Can the NRTL parameters for liquid-liquid phase equilibrium and gas-liquid phase equilibrium be used interchangeably?

How can the gas-phase results be inferred when only liquid-liquid experiments are conducted?

NRTL is used to calculate the liquid-phase activity coefficient; of course, the parameters remain the same for the same system

Is that the case? In other words, are the NRTL parameters derived from gas-liquid equilibrium and liquid-liquid equilibrium applicable universally?

2# lxj1015 The binary parameters obtained from liquid-liquid phase equilibrium are related to those obtained from vapor-liquid phase equilibrium. A+ can use LL binary data, such as LL-ASPEN parameter data, to generate the TXY data for this binary system. Then, the VL equilibrium binary parameter data is obtained through regression of these TXY data. For more details, you can refer to this post: http://bbs.hcbbs.com/thread-641282-1-3.html I encountered the same problem as you before.

I’d also like a detailed explanation: does this calculation necessarily reflect the actual properties of the gas-liquid system? My data is used for production purposes, so the requirements are very strict

7# lxj1015 Your question is whether it is possible to derive the vl binary parameters. As for whether the parameters are accurate, the next step is verification. There are two methods for verification: one is to compare with data from literature, another is to find the azeotrope (if the azeotrope obtained by using the binary parameters you estimated is close to the one resulting from your calculations, it proves that the parameters are valid); and then there is comparison with the actual conditions in production. Explore slowly. Mine is also used for production. If you want to use it directly in production, thorough preparation in the form of literature research is required in the initial stage. Also, collect actual production data from the site and prepare detailed material balance sheets on your own. Parameter verification is a preliminary step; only if the subsequent simulations match the actual field data can it be considered a successful simulation.

Theoretically speaking. For low-pressure non-ideal systems, using the LL parameters of the locally compositional equations NRTL/UNIQUAC/WISON to estimate VL equilibrium data, that is, to calculate the activity coefficients of the liquid phase and the behavior under ideal gas conditions, yields a good degree of correlation. It is mentioned in chemical engineering thermodynamics; those few pages are the best ones. I can’t remember exactly; go and check it in Zhu Ziqiang’s book.

Theoretically, models such as Wilson NRTL can calculate multicomponent interaction parameters using the binary interactions themselves; they can make use of calculated gas-liquid equilibrium data for liquid-liquid systems, and all that is needed is a thermodynamic equation of state that can accurately describe gases. Many of the equation of state options available in Aspen are capable of providing reasonable agreement with experimental values! Just my personal opinion: loveliness:

I still don’t quite understand what’s going on

Let me add some experimental conditions: The LL experimental method is a low-pressure experiment, with the system temperature kept constant while the pressure is varied to measure the LL system. However, in the VL experiment, a fixed pressure is used while the temperature is varied for testing. Is there a difference in the correlation of the model parameters obtained from these two methods compared to those in the software?