Permeability Theory for Carbonized Membranes
Permeability was calculated with Dagan theory for membranes in this post. Pore processing was performed and r.mat files were run through the programs used to calculated permeability in submitted Science paper.
No Carbonization
Experimental: 591 ul/(min*PSI*cm^2)
Theoretical: 801 ul/(min*PSI*cm^2)
Carbonized (6.2 nm taken out of diameters)
Experimental: 289 ul/(min*PSI*cm^2)
Theoretical: 438 ul/(min*PSI*cm^2)
It seems that there is some disconnect between theory and experimental for these membranes. This could be due to pore processing or variability from TEM image. However let’s check the ratios:
Expermental untreated:carbonized = 2.0
Theory untreated:carbonized= 1.8
To me this shows that the carbonized samples have smaller pores and match about the amount of pore occlusion that we expect. Something is just off with the original pore characteristics, or not all pores are being filled with water.
UPDATE:
Previously, to calculate the carbonized theoretical permeability I had simply reduced the pore sizes by 6.2 nm. It was pointed out by Chris that the porosity drops much more than the pore size adjustment. Dave sent me a processed image of the carbonized sample and here is the new theory:
Carbonized Theoretical: 124 ul/(min*PSI*cm^2)
Carbonized Experimental: 289 ul/(min*PSI*cm^2)
Now the theory expects less permeability than the experimental data. The ratios change in the following manner:
Experimental untreated:carbonized = 2.0
Theory untreated:carbonized= 6.4
Now the drop between untreated and carbonized is not the same for both experiment and theory. I tried to reprocess the original image and got about the same theoretical permeability. Why is the prediction and experimental so dissimilar? Do we need more repeats of this?