ABSTRACT
A novel unified implicit computational framework and a unique pore-to-cell meshing method is used to predict, by coupling the inside-drying outside-flow processes, the slow drying of a pore-network-represented porous medium placed adjacent to a laminar flow of air in a slit. The effects of numerical algorithms, spatial and temporal discretization schemes, and meshing methods in minimizing computational effort are studied. The numerical solutions of the outer and inner processes are tested against several benchmark studies. The external air velocity affects drying during the initial stages. The microstructure of the pore network (P-N is found to have a strong influence on drying.
Nomenclature
A, Ai, j | = | area of throat cross-section or cell faces |
C, Csat, C∞ | = | liquid vapor concentration/saturated concentration/environment concentration |
D, D* | = | liquid mass diffusivity |
H | = | height |
l, l* | = | throat length |
= | mass flux between two pores | |
M, Md, Mv | = | molar mass/molar mass of dry air/molar mass of vapor |
MVP | = | maximum variation percentage |
p, pa, pv, pl | = | pressure/atmosphere pressure/vapor pressure/liquid pressure |
Pe | = | Peclet number |
R | = | radius of throats |
RSN | = | random seed number |
ℛ | = | universal gas constant |
Re | = | Reynolds number |
t | = | time |
T | = | temperature |
u, Uavg, U* | = | velocity/average velocity |
x, y | = | voordinate |
x0 | = | venter of the spike |
γ | = | surface tension |
δi, j | = | distance between two cells or pores |
ρg, ρair, ρ* | = | gas density/air density |
μg,μair, μ* | = | gas viscosity/air viscosity |
α | = | width parameter of the spike |
Subscript | = | |
i, j | = | index number of cell/pore/throat |
Superscript | = | |
n | = | time step |
Nomenclature
A, Ai, j | = | area of throat cross-section or cell faces |
C, Csat, C∞ | = | liquid vapor concentration/saturated concentration/environment concentration |
D, D* | = | liquid mass diffusivity |
H | = | height |
l, l* | = | throat length |
= | mass flux between two pores | |
M, Md, Mv | = | molar mass/molar mass of dry air/molar mass of vapor |
MVP | = | maximum variation percentage |
p, pa, pv, pl | = | pressure/atmosphere pressure/vapor pressure/liquid pressure |
Pe | = | Peclet number |
R | = | radius of throats |
RSN | = | random seed number |
ℛ | = | universal gas constant |
Re | = | Reynolds number |
t | = | time |
T | = | temperature |
u, Uavg, U* | = | velocity/average velocity |
x, y | = | voordinate |
x0 | = | venter of the spike |
γ | = | surface tension |
δi, j | = | distance between two cells or pores |
ρg, ρair, ρ* | = | gas density/air density |
μg,μair, μ* | = | gas viscosity/air viscosity |
α | = | width parameter of the spike |
Subscript | = | |
i, j | = | index number of cell/pore/throat |
Superscript | = | |
n | = | time step |
Acknowledgments
We would like to thank Dr Marc Prat for providing some insightful guidance during the implementation of our simulation.