Solid/Liquid/Vapour Water Properties (Gri30, liquid/vapor, and Nasa Polynomials)

[matteo...@gmail.com]

Good morning everybody,

I am a PhD student in Bio-energy and Thermodynamics.

I am struggling with the use of NASA Polynomials.

My final goal was to model the equilibrium composition of ash, given the ash composition, creating a Cantera object based on NASA polynomial for the different species. Also melting should have been taken into account.

Since I couldn't get how Cantera handles these polynomials, I tested these features with water, which properties are well known.

In the code I created (attached as Test_WaterProp) the specific enthalpy of water is modeled through different models in a temperature range from (-50 to 150 C).

The models are:

1. Gri30
2. liquid_vapor
3. NASA polynomials.
   

For the NASA polynomials, three different versions are created:

1. Only the solid ice is given as a possible species
2. Only the liquid water is given as a possible species
3. Both solid ice and liquid water are given as possible species

Every model is evaluated inside it's temperature validity.

The liquid vapor model behave perfectly from 0 to 150 °C at P=ct.one_atm.

Gri30 and liquid_vapor models have the same behavior in the vapor zone.

NASA Pure liquid model (so no solid ice is defined here) behaves as  the liquid_vapor model in the liquid zone (this makes me think they are correct here)

NASA Pure solid model (no liquid water species is defined here) starts at 0°C with enthalpy equal to the one of the liquid water minus the standard heat of fusion for ice. I therefore think also this part is correct.

NASA Solid/Liquid model stays in the middle, this shouldn't happen.

The liquid and solid fraction vs temperature are here reported for the NASA solid/liquid model only.

My questions are mainly:

1. How Cantera handles the state transition with condensed phases defined by means of NASA polynomials?
2. Why the NASA solid liquid model(yellow) doesn't follow the pure solid (blue) below 0 C, and then follows the pure liquid (green) from 0 to 100?
3. Why the transition between solid and liquid water is not sharp?

Thank you very much for any help. 

 

[Bryan Callaway]

GRI30 as implemented in Cantera assumes ideal gasses.  This poses obvious problems for anything to do with phase changes.  I've never tried to model a non-ideal gas in Cantera, so I can't tell you much more.

[Giel Ramaekers]

Dear Matteo, 

I am currently running into very similar issues as you ran into in 2018. 

I would like to model these phase transitions in Iron (melting/solidification), as well as oxidation making use of chemical equilibrium for given temperatures.

Have you found a solution to your problem? If so: would you like to share your solution?

An alternative way might be to use the NASA CEA equilibrium solver: I found that for iron this solver does obey the limits for the thermodynamic polynomials. In other words, species are not allowed to exist outside the limits of their thermodynamic polynomials. For iron I do observe very sharp changes between different solid lattices and phases.

Looking forward to hear whether you have been able to crack this nut using Cantera ;-)

Kind regards,

Giel

Op maandag 16 juli 2018 om 18:16:51 UTC+2 schreef bryan.a....@gmail.com: