Page 113 - gas transport in porous media
P. 113

106
                           we can impose the simplification
                                                           γ
                                                                                        (6.186)
                                                   c γ = c γ   = constant              Whitaker
                           Imposition of this condition means that there are only N − 1 independent transport
                           equations of the form given by
                                             E=N−1
                                   ∂c Aγ
                                        =∇ ·       D AE ∇c Eγ ,  A = 1, 2, ... , N − 1  (6.187)
                                    ∂t
                                              E=1
                           and for the case of passive transport the boundary conditions at the fluid-solid
                           boundary are given by

                                              E=N−1

                                        − n γκ ·    D AE ∇c Eγ = 0,  at the γ –κ interface  (6.188)
                                               E=1
                                           A = 1, 2, ... , N − 1,

                           One can make use of the averaging theorem and the no-flux boundary condition to
                           show that the volume averaged form of Eq. (6.187) is given by

                                        γ      E=N−1
                                  ∂ c Aγ
                                ε γ      =∇ ·        D AE ∇c Eγ  ,  A = 1, 2, ... , N − 1  (6.189)
                                     ∂t
                                                E=1
                           At this point we decompose the elements of the diffusion matrix according to
                                                              γ   ˜
                                                   D AE = D AE   + D AE                 (6.190)
                                                      γ
                                     ˜
                           and neglect D AE relative to  D AE   so that Eq. (6.189) simplifies to
                                       γ      E=N−1
                                  ∂ c Aγ                 γ
                                ε γ      =∇ ·        D AE    ∇c Eγ  ,  A = 1, 2, ... , N − 1  (6.191)
                                    ∂t
                                               E=1
                           We can represent this simplification as,

                                                       ˜
                                                      D AE   D AE   γ                   (6.192)
                           and when it is not satisfactory it may be possible to develop a correction based on
                                                         ˜
                           the retention of the spatial deviation, D AE . The inequality represented by Eq. (6.192)
                           is equivalent to ignoring variations of D AE within the averaging volume.
                             The volume averaging theorem can be used with the average of the gradient in Eq.
                           (6.191) in order to obtain

                                                               1
                                              ∇c Eγ  =∇ c Eγ  +    n γκ c Eγ dA         (6.193)
                                                              V
                                                                A γκ
   108   109   110   111   112   113   114   115   116   117   118