Let
Oxyz be a reference frame with fundamental unit vectors
\(\mathbf{i},\mathbf{j},\mathbf{k}\) (
\(\mathbf{k}\) pointing vertically upward). Let
L be a bi-disperse porous layer of thickness
d uniformly heated from below and rotating about the vertical axis
z, with constant angular velocity
\(\mathbf{\Omega }=\Omega \mathbf{k}\). Let
L be saturated by a homogeneous incompressible fluid at rest state, and let us assume the validity of the local thermal equilibrium between the f-phase and the p-phase, i.e.
\(T^f=T^p=T\). The saturated bi-disperse porous medium is also supposed to be
horizontally isotropic. Let the axes (
x,
y,
z) be the
principal axes of the permeability, so the macropermeability tensor and the micropermeability tensor may be written as:
$$\begin{aligned} \begin{aligned} \mathbf{K}^f&= \text {diag}(K^f_x,K^f_y,K^f_z)=K^f_z \ \mathbf{K}^{f*}, \\ \mathbf{K}^p&= \text {diag}(K^p_x,K^p_y,K^p_z)=K^p_z \ \mathbf{K}^{p*}, \end{aligned}\\ \mathbf{K}^{f*}=\text {diag}(k,k,1),\quad \mathbf{K}^{p*}=\text {diag}(h,h,1), \end{aligned}$$
where
$$\begin{aligned} k=\frac{K^f_x}{K^f_z}=\frac{K^f_y}{K^f_z}, \quad h=\frac{K^p_x}{K^p_z}=\frac{K^p_y}{K^p_z}. \end{aligned}$$
A Boussinesq approximation is used, whereby the density is constant except in the buoyancy forces, which are linear in temperature. Taking into account the Coriolis terms due to the uniform rotation of the layer about the vertical axis
z for the micropores and the macropores [
6] and extending the Brinkman model for a simple porous medium to BDPM [
21], the relevant equations are:
$$\begin{aligned} \!\!\!\! {\left\{ \begin{array}{ll} \mathbf{v}^f\! =\! \mu ^{-1} \mathbf{K}^f \! \cdot \! \Bigl [ - \zeta (\mathbf{v}^f\! -\! \mathbf{v}^p) \!-\! \nabla p^f \! + \! \varrho _F \alpha g T \mathbf{k}\! -\! \dfrac{2 \varrho _F \Omega }{\varphi } \mathbf{k} \times \mathbf{v}^f \!+\! {\tilde{\mu }}_f \varDelta \mathbf{v}^f \Bigr ], \\ \mathbf{v}^p \!=\! \mu ^{-1} \mathbf{K}^p \! \cdot \! \Bigl [ - \zeta (\mathbf{v}^p \!-\! \mathbf{v}^f) \!-\! \nabla p^p \! + \! \varrho _F \alpha g T \mathbf{k} \!-\! \dfrac{2 \varrho _F \Omega }{\epsilon } \mathbf{k} \times \mathbf{v}^p \!+\! {\tilde{\mu }}_p \varDelta \mathbf{v}^p \Bigr ], \\ \nabla \cdot \mathbf{v}^f = 0, \\ \nabla \cdot \mathbf{v}^p = 0, \\ (\varrho c)_m T_{,t} + (\varrho c)_f (\mathbf{v}^f + \mathbf{v}^p) \cdot \nabla T = k_m \varDelta T, \end{array}\right. } \end{aligned}$$
(1)
where
$$\begin{aligned} p^s=P^s-\frac{\varrho _F}{2} \vert \mathbf{\Omega } \times \mathbf{x} \vert ^2, \quad s=\{f,p\} \end{aligned}$$
is the reduced pressure, with
\(\mathbf{x}=(x,y,z)\),
\(\mathbf{v}^s\) = seepage velocity,
\(P^s\)= pressure,
\(\varrho \)= density,
\(\zeta \) = interaction coefficient between the f-phase and the p-phase,
\(\mathbf{g}=-g \mathbf{k}\) = gravity,
\(\mu \) = fluid viscosity,
\({\tilde{\mu }}_s\) = effective viscosity,
\(\varrho _F\) = reference constant density,
\(\alpha \) = thermal expansion coefficient,
c = specific heat,
\(c_p\) = specific heat at a constant pressure,
\((\varrho c)_m=(1-\varphi )(1-\epsilon )(\varrho c)_{sol}+\varphi (\varrho c)_f+\epsilon (1-\varphi )(\varrho c)_p\),
\(k_m=(1-\varphi )(1-\epsilon )k_{sol}+\varphi k_f+\epsilon (1-\varphi )k_p\) = thermal conductivity (The subscript
sol refers to the solid skeleton).
To (
1), the following boundary conditions are appended:
$$\begin{aligned} \begin{array}{l} \mathbf{v}^s \cdot \mathbf{n}=0\,, \quad s=\{f,p\}\qquad \text{ on }\quad z=0,d,\\ T=T_L\,,\quad \text{ on }\quad z=0\,,\qquad T=T_U\,, \quad \text{ on }\quad z=d \end{array} \end{aligned}$$
(2)
where
\(\mathbf{n}\) is the unit outward normal to the impermeable horizontal planes delimiting the layer and
\(T_L>T_U\).
The problem (
1)–(
2) admits the stationary conduction solution:
$$\begin{aligned} \mathbf{{\overline{v}}}^f=\mathbf{0},\quad \mathbf{{\overline{v}}}^p=\mathbf{0}, \quad {\overline{T}}=- \beta z + T_L, \end{aligned}$$
where
\(\beta =\dfrac{T_L-T_U}{d}\) is the temperature gradient. Denoting by
\(\{ \mathbf{u}^f, \mathbf{u}^p, \theta , \pi ^f, \pi ^p \}\) a perturbation to the steady solution, one recovers that the evolutionary system governing the perturbation fields is given by
$$\begin{aligned} {\left\{ \begin{array}{ll} \mathbf{u}^f\! =\! \mu ^{-1} \mathbf{K}^f \!\cdot \! \Bigl [ -\! \zeta (\mathbf{u}^f \!-\! \mathbf{u}^p) \!-\! \nabla \pi ^f \!+\! \varrho _F \alpha g \theta \mathbf{k} \!-\! \dfrac{2 \varrho _F \Omega }{\varphi } \mathbf{k} \times \mathbf{u}^f \!+\! {\tilde{\mu }}_f \varDelta \mathbf{u}^f \Bigr ], \\ \mathbf{u}^p \!=\! \mu ^{-1} \mathbf{K}^p \!\cdot \! \Bigl [ - \!\zeta (\mathbf{u}^p \!-\! \mathbf{u}^f)\! -\! \nabla \pi ^p \!+\! \varrho _F \alpha g \theta \mathbf{k}\! -\! \dfrac{2 \varrho _F \Omega }{\epsilon } \mathbf{k} \times \mathbf{u}^p \!+\! {\tilde{\mu }}_p \varDelta \mathbf{u}^p \Bigr ], \\ \nabla \cdot \mathbf{u}^f = 0, \\ \nabla \cdot \mathbf{u}^p = 0, \\ (\varrho c)_m \theta _{,t} + (\varrho c)_f (\mathbf{u}^f + \mathbf{u}^p) \cdot \nabla \theta = (\varrho c)_f \beta (w^f+w^p) + k_m \varDelta \theta , \end{array}\right. } \end{aligned}$$
(3)
where
\(\mathbf{u}^f=(u^f,v^f,w^f),\,\mathbf{u}^p=(u^p,v^p,w^p)\). Introducing the non-dimensional parameters
$$\begin{aligned} \! \mathbf{x}^{*}= & {} \displaystyle \frac{\mathbf{x}}{d}, \ t^{*}=\displaystyle \frac{t}{\tilde{t}}, \ \theta ^{*} = \displaystyle \frac{\theta }{\tilde{T}}, \ \mathbf{u}^{s*} = \displaystyle \frac{\mathbf{u}^s}{\tilde{u}}, \ \pi ^{s*} = \displaystyle \frac{\pi ^s}{\tilde{P}}, \quad \text {for} \ s=\{f,p\},\\ \eta= & {} \displaystyle \frac{\varphi }{\epsilon }, \ \sigma =\displaystyle \frac{{\tilde{\mu }}_p}{{\tilde{\mu }}_f}, \ \gamma _1=\displaystyle \frac{\mu }{K^f_z \zeta }, \ \gamma _2=\displaystyle \frac{\mu }{K^p_z \zeta }, \end{aligned}$$
where the scales are given by
$$\begin{aligned} \tilde{u}=\frac{k_m}{(\varrho c)_f d}, \ \tilde{t} = \frac{d^2 (\varrho c)_m}{k_m}, \ \tilde{P} = \frac{\zeta k_m}{(\varrho c)_f}, \ \tilde{T} = \sqrt{\frac{\beta k_m \zeta }{(\varrho c)_f \varrho _F \alpha g}}, \end{aligned}$$
and setting
$$\begin{aligned} \mathcal {T}= \dfrac{2 \varrho _F \Omega K^f_z}{\varphi \mu }, \qquad Da_f=\dfrac{{\tilde{\mu }}_f K^f_z}{d^2 \mu }, \qquad R = \sqrt{\dfrac{\beta d^2 (\varrho c)_f \varrho _F \alpha g}{k_m \zeta }}, \end{aligned}$$
which are the Taylor number
\(\mathcal {T}\), the Darcy number
\(Da_f\), and the thermal Rayleigh number
R, respectively, and the resulting non-dimensional perturbation equations, dropping all the asterisks, are
$$\begin{aligned} {\left\{ \begin{array}{ll} \gamma _1 (\mathbf{K}^f)^{-1} \mathbf{u}^f\! +\! (\mathbf{u}^f \!-\! \mathbf{u}^p) \!=\! -\! \nabla \pi ^f \!+\! R \theta \mathbf{k} \!-\! \gamma _1 \mathcal {T} \mathbf{k} \times \mathbf{u}^f \!+\! Da_f \gamma _1 \varDelta \mathbf{u}^f, \\ \gamma _2 (\mathbf{K}^p)^{-1} \mathbf{u}^p \!-\! (\mathbf{u}^f \!-\! \mathbf{u}^p) \!=\! -\! \nabla \pi ^p \!+\! R \theta \mathbf{k} \!-\! \eta \gamma _1 \mathcal {T} \mathbf{k} \times \mathbf{u}^p \!+\! Da_f \gamma _1 \sigma \varDelta \mathbf{u}^p, \\ \nabla \cdot \mathbf{u}^f = 0, \\ \nabla \cdot \mathbf{u}^p = 0, \\ \theta _{,t} + (\mathbf{u}^f + \mathbf{u}^p) \cdot \nabla \theta = R (w^f+w^p) +\varDelta \theta , \end{array}\right. } \end{aligned}$$
(4)
under the initial conditions
$$\begin{aligned} \mathbf {u}^s(\mathbf {x},0)=\mathbf {u}^s_0(\mathbf {x})\,,\qquad \varphi ^s(\mathbf {x},0)=\varphi _0(\mathbf {x})\,,\qquad \theta (\mathbf {x},0)=\theta _0(\mathbf {x}) \end{aligned}$$
with
\(\nabla \cdot \mathbf {u}_0^s=0,\, s=\{f,p\}\), and the stress-free boundary conditions [
12]
$$\begin{aligned} u^f_{,z}=v^f_{,z}=u^p_{,z}=v^p_{,z}=w^f=w^p=\theta =0 \quad \text {on} \ z=0,1. \end{aligned}$$
(5)
Moreover, according to experimental results, let us assume that the perturbation fields are periodic functions in the
x,
y directions and denote by
$$\begin{aligned} V=\Big [ 0,\frac{2 \pi }{l} \Big ] \times \Big [ 0,\frac{2 \pi }{m} \Big ] \times [0,1] \end{aligned}$$
the periodicity cell.