Episode #83: Which diffusion coefficient is calculated from Warburg circuit fitting?

[@clarem22]

Thank you so much for addressing my questions! To clarify/follow-up: 1) At 1:14:10 - I did not use real-time iR compensation during the experiment. Instead, I applied iR compensation post-run during analysis, using the PEIS measurement recorded just before the run. I was under the impression that the timing of iR compensation (whether during or after a run) didn’t matter, but based on further reading, it seems that potentiostats exhibit a time lag between potential application and current measurement, which can introduce errors when applying post-run iR compensation. Therefore, is it fundamentally the time lag inherent to the potentiostat that distorts data when post-run iR compensation is applied, particularly at higher scan rates and currents? 2) At 1:17:30 - I actually used the same resistance value for all three replicates ("r" stands for "rep" in these plots). The resistance was determined via PEIS before the entire sequence of runs: OCV-PEIS-OCV-LSV1-OCV-LSV2-OCV-LSV3. Since I’ve found that resistance typically remains stable between consecutive runs, I only measured it once for the set of triplicate scan rates.

[@Pineresearch]

Thanks for the clarification. A couple things: 1) Post-run iR compensation should only really be done if the iR drop is not very large. In some ways, it is a classic Catch-22. That is, if the iR drop is low, you can correct during the experiment or after the fact, but either way since the iR drop is pretty small it may not make much of a difference anyway. However, if it's sizable enough that it does impact your measured potential, correcting afterwards can be done but it is not the same as correcting with the potentiostat/software during your experiment. I can talk about this and show some graphics to illustrate it for you on our next livestream this coming Friday, but in short what it means is you are correcting your E axis but your scan rate was not constant. To address your last point, I am not aware of any such time lag with the potentiostat circuitry itself, though I suppose such a thing might be possible. But in general, the feedback loop of a potentiostat (assuming it is not oscillating or stuck) is extremely fast and on a quicker timescale than your electrochemistry experiment so I would be skeptical that instrument circuitry is impacting this process. 2) The only thing I will add about this part with your HER curves is that you are working against two competing phenomena in my estimation: first, the potential metric for whatever certain current density is really important and the kind of thing HER researchers are usually very focused on; and second, variation in any set of electrochemistry experiments (HER included) is ubiquitous. So, all I mean to say here is I understand your scrutiny and concern over the apparent drifting of your curves from test to test, even if as you say you are using the same Ru value each time. But I think variance is part of electrochemistry, unfortunately. I suppose that is partly the name of the game, repetition until you can obtain the best data amidst all the noise, but you will likely never eliminate some degree of variation even between tests where you did everything "the same" in your view.

[@clarem22]

​@@Pineresearch Thank you for the detailed response! And an explanation of real-time iR compensation at the next livestream would be much appreciated!

[@Pineresearch]

@@clarem22 My pleasure! I have added it to our saved queue and will plan to discuss Friday, Jan 24 around 1pm EST, it will be our livestream #84 if you need to watch it later.