tration of the nanofluid is a crucial determinant of these characteristics, thereby enhancing heat transfer [20]. This study investigates the ther­
and it can be calculated using the following equation [47]: mal improvements of these mechanisms at weight concentrations of 0.1
mnp
%, 0.5 %, and 1 %.
(18)
ρnp
φ = mnp mbf
ρnp + ρbf
4.1.1. Heat transfer enhancement in longitudinal direction
The study aims to examine heat transfer in both the longitudinal and
In this equation m denotes mass, while the subscripts nf and bf circumferential directions of the tube. In this section, the focus is on
represent nanofluid and base fluid, respectively. The study conducted by investigating the Nusselt number and the convective heat transfer co­
Pak and Cho [48] presents an equation for calculating density as follows: efficient (HTC) along the longitudinal direction. The results of this study
will help to understand the flow behavior of nanofluid in the laminar
ρnf = φρnp + (1 − φ)ρbf (19)
regime and obtain the overall and local Nusselt numbers. A comparison
Also, the specific heat of nanofluid can be calculated as [49]: with the passage of distilled water through a simple tube has confirmed
the effectiveness of this method in improving heat transfer. The local
φρnp Cp.np + (1 − φ)ρbf Cp.bf
Cp.nf = (20) Nusselt numbers at the points introduced in the previous section in the
ρnf tube with deep dimple and passage of nanofluid with theoretical values
Einstein’s equation [50] can estimate the viscosity of nanofluid with at Reynolds numbers from 500 to 2250 are compared in Fig. 9(a)–(e),
spherical particles and low volume concentrations (less than five respectively.
percent): Fig. 9(a) and (b) demonstrate that enhancing heat transfer for Rey­
nolds numbers less than 1000 is impossible in this tube type. Never­
μnf = (1 + 2.5φ)μbf (21) theless, for Reynolds numbers between 1500 and 2250, there is a
notable improvement in heat transfer performance, as indicated in Fig. 9
Finally, the thermal conductivity coefficient can be determined by
(c) to (e). At a Reynolds number of 2250, the local Nusselt number is
Maxwell’s equation [21]:
elevated by 6.35 times (536 %), as shown in the figures.
( )
knf kbf + knp + 2φ knp − kbf As Fig. 9(a) illustrates, nanofluid enhances the thermal performance
= ( ) (22)
kbf kbf + knp − φ knp − kbf compared to distilled water, but the heat transfer rate remains lower
than the simple tube in the initial section. The error values are not too
As Table 4 shows, the type of material and concentration affect the close in different nanofluid concentrations. So, results suggest that an
thermophysical properties of the nanoparticle and working fluids. These increase in concentration usually results in an improvement in heat
properties are calculated at a temperature of 20◦ C, which is the ambient transfer. Moreover, the results reveal that the improvement mechanisms
condition for the study. The thermophysical properties of Al2O3 − γ become more effective near the end of the tube and result in better heat
nanoparticle was obtained from the nanofluid research institute, which transfer than the simple tube. Furthermore, the local Nusselt number of
supplied the nanofluids. The details for different nanofluid concentra­ the simple tube decreases with a downward trend, while the new ge­
tions are calculated using equations (18-22). ometry shows a better slope and improved heat transfer when the
nanofluid flows through it. In addition, most measurement points do not
4. Results and discussions show fluctuations in the values of interest (except for the endpoints
where the improvement mechanisms are intensified). The utilization of
4.1. Thermal performance points with average temperature has resulted in the minimization of
temperature fluctuations in the calculations. This is accompanied by
As mentioned before, the current research examines the impact of computing the local Nusselt average between two dimples instead of the
nanofluid application in a deep dimple tube on heat transfer enhance­ fluctuations between them. The tube’s downward slope is evident, and it
ment and pressure drop. Cheraghi et al. [16], previously in a numerical can be utilized as a benchmark to monitor the fluctuations induced by
simulation, determined that a deep dimple tube with specific geometric the enhancement mechanisms. This implies that the fluctuations in other
characteristics augments heat transfer via three primary mechanisms: Reynolds numbers are due to the improvement mechanisms rather than
local velocity augmentation near the dimple edge, vortex generation the temperature fluctuations from measurement. It is worth mentioning
post-dimple, and axial swirling due to the alternating dimple presence that the inclusion of average temperature points in the calculations helps
along the tube. Additionally, the previous study [51] investigated the to minimize errors associated with overall values, such as the overall
impact of distilled water; however, this study reveals that nanofluid Nusselt number of the tube.
behaves slightly differently. Fig. 9(b) illustrates that the heat transfer performance of the con­
Furthermore, nanoparticle utilization enhances heat transfer by ventional tube is superior to that of the test tube. The improvement
mechanisms are efficient, as indicated by the increased fluctuations of
Table 4 the local Nusselt number. Some parts of the novel geometry transfer heat
Thermophysical properties of different working fluids in inlet temperature. better than a simple tube for Re = 500. Therefore, the conventional tube
Materials H2O Al2O3 0.1 wt% 0.5 wt% 1 wt% is expected to have a much higher overall Nusselt number than the test
properties [32] − γ Al2O3/H2O Al2O3/H2O Al2O3/H2O tube. Due to the proximity of the error bars of different concentrations at
ρ(kg/m3) 998 3690 1005.3 1035.8 1076.5 Re = 1000, a point-by-point comparison is not practicable. However, it
Cp(J/Kg.K) 4182 880 4149 4016.9 3851.8 can be argued that an elevation in concentration leads to an enhance­
k(W/m.K) 0.598 18 0.6026 0.6219 0.6483 ment in thermal performance. At this Reynolds number, the nanofluid
μ(kg/m.s) 10− 3 - 1.007*10− 3
1.035*10− 3
1.073*10− 3
demonstrates better heat transfer capability than the pure base fluid.

9
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

Fig. 9. The local Nusselt number of the nanofluid-filled tube with deep dimples versus a water-filled simple tube along the tube length for various flow rates.

The tube shows a decreasing slope as expected, indicating the laminar experimental error since this phenomenon occurs in all concentrations
flow of the nanofluid, as evidenced by the incomplete coverage of and repetitions.
fluctuations in the test section. Fig. 9(d) illustrates a notable enhancement in thermal performance
As Fig. 9(c) demonstrates, the tube with deep dimples outperforms at Re = 2000. The tube exhibits intense fluctuations, suggesting that the
the conventional tube in heat transfer for the first time. Moreover, the flow regime is turbulent. Despite the proximity of diagrams corre­
heat transfer rate rises with increasing concentration, as indicated by the sponding to various concentrations and distilled water and their asso­
variation of the error bars. However, at Re = 1500, distilled water ciated errors, establishing the supremacy of one concentration over
performs better than nanofluid, unlike other Reynolds numbers. It another is difficult. However, a slight enhancement in thermal perfor­
should be noted that at this Reynolds number, the fluctuations are close mance can be observed with an increase in nanoparticle proportion. It is
to the beginning of the tube. Furthermore, from other figures, it appears important to note that due to the similar values across different con­
that the flow reaches the turbulent regime from Re = 2000 due to the centrations, the local Nusselt numbers of the simple tube are depicted in
severe fluctuations throughout the tube. Therefore, the flow at Re = a minor diagram for easier comparison. At this Reynolds number, the
1500 is in the transient region. Based on this observation, one can infer local Nusselt number has generally increased between two to five times
that the complex flow processes in the transient region cause the in most sections. Also, there is a decreasing slope at this Reynolds
opposite performance of nanofluid compared to other Reynolds number, as shown by the initial and final points, but it is hard to observe
numbers. The superiority of pure base fluid over nanofluid is also re­ due to significant fluctuations.
ported by Briclot et al. [27], who investigated the transient region. Thus, As in Fig. 9(e), the flow at this Reynolds number exhibits severe
it is more likely that the flow at Re = 1500 is in the transient region. It fluctuations, indicating turbulent regime. The local Nusselt numbers for
should be considered that this performance is not attributed to the different concentrations are close to each other, making it hard to

10
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

compare them. However, the general trend of increasing heat transfer that heat transfer improvement is higher from Reynolds numbers 1000
with increasing nanoparticle proportion persists. Also, the heat transfer to 2000 and then slightly lower. The trend of heat transfer increase
rate is comparatively higher than that of lower velocities and is observed suggests that the nanofluid flow in the tube with deep dimples un­
to be 2 to 6 times greater than that of a simple tube in most sections. The dergoes all three flow regimes: laminar, transient, and turbulent. This is
local Nusselt number decreases slightly from the beginning to the end of consistent with the studies of Briclot et al. [27], who report a decrease in
the tube, but this is not evident in the diagram due to the fluctuations. the thermal performance of nanofluid compared to distilled water in the
According to Fig. 9, it can be observed that the variation in con­ transient region. Their results imply that the flow regime changes and
centrations diminishes as the velocity increases. This phenomenon can enters the transient region around Re = 1500. Finally, the overall Nus­
be explained by the fact that the transfer of heat through conduction is selt number reaches its maximum value of 41.8 at Re = 2250 and the
directly proportional to the proximity of the particles, which results in a highest concentration.
longer duration for the heat transfer process. Therefore, conduction heat As illustrated in Fig. 11, nanofluid use at Re = 500 improved the
transfer is lower at high Reynolds numbers and higher at low Reynolds thermal performance of the tube with deep dimples. The Nusselt number
numbers, as evidenced by the results of different concentrations. increased by about 1.15 times with 1wt% Al2O3 nanofluid at this ve­
Moreover, the increase in nanofluid heat transfer compared to distilled locity. This improvement persisted at Re = 1000, where the thermal
water decreases with increasing velocity. In addition, for a more performance of the tube with nanofluid was similar to that of the simple
comprehensive comparison, the overall Nusselt number of the tube, one with distilled water. However, lower concentrations and distilled
which indicates the improvement in thermal performance, is presented water resulted in lower performance than the simple tube at this ve­
in Fig. 10. locity. Despite the error bars’ closeness, Fig. 11 shows a direct correla­
Fig. 10 shows that the overall Nusselt number is lower for Reynolds tion between nanofluid concentration and thermal performance
numbers below 1000, as expected from the local Nusselt number dia­ improvement. This diagram suggests that heat transfer always increases
grams. The enlarged part of the diagram reveals that only the nanofluid with increments in nanoparticle proportion. This increase is more
with the highest concentration has a thermal performance similar to the noticeable at Re = 2250, where the overall Nusselt number is about 3.42
simple tube. The error values are close to each other, indicating that heat times that of the simple tube.
transfer improves with increasing nanoparticle proportion compared to A notable observation from Fig. 9 is that the local Nusselt number at
distilled water. However, other studies, such as Gürdal et al. [32] and Re = 1000 is higher than at Re = 500. However, the opposite is true in
Cruz et al. [33], also report no effect of improvement methods and lower Fig. 11. This implies that nanofluid at Re = 1000 did not improve the
performance than the simple tube. In the worst case, i.e., pure base fluid thermal performance of the tube as much as at Re = 500. The flow
in the tube with deep dimples, the Nusselt number at Re = 1000 is only behavior also differs between distilled water and nanofluid at Re = 1500
0.92 times that of the simple tube, which is insignificant. Moreover, the in both diagrams, which warrants further investigation around this
overall Nusselt values of the nanofluid with the lowest concentration Reynolds number. The flow regime may transition from laminar to
approach those of distilled water as the flow velocity increases, reducing transient near Re = 1500 and then from transient to turbulent before Re
the positive thermal effect of nanoparticles. = 2000. However, further studies are required to investigate this tran­
As anticipated, the diagram shows an increasing trend of the Nusselt sition in detail for Reynolds numbers between 1000 and 2000.
number with an increase in the Reynolds number. The overall Nusselt
number has a considerable jump from 12.85 to 33.28 in the highest 4.1.2. Heat transfer characteristics around the dimpled tube
concentration between Reynolds numbers 1500 and 2000. In this range, This section investigates the heat transfer of the tube with deep
the overall Nusselt number increases by about 2.6 times, increasing by dimples in the circumferential direction. Previous sections have studied
about 1.7 times in the previous Reynolds numbers (1000 to 1500) and by the heat transfer along the tube length, but the dimples also affect the
nearly 1.3 times in the subsequent range (2000 to 2250). This indicates temperature distribution around the tube. Six thermocouples at 60-

Fig. 10. Overall Nusselt number in dimpled and smooth tubes at various Reynolds numbers.

11
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

Fig. 11. Comparing the Nusselt number for various working fluids in the dimple tube and distilled water in a simple tube.

degree intervals measured the temperature in three sections, as depicted coefficient corresponds to a lower temperature and vice versa. There­
in Fig. 5(a). By considering the tube length and dimple pitch, it reveals fore, hot spots are formed between 90 and 210 degrees. These hot spots
that there are 14 dimple-sets along the tube. Three thermocouples were decrease as the flow moves towards the end of the tube due to the
on the dimple, and three were at the outermost point of the test section. improvement mechanisms and the secondary flow and mixing caused by
Also, four thermocouples at 90-degree intervals measured the temper­ periodic dimples. However, there is a persistent hot spot at 150 degrees.
ature at the beginning and end of the fourth dimple-set to observe the At this Reynolds number, the flow deviation and the formation of
effect of a dimple on the flow and surface temperature. For this purpose, vortices impede the flow from following the streamlines and wetting
numerical simulations usually investigate heat transfer in the circum­ behind the dimple well. Moreover, the vortices exhibit inadequate in­
ferential direction because they can access the temperature of all points tensity to envelop the regions that are not adequately traversed by the
in the domain. In addition, experimental studies need many measure­ fluid at this particular velocity. This results in the existence of certain
ments for such an investigation. However, this study used a practical areas between 60 and 165 degrees that have not been sufficiently
method to experimentally investigate the temperature around the tube crossed by the flow, leading to the formation of hot spots. The presence
and study the effect of dimples on the flow temperature. To avoid of these hot spots significantly reduces the heat transfer and negatively
cluttering the diagrams and for better understanding, only the best result affects the tube’s performance.
(according to the Nusselt number diagrams along the tube), i.e., 1wt% Fig. 12(b) also illustrates the presence of hot spots, although the
alumina nanofluid, is shown, considering that four working fluids were intensity levels are lower than those in Fig. 12(a). As the velocity in­
used. creases, the improvement mechanisms enhance the temperature uni­
Fig. 12 depicts the convective heat transfer coefficient in different formity along the flow path. The periodic dimples create and intensify
dimple-sets at various Reynolds numbers. These diagrams display the vortices and secondary flow, which increase the mixing of the flow.
circumferential temperature of the tube from a cross-sectional view. It Furthermore, a symmetrical maximum point at 330 degrees and a
shows three sections of the tube for each Reynolds number. The points at minimum value at 150 degrees are noticeable. The convective heat
90, 210, and 330 degrees correspond to the centers of the dimples in a transfer coefficient values slightly rise compared to the previous Rey­
dimple-set. The convective heat transfer coefficient decreases as the nolds number, while the simple tube has a much higher coefficient at
distance from the origin of the tube increases, according to the theo­ this velocity. Therefore, Fig. 10 shows a higher overall Nusselt number
retical relations. This is due to the higher bulk fluid temperature and for Re = 1000 than Fig. 11, which has a lower value for Re = 500.
lower temperature difference near the end of the tube. To enhance the Fig. 12(c) shows that the novel geometry has better thermal per­
visual clarity of the diagrams and reduce visual complexity, the values formance than the plain tube in all circumference points at Re = 1500.
that are plotted represent the highest convective heat transfer coefficient This implies that as the velocity increases, the mechanisms responsible
at the closest distance from the origin of the simple tube. for enhancing heat transfer become more intense and lead to turbulence
Fig. 12(a) illustrates the convective heat transfer coefficient at generation. So, the dimples induce vortices and turbulence at a lower
various circumferential points of the tube at Reynolds 500. The diagram Reynolds number than regular tubes. The highest thermal performance
has line breaks because it only has data from 6 points of the tube is achieved on the dimple points, as expected. However, the temperature
circumference. More data would make the figures smoother. The figure around 150 degrees is still lower than the other values at this velocity.
indicates that the heat transfer coefficient on the dimples increases as This problem is solved at higher speeds, as illustrated in Fig. 12(d) and
the distance from the origin of the tube increases, unlike the simple tube, (e). These two figures have similar trends and only differ in the values
where it decreases. This suggests that the mechanisms responsible for related to Re = 2250. These figures show that the temperature distri­
enhancing heat transfer, as previously mentioned, are more efficient bution becomes more uniform with increasing velocity and approaching
towards the tube’s outlet. These points also match the maximum values the end of the tube. This indicates that the performance improvement
of Fig. 9. According to the theoretical relations, a higher heat transfer mechanisms find better conditions and intensify each other’s effects.

12
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

Fig. 12. Circumferential distribution of the convective heat transfer coefficient (HTC) for different dimple-sets and Reynolds numbers for Al2O31 wt%.

The location of the point on a dimple of the first dimple-set coincides in most points, but its overall Nusselt number is lower because of the hot
with the outermost point of the subsequent dimple-set. So, its periodic spots. This implies that the hot spots have a detrimental impact on the
repetition facilitates efficient mixing of the flow. These data not only heat transfer.
expand our knowledge for numerical simulations, particularly as the Four thermocouples were installed before and after the fourth
flow becomes turbulent at lower Reynolds numbers and cannot over­ dimple-set to examine the dimple effect on the flow downstream of the
come viscous forces when Re < 1000, but also help us understand the dimples, as Fig. 14 illustrates. The convective heat transfer coefficient
main mechanisms that enhance thermal performance. As the flow be­ before the dimples is depicted in Fig. 14(a). Cheraghi et al. [16] reported
comes turbulent near Re = 2000, it is essential to adjust the correlation that the flow velocity increases due to the cross-sectional area reduction
according to the flow type. when passing through the dimple, which enhances heat transfer. After
The heat transfer coefficient distribution around the tube for the dimple, the velocity decreases due to the cross-sectional area in­
different dimple-sets at various Reynolds numbers is presented in crease, which reduces heat transfer and restores the initial value. This is
Fig. 13. It reveals that the temperature distribution becomes more uni­ one of the heat transfer improvement mechanisms. This mechanism can
form with increasing velocity at any point of the tube. Also, the heat be experimentally observed by comparing Fig. 14(b) with (a). This
transfer enhancement increases from Fig. 13(a) to (c). These results section’s figures helped to examine three mechanisms of velocity in­
indicate that the dimples generate vortices along the tube, which crease, vortex generation, and flow rotation experimentally.
enhance the flow mixing along the flow path. In addition, this effect is
amplified by increasing the velocity.
4.2. Pressure drop penalty
Fig. 13(b) shows a symmetric temperature distribution, which is
absent in Fig. 13(a) and (c). This is due to the change of angle and pitch
In most methods of enhancing heat transfer, a penalty is typically
of the dimples, which induce a flow rotation in one direction for each
present. In the current study, this drawback is evident as a rise in
dimple-set. It gets balanced in the tube’s middle. Additionally, Fig. 13
pressure drop. This increment in pressure drop results in an increased
indicates that Re = 500 has a better radial performance than Re = 1000
pumping power requirement to maintain the desired flow rate. A

13
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

Fig. 13. Convective heat transfer coefficient (HTC) circumferential variation at Selected Reynolds numbers for Al2O31wt%.

Fig. 14. Circumferential distribution of the convective heat transfer coefficient (HTC) for the inlet and outlet of dimple-set number four at various Reynolds numbers
for Al2O3 1 wt%.

flowmeter was installed after the test tube to prevent fluctuations inside concentration leads to a higher pressure drop, except at Re = 1500. The
the test section. The flow rates obtained are precisely the desired ve­ data at Re = 1500 deviates from the general trend, as does the data
locity downstream, which is advantageous for industrial applications. related to thermal performance. The friction factor changes after Re =
Furthermore, the pattern of pressure drop and the friction factor of the 1500, as shown in Fig. 15. It rises until Re = 2000 and then falls after­
flow can serve as valuable tools for examining various flow regimes ward. Therefore, the friction factor changes in a pattern of decrease,
within the tube. increase, and decrease again in the studied Reynolds numbers range.
Fig. 15 provides a detailed analysis of the friction factor, comparing This pattern is similar to the flow regimes in the tube (laminar, transi­
its values across different working fluids and specifically contrasting its tional, and turbulent) presented by Poiseuille, Blasius, and Hrycak et al.
behavior in plain versus dimpled tubes. The friction factor varies [53]. Also, the diagrams related to heat transfer along the tube suggest
significantly at Reynolds numbers 500 and 1000 due to the low velocity that the flow undergoes all three regimes in the Reynolds range under
and the obstacles in the fluid path created by the dimples. The dimples study. The flow enters the transitional regime around Re = 1500 and
have a negative effect on the friction factor at low Reynolds numbers, becomes fully turbulent after Re = 2000. Other studies, such as [45],
which diminishes as the velocity increases. The error bar is close for all reported similar pressure drop values for nanofluid and pure water.
cases, so it is only possible to state generally that increasing Fig. 16 provides a detailed illustration of the trend associated with

14
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

Fig. 15. Experimental and theoretical analysis of friction factor in tubes with and without dimpled Surfaces for different Reynolds numbers.

Fig. 16. Friction factor comparison between dimpled tubes with various nanofluid concentrations and smooth tube.

varying flow regimes. It presents the comparative ratio of the friction pressure drop increase in other thermal performance improvement
factor in a tube featuring deep dimples and diverse working fluids methods. Therefore, the tube with deep dimples can achieve higher heat
relative to the corresponding values for a plain geometry. The ratio transfer with lower pressure drop, which can be advantageous for in­
decreases from Reynolds numbers 500 to 1000, increases from Reynolds dustrial applications and pump power consumption.
numbers 1000 to 2000, and declines again after Re = 2000. The figure
also indicates that the friction factor and pressure drop increase with
4.3. Overall performance of the dimpled tube
nanofluid concentration, except at Re = 1500. The reason for this
exception needs further investigation, but it may be related to the
The thermal and hydraulic characteristics of the tube have been
transitional regime. The friction factor increases by about 3.3 times at Re
investigated independently, revealing both the merits and penalties of
= 500 and approximately 2.6 times at Re = 2000 for the highest nano­
this approach. However, the practicality of this tube, considering its
fluid concentration compared to the plain tube. The friction factor in­
intrinsic characteristics and possible drawbacks, is yet to be determined.
creases nearly 1.16 times at Re = 500 for the nanofluid compared to
Consequently, we employ the Performance Evaluation Criteria (PEC)
distilled water in the dimpled tube. The pressure drop increase in
and Equation 14 to evaluate the usefulness of utilizing nanofluid in a
nanofluid is due to the higher viscosity and the enhanced turbulence by
tube featuring deep dimples. This parameter allows for a simultaneous
the nanoparticles [24]. However, these values are lower than the
investigation of the effects on heat transfer enhancement and pressure

15
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

Fig. 17. Performance evaluation criteria (PEC) for the modified tube across various Reynolds numbers.

drop. significant effect on overall performance in the studied range, even

As Fig. 17 illustrates, nanofluid use in deep-dimpled tubes is ineffi­ though overall performance improves. At a Reynolds number of 2250,
cient for Reynolds numbers below 1000, so the details of different the use of 1wt% of alumina nanofluid in deep-dimpled tubes shows a
concentrations in these Reynolds numbers are not precisely shown. maximum PEC value of 2.80, which is slightly higher than the PEC value
However, nanofluid use and optimized tube geometry are beneficial of 2.71 obtained with distilled water under the same conditions. PEC
from Re = 1000 onwards. Generally, higher nanofluid concentration also increases with Reynolds number directly. It is worth mentioning
enhances overall performance (PEC) despite the small error bar values. that at Re = 2250, a concentration of 0.1wt% exhibits poorer perfor­
At Re = 1500, unlike other Reynolds numbers, nanofluid utilization does mance than distilled water. The observed data indicates a reduction in
not improve performance, and distilled water is more efficient, as the conductive heat transfer corresponding to an increase in velocity at
thermal and hydraulic performance suggests. The transient region cau­ lower concentrations. The data implies that conductive heat transfer
ses this trend in the figure, based on the study [27], and the lower diminishes as velocity increases at lower concentrations. As a result, an
thermal performance of nanofluid compared to distilled water in that elevated nanofluid concentration becomes necessary to compensate for
region. In higher Reynolds numbers, nanofluid and pure base water the increased velocities.
values are similar, indicating that concentration increase has no This study examined the effect of nanofluid concentrations on the

Fig. 18. A relative assessment of the performance evaluation criteria for the dimpled tube using distilled water as a reference.

16
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

performance of a tube with deep dimples using the PECNano parameter. • Heat transfer enhancement and pressure drop values increase with
This parameter is analogous to the conventional PEC, but it uses the nanofluid concentration in all cases except Re = 1500. However, this
properties of distilled water in the dimpled tube instead of the plain trend only applies to PEC for Reynolds numbers 500 and 1000. Re =
tube. The diagram shows the improvement in the tube performance for 1500 behaves differently, and concentrations of 0.1 and 0.5 % values
each Reynolds number due to different nanofluid concentrations. vary at Reynolds numbers 2000 and 2250. The highest PEC of 2.80 is
The effect of nanofluid on the overall performance of the tube with achieved at Re = 2250 and 1wt%
deep dimple, regardless of the tube shape, is illustrated in Fig. 18. A • Al2O3 − γ nanofluid can enhance the Nusselt number by 1.07 to 3.42
deviation from the general trend is observed at Re = 1500 in this dia­ times in a tube with deep dimples. So, the maximum Nusselt number
gram and others. The figure indicates that the nanofluid effect di­ of 41.80 occurs at the highest concentration and Reynolds number.
minishes as the flow velocity increases. A similar decrease in thermal • Implementing this passive heat transfer enhancement method comes
performance improvement with increasing Reynolds number is reported at the cost of an increment in pressure drop. The additional pumping
in the study [28], attributed to the reduced Brownian motion of parti­ power needed for this increase should be taken into account. The
cles. This implies that conductive heat transfer requires more time to friction factor of the tube also rises, going from about 1.38 times at
take effect. In addition, in Re = 2000, the overall performance for 0.5 % Re = 1500 to almost 3.32 times at a concentration of 1 % for Re =
concentration is lower than that for 0.1 %, suggesting that the pressure 500, although this value is still considered acceptable. However, at a
drop increment outweighs the heat transfer enhancement. The same concentration of 0.5 % and the Reynolds number of 1500, the friction
phenomenon occurs for 0.1wt% at Re = 2250. The maximum PEC of 1.1 factor experiences a minimum value of 0.065.
is achieved at Re = 500 with 1 % concentration. When only the nano­ • The heat transfer enhancement in the tube with deep dimples was
fluid effect on overall performance is considered, low PEC improvement examined through the measurement of longitudinal and circumfer­
values are also found in [37] for low nanofluid concentrations, showing ential heat transfer, revealing three primary mechanisms: local ve­
consistency with the results. locity augmentation, vortex generation, and flow rotation and
Given the specified PEC values and pressure drop for this particular mixing. These mechanisms were explored using longitudinal and
tube geometry, it is deduced that the deep-dimpled tube presents a circumferential figures. The diagram corresponding to the dimple-
feasible substitute for conventional tubes in standard heat exchangers. set’s inlet and outlet demonstrated the effect of velocity augmenta­
These tubes maintain low pump power increment by reducing pressure tion near the dimple tip. Furthermore, figures depicting heat transfer
drop. These tubes enable the production of a heat exchanger with either on the dimple at various tube lengths revealed the impact of vortex
equal size and higher heat duty or smaller size and equal heat duty. A generation and flow rotation. Longitudinal diagrams were employed
higher heat capacity exchanger directly lowers fuel consumption and to observe the effects associated with the conductive heat transfer of
pollutant emissions. Conversely, a smaller heat exchanger indirectly nanoparticles and other mechanisms that contribute to nanofluid
lowers pollutant emissions by reducing investment costs and the con­ heat transfer enhancement.
struction process. Higher Reynolds numbers require further study of the • Based on different figures of heat transfer along the geometry and
deep-dimpled tube for industrial applications. Other studies on research [27], the flow in the tube with deep dimples undergoes
improving the tube’s heat transfer and the present study’s characteris­ laminar, transitional, and turbulent regimes. The local Nusselt
tics suggest that the highest PEC occurs in the Reynolds numbers range number figures show vortex formation with increasing velocity,
of 2500 to 3500. Additionally, the results for Re = 1500 recommend a causing turbulence in the flow. The heat transfer coefficient figures
more accurate investigation of the Reynolds numbers range of 1000 to around the tube indicate a more uniform temperature distribution at
2000 for future research. the end of the tube and higher velocity. This suggests the occurrence
of turbulence and the formation of vortices, which intensify and
5. Conclusion rotate within the flow. The flow reaches the transitional regime 1000
< ReCr < 1500 around Re = 1500 and the turbulent regime around Re
This research aims to assess the efficiency of a tube featuring deep = 2000.
dimples, representing a novel tube geometry when gamma alumina
nanofluid of 0.1 wt%, 0.5 wt%, and 1 wt% flows through it. The The study demonstrated that utilizing a nanofluid and a tube with
enhanced thermophysical properties of nanofluids facilitate the poten­ deep dimples resulted in superior performance in Reynolds numbers
tial for novel tube designs to supplant traditional heat exchangers, ranging from 1000 to 2250 compared to a smooth tube. The results
thereby augmenting their heat capacity. This research conducts an indicated that the modified tube geometry enhanced heat transfer and
experimental examination of various mechanisms that enhance heat had a low pressure drop penalty, making it more suitable for industrial
transfer, focusing on the less explored laminar flow range for the given applications and reducing energy consumption and environmental pol­
tube dimensions. The circumferential heat transfer of the tube is also lutants such as CO2, a current topic of discussion. Another benefit of this
considered in this study. The experiments are carried out under constant method is that it creates and intensifies turbulence at both low and high
heat flux and at Reynolds numbers ranging from 500 to 2250. This range velocities, respectively, which improves heat transfer. An unusual trend
covers only the laminar regime for conventional tubes. However, was observed in Re = 1500 when using nanofluid, which requires further
research on this unique geometry has identified the existence of laminar, investigation in the range of 1000 to 2000. Future research can also
transitional, and turbulent flow regimes within the Reynolds number consider the effect of higher concentrations, constant and alternating
range. The tube’s geometry is derived from the research conducted by magnetic field, Re = above 2500, and hybrid nanofluids on the heat
Cheraghi et al. [16]. A smooth tube is employed to validate the findings transfer improvement and the PEC. Future research could focus on the
from the experimental apparatus. Re > 6000, the transient flow regime, i.e., Re ≈ 1500, higher nanofluid
The following results have been obtained from this study: concentrations, and two-phase flow.

• In the simple tube, the Nusselt numbers and friction factor obtained Declaration of generative AI and AI-assisted technologies in the
from experimental data and reference equations showed good writing process
agreement for all Reynolds numbers examined. Furthermore, the
analysis indicates that using the nanofluid and the tube with deep During the preparation of this work the authors used Bing AI in order
dimples is ineffective for Reynolds numbers below 1000. However, to improve readability of the text. After using this tool, the authors
substituting nanofluid for distilled water marginally enhances over­ reviewed and edited the content as needed and take full responsibility
all performance. for the content of the publication.

17
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

CRediT authorship contribution statement [17] L. Zhang, W. Xiong, J. Zheng, Z. Liang, S. Xie, Numerical analysis of heat transfer
enhancement and flow characteristics inside cross-combined ellipsoidal dimple
tubes, Case Stud. Therm. Eng. 25 (2021) 100937, https://doi.org/10.1016/j.
Alireza Khashaei: Writing – original draft, Validation, Software, csite.2021.100937.
Methodology, Investigation, Formal analysis, Data curation. Moham­ [18] S. Xie, Z. Guo, Y. Gong, C. Dong, J. Liu, L. Ren, Numerical investigation of thermal-
mad Ameri: Writing – review & editing, Visualization, Supervision, hydraulic performance of a heat exchanger tube with helical dimples, Int. J.
Therm. Sci. 177 (2022) 107530.
Resources, Methodology, Investigation, Formal analysis, Data curation, [19] T. Wang, Q. Zhang, K. Song, K. Zhang, M. Su, X. Wu, Case Studies in Thermal
Conceptualization. Shahram Azizifar: Validation, Supervision, Meth­ Engineering Thermodynamic characteristics of a novel combination of three-start
odology, Investigation, Conceptualization. twisted tube and oval dimples, Case Stud. Therm. Eng. 37 (2022) 102284, https://
doi.org/10.1016/j.csite.2022.102284.
[20] S. Kia, S. Khanmohammadi, A. Jahangiri, Experimental and numerical
investigation on heat transfer and pressure drop of SiO2 and Al2O3 oil-based
Declaration of competing interest nanofluid characteristics through the different helical tubes under constant heat
fluxes, Int. J. Therm. Sci. 185 (2023) 108082, https://doi.org/10.1016/j.
The authors declare that they have no known competing financial ijthermalsci.2022.108082.
[21] J.C. Maxwell, A Treatise on Electricity and Magnetism, Clarendon Press, Oxford,
interests or personal relationships that could have appeared to influence 1873.
the work reported in this paper. [22] S.U.S. Choi, J.A. Eastman, Enhancing Thermal Conductivity of Fluids with
Nanoparticles, Argonne National Lab.(ANL), Argonne, IL (United States), 1995.
[23] A.S. Hatwar, V.M. Kriplani, A review on heat transfer enhancement with nanofluid,
Data availability Int. J. Adv. Res. Sci. Eng. 3 (2014) 175–183.
[24] E. Gürsoy, M. Gürdal, H. Kadir Pazarlıoğlu, A. Dağdeviren, M. Tekir, K. Arslan,
The data that has been used is confidential. E. Gedik, A. Adriana Minea, H. Kurt, Effect of magnetic field locations on thermo-
magnetic convection performance of Fe3O4/H2O ferrofluid flowing in a novel
dimpled tube: An experimental study, Appl. Therm. Eng. 226 (2023), https://doi.
org/10.1016/j.applthermaleng.2023.120305.
References [25] T.T. Loong, H. Salleh, A. Khalid, H. Koten, Thermal performance evaluation for
different type of metal oxide water based nanofluids, Case Stud. Therm. Eng. 27
(2021) 101288, https://doi.org/10.1016/j.csite.2021.101288.
[1] REN21, Renewables 2022 Global Status, 2022. https://www.ren21.net/gsr-2022/.
[26] M.S. Bretado de los Rios, C.I. Rivera-Solorio, A.J. García-Cuéllar, Thermal
[2] R. Barzegarian, A. Aloueyan, T. Yousefi, Thermal performance augmentation using
performance of a parabolic trough linear collector using Al2O3/H2O nanofluids,
water based Al2O3-gamma nanofluid in a horizontal shell and tube heat exchanger
Renew. Energy. 122 (2018) 665–673, https://doi.org/10.1016/j.
under forced circulation, Int. Commun. Heat Mass Transf. 86 (2017) 52–59,
renene.2018.01.094.
https://doi.org/10.1016/j.icheatmasstransfer.2017.05.021.
[27] A. Briclot, J.F. Henry, C. Popa, C.T. Nguyen, S. Fohanno, Experimental
[3] K.L. Liaw, J.C. Kurnia, A.P. Sasmito, Laminar convective heat transfer in helical
investigation of the heat and fluid flow of an Al2O3-water nanofluid in the laminar-
twisted multilobe tubes, Case Stud. Therm. Eng. 39 (2022) 102459, https://doi.
turbulent transition region, Int. J. Therm. Sci. 158 (2020) 106546, https://doi.org/
org/10.1016/j.csite.2022.102459.
10.1016/j.ijthermalsci.2020.106546.
[4] A.H. Altun, H. Nacak, E. Canli, Effects of trapezoidal and twisted trapezoidal tapes
[28] S. Zhang, L. Lu, T. Wen, C. Dong, Turbulent heat transfer and flow analysis of
on turbulent heat transfer in tubes, Appl. Therm. Eng. 211 (2022) 118386, https://
hybrid Al2O3-CuO/water nanofluid: An experiment and CFD simulation study,
doi.org/10.1016/j.applthermaleng.2022.118386.
Appl. Therm. Eng. 188 (2021) 116589, https://doi.org/10.1016/j.
[5] M. Gürdal, H.K. Pazarlıoğlu, M. Tekir, K. Arslan, E. Gedik, Numerical investigation
applthermaleng.2021.116589.
on turbulent flow and heat transfer characteristics of ferro-nanofluid flowing in
[29] M. Gürdal, H.K. Pazarlıoğlu, M. Tekir, K. Arslan, E. Gedik, Numerical investigation
dimpled tube under magnetic field effect, Appl. Therm. Eng. 200 (2022) 117655,
on turbulent flow and heat transfer characteristics of ferro-nanofluid flowing in
https://doi.org/10.1016/j.applthermaleng.2021.117655.
dimpled tube under magnetic field effect, Appl. Therm. Eng. 200 (2022), https://
[6] S. Rabiei, R. Khosravi, M. Bahiraei, M. Raziei, A. Ahmadian Hosseini, Thermal and
doi.org/10.1016/j.applthermaleng.2021.117655.
hydraulic characteristics of a hybrid nanofluid containing graphene sheets
[30] F.M. Altunay, H.K. Pazarlıoğlu, M. Gürdal, M. Tekir, K. Arslan, E. Gedik, Thermal
decorated with platinum through a new wavy cylindrical microchannel, Appl.
performance of Fe3O4/water nanofluid flow in a newly designed dimpled tube
Therm. Eng. 181 (2020) 115981, https://doi.org/10.1016/j.
under the influence of non-uniform magnetic field, Int. J. Therm. Sci. 179 (2022)
applthermaleng.2020.115981.
107651.
[7] R.B. Manoram, R.S. Moorthy, R. Ragunathan, Investigation on influence of dimpled
[31] F. Ahmed, M.A. Abir, A.S.M. Redwan, A.A. Bhuiyan, A.S. Mollah, The impact of D-
surfaces on heat transfer enhancement and friction factor in solar water heater,
shaped jaggedness on heat transfer enhancement technique using Al2O3 based
J. Therm. Anal. Calorim. 145 (2021) 541–558.
nanoparticles, Int. J. Thermofluids. 10 (2021) 100069, https://doi.org/10.1016/j.
[8] M. Amani, M. Ameri, A. Kasaeian, Investigating the convection heat transfer of
ijft.2021.100069.
Fe3O4 nanofluid in a porous metal foam tube under constant magnetic field, Exp.
[32] M. Gürdal, H.K. Pazarlıoğlu, M. Tekir, F.M. Altunay, K. Arslan, E. Gedik,
Therm. Fluid Sci. 82 (2017) 439–449.
Implementation of hybrid nanofluid flowing in dimpled tube subjected to magnetic
[9] M. Chandrasekar, S. Suresh, A.C. Bose, Experimental studies on heat transfer and
field, Int. Commun. Heat Mass Transf. 134 (2022), https://doi.org/10.1016/j.
friction factor characteristics of Al2O3/water nanofluid in a circular pipe under
icheatmasstransfer.2022.106032.
laminar flow with wire coil inserts, Exp. Therm. Fluid Sci. 34 (2010) 122–130,
[33] G.G. Cruz, M.A.A. Mendes, J.M.C. Pereira, H. Santos, A. Nikulin, A.S. Moita,
https://doi.org/10.1016/j.expthermflusci.2009.10.001.
Experimental and numerical characterization of single-phase pressure drop and
[10] S. Suresh, M. Chandrasekar, S.C. Sekhar, Experimental studies on heat transfer and
heat transfer enhancement in helical corrugated tubes, Int. J. Heat Mass Transf.
friction factor characteristics of CuO/water nanofluid under turbulent flow in a
179 (2021) 121632, https://doi.org/10.1016/j.ijheatmasstransfer.2021.121632.
helically dimpled tube, Exp. Therm. Fluid Sci. 35 (2011) 542–549, https://doi.org/
[34] F. Andrade, A.S. Moita, A. Nikulin, A.L.N. Moreira, H. Santos, Experimental
10.1016/j.expthermflusci.2010.12.008.
investigation on heat transfer and pressure drop of internal flow in corrugated
[11] M. Moravej, M.V. Bozorg, Y. Guan, L.K.B. Li, M.H. Doranehgard, K. Hong,
tubes, Int. J. Heat Mass Transf. 140 (2019) 940–955, https://doi.org/10.1016/j.
Q. Xiong, Enhancing the efficiency of a symmetric flat-plate solar collector via the
ijheatmasstransfer.2019.06.025.
use of rutile TiO2-water nanofluids, Sustain. Energy Technol. Assessments. 40
[35] T. Dagdevir, V. Ozceyhan, An experimental study on heat transfer enhancement
(2020), https://doi.org/10.1016/j.seta.2020.100783.
and flow characteristics of a tube with plain, perforated and dimpled twisted tape
[12] S. Suresh, K.P. Venkitaraj, P. Selvakumar, M. Chandrasekar, Effect of Al 2O 3-Cu/
inserts, Int. J. Therm. Sci. 159 (2021), https://doi.org/10.1016/j.
water hybrid nanofluid in heat transfer, Exp. Therm. Fluid Sci. 38 (2012) 54–60,
ijthermalsci.2020.106564.
https://doi.org/10.1016/j.expthermflusci.2011.11.007.
[36] T. Dagdevir, M. Uyanik, V. Ozceyhan, The experimental thermal and hydraulic
[13] A. García, J.P. Solano, P.G. Vicente, A. Viedma, The influence of artificial
performance analyses for the location of perforations and dimples on the twisted
roughness shape on heat transfer enhancement: corrugated tubes, dimpled tubes
tapes in twisted tape inserted tube, Int. J. Therm. Sci. 167 (2021) 107033.
and wire coils, Appl. Therm. Eng. 35 (2012) 196–201, https://doi.org/10.1016/j.
[37] B. Sun, Y. Guo, D. Yang, H. Li, The effect of constant magnetic field on convective
applthermaleng.2011.10.030.
heat transfer of Fe3O4/water magnetic nanofluid in horizontal circular tubes,
[14] R. Sabir, M.M. Khan, N.A. Sheikh, I.U. Ahad, D. Brabazon, Assessment of thermo-
Appl. Therm. Eng. 171 (2020) 114920, https://doi.org/10.1016/j.
hydraulic performance of inward dimpled tubes with variation in angular
applthermaleng.2020.114920.
orientations, Appl. Therm. Eng. 170 (2020) 115040, https://doi.org/10.1016/j.
[38] L.S. Sundar, M.K. Singh, V. Punnaiah, A.C.M. Sousa, Experimental investigation of
applthermaleng.2020.115040.
Al2O3/water nanofluids on the effectiveness of solar flat-plate collectors with and
[15] S. Xie, Z. Liang, J. Zhang, L. Zhang, Y. Wang, H. Ding, Numerical investigation on
without twisted tape inserts, Renew. Energy. 119 (2018) 820–833, https://doi.org/
flow and heat transfer in dimpled tube with teardrop dimples, Int. J. Heat Mass
10.1016/j.renene.2017.10.056.
Transf. 131 (2019) 713–723, https://doi.org/10.1016/j.
[39] S. Azizifar, M. Ameri, I. Behroyan, Experimental study of the effect of metal foams
ijheatmasstransfer.2018.11.112.
on subcooled flow boiling heat transfer of water and developing a correlation for
[16] M.H. Cheraghi, M. Ameri, M. Shahabadi, Numerical study on the heat transfer
predicting heat transfer, Chem. Eng. Sci. 262 (2022) 118032.
enhancement and pressure drop inside deep dimpled tubes, Int. J. Heat Mass
Transf. 147 (2020) 118845, https://doi.org/10.1016/j.
ijheatmasstransfer.2019.118845.

18
A. Khashaei et al. International Journal of Thermo uids 24 (2024) 100827

[40] R.L. Webb, Performance evaluation criteria for use of enhanced heat transfer [47] P.K. Kanti, K.V. Sharma, Z. Said, M. Gupta, Experimental investigation on thermo-
surfaces in heat exchanger design, Int. J. Heat Mass Transf. 24 (1981) 715–726, hydraulic performance of water-based fly ash–Cu hybrid nanofluid flow in a pipe at
https://doi.org/10.1016/0017-9310(81)90015-6. various inlet fluid temperatures, Int. Commun. Heat Mass Transf. 124 (2021)
[41] S.J. Kline, Describing uncertainty in single sample experiments, Mech. Eng. 75 105238, https://doi.org/10.1016/j.icheatmasstransfer.2021.105238.
(1953) 3–8. [48] B.C. Pak, Y.I. Cho, Hydrodynamic and heat transfer study of dispersed fluids with
[42] A. Bejan, A.D. Kraus, Heat Transfer Handbook, John Wiley & Sons, 2003. submicron metallic oxide particles, Exp. Heat Transf. an Int. J. 11 (1998) 151–170.
[43] Y.A. Cengel, M.A. Boles, M. Kanoğlu, Thermodynamics: an Engineering Approach, [49] M.J. Babu, N. Sandeep, S. Saleem, Free convective MHD Cattaneo–Christov flow
McGraw-hill, New York, 2011. over three different geometries with thermophoresis and Brownian motion,
[44] S. Kim, B. Tserengombo, S.H. Choi, J. Noh, S. Huh, B. Choi, H. Chung, J. Kim, Alexandria Eng. J. 56 (2017) 659–669.
H. Jeong, Experimental investigation of heat transfer coefficient with Al2O3 [50] D.A. Drew, S.L. Passman, Theory of Multicomponent Fluids, Springer Science &
nanofluid in small diameter tubes, Appl. Therm. Eng. 146 (2019) 346–355, https:// Business Media, 2006.
doi.org/10.1016/j.applthermaleng.2018.10.001. [51] A. Khashaei, M. Ameri, S. Azizifar, M.H. Cheraghi, Experimental investigation on
[45] W. Ajeeb, R.R.S. Thieleke da Silva, S.M.S. Murshed, Experimental investigation of the heat transfer augmentation and friction factor inside tube enhanced with deep
heat transfer performance of Al2O3 nanofluids in a compact plate heat exchanger, dimples, Int. Commun. Heat Mass Transf. 149 (2023) 107149, https://doi.org/
Appl. Therm. Eng. 218 (2023) 119321, https://doi.org/10.1016/j. 10.1016/j.icheatmasstransfer.2023.107149.
applthermaleng.2022.119321. [52] S. Aberoumand, A. Jafarimoghaddam, Mixed convection heat transfer of
[46] N. Acharya, A.J. Chamkha, On the magnetohydrodynamic Al2O3-water nanofluid nanofluids inside curved tubes: An experimental study, Appl. Therm. Eng. 108
flow through parallel fins enclosed inside a partially heated hexagonal cavity, Int. (2016) 967–979, https://doi.org/10.1016/j.applthermaleng.2016.06.032.
Commun. Heat Mass Transf. 132 (2022) 105885, https://doi.org/10.1016/j. [53] ASM, Materials Selection and Design, American Society for Metals, 1997.
icheatmasstransfer.2022.105885.

19