Studying zinc coatings with EQCM-D and EIS

In the first instance, cyclic voltammetry was conducted using the parameters listed in Table 1. The results are shown in Figure 1.

In the case of A, on the cathodic scan a single peak is detected at -1.23 V corresponding to the uninhibited deposition of Zn²⁺ to form a layer of Zn. 

On the reverse scan there is a stripping peak at -0.8 V corresponding to removal of the Zn layer. In this case, during the cycle, the deposition of the Zn is accompanied by a change in resonance frequency of about -44000 Hz, from the Sauerbrey Equation (below), corresponding to about 190,000 ng/cm² loaded on to the crystal.
 

Δφ = -C_f · Δf
 

Here, the term C_f = 4.3 ng · cm^-2 · Hz^-1 is the sensitivity coefficient for this crystal, and Δφ = Δm / A_q represents the area density. The layer thickness can then be calculated by dividing Δφ by the material density ρ.

In the case of B, the introduction of PEG into the electrolyte shrinks the first cathodic peak to a height that is about half of the original. Some small shift is seen, now being closer to -1.24 V. A second peak is observed at a more negative potential of -1.63 V. The stripping peak is at -0.78 V in this case. The change in the cyclic voltammogram upon the addition of PEG is also noted in the literature [2]. The new and changed features are attributed the adsorption of PEG on the surface of the electrode. The initial cathodic step is hampered by the PEG, which blocks some of the active sites – hence the lower peak height. The second cathodic step corresponds to the filling of active sites that are freed when the desorption of PEG occurs at this negative potential, as well as bulk Zn deposition.

Figure 2 shows the change in resonance frequency as a function of potential. This indicates that Zn deposition is slower in the presence of PEG, as evidenced by the reduced slope of the frequency change (green vs. purple curve).

The change in the CV is also reflected in the damping signal, shown in Figure 3. The damping signal corresponding for the CV conducted in A shows a double peak, while in B it shows a triple peak. A double peak is consistent first with increased roughness due to the Zn deposition, through island formation, and then a subsequent fall as the islands are converted into layers, consistent with previously reported findings [3]. Regarding B, the deposition occurs in two parts. When the PEG is desorbed from the electrode in the second step, this also increases the roughness as the remaining sites are exposed, triggering a rise in the damping signal (middle peak), which then consequently falls when the Zn deposition is completed. In both cases, the last peak corresponds to the removal of Zn and cleaning of the surface. 

Potentiostatic deposition (chronoamperometry) in conjunction with EQCM-D monitoring was also performed on solutions A and B. In the case of A, only one deposition experiment was made behind the cathodic peak, while in B, deposition was done behind both peaks in turn. See Table 1 for the details of the parameters. The results are shown in Figure 4.

Most interesting is to first consider the top green and purple curves. These correspond to A and B at the same deposition potential (-1.35 V). In the first 20 seconds of the deposition, the PEG very obviously impacts the rate of deposition as expected. The gradient is much higher without PEG than with PEG, corresponding to unrestricted PEG deposition. Eventually the gradient stabilizes and matches, which probably corresponds to bulk Zn deposition.

In Figure 4 it is also interesting to note that the deposition conducted in solution B, but at -1.8 V (bottom green curve), matches very well with the curve measured for solution A. This suggests that at  this lower potential the PEG is no longer adsorbed onto the surface of the electrode, and the deposition can proceed at pace with no restriction of active sites. 

EIS following the deposition was also conducted. The Nyquist plots are overlaid and shown in Figure 5. In green is the Nyquist plot recorded after the deposition in B at -1.35 V, and in purple the deposition in A at -1.35 V. Note that the spectrum recorded for A (purple), and for B at -1.8 V (black), show a much higher degree of similarity than the two plots recorded for solution B (green and black). The Nyquist plot recorded at -1.35 V in B (green) shows a larger charge-transfer reaction because the deposition is incomplete due to the blocking of PEG that leaves some of the surface exposed.