How to obtain phase equilibrium ternary data such as NRTL from two-phase equilibrium data

I don’t know how to do this either; at most I use the known NRTL parameters to calculate Wilson’s parameters, using two-component data to determine those for three components. I’m looking forward to learning it! (⊙ o ⊙)

This should be calculable. Three sets of data (binary phase equilibrium data among the three substances) are established respectively, and regression R-1 is developed. Using the three sets of data established, regression is performed on the ternary parameters; there are approximately 8 NRTL ternary parameters.

I think what the original poster referred to as “ternary parameters” should more accurately be called “binary interaction parameters in a ternary system”. The main purpose of activity parameter models such as NRTL is to use binary interaction parameters to predict the properties of multicomponent systems. Generally, if it’s possible to estimate the binary interaction parameters using data from a binary system, there’s no need to use data from a ternary system for that purpose – unless it’s not possible to accurately estimate the binary interaction parameters using binary system data (in other words, such parameters cannot describe the ternary system). In that case, it seems appropriate to conduct regression analyses for the pairwise binary interaction parameters among the three components (AB, BC, AC). If that still doesn’t work – for example, if it’s still not possible to predict things like ternary azeotropes – then regression analysis using ternary system data can be attempted. There are many discussions on data regression on forums; you can also check out the posts I’ve posted at http://bbs.hcbbs.com/thread-565911-1-1.html. I’m not sure if I’ve explained this clearly enough; please feel free to discuss any issues. *aohaiyi8080@yahoo.com.cn

Well, I agree with what was said above; it should be the binary interaction parameter in a ternary system

2nd floor: Use the known NRTL parameters to determine Wilson’s parameters? Could you please introduce it? I’d like to learn * about it. Thank you.

6# lxj1015: I really haven’t thought about this issue before. I have an idea, but I’m not sure if it’s the same as what others have in mind: use the existing NRTL parameters to estimate the phase equilibrium data (I’ve posted a thread specifically on how to obtain phase equilibrium data), and then use the phase equilibrium data obtained through this method as “experimental data” to regress the Wilson model parameters using regression analysis

Good idea, I’ve learned from it. Thank you.

Reply to 7# *aohaiyi8080: Hello, moderator. Could you please tell me how to use NRTL to estimate phase equilibrium data? I want to use NRTL parameters derived from binary phase equilibrium data to calculate triple phase equilibrium – how should I go about doing that?

Hello, I have a question: Have you managed to solve the problem of using binary phase equilibrium parameters to calculate ternary phase equilibrium data? Could you give some guidance on how to do it? Thank you for your help:handshake