Warburg parameters

[Dave S]

Hi, I am measuring the impedance spectrum of various ion exchange resins and I am trying to model my data using LEVMW with a simple circuit that includes a Warburg element. Attached is a schematic of the circuit. There are a possible five parameters that can be specified for the Warburg element; RDE, TDE, UDE, PDE and NDE, according to the LEVM manual. I don't know what these parameters specify or how many parameters to use or what values to initially assign to the parameters. Can you possibly offer some suggestions as a starting point? I can get a pretty good fit of the semicircle by deleting the Warburg element but would like to include it because some of the data shows the 45° line at low frequencies before the semicircle portion.

 

Thanks for any suggestions you may be able to offer.

 

Dave

[Fabien]

Hi,
I am actually trying to use Warburg impedance for my equivalent circuit. I can tell you that :

• NDE is the NELEM number you want to associate to the DE component : for a warburg impédance NDE should be 9, 15 or 16.
• UDE depend of the equation you want, for NDE=9 :
• if UDE = 1 or 2, Phi will be use as exponent, otherwise, Phi is not use (=1)
  
• if UDE = 2, change the tanh() of the equation 1 (page 4-9 of the manual), to ctnh()
• PDE correspond to Phi
• I don't know what RDE and CDE represent.

To simulate your equivalent circuit, you need to choose a circuit (for exemple circuit A, page 5-6) and associate non-zero value to parameters you want to use : if you want to use R1, R5, C4 and DE3 you will respectivelly associate value to P(1), P(26), P(25) and [P(16) to P(20)].

For now I have not understood the good use of UDE parameter. The manual say the parameter P(8) must be used to define UDE instead of the UDE corresponding to the DE choosen (P(18) in my last paragraph). You can see the difference between the two enclosed files, which represent a single Warburg element.

I don't know if I am really clear. I hope it will help you.
If someone know somethink about the UDE parameter number

[Søren Koch]

Hi Dave.


One thing you can try (at least in Elchemea Analytical) is to fit the semi circle as a normal RC parallel connection first, and then use those values for the C and the R (the one in series with the Warburg). Once you have those parameters, build the complete model and put in the values determined as start guesses and the fit should converge (at least if the low frequency tail is visible).

based on the circuit shown you should get a semi circle at medium to high frequencies and then the 45 degree at low frequencies.

In Elchemea Analytical the model you show only has 4 parameters: 2 R's, one Capacitance and one admittance parameter (for the W element)


Yours
Søren Koch