EIS on PEMs is not an easy task. Most results in the literature indeed measure with bias current as the cells are more stable under these conditions.

So lengthening the integration time might not be the perfect solution (although it's worth a trial). Make sure to measure the same spectrum twice to exclude a degradation of the cell over time (one of my favorite tips by A. Lasia). If the spectrum under bias looks completely different, maybe there is also an issue with the measurement device. Which one are you using? What is the bias compared to the amplitude?

Further, I recommend three papers on that topic. Hope this helps.

Article Change of electrochemical impedance spectra (EIS) with time ...

Article Interpretation of Low-Frequency Inductive Loops in PEM Fuel Cells

Article A Physics-Based Impedance Model of Proton Exchange Membrane ...

EIS on PEMs is not an easy task. Most results in the literature indeed measure with bias current as the cells are more stable under these conditions.

So lengthening the integration time might not be the perfect solution (although it's worth a trial). Make sure to measure the same spectrum twice to exclude a degradation of the cell over time (one of my favorite tips by A. Lasia). If the spectrum under bias looks completely different, maybe there is also an issue with the measurement device. Which one are you using? What is the bias compared to the amplitude?

Further, I recommend three papers on that topic. Hope this helps.

Article Change of electrochemical impedance spectra (EIS) with time ...

Article Interpretation of Low-Frequency Inductive Loops in PEM Fuel Cells

Article A Physics-Based Impedance Model of Proton Exchange Membrane ...

Cyclic voltametry could provide information about the adsorption-desorption processes, oxide formation both in the anodic and cationic sections of the profile in  electro catalyst you could also study an electrochemical pre action. In addition, the impedance spectra at different frequencies using the developed Theory and the different graphical representations. In my book Physical Chemistry of Materials, CRC Press, FL, USA, 2009 a complete chapter is dedicated to Impdance Spctometry

Dear Sarah,

I do not know what PEM cell is and therefore I can not give you any specific suggestions concerning this particular System Under Test (SUT). 

Your question is however also about Electrical Impedance Spectroscopy and here I can come with some general comments :

1. The plots of the kind you give(Nyquist plot) are more or less useless if one wants to discuss the general characteristics of the measured impedance. This way of "visualising" EIS data  does not tell you much more than just looking at the raw numbers (frequency and complex impedance). My first suggestion for you is therefore to plot the real and imaginary parts of the measured impedance in the log-log plot, as the log of the real and imaginary part of the impedance versus log of the frequency. Once plotted in this way, it is easier to make judgement about possible sources of experimental errors (systematic  and/or random). 

2. Any instrument has a finite precision and averaging the measurement over number of elementary runs (elementary run is a full frequency sweep) is the way to improve the quality of data. In the instruments (Impedance Analyzers) supplied for example by HP/Agilent/Keysight there is even a push button which enables you to choose the number of times you want to run your measurement (I think the maximum is about 250 !!).

3. Do not use different dc bias voltages/currents until you know that your measurements are "clean" and correct. As you self point out, the results are different.  The reason is clear, almost any SUT has some built-in non-linearity and the measured impedance might be therefore strongly dc voltage/current dependent. This type of the measurement can be very useful, but not before you can trust your experiment. Here I would suggest to calibrate your impedance measurement system carefully ( again, HP/Agilent/Keysight give a very good discussion in their "Agilent Technologies Impedance Measurement Handbook 2006".

4. Last but not least, DO NOT USE large measurement signal (oscillation level/amplitude of your input) . Keep it  less than ~kT (in volts).

I hope this will help you.

With best regards

Petr