Dear All,
I'm trying to make an easy demonstration experiment. A cyclic voltammogram of a gold SPE in 0.5 M H2SO4 between 0 and 1.4 V (versus the internal Ag electrode) is recorded for multiple scan rates 5 mV/s, 10 mV/s, 25 mV/s, 50 mV/s, 100mV/s, 250 mV/s and 500 mV/s). The CVs show the expected shape.
Now come the bad part: If I plot the peak current of the reduction peak versus the scan rate, I don't get a good line. A linear fit for a plot vs the scan rate always delivers a R² that is smaller than a plot versus the square root of the scan rate.
I thought, a gold surface is the ideal demonstration for the behavior of a surface bound species, but it seems I'm wrong or do something not so smart during the experiment.
I appreciate any thoughts on this.
Dear Lutz, I am not a big expert in this topic. Please refer to this post:
https://www.researchgate.net/post/Why_is_the_current_density_during_cyclic_voltammetry_proportional_with_the_scan_rate
Did you clean the Au surface properly? Here some methods: doi:10.1016/j.mee.2008.11.045
Did you consider roughness? Here something interesting from EIS point of view: http://www.consultrsr.net/resources/eis/cpe2.htm
I´m not an expert for gold electrodes, but are you sure your reaction is truly surface bound and quasi-reversible? Maybe your process gets limited by counter ion diffusion at higher scan rates? How is the ratio between Peak current (cathodic) and peak current (anodic) (should be the same if quasi reversibel without any follow up reactions)? Could there be a follow up reaction after oxidiation or a change in surface geometry like Kirill mentioned (roughness)?
You must first make sure that the experimental conditions. There should be no any pollution or electrical leakage. Also, At least 5 times higher than that of CV concentrations compared to other voltammetric method. Then whether there is reversible.In the reversible reactions, R value is not always a good.
I attached a series of CVs with different scan rate for those, who want to know how the CV looks like.
Thanks for the comments so far.
I did some CVs in 0.5 M H2SO4 before the series of scans to clean the surface. A screen printed electrode (SPE) cannot be polished mechanically because you would just polish away your Au film. Surface roughness should not be a major issue, since the capacitive current and the Faraday current for an adsorbed species should increase linear... at least that are my theoretical expectations.
Sascha as always you ask the uncomfortable questions. Although the gold oxidation is widely used for surface determination it is a rather complicated process. During the oxidation the first two layers of gold are oxidized and during reduction the oxygen is released, as I understood the publications. I'm looking at the reduction peak. Assuming that the oxide layers is fully grown, I should have the same charge to reduce the surface again. As I increase the scan rate this charge flows in a shorter time. Current is charge per time, so I expect a linear increase with v. The counterions would be the release of oxygen. I was actually thinking if this could be a limitation, but it seems to my that a layer of gold oxide should not be diffusion limited. The peak current for the oxidation and reduction are very different, because the width of the peaks is very different. This is due to the fact that the Oxidation is not one single step but there are some processes involving gold hydroxides. Because the oxidation peak overlaps with the beginning of water splitting, it is difficult to compare the charge of oxidation and reduction. At metal disc it is often observed that the electropolishing causes a change in surface area, but I haven't observed this for SPEs. Also the repeating of multiple measurements with the same electrode didn't show any changes.
Ersin your comment about pollution is also interesting. I tried to keep chloride out of the solution and don't see the shoulder in the reduction peak or the growth of the gold chloride oxidation peak, so I expect this to have little influence. Of course, I try to make my solutions as pure as possible, but what species could be an issue in this experiment? The leakage of the instrument itself is below 20 pA/V, it is around 2 pA/V, I think.
Concerning Amal's remark: This sounds like a reasonable assumption. But how would I complete the oxide formation without having violent gas evolution at the surface, which might lead to further damage of the electrode? I use right know the internal silver reference electrode of the SPE and the potential range is 0 to 1.4 V.
Theoretical equations assume something. Once you have solids and additional reactions, they dont. So, search in that direction. Ex. when you have solids, teh dependence is not with square root of speed, but tends to be with the rate. But then, gold adheres sulfur, gets also oxygenated, and the effective area changes with each cycle