Four electrode impedance measurement for electrolyte-membrane experiment

[rtorsvik]

Hello!

I am looking into making a 4 wire setup for impedance measurement over a membrane as shown below (left).

I was thinking I could use the BNC adapter and connect wavegen 1 and channels 1 and 2 as follows (right) and then use the WaveForms Impedance Analysis tool to do the measurements. However, after some experiments with the software, it seems that WaveForms bases its impedance calculations on what reference resistors you chose, which I don't have in my setup. Is there any way to get around this?

The four-electrode system is interesting because the measuring electrodes ch1+ and ch2+ do not carry any current as they are separated from the electrodes applying a voltage to the electrolyte solution. Hence there should be no buildup of material, electrolysis, or electrode polarization impedance at the measuring electrodes.

 

Some similar topics were touched upon in this thread https://forum.digilent.com/topic/16915-ad2-as-a-electrochemical-impedance-meter-is-possible/

 

Best regards
rtorsvik

 

 

[attila]

Hi @rtorsvik

The device AWG output controlled in voltage and the "Constant Current" in Impedance Analyzer is solved by software adjusting the output according the measurements, similar to the "Custom" mode examples:

 

With AD3, ADP3X50 and newer devices the AWG's digital output can be captured but for the precision needed in the IA the actual AWG output or DUT input is captured with Scope channel 1.
The IA interfaces and API simplifies the usage by hiding the AWG control and raw Scope data, but custom NA/IA measurement can be implemented with script or custom app using the raw AWG control and Scope data.

[attila]

Hi @rtorsvik

The 1+ should be connected to the W1, this is done in the IA adapter, to measure the exact output of the W1 for reference, since precise phase and amplitude are important when measuring the second node with 2+
The scope inputs are 1M, so if you are not using additional reference resistor you can enter a huge number, like 1T Ohm in the Calibration dialog.

See a follow-up on related post:

 

[attila]

Hi @rtorsvik

The setup should work since the measurements are done on the Scope inputs. Take in consideration the 1M of the inputs.

[rtorsvik]

Hi @attila, thank you so much for your feedback.

Based on your answer I have done some more research and now I have some follow-up questions.

I added some buffer stage opamps with high input impedance such that I should not have to take voltage drop over the 1M internal pulldown resistors into account. Also, this should limit the current through the pick-up electrodes in the middle. Now, all the current going through the current carrying should go to GND and I should be able to calculate impedance Z as Z(ω) = ΔV(ω) / I(ω), where ΔV(ω) is the Fourier transform of ΔV(t), the voltage drop over the membrane, and I(ω) is the Fourier transform of I(t)

So the questions regarding this solution:

1. If you set the WaveForms impedance tool mode to "Constant Current" and a current to for instance 1mÃ, this will give me a sinusoidal excitation where the peak or RMS currents are 1 mA? ( i.e. W1 = 1mA*sin(ωt) )
2. Are you able to get/read the internal data from the wave generator so that I can actually process the data? (using the script tool maybe)