Simulate a mixed solution containing dimethyl carbonate, methanol, ammonia, carbon dioxide, and water in ProII
Thread Content
A mixed solution containing dimethyl carbonate, methanol, ammonia, carbon dioxide, and water was simulated using ProII; this flow stream mainly consisted of dimethyl carbonate and methanol, with very low amounts of ammonia, carbon dioxide, and water. The temperature was 180 degrees and the pressure was 35 kilograms. The NRTL thermodynamic model was used, and no gas phase was detected. However, ammonia is a non-condensable gas under such pressure and temperature conditions. Are the previous simulation results then valid?
In the absence of actual experimental data, the first step in roughly assessing the accuracy of a simulation is to determine whether the thermodynamic methods used are appropriate: 1. Whether there are binary interaction parameters between the components in the system, and whether those parameters fall within their applicable range (temperature, pressure). 2 Does the selected thermodynamic method conform to the general selection rules? 3. Is the setup of the thermodynamic method accurate (Henry constant setting, selection of parameter databases, parameter estimation, etc.)? Personally, I believe that at 35 atmospheres, the NRTL method is no longer suitable; it is recommended to use more advanced equations of state such as SRKM that can handle polar systems, and to check their binary interaction parameters as well.
Logically, the activity coefficient method is indeed not necessarily applicable under high pressure; after checking SRKM, it seems that there is no way to input azeotrope data. This post was last edited by robinbird on 2008-6-29 18:47]
I checked the two methods; regardless of their scope of application, neither of them provides parameters for the binary interactions between DMC and other components. Is there any data on the azeotropy between DMC and methanol at 3.5 MPa in the upgraded version?
You can use ASPEN for simulation; ASPEN has the physicochemical properties of relevant compounds
Should we consider the issue of the Henry component for this problem? I think that’s fine. . .