In situ Electrochemical Dissolution of Platinum and Gold in Organic-based Solvent
Primož Jovanovič†, Vid Simon Šelih*†, Martin Šala†, Nejc Hodnik*‡
†Department of Analytical Chemistry, National Institute of Chemistry, Hajdrihova 19, SI-1000 Ljubljana, Slovenia.
‡Department of Catalysis and Chemical Reaction Engineering, National Institute of Chemistry, Hajdrihova 19, SI-1000 Ljubljana, Slovenia.
Supporting information
S1: EFC ICP-MS experiments
In order to synchronize electrochemical and ICP-MS response (putting both on the same time scale) the time delay should be determined. This can be achieved by causing an instant dissolution of the electrode by jumping from the OCP value to a significantly higher or lower potential by a potentiostatic pulse. By carefully monitoring the exact time that ICP-MS signal is increased the time lag of the ICM-MS can be determined (Fig.S1.1).
Figure S1.1: Determination of time delay between ICP-MS measurement response and applied potential.
Figure S1.2: Potential- and time-resolved Pt dissolution in0.1 mol/L NH4NO3aqueous based electrolyte when cycling to: a) 1.1 V and, b) 1.4 V vs. Ag/AgCl with a scan rate of 20mVs-1.
The dissolution profiles in the Fig. S1.2 exhibit predominantly cathodic dissolution.It must be noted that at the beginning of the experiment (time 0 seconds S1.2a) the dissolution also starts from zero at the same time as anodic scan starts. One might think that this is an anodic dissolution. However, due to the fact that there is always some oxide present on the surface of noble metals besides gold1 before the experiment, we can safely say that this is still cathodic dissolution. Interestingly, in the Figure S1.3a gold does not exhibit cathodic dissolution at the beginning of the experiment.
Careful inspection reveals minor shoulder after the dominant dissolution peak.This can be ascribed to the anodic dissolution due to Pt oxidation that passivates the surface. It is intriguing that even at 1.4 V passivation of Pt is preventing the corrosion. At the Au (Fig. S1.2a) anodic dissolution is observed from the beginning, however still less pronounced compared to acidic and alkaline media.2 This is most likely because gold oxidation has lower ability to passivate and prevent the anodic dissolution compared to Pt.2
Minor inconsistencies in the shapes of the dissolution profiles of the first cycles (Fig. S1.2a) are presumably due to reconstruction and maybe also due to the removal of some impurities at the freshly polished Pt surface. Nevertheless, two anodic peaks were already observed before.2 At higher potential (1.8 V in Figure 1) polycrystalline Pt clearlyexhibits two characteristic peaks (anodic and cathodic).
Figure S1.3: Potential- and time-resolved Au dissolution in0.1 mol/L NH4NO3 aqueousbased electrolyte when cycling to a) 1.1 V and b) 1.4 V vs Ag/AgCl with a scan rate of 20mVs-1.
Figure S1.4:Voltammograms of a) Au electrode and b) Pt electrode in 0.1 M NH4NO3 aqueous solution to selected upper potential limits (UPL).
Figure S1.5:Voltammograms of a) Au electrode and b) Pt electrode in 0.1 M NH4NO3 methanol solution to selected upper potential limits (UPL).
In comparison to aqueous electrolyte experiments, several ICP-MS hardware and instrumentalsettings (listed below) had to be changedin order tosuccessfully perform the experiments in organic electrolyte:
Hardware/settingorganicaqueous
sampler conePlatinumNickel
skimmer conePlatinumNickel
skimmer baseBrassStainless steel
torch injector ID1.5 mm2.5 mm
drain kitfor organicsfor aqueous
RF power (W)16001500
optional gas setting15%0%
optional gas flow0.18 Lmin-10.0 Lmin-1
spray chamber temperature- 5 °C4 °C
Optional gaswe used consisted of 20% oxygen in argon. It was introduced via the 3rd mass flow controller of the ICP-MS instrument directly to the tee-of the transfer line between the spray chamber and the torch. To find the correct optional gas flow (final: 15% setting, approx. 0.18 Lmin-1) to completely burn away the carbon in the organic electrolyte and prevent soot build upand clogging of the sampler cone orifice, the optional gas flow was slowly ramped up as plasma was observed. In a too-low-oxygen plasma an intense green light emission of C2species appears. As the flow of optional gas increases, the green glow fades. The setting we used provided that the green C2 emission “tongue” faded well before the sampler cone, however, the flow was not increased beyond that in order to protect the cones from excessively corrosive nature of oxygen containing plasma.
To further prevent the possible carbon and oxygen isobaric spectral interferences from (carbon and oxygen containing) plasma, the octopole reaction system with He as collision gas in kinetic energy discrimination mode (flow: 5 mLmin-1, barrier voltage: 16 V) was used.According to the noise ratio (S/N) limit of detection was estimated to be 15 ppb for Au and above 60 ppb forPt. Further optimization is needed forthesignal to become quantitatively viable.
S2: Electrochemical dissolution of Au in methanol based electrolyte
Figure S2: Potential- and time-resolved Au dissolution in methanol based electrolyte (0.1 mol/L NH4NO3) when cycling to a) 1.1 V, b) 1.2 V, c) 1.4 V and d) 1.6 V vs Ag/AgCl with a scan rate of 20mVs-1.
S3: Comments on the applied potentials in terms of reversible hydrogen electrode (RHE)
Potentials can be transformed to the RHE values. This is useful when pH is affecting the investigated electrochemical phenomena. The potential of the RHE is linearly dependent on the pH (ERHE = 0.059pH). Thus, by converting the potential scale of conventional reference electrodes like Ag/AgCl to the RHE potentials, one can elegantly compare the results obtained at different pH conditions with each other. In the case of the electrolytes used in this study, the potential scale could be converted to RHE scale for the aqueous electrolyte. However, for the organic media, similar to the pH, RHE is by definition not possible. Nevertheless, since the measured pH values of both electrolytes were very similar potential scale for both electrolytes should also be similar – increase the values of AgAgCl for approximately 550 mV. This is not correct. However, in our opinion, it can be taken for qualitative comparison between the reactions occurring in both systems. We provide this discussion in order to provide the reader with more insights.
When examining the Au and Pt electrodes stability in different solvents, our main argument for controlling the dissolution is the oxide passivation of the surface.This is clearly apH-dependent reaction. However, as can be seen below some dissolution reactions are pH dependent and some not. For that reason, it is also more appropriate to compare the dissolution behaviours in different electrolytes against the same reference electrode at the same pH conditions. Further studies are needed to examine in detail the effect of different surface passivations on Au and Pt in organic solvents.
Electrochemical reactions of Au dissolution:
6;
3;
3;
Electrochemical reactions of Pt dissolution:
PtO;
=
2;
PtO2;
1.Cherevko, S., Electrochemical dissolution of noble metals native oxides. Journal of Electroanalytical Chemistry 2017, 787 (Supplement C), 11-13.
2.Cherevko, S.; Zeradjanin, A. R.; Keeley, G. P.; Mayrhofer, K. J. J., A Comparative Study on Gold and Platinum Dissolution in Acidic and Alkaline Media. Journal of The Electrochemical Society 2014, 161 (12), H822-H830.