1. Introduction
Convective transport has had a noteworthy impact io many real applications such as energy-producing plants, energy distribution systems, and environmental problems [
1]. The convective transport can be induced by diffusion and advection where the diffusion refers to the random Brownian motion, whereas the advection refers to the transportation of heat by a larger-scale motion. Due to the substance of its theory and concepts, the ideas of convective transport have been extended to many fluid models that cover Newtonian and non-Newtonian fluid types. Additionally, the model of Eyring Powell fluid has been investigated by Al Jabali et al. [
2] and Khalil et al. [
3], whereas the Jeffrey model was considered by Shehzad et al. [
4] and Kasim et al. [
5]. Other interesting models such as viscoelastic can be found in Kasim et al. [
6]. Additional previously studied models include the viscous model [
7], Casson model [
8], and micropolar model [
9]. The development of the fluid flow modelling is continued with the discovery of the upgraded thermal fluid. In the late 20th century, nanofluids were introduced in an effort to host the concept of the diffusion of nanoscale particles into fluids. The presence of nanofluids has been shown to increase the capacity for thermal conductivity. A pioneering study on nanofluids was conducted by Choi and Eastman [
10]. Since then, this field of research has continued to receive high attention from researchers both experimentally and theoretically. This situation can be seen from the increase in publications every year where the contribution is based on the analysis of the fluid characteristics induced by the different geometries such as rectangular enclosure [
11,
12,
13], stretching/shrinking sheet [
14,
15,
16,
17], thin needle [
18,
19], rotating disk [
20], asymmetric wavy channel [
21], vertical plate [
22], and moving inclined [
23]. These investigations also involve several significant effects including chemical reactions, magnetic fields, viscous dissipation circumstance, Dufour and Soret consequence, activation energy, and Joule heating. Simulations investigating the convective transport and heat transfer in cryogenic fuel tanks have been documented in refs. [
24,
25] where the improved algorithms determining the output were proposed. The analysis of heat transfer characteristics in a thermal energy storage system using single and multi-phase cooled heat sinks is presented in Alireza [
26] where the evolution of the experimental, numerical, and computational efforts on energy storage was presented in detail.
The hybrid nanofluid offers a higher heat transfer rate in comparison to conventional nanofluid. This fluid is widely used in several applications and mainly can be found in heat exchangers activities, vehicle brake systems, solar industries, and also in refrigerator production [
27]. The earlier studies which applied the hybrid nanoparticles in their investigations were Turcu et al. [
28] and Jana et al. [
29]. The composition of Cu-Al
2O
3/water was successfully studied by Suresh et al. [
30] in discovering fluid with good thermal conductivity. The input stated, that despite having low thermal conductivity, Al
2O
3 offers a good chemical motionlessness in alumina which compromises a stable composition. A further study on combining Al
2O
3 with supplementary nanoparticles was established by Singh and Sarkar [
31], Farhana et al. [
32], and Takabi and Salehi [
33]. Some other interesting publications on the development of the theoretical study of hybrid nanofluid can be viewed in refs. [
34,
35,
36,
37,
38,
39,
40,
41,
42] where the investigations were performed using different particles, approaches, and various surfaces. Furthermore, some other publications reviewing the progress of hybrid nanofluids are documented in refs. [
43,
44,
45,
46].
Taking advantage of the progress of nanotechnology, scientists first created ferrofluid to counter the problem of logistics of bringing rocket fuel. The idea of this innovation is based on how to direct the fluid in space, and it is supported by the concept of the magnetic fluid can be well-ordered by a magnetic field. Then, the liquid fuel is mixed with the ferrofluid in the system together with the external magnetic field. The liquid with particles of Fe3O4 in nanometres size that happened in conventional base fluids is referred to as ferro-nanofluids or sometimes it is just called ferrofluid. Similar to nanofluids, ferrofluid also has high thermal conductivities and a great heat transfer rate, which is very important in device production. In fact, this technological fluid can be found in the production of hard drives where it is used to seal the interior of devices in order to avoid dust or external source damage to the delicate plate.
Due to the importance and many applications, interest in investigating ferrofluid came into demand for both experimental and simulated procedures. Khan et al. [
47] provided a solution for heat transfer analysis of ferrofluid in the presence of viscous dissipation and concluded that kerosene-based ferrofluids developed high skin friction and Nusselt numbers compared to water-based ferrofluids. The investigation of ferrofluid, later on, was extended by other researchers by considering different surfaces. For instance, Qasim et al. [
48] consider the stretching cylinder as the surface where the fluid is moving while Mehrez and Cafsi [
49] proposed the flow in a rectangular channel. Further research on the topic of ferrofluid has been proposed by Sekar and Raju [
50] and Hamid et al. [
51] where the ferrofluid is considered together with dust particles and was studied as a two-phase flow. Goshayeshi et al. [
52] considered the effect of particle size and types in their investigation while Rashad [
53] investigated the anisotropic impact on the flow field. In the same year, the same author documented the analysis of ferrofluid under thermal radiation and MHD circumstances [
54]. The demand for investigating ferrofluid led to the development of new ideas and this progress is detailed out in refs. [
55,
56,
57,
58,
59,
60,
61].
The interest in innovating the effective fluid has led to the discovery of hybrid ferrofluids where the investigations have involved multiple nanoparticles scattered in ferrofluid. Chu et al. [
62] provided a solution for hybrid ferrofluids along with multi-wall carbon nanotubes (MWCNT) in a cavity for natural convection by applying a finite element scheme. It is revealed that the hybridization of ferrite nanoparticles uplifts in a physical thermos to see the studied flow. In the same year, Kumar et al. [
63] proposed an investigation under the topic of MHD hybrid ferrofluid together with the effect of irregular heat source/sink towards the flow of the radiative thin film. The output declared from the hybrid ferrofluid intensifies the heat transfer rate in comparison with the conventional ferrofluid. A very recent study on this topic was presented by Anuar et al. [
64] where their study of hybrid ferrofluid (CoFe
2O
4–Fe
3O
4/water) was focused on exponentially deformable sheets in stagnation point region. Apart from these mentioned studies, other interesting reports on hybrid ferrofluids can be found in refs. [
65,
66,
67]. One real application that can be highlighted from the uses of hybrid ferrofluids is that it can be used as seeds for acid mine drainage (AMD) treatment. The use of a hybrid ferrofluid will provide a sustainable remedy for wastewater treatment, especially for AMD since the presence of AMD will cause severe destruction to the environment and predominantly disturb human life, aquatic organisms, animals, and also plants. The special properties of hybrid ferrofluids are unique magnetized characteristics with additional low toxicity and strong capacity for contaminant removal, which have led to enhancing the chemical reactivities [
68].
As mentioned previously, the study on fluid flow normally deviates from its geometries. The pioneering study on infinite rotating disk was performed by Von Kármán [
69] where the flow is taken as a viscous flow in which the disk rotates through uniform angular velocity. The Navier–Stokes equations representing the mathematical model are reduced to ordinary differential equations by adopting the similarity variables. The ideas from Von Kármán’s work have been extended by Fang [
70] where the investigation was focused on the flow over the stretchable rotating disk and later Fang and Zhang [
71] extend to two infinite stretchable disks. Turkyilmazoglu [
72] considered the presence of a magnetic field together with viscous dissipation and joule heating in the investigation after considering the unsteady flow over a rotating disk with outer radial flow [
73]. The unsteady flow over a decelerating rotating sphere was then established by Turkyilmazoglu [
74]. Another analysis of flow over a rotating disk was documented in refs. [
75,
76,
77,
78]. The study of flow over a rotating disk is still in demand due to its available applications in industries and engineering activities such as the viscometers instrument, turbines industries, rotating disk electrodes, mechanical devices, and Brake rotors [
79,
80].
Motivated by the studies available in the literature and fulfilling the gap in the study of hybrid ferrofluids, this study discusses the investigation on the unsteady flow of hybrid ferrofluid flow over a rotating disk. It is worth mentioning that the non-unique solutions are available but are restricted to pertinent parameters and the stability analysis affirms the physical solution.
2. Mathematical Formulation
Consider the unsteady flow of hybrid ferrofluids with different base fluids (Fe
3O
4-CoFe
2O
4/H
2O-EG and Fe
3O
4-CoFe
2O
4/H
2O) induced by a rotating disk as illustrated in
Figure 1 with the imposition of suction/injection effect. The disk rotates with the velocity,
while being stretched/shrunk with velocity,
. Moreover, it is considered that the suction/injection effect is embedded with the mass flux velocity,
. The constant surface and ambient temperatures are symbolized as
and
, respectively. A further assumption is that the magnetic field is applied normally to the disk such that
with constant magnetic strength
(see
Appendix A for the derivation of magnetic field).
Thus, the governing equations in cylindrical coordinates
are [
77,
80,
81]:
subject to:
where
are the velocities for
directions while
is the temperature. It is further assumed that
where
and
are the respective time and angular velocity (constant), accordingly. Furthermore,
represents the unsteadiness strength (constant) while
is the mass velocity with constant
. Moreover,
represents a static disk while
stand for the stretching/shrinking disk, respectively.
The thermophysical properties of the base fluids (water and water–ethylene glycol), magnetite, and cobalt ferrite nanoparticles are described in
Table 1 [
66]. Meanwhile, the correlations for hybrid nanofluid are shown in
Table 2 [
33].
Now, consider the following similarity transformations [
77,
80,
81]:
Equation (1) is fully satisfied by considering the similarity variables in Equation (8). Furthermore, the reduced Equations (2)–(5) are:
subject to:
The parameters relevant to this problem are the unsteadiness parameter
, Prandtl number
, suction/injection parameter
, and magnetic parameter
. Meanwhile, the skin friction coefficients are
(radial direction) and
(azimuthal direction), and the local Nusselt number for evaluating heat transfer performance is
(see Waini et al. [
81]):
Using Equations (8) and (13), one obtains (see Waini et al. [
81]):
where
is the local Reynolds number.
4. Results and Discussion
Discussions of the obtained results are provided in this section. The bvp4c solver was used for the computation. This solver occupies a finite difference method that employs the three-stage Lobatto IIIa formula, see Shampine et al. [
85,
86]. The present results were validated by conducting a comparison with previously published data. In this respect,
Table 3 provides the validation of test values when
with different values of
, and an excellent agreement can be observed from the comparison. Therefore, the present results are acceptable and accurate.
Furthermore, the total composition of Al
2O
3 and Cu concentrations in this study are considered as
of Fe
3O
4 and
of CoFe
2O
4 The Prandtl number subject to the water base fluid is
and
for water–ethylene glycol. The dual solution’s availability is possible, as displayed in
Figure 2,
Figure 3,
Figure 4,
Figure 5,
Figure 6,
Figure 7,
Figure 8,
Figure 9,
Figure 10,
Figure 11,
Figure 12 and
Figure 13 when the parameters are used within the allocated interval; unsteadiness decelerating parameter
, magnetic parameter
, suction/injection parameter
while other parameters are fixed such that
and
.
The new numerical results obtained from this present study are presented in
Table 4 and
Table 5. Here, the values of the physical quantities such as
,
, and
are obtained for varied values of
with different base fluids (H
2O-EG and H
2O). For the first solution, the results show a decrement in these physical quantities as
increases. Quantitatively, a 15.98% decrement is observed for the values of
when
increases up to 30%
. Meanwhile, the reduction in
is prominent with a 61.66% decrement and the values of
decrease with 1.88%. However, a significant increment is observed for
when H
2O-EG is considered with 107.60% compared to H
2O as the base fluid. Furthermore, the numerical results provided could be important to other researchers for future reference. Interestingly, multiple solutions are obtained under certain circumstances. To determine a realistic or stable solution,
Table 6 shows the smallest eigenvalues
that were obtained from the eigenvalue problems (see
Section 3). Surprisingly, both solutions give positive values
that reveal the stability of both solutions and the possibility of another solution with a negative eigenvalue [
81].
The effects of magnetic
and unsteadiness
parameters on the physical quantities (
,
, and
) are elucidated in
Figure 2,
Figure 3 and
Figure 4. The declining behaviour is noticed when a larger
value is considered. However, the opposite trend is shown for stronger deceleration strength
. Physically, the Lorentz force is created when the magnetic field is imposed on the boundary layer, and it is intensified for stronger magnetic field strength. This force possesses the flow and increases the shear stress on the surface, and consequently increases the thermal rate. It should be noted that the outcomes of this study are contradicted by its physical phenomenon due to the interaction between the magnetic field and the unsteadiness of the flow. This observation is supported by a previous study, see Waini et al. [
81], which stated that the unsteadiness condition creates an obstacle to the flow and thermal behaviour of the fluid.
Figure 5,
Figure 6 and
Figure 7 displayed the values of
,
, and
for different values of
(mass flux parameter). The quantities of
, and
are lower for
(suction case). Physically, the wall suction causes a higher drag force and torque on the revolving disk and thus raising the shear stress. However, the opposite observation was seen in this study (see
Figure 5 and
Figure 6). This is due to the simultaneous effect of the physical parameter on the flow. However, the behaviour is contradicted for
where
(suction case) gives higher values of
.
Figure 8,
Figure 9 and
Figure 10 are provided in order to investigate the impact of the different base fluids (H
2O and H
2O-EG) on the flow and thermal behaviour. Here, H
2O-EG contributes to enhancing the values of
, and
. Meanwhile, the values of
decline. This observation can be explained from the values of their Prandtl number, Pr. Note that H
2O-EG has larger Pr, i.e.,
if compared to H
2O
. Larger
means the convection process for the flow is dominant. As a result, fluid momentum, rather than fluid conduction, is the best way to transmit heat. Additionally, the velocity and the temperature profiles of Fe
3O
4-CoFe
2O
4/H
2O-EG for several
are given in
Figure 11,
Figure 12 and
Figure 13. The thickness of the respective boundary layers is lessened as S becomes smaller. This implies that the gradient of these profiles near the wall is increased for stronger deceleration strength
and gives higher values of the shear stress and heat transfer on the wall, as reported in
Figure 4,
Figure 5 and
Figure 6.