How to calculate vapor-liquid-liquid equilibrium

This post was last edited by phychemlxd_HON on 2010-7-11 20:11. If you just want to know the result, it should be easy to calculate it using UniSim Design or other software. If you write the program yourself, it’s a relatively simple process. 1. Choose the distribution coefficient as the independent variable, such as the well-known Wilson initial value method. If you are engaged in research or something similar, there should be better methods for providing initial values. 2. Perform mass balance calculations to determine the phase fractions and phase compositions. 3. Calculate various fugacities or fugacity coefficients. 4. In simple cases, the fugacity coefficient can be directly assigned as the distribution coefficient. 5. Assess the change in the distribution coefficient between two measurements; if the change is small, it is considered that convergence has been achieved ； Otherwise, repeat steps 2 to 5. Step 2 can be explained as follows: Since there are three phases, there are two sets of distribution coefficients, denoted as K1i and K2i. Zi = (1-L1-L2)*yi + L1*K1i*yi + L2*K2i*yi, where L1 and L2 represent the fractions of the two liquid phases. Considering that the sum of the components is 1, we obtain the following two equations: sum(Zi*(1-K1i)/((1-L1-L2) + L1*K1i + L2*K2i)) = 0 and sum(Zi*(1-K2i)/((1-L1-L2) + L1*K1i + L2*K2i)) = 0. Using these equations, it is possible to determine L1 and L2, and thus determine yi for the two liquid phases. (This is roughly how it works; I don’t have any reference books available, so it may not be entirely correct.) In the fourth step, more actions can be taken, such as using the fugacity difference to update the distribution coefficients in order to make the iterative process more stable.

I’m not quite sure; please discuss it further. Calculating the gas-liquid-liquid three-phase system is indeed a bit tricky.