[hooman.esl]

hi every body,
I am simulating hydrate dissociation in porous media.
I have written a code using DEFINE_SOURCE for mass generation of methane and hydrate consumption(for when the pressure is below the equilibrium pressure).
I compiled it successfully, but I have a problem:

i start the simulation with deltat of 1e-05 for time step and it is converging every 4 or 5 iterations but when i want to increase the time step to 1e-04 the message of reverse flow appears and influence all the result.
what should i do? because i want to run the simulation for 40 min.

here is my code:

#include "udf.h"
#define deltaE 77330.0 // activation energy (j/kmol)
#define mwg 16. // molecular weight (kg/kmol)
#define mwh 119.5 // molecular weight (kg/kmol)
#define R 8.314 // gas constant
#define ahs 300000. // m^-1
#define phi 0.181 // porosity
#define Kd 8.06 // intrinsic dissociation constant (kmol/pa.s.m2)

/****************************************source term for methane generation**************************************** ****/

DEFINE_SOURCE(Gas_Generation,c,t,dS,eqn)
{
real source, mg, sh, T, P, Pe, Kb;
T = C_T(c, t);
P = C_P(c, t);
Pe = pow(10, (0.0342 * (T - 273.15) + 0.0005 * pow((T - 273.15),2) + 6.4804));
if (P < Pe)
{
Kb = Kd * exp(-deltaE / (R * T));
mg = Kb * mwg * ahs * phi * sh * (Pe - P);
source = mg; // it is a positive term
dS[eqn] =-1* Kb * mwg * ahs * phi * sh ;
}
else
{
source = 0;
dS[eqn] =0;
}
return source;
}
/*************************************source term for hydrate consumption*************************************** ***/

DEFINE_SOURCE(Hydrate_Consumption,c,t,dS,eqn)
{
real source, mg, sh, T, P, Pe, Kb;
T = C_T(c, t);
P = C_P(c, t);
Pe = pow(10, (0.0342 * (T - 273.15) + 0.0005 * pow((T - 273.15),2) + 6.4804));
if (P < Pe)
{
Kb = Kd * exp(-deltaE / (R * T));
mg = Kb * mwg * ahs *phi * sh * (Pe - P);
source = -1* mwh * (Kb * mwg * ahs *phi * sh * (Pe - P)) / mwg;
dS[eqn] = mwh * (Kb * mwg * ahs *phi * sh ) / mwg;
}
else
{
source = 0;
dS[eqn] =0;
}
return source;
}