Henry's constant calculations

In this notebook we'll calculate the characteristic Henry's constant at zero loading. Unlike fitting the entire isotherm with a Henry model, these methods will attempt to fit a straight line only on the initial part of the isotherm. First, make sure the data is imported.



In [1]:

    
%run import.ipynb









    



Selected 5 isotherms with nitrogen at 77K
Selected 2 room temperature calorimetry isotherms
Selected 2 isotherms for IAST calculation
Selected 3 isotherms for isosteric enthalpy calculation

Slope method

The slope method of calculating Henry's fits a linear henry model to the isotherm points. If the model does not fit, it goes to progresivelly lower pressures until it finds a good fit. For data that is very non-linear, this fit might be between p=0 and the first isotherm point.



In [2]:

    
# Slope method
isotherm = next(i for i in isotherms_n2_77k if i.material=='MCM-41')
h1 = pygaps.initial_henry_slope(isotherm, max_adjrms=0.01, logx=True, verbose=True)

isotherm = next(i for i in isotherms_n2_77k if i.material=='UiO-66(Zr)')
h2 = pygaps.initial_henry_slope(isotherm, max_adjrms=0.01, logx=True, verbose=True)

print(h1,h2)









    



Calculated K = 5.55e+04
Starting points: 42
Selected points: 2
Final adjusted RMSE: 5.32e-13
Calculated K = 6.98e+05
Starting points: 91
Selected points: 7
Final adjusted RMSE: 6.10e-03
55485.962663510676 697654.2592115065

Virial method

The virial method uses a virial model to fit the data and then obtain the Henry constant from the value of the virial function at n=0. If the data can be fit well by a virial model, the resulting Henry constant will be very accurate.



In [3]:

    
# Virial method
isotherm = next(i for i in isotherms_n2_77k if i.material=='MCM-41')
h1 = pygaps.initial_henry_virial(isotherm, verbose=True)

isotherm = next(i for i in isotherms_n2_77k if i.material=='UiO-66(Zr)')
h2 = pygaps.initial_henry_virial(isotherm, verbose=True)

print(h1,h2)









    



Attempting to model using Virial
Model Virial success, rmse is 0.21735243185170605
Attempting to model using Virial
Model Virial success, rmse is 0.49400079169123784
194297.38673183217 1365637.3640787776

More information about the functions and their use can be found in the manual.