Professional Documents
Culture Documents
Heliyon
journal homepage: www.cell.com/heliyon
Research article
A R T I C L E I N F O A B S T R A C T
Keywords: The penetrative ferrothermal convection (FTC) in a ferrofluid (FF) saturated high porosity anisotropic porous
Anisotropic porous layer layer via uniform internal heating is investigated. The Brinkman-extended Darcy equation is applied to describe
Ferrofluid the flow in the porous medium. The permeability in the vertical direction is taken to be twice that of the
Internal heating
permeability in the horizontal direction while the ratio of horizontal to vertical effective thermal diffusivity is
Linear instability
Thermal convection
allowed to vary. The Galerkin method is applied to solve numerically the stability eigenvalue problem for
different boundary combinations namely, (i) rigid-paramagnetic (R–P) with large and low magnetic susceptibility,
(ii) rigid-ferromagnetic (R–F), and (iii) free-ferromagnetic (F–F). The R–P boundaries with large magnetic sus-
ceptibility offer most, while F–F boundaries offer least stabilizing effect against FTC. Besides, the effect of
increasing the magnetic number, non-linearity of fluid magnetization parameter, Darcy number and internal heat
source strength is to speed up FTC, while the thermal anisotropy and magnetic susceptibility parameter indict a
contradictory effect on FTC.
1. Introduction experimental and theoretical studies on FFs for guiding and detecting
fluids in the subsurface using geophysical methods. Oldenburg et al. [9]
Ferrofluids (FFs) are synthesized by suspending single domain developed capabilities of simulation for both miscible and immiscible of
ferromagnetic nanoparticles stabilized in various nonmagnetic carrier FF-saturated porous media. Borglin et al. [10] described simple
fluids, which exhibit both magnetic and fluid properties [1, 2]. These laboratory-type investigations on penetrative convection in a
fluids are now termed as magnetic nanofluids and the study of such fluids FF-saturated porous medium.
has been a subject of intensive investigations over decades due to their To understand the heat transfer characteristics in a FF-saturated
potential applications in magnetically heat controlled thermosiphons for porous medium, FTC in a porous medium has also been studied exten-
technological purposes [3, 4]. Thermal convection in a horizontal FF sively (Sekar et al. [11], Shivakumara et al. [12, 13], Nanjundappa et al.
layer in the existence of magnetic field, called ferro-thermal-convection [14, 15, 16]). Besides, the implications of additional effects such as
(FTC), has been studied extensively both theoretically and experimen- magnetic field dependent viscosity (Nanjundappa et al. [17, 18], Shiva-
tally over the years to understand the heat transfer mechanisms and the kumara et al. [19, 20]), rotation (Shivakumara et al. [21], Nanjundappa
details are amply documented in the review article by Nkurikiyimfura et al. [22]), internal heating (Nanjundappa et al. [23, 24]), throughflow
et al. [5]. (Nanjundappa et al. [25]), non-uniform basic temperature gradients
The study of FFs through porous media has also captivated the (Nanjundappa et al. [26]), local thermal nonequilibrium (Shivakumara
attention of researchers because of their potential utility in subsurface et al. [27, 28, 29, 30]), Cattaneo heat flux law (Shivakumara et al. [31,
environmental engineering wherein externally applied magnetic field 32]) have also been investigated on FTC in porous media.
are used to control and direct the FFs flow. Rosensweig et al. [6] In many engineering applications the porous matrix used turns
experimentally demonstrated the FFs penetration in the Hele-Shaw cell, out to be anisotropic both mechanically and thermally. Shiina and
while Zahn et al. [7] studied the FTC in a porous medium in the exis- Inagaki [33] reported the improvement of heat transfer by the in-
tence of oblique magnetic field. Moridis et al. [8] carried out clusion of high thermal conductivity porous medium. But the
* Corresponding author.
E-mail address: nanjundappace@gmail.com (C.E. Nanjundappa).
https://doi.org/10.1016/j.heliyon.2021.e06153
Received 1 July 2020; Received in revised form 15 December 2020; Accepted 27 January 2021
2405-8440/© 2021 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Y.L. Savitha et al. Heliyon 7 (2021) e06153
∂T dTb ∂2 T
A þ w ¼ κ h r1 2 T þ κ z 2 þ Q (15)
∂t dz ∂z
M0 ∂2 φ ∂T
1þ r21 φ þ ð1 þ χ Þ 2 K2 ¼ 0 (16)
H0 ∂x ∂z
2
Y.L. Savitha et al. Heliyon 7 (2021) e06153
We seek the solution of Eqs. (14), (15), and (16) using the method of X
n
normal modes of the form (noting that the principle of exchange of sta- fW; Θ; ΦgðzÞ ¼ fAm Wm ðzÞ; Bm Θm ðzÞ; Cm Φm ðzÞg (29)
m¼1
bility holds)
where Am ; Bm , Cm (m ¼ 1; 2;:::::::;n) are constants and basis functions Wm ;
fw; T; Φgðx; y; zÞ ¼ fW; Θ; ΦgðzÞexpiða1 xþa2 yÞ (17) Θm , Φm are the trial functions chosen usually satisfying (25)–(28).
where
D2 a2 M3 Φ DΘ ¼ 0: (24)
Z 1
The following boundary combinations are considered: Cnm ¼ Da D2 Wn D2 Wm þ 2a2 DWn DWm þ a4 Wn Wm dz
0
Case (i): Lower and upper rigid surfaces are ferromagnetic (R–F), and Z 1
isothermal a2
þ DWn DWm dz þ Wn Wm dz
0 2
W ¼ DW ¼ 0; Θ ¼ 0; Φ¼0 at z ¼ 0; 1 (25)
Z 1
Case (ii): Lower and upper free surfaces are ferromagnetic (F–F), and Mnm ¼ a2 Da f½Rt Rt M1 ½Nsð1 2zÞ 1Wm Θn g dz;
isothermal 0
Z
W ¼ D2 W ¼ 0; Θ ¼ 0; DΦ ¼ 0 at z ¼ 0; 1 (26) 1
Fnm ¼ a2 Rt M1 Da ½Nsð1 2zÞ 1DΦn Wm dz
0
Case (iii): Lower and upper rigid surfaces are paramagnetic (R–P),
and isothermal Z 1
Gnm ¼ f½Nsð1 2zÞ 1Θm Wn g dz;
W ¼ DW ¼ 0; Θ¼0 at z ¼ 0; 1 (27) 0
8 aΦ Z 1
< 1 þ χ at z ¼ 0
> Hnm ¼ DΘn DΘm þ a2 η Θn Θm dz;
0
DΦ ¼ : (28)
>
: a Φ at z ¼ 1 Z
1þχ 1
gnm ¼ Φm DΘn dz
0
3. Numerical solution
a
Z 1
Eqs. (22), (23), and (24) subject to any choice of boundary conditions Jnm ¼ ½Φm ð1ÞΦn ð1Þ Φm ð0ÞΦn ð0Þ þ DΦm DΦþa2 M3 Φm Φn dz:
1þχ 0
cases form an eigenvalue problem and solved numerically by the Galer-
kin technique. We expand the unknown variables in power series of the Eq. (33) possesses a non-trivial solution provided
form
3
Y.L. Savitha et al. Heliyon 7 (2021) e06153
Table 1. Process of convergence of results for different orders of approximations for different Da and χ when M1 ¼ M3 ¼ 1, Λ ¼ η ¼ 1 and Ns ¼ 2
Boundaries Da m ¼n ¼1 m ¼n ¼3 m ¼n ¼5 m ¼n ¼7 m ¼n ¼8
Cnm Mnm Fnm f ðRt ; Rm ; Da; Ns; M1 ; M3 ; η; a Þ ¼ 0: (35)
Gnm Hnm 0 ¼ 0: (34)
From Eq. (35), the critical value of Rt (Rtc ) or Rm (Rmc ) is computed
0 gnm Jnm
with respect to ‘a’ (wave number) for fixed governing physical parame-
The eigenvalues are extracted from Eq. (34). Substitution of trial ters Da, M1 , η;Ns and M3 . A symbolic algebraic package of Microsoft
functions into Eq. (35) for each cases of boundary conditions leads to a Mathematica 12.0 (see [35]) is used for this purpose.
characteristic equation of the form
4
Y.L. Savitha et al. Heliyon 7 (2021) e06153
Rtc
Da = 0.01
Da = 10
Da = 0.01
Da = 10
ac
Figure 2. Variation of (a) Rtc and (b) ac against η for different Da when M1 ¼ Ns ¼ 2 and M3 ¼ 1 .
5
Y.L. Savitha et al. Heliyon 7 (2021) e06153
Rtc
=0
=2
M1 =2
ac
=0
Da
Figure 3. Variation of (a)Rtc and (b) ac against Da for different M1 when η ¼ M3 ¼ 1 and Ns ¼ 2
4. Results and discussion The process of convergence of results for representative values of gov-
erning parameters are shown in Table 1 and note that eight terms (m ¼
The penetrative FTC instilling due to volumetric heating in a FF- n ¼ 8) in the Galerkin expansion are needed to achieve the convergence,
saturated high porosity anisotropic porous layer has been investigated. in general. In Table 2, the results obtained for R–F surfaces when M3 ¼ 1
6
Y.L. Savitha et al. Heliyon 7 (2021) e06153
Rigid-paramagnetic with a
Rigid-paramagnetic with
Rigid-ferromagnetic
Free-Ferromagnetic
Rtc
=1
= 0.01
b
= 0.01
ac =1
Figure 4. Variation of (a)Rtc and (b) ac against M1 for different η for M3 ¼ 1, Da ¼ 10 and Ns ¼ 2.
and Ns ¼ 0 are compared with Shiina and Hishida [34] for different critical Rayleigh number Rtc plotted against thermal anisotropy param-
values of η and Da, and the results are in excellent agreement. eter η for Da ¼ 0:01; 10, M3 ¼ 1 and Ns ¼ M1 ¼ 2. The effect of
The results obtained covering a wide range of governing parameters increasing η is to increase Rtc indicating its effect is to postpone the onset
for R–P boundaries with χ 6¼ 0 (solid curves) and χ ¼ 0 (dashed curves), of FTC. This may be attributed to the fact that, as η decreases, a heated
R–F boundaries (dotted curves) and F–F boundaries (dashed-dotted fluid parcel carries a lesser amount of heat in the horizontal direction.
curves) are exhibited in Figures 2, 3, 4, 5, 6, and 7. Figure 2(a) show the Further, R–P surfaces with large magnetic susceptibility (1 þ χ ¼ 102 )
7
Y.L. Savitha et al. Heliyon 7 (2021) e06153
Da = 0.01 a
0.01
0.01
0.01
Rtc
Da = 10
10
10
10
M3
Da = 0.01
0.01
0.01
0.01
ac
10
10
10
10
M3
Figure 5. Variation of (a)Rtc and (b) ac against M3 for different Dafor M1 ¼ Ns ¼ 2 and η ¼ 1.
show more stabilizing effect against FTC when compared to the low the convection cell size. Further, the values of ac for F–F surfaces are the
magnetic susceptibility ( χ ¼ 0) case, while F–F surfaces show more least as compared to R–F surfaces and highest for R–P surfaces.
destabilizing influence on the same. Thus the impact of R–F surfaces on The influence of M1 (magnetic number) and Da (Darcy number) on Rtc
the stability of the system lies between those of R–P and F–F surfaces. is demonstrated in Figure 3 (a) for M3 ¼ 1, η ¼ 1 and Ns ¼ 2: Increasing
That is, the R–P surfaces have sturdy effect in stabilizing the thermally Da and M1 is to decrease Rtc representing their effect is to hasten the
unstable FF-saturated anisotropic porous layer than R–F and F–F surfaces onset of FTC. For a fixed porous layer thickness, increase in Da amounts
because the rigid boundaries with an increase in the magnetic induction to increase in the permeability of the porous medium which in turn speed
suppress the disturbances more effectively. These results are in accor- up the flow of FF in porous media and hence lower Rtc values are required
dance with those of Gotoh and Yamada [36]. The critical wave number ac for the onset of FTC. Moreover, the convection cell size becomes widened
decreases with increasing η and Da indicating their effect is to broaden and contracted with increasing Da and M1 , respectively.
8
Y.L. Savitha et al. Heliyon 7 (2021) e06153
=1
= 0.8 1 a
0.8
1
0.8
Rtc
0.8 1
Ns
b
=1
= 0.8
0.8
1
ac
0.8
0.8 1
Ns
Figure 6. Variation of (a) Rtc and (b) ac against Ns for different η for M1 ¼ 2, Da ¼ 10 and M3 ¼ 1.
Figure 4(a) illustrates the variation of Rtc as a function of M1 for increase in M1 (Figure 4b). With increasing value of M1 , the curves of ac
different η values when M3 ¼ 1, Da ¼ 10 and Ns ¼ 2. The values of Rtc for two values of η coalesce for different bounding surfaces.
decrease quite rapidly at first then slowly and finally the curves of The variation of M3 (nonlinearity of fluid magnetization parameter)
different η remain invariant as M1 increases and this is due to the satu- on Rtc is exhibited in Figure 5(a) for M1 ¼ 2, η ¼ 1 and Ns ¼ 2. The figure
ration of destabilizing magnetic force. The critical wave numbers in- demonstrates that increasing M3 has a destabilizing effect on the system
creases with M1 , reach maximum, and thereafter decrease with further but it is only marginal. Thus, the system becomes unstable with a smaller
9
Y.L. Savitha et al. Heliyon 7 (2021) e06153
Ns =0
0
Rtc
0
0 2
Rmc
Figure 7. Locus of Rtc against Rmc for different Ns for M1 ¼ 2;η ¼ M3 ¼ 1 and Da ¼ 10.
temperature gradient as M3 increases. From the Figure 5 (b), we note that The effect of increasing thermal anisotropy (η) and internal heat
an increase in M3 is to decrease ac and thus to increase the convection cell source (Ns) is to delay and hasten the onset of FTC in an aniso-
size. tropic porous medium, respectively. Thus the porous medium
Figure 6(a) represents Rt c against the internal heat source parameter anisotropy can be used effectively to control (suppress or
Ns for different η when M1 ¼ 2, Da ¼ 10 and M3 ¼ 1. It is seen that Rt c augment) FTC.
decreases monotonically with increasing Ns and thus hastens the onset of The critical stability parameters satisfy the following inequality sug-
FTC. This is because; increasing Ns amounts to increase in the energy gesting different boundaries significantly influence the instability
supply to the system. Figure 6(b) shows the variation of ac and note that it characteristics of FTC. That is,
increases as Ns increases and thus its effect is to decrease the convection
cells size. ðRtc or Rmc ÞFF < ðRtc or Rmc ÞRF < ðRtc or Rmc ÞRPðχ¼0Þ
Figure 7 shows the locus of Rtc versus Rmc for different Ns and < ðRtc or Rmc ÞRPðχ >>1Þ :
bounding surfaces when M1 ¼ 2;M3 ¼ 1, η ¼ 1 and Da ¼ 10. It seen that
there is a strong coupling between Rtc and Rmc , and an increase in Rmc has
a destabilizing effect on the onset of FTC. Thus, the gravitational force The increase in magnetic susceptibility χ is to delay the penetrative
(Rtc ) becomes negligible when the magnetic force (Rmc ) is predominant FTC.
and vice versa. Moreover, the stability curves are slightly convex. The Increase in nonlinearity of fluid magnetization and the strength of
instability sets in at higher Rmc representing the system is more stable magnetic parameters tend to destabilize the system.
when magnetic forces alone are present. In all these figures, the results The magnetic ðRm Þ and the gravitational ðRt Þ forces are comple-
for F–F, R–F and R–P boundaries with χ ¼ 0 and χ >> 1 are presented. It mentary with each other and the buoyancy force has more stabi-
is observed that lizing effect on the system compared to magnetic force. i.e.
Rtc < Rmc .
ðRtc or Rmc ÞFF < ðRtc or Rmc ÞRF <ðRtc or Rmc ÞRPðχ ¼ 0Þ <ðRtc or Rmc ÞRPðχ >>1Þ : The effect of increasing η; Da as well as decrease inM1 , M3 and Nsis to
increase the critical wave number and hence their effect is to contract
Thus we note that the system is more stabilizing against FTC if the the convection cells size.
surfaces are rigid with high magnetic susceptibility (χ > > 1) and least
stable if the surfaces are F–F. Declarations
The onset of penetrative FTC in a FF-saturated high porosity aniso- C. E. Nanjundappa: Conceived and designed the analysis; Analyzed
tropic porous layer is investigated by uniformly heating the porous me- and interpreted the data; Wrote the paper.
dium internally. The Galerkin method is applied to solve the stability Savitha Y. L. & I.S. Shivakumara: Conceived and designed the anal-
eigenvalue problem numerically. The results of the present investigations ysis; Analyzed and interpreted the data; Contributed analysis tools or
can be summarized as follows: data; Wrote the paper.
10
Y.L. Savitha et al. Heliyon 7 (2021) e06153
Funding statement [15] C.E. Nanjundappa, M. Ravisha, Jinho Lee, I.S. Shivakumara, Penetrative
ferroconvection in a porous layer, Acta Mech. 216 (2011) 243–257.
[16] C.E. Nanjundappa, I.S. Shivakumara, Jinho Lee, M. Ravisha, The onset of Brinkman
This research did not receive any specific grant from funding agencies ferroconvection in an anisotropic porous medium, Int. J. Eng. Sci. 49 (2011)
in the public, commercial, or not-for-profit sectors. 497–508.
[17] C.E. Nanjundappa, I.S. Shivakumara, K. Srikumar, Effect of MFD viscosity on the
onset of ferromagnetic fluid layer heated from below and cooled from above with
Data availability statement constant heat flux, Meas. Sci. Rev. 9 (2009) 75–80.
[18] C.E. Nanjundappa, I.S. Shivakumara, M. Ravisha, The onset of ferroconvection in a
Data included in article/supplementary material/referenced in horizontal ferrofluid saturated porous layer heated from below and cooled from
above with constant heat flux subject to MFD viscosity, Int. Commun. Heat Mass
article. Transf. 37 (2010) 1246–1250.
[19] I.S. Shivakumara, J. Lee, C.E. Nanjundappa, Onset of thermogravitational
Declaration of interests statement convection in a ferrofluid layer with temperature dependent viscosity, J. Heat Tran.
134 (2012) 12501-1–12501-7.
[20] I.S. Shivakumara, J. Lee, C.E. Nanjundappa, M. Ravisha, D.H. Lee, Brinkman
The authors declare no conflict of interest. ferromagnetic convection in a porous layer: effect of MFD viscosity and
ferromagnetic boundaries, Int. J. Mech. Sci. Tech. 27 (2013) 3875–3884.
[21] I.S. Shivakumara, J. Lee, C.E. Nanjundappa, M. Ravisha, Ferromagnetic convection
Additional information in a rotating ferrofluid saturated porous layer, Transp, Porous Medium 87 (2011)
251–273.
No additional information is available for this paper. [22] C.E. Nanjundappa, I.S. Shivakumara, H.N. Prakash, Effect of Coriolis force on
thermomagnetic convection in a ferrofluid saturating porous medium: a weakly
nonlinear stability analysis, J. Magn. Magn Mater. 370 (2014) 140–149.
Acknowledgements [23] C.E. Nanjundappa, I.S. Shivakumara, J. Lee, M. Ravisha, Effect of internal heat
generation on the onset of Brinkman–Benard convection in a ferrofluid saturated
The authors wish to thank the reviewers for their constructive com- porous layer, Int. J. Therm. Sci. 50 (2011) 160–168.
[24] C.E. Nanjundappa, I.S. Shivakumara, H.N. Prakash, Penetrative ferroconvection via
ments which helped in improving the paper substantially. internal heating in a saturated porous layer with constant heat flux at the lower
boundary, J. Magn. Magn Mater. 324 (2012) 1670–1678.
References [25] C.E. Nanjundappa, I.S. Shivakumara, R. Arunkumar, R. Tadmor, Ferroconvection in
a porous medium with vertical throughflow, Acta Mech. 226 (2015) 1515–1528.
[26] C.E. Nanjundappa, I.S. Shivakumara, R. Arunkumar, Effect of cubic temperature
[1] R.E. Rosensweig, Ferrohydrodynamics, Cambridge Univ. Press, Cambridge, 1985. profiles on ferroconvection in a Brinkman porous medium, J. Appl. Fluid Mech. 9
[2] M.I. Shliomis, Magnetic Fluids,, Sov. Phys. Usp., 17 (1974) 153–169. (2016) 1955–1962.
[3] S.W. Charles, The preparation of magnetic fluids, in: S. Odenbach (Ed.), Ferrofluids: [27] I.S. Shivakumara, J. Lee, M. Ravisha, R. Gangadhara Reddy, The onset of Brinkman
Magnetically Controllable, Springer-Verlag, Berlin–Heidelberg, 2002. ferroconvection using a thermal non-equilibrium model, Int. J. Heat Mass Tran. 54
[4] E. Blums, Heat and mass transfer phenomena, in: S. Odenbach (Ed.), Ferrofluids: (2011) 2116–2125.
Magnetically Controllable Fluids and Their Applications, Springer-Verlag, Berlin, [28] I.S. Shivakumara, J. Lee, M. Ravisha, R. Gangadhara Reddy, Effects of MFD viscosity
Heidelberg, New York, 2002. and LTNE on the onset of ferromagnetic convection in a porous medium, Int. J. Heat
[5] I. Nkurikiyimfura, Y. Wanga, Z. Pan, Heat transfer enhancement by magnetic Mass Tran. 54 (2011) 2630–2641.
nanofluids—a review, Renew. Sustain. Energy Rev. 21 (2013) 548–561. [29] I.S. Shivakumara, C.O. Ng, M. Ravisha, Ferromagnetic convection in a
[6] R.E. Rosensweig, M. Zhan, T. Volger, Stabilization of fluid penetration through a heterogeneous Darcy porous medium using a local thermal non-equilibrium (LTNE)
porous medium using magnetizable fluids, in: Berkovsky, B. Thermomechanics of model, Transp. Porous Media 90 (2011) 529–544.
Magnetic Fluids, Hemisphere, Washington, 1978. [30] I.S. Shivakumara, J. Lee, M. Ravisha, R. Gangadhara Reddy, The effects of local
[7] M. Zahn, R.E. Rosensweig, Stability of magnetic fluid penetration through a porous thermal nonequilibrium and MFD viscosity on the onset of Brinkman
medium with uniform magnetic field oblique to the interface, IEEE Trans. Magn. 16 ferroconvection, Meccanica 47 (2012) 1359–1378.
(1980) 275–282. [31] I.S. Shivakumara, M. Ravisha, C.O. Ng, V.L. Varun, A thermal non-equilibrium
[8] G.J. Moridis, S.E. Borglin, C.M. Oldenburg, A. Becker, Theoretical and Experimental model with Cattaneo effect for convection in a Brinkman porous layer, Int. J. Non
Investigations of Ferrofluids for Guiding and Detecting Liquids in the Subsurface, Lin. Mech. 71 (2015) 39–47.
Ernest Berkeley Orlando Lawrence National Laboratory, Earth Sci. Div, 1998. [32] I.S. Shivakumara, M. Ravisha, C.O. Ng, V.L. Varun, Porous ferroconvection with
[9] C. Oldenburg, S.E. Borglin, G.J. Moridis, Numerical simulation of ferrofluid flow for local thermal nonequilibrium temperatures and with Cattaneo effects in the solid,
subsurface environmental engineering, Appl. Transp. Porous Media 38 (2000) Acta Mech. 226 (2015) 3763–3779.
319–344. [33] Y. Shiina, T. Inagaki, Study on the efficiency of effective thermal conductivities on
[10] S.E. Borglin, J. Mordis, C.M. Oldenburg, Experimental studies of the flow of melting characteristics of latent heat storage capsules, Int. J. Heat Mass Tran. 48
ferrofluid in porous media, Transp. Porous Media 41 (2000) 61–80. (2005) 373–383.
[11] R. Sekar, G. Vaidynathan, A. Ramathan, Ferrocnvection in an anisotrophic porous [34] Y. Shiina, M. Hishida, Critical Rayleigh number of natural convection in high
medium, Int. J. Eng. Sci. 34 (4) (1996) 399–405. porosity anisotropic horizontal porous layers, Int. J. Heat Mass Tran. 53 (2010)
[12] I.S. Shivakumara, C.E. Nanjundappa, M. Ravisha, Effect of boundary conditions on 1507–1513.
the onset of thermomagnetic convection in a ferrofluid saturated porous medium, [35] Stephen Wolfram, The Mathematica Book, Fourth ed., Cambridge University Press,
J. Heat Tran. 131 (2009) 101003-1–101003-9. New York, 1999.
[13] I.S. Shivakumara, C.E. Nanjundappa, H.N. Prakasha, M. Ravisha, Effect of vertical [36] K. Gotoh, M. Yamada, Thermal convection in a horizontal layer of magnetic fluids,
heterogeneity on the onset of ferroconvection in a Brinkman porous medium, Ain J. Phys. Soc. Jpn. 51 (1982) 3042–3048.
Shams Eng. J. 6 (2015) 649–655.
[14] C.E. Nanjundappa, I.S. Shivakumara, M. Ravisha, The onset of buoyancy-driven
convection in a ferromagnetic fluid saturated porous medium, Meccanica 45 (2010)
213–226.
11