Greetings!
I have prepared a several UME exemplars for the project I am working on. These were prepared from an etched Au wire (d = 0.25 mm) placed inside a PTFE tubing, with wire protruding from the tubing cca 3 mm on etched side. This setup was then sealed in a polyethylene coating in a way that only the very apex of the etched Au tip is exposed.
After visual checking the UMEs under the optical microscope, I wanted to evaluate them with cyclic voltammmetry. My plan was to check the CV response in a solution of 1 mM K4[Fe(CN)6] in 0.1 M Na2SO4 aqueous electrolyte. I prepared the solution and assembled the cell. Used the Ag/AgCl as my reference electrode and Au wire (d = 0.5mm) as the counter electrode. Counter electrode was rinsed with acetone and distilled water, flame annealed and then let cool in an argon atmosphere. Solution was sparged with argon gas for cca 30 mins before performing the CV.
Test run with Au macroelectrode at 25 mV/s looked fine in my opinion (see Fig. 1), so I proceeded with my UME exemplars. None of them showed the expected steady-state voltammetric response, but rather a highly distorted one with one or more crossovers during the cycle. Black lines on Figs 2-4 show the initial voltammetric responses of selected three different UMEs at 10 mV/s. Voltammograms of other exemplars at the same scan rate more or less resemble one of the three shapes presented in Figs 2-4. Additional electrochemical cleaning, performed by CV polarization of the UME/electrolyte interface with the cathodic limit set to -1 V vs. Ag/AgCl at 10 mV/s, had no improving effect on the voltammogram shape. Red lines in Figs 2-4 show the voltammetric response of presented UME exemplars after cleaning.
Any idea why those voltammograms look so terrible and far away from expected steady-state shape? Could be an effect of electrolyte leakage caused by imperfect seal between Au wire and polyethylene insulation, or due to some microcracks development? Or is it the effect of capacitive contribution to measured current? Or perhaps my UME design is flawed?
Hi František Vavrek
The signal you're measuring for the UMEs looks like it has a very periodic oscillation, to me that suggests external electrical noise. There are a few things I can think of that may help. If you haven't already done so I would suggest reducing the current range you are working at on moving from the macro electrode to the UME. For the macro electrode you will be ok working with something like 10-100 uA/V whereas for your UME you will probably want something more like 1-100 nA/V to get the best signal. You may also need to change the bandwidth option you are using, what works fine for a macro electrode is not always ok for a UME. Finally if you can't remove the oscillation with these sorts of filters, you may need to use a Faraday cage when working with the UME.
This article and the links within might help: https://www.biologic.net/topics/how-to-get-quality-measurements/
I hope this helps!
Samantha
Hello Samantha Catarelli
Thank you for your reply as well as for your suggestions. In the meantime I was discussing it with my colleague. She suggested it is indeed a matter of noise, since very small curents are measured here. Also pointed out that my sampling rate was set too low. I used a scan rate of 10 mV/s and potential interval of 5 mV hence the sampling rate of 2 Hz. That left all the sub-hertz noise components in my signal, causing those long-period sine-wave oscillations I observed. In other words, my signal was masked by these sub-hertz oscillations (by looking at data on Fig 3 I estimated the period of these oscillations being about 50 s). She also suggested using Faraday cage to reduce the noise as well. As for the applied current range, I switched it into proper value when replacing Au macroelectrode with my UMEs.
Last Friday I repeated the measurement, using the same setup placed in Faraday cage. Even tried a different bandwith regime as you suggested - that one had no improving effect though. However, using higher sample rate was the charm - despite of having my current signal more noisy, I was able to retrieve the "baseline" from it. That is my useful information signal. For comparison, I attached the voltammetric response from the UME exemplar that was depicted on Fig 3 in my original post. The difference is significant.
Dear František Vavrek
the range of your graph means that you have mis-connections of circuit
Khaled Faisal Qasim which particular graph (or graphs) are you referring to?
Khaled Faisal Qasim Those are all signals from ultramicroelectrodes. Currents in the range up to one or a few nanoamps are expected. The very first graph (duck-shaped CV curve) was made using the same experimental setup, connections and wiring, only WE was a gold macroelectrode in that case. The graph I posted on March 14th is of the expected shape, with steady-state current being up to cca 1-1.5 nA at max.
- Badges
- Science topic
- Similar topics
- Engineering
- Petroleum Engineering
- Oil