American Journal of Fluid Dynamics
2012; 2(4): 65-70
doi: 10.5923/j.ajfd.20120204.06
Pardeep Kumar
Department of Mathematics, ICDEOL, Himachal Pradesh University, Shimla-171005, India
Correspondence to: Pardeep Kumar , Department of Mathematics, ICDEOL, Himachal Pradesh University, Shimla-171005, India.
| Email: | ![]() |
Copyright © 2012 Scientific & Academic Publishing. All Rights Reserved.
The present paper considered the thermosolutal convection in a compressible Walters B' viscoelastic fluid layer heated and soluted from below in the presence of uniform rotation. Following the linearized stability theory and normal mode analysis, the dispersion relation is obtained. For the case of stationary convection, Walters B′ viscoelastic fluid behaves like a Newtonian fluid and compressibility, rotation and stable solute gradient have stabilizing effect on the system for G >1. Graphs have been plotted by giving numerical values to the parameters to depict the stability characteristics. The stable solute gradient, rotation and viscoelasticity introduce oscillatory modes in the system which were non-existent in their absence. The sufficient conditions for the non-existence of overstability are also obtained.
Keywords: Thermosolutal Convection, Walters B' Viscoelastic Compressible Fluid, Uniform Rotation
and a uniform rotation
. This layer is heated and soluted from below such that a steady adverse temperature gradient
and a solute concentration gradient
are maintained. For thermosolutal convection problem, the Boussinesq approximation has been used, which is well justified in the case of incompressible fluids.The equations governing the system become quite complicated when the fluids are compressible. To simplify them, Boussinesq try to justify the approximation for nearly incompressible fluids when the density variations arise principally from thermal effects by noting that atmospheric pressure fluctuations are much too small to produce the observed density changes. Spiegel and Veronis'[17] have simplified the set of equations governing the flow of compressible fluids under the following assumptions;(1) The vertical dimension of the fluid is much less than any scale height, as defined by them, if only motions of infinitesimal amplitudes are considered and (2) The motion - induced fluctuations in density and pressure do not exceed in order of their total static variations.Under the above approximations, Spiegel and Veronis'[17] have shown that the equations governing convection in a perfect gas are formally equivalent to those for an incompressible fluid if the static temperature gradient is replaced by its excess over the adiabatic and
is replaced by
denote respectively the fluid pressure, density, temperature, solute concentration and velocity. The equations expressing the conservation of momentum, mass, temperature, solute concentration and equation of state of compressible Walters B' viscoelastic fluids in presence of rotation are ![]() | (1) |
![]() | (2) |
![]() | (3) |
![]() | (4) |
![]() | (5) |
![]() | (6) |
![]() | (7) |
![]() | (8) |
![]() | (9) |
![]() | (10) |
denote respectively the perturbations in temperature T, solute concentration C, fluid velocity, pressure p and density
Here
stand for kinematic viscosity, kinematic viscoelasticity, thermal diffusivity, solute diffusivity and specific heat at constant pressure.The change in density
caused by the perturbations and in temperature and solute concentration is given by ![]() | (11) |
![]() | (12) |
![]() | (13) |
![]() | (14) |
![]() | (15) |
denotes the z-component of the vorticity and
We consider the case in which both the boundaries are free and here temperatures, solute concentrations at the boundaries are kept constant. Then the boundary conditions appropriate to the problem are ![]() | (16) |
![]() | (17) |
are the wave numbers along the x- and y- directions respectively.
is the resultant wave number and n is, in general, a complex constant.Using (17), equations (12)–(15) in non–dimensional form become ![]() | (18) |
![]() | (19) |
![]() | (20) |
![]() | (21) |
is the Prandtl number,
is the Schmidt number,
is the Taylor number,
is the dimensionless compressibility parameter,
and
. Eliminating
between equations (18) to (21), we obtain 
![]() | (22) |
is the Rayleigh number,
is solute Rayleigh number.The boundary conditions (16) transform to ![]() | (23) |
characterizing the lowest mode, equation (22) yields the dispersion relation as![]() | (24) |
.
, equation (24) reduces to ![]() | (25) |
viscoelastic fluid behaves like an ordinary compressible Newtonian fluid.If the non-dimensional numbers
accounting for the rotation, solute gradient and compressibility effects be kept as fixed in equation (25), then we find that ![]() | (26) |
denote, respectively, the critical Rayleigh numbers in the presence and absence of compressibility. Since critical number is positive and finite, so G > 1 and we obtain a stabilizing effect of compressibility as its result is to postpone the onset of double-diffusive convection. The cases G < 1 and G =1 correspond to negative and infinite values of critical Rayleigh numbers in the presence of compressibility, that are not relevant in the present study. To investigate the effect of stable solute gradient and rotation, we examine the behaviour of
analytically. It is evident from equation (25) that![]() | (27) |
![]() | (28) |
against wave numbers x for fixed values of G = 10 and
= 50 and for
= 10, 30 50, 80, 100. It is clear from Figure 1 that the stable solute gradient has stabilizing effect on the system. Figure 2 depicts
against x for fixed values of G = 10,
= 100 and variable values of
= 10, 30, 50, 70, 90. It is evident from Figure 2 that Rayleigh number is increasing with the increase in the value of
, thus implying the stabilizing effect of rotation.![]() | Figure 1. Variation of Rayleigh number R1 against wave number x for G = 10 and T1 =50 |
![]() | Figure 2. Variation of Rayleigh number R1 against wave number x for G = 10 and S1 =100 |
![]() | (29) |
,![]() | (30) |
is the complex conjugate of
.Putting
and then equating real and imaginary parts of (29), we obtain
![]() | (31) |
![]() | (32) |
may be positive or negative, that is, there may be instability or stability in the presence of solute gradient. Equation (32) yields that
or
which means that the modes may be non-oscillatory or oscillatory. The oscillatory modes are introduced due to the presence of rotation and solute gradient, which were non-existent in their absence.
real. Separating real and imaginary parts of equation, we have by eliminating 
where
.Thus
are the sufficient conditions for non-existence of overstability, the violation of which does not necessarily imply occurrence of overstability.
are the sufficient conditions for the non-existence of overstability, the violation of which does not necessarily imply occurrence of overstability.