For decades, the use of extremely toxic and environmentally detrimental cyanide in the hydrometallurgical extraction of gold has been a target of research looking for greener alternatives. One promising alternative for cyanide is non-toxic thiosulfate solutions. In fact, for carbonaceous gold ores, the recovery of gold has been shown to improve with thiosulfate solutions versus cyanide. This is attributed to the thiosulfate compound with gold not adsorbing on the carbonaceous ore, unlike that of cyanide compound with gold. In addition, hydrometallurgical leaching with thiosulfate is more economically viable versus cyanide due to the lower costs of thiosulfate salts versus that of cyanide.
By Ibrahim M. Gadala, Ph.D., P.Eng., P.E., Senior Materials Engineer, Powertech Labs (BC Hydro)
Although thiosulfate leaching poses significant promise over traditional methods, notable risks stand as obstacles to its widespread use in industrial processes. Mainly, the material degradation of assets caused by thiosulfate is not well understood. Corrosion and electrochemically induced structural failures of materials used in hydrometallurgical plants would not be surprising with the increased use of thiosulfate and therefore would be an issue for the structural materials used in this process. Naturally, stainless steels are an attractive metallurgy for use in such environments, due to their corrosion resistance combined with adequate mechanical properties, all with a relatively affordable price tag when compared to exotic materials or superalloys.
The corrosion resistance of stainless steels hinge on the passive layer which forms on its surface when exposed to the surrounding environment. The thickness is one of the most important aspects of this passive layer, which is highly consequential in determining its protectiveness. Surprisingly, the thickness of this passive layer providing the widely recognized corrosion protectiveness of stainless steels can be as thin as 1 nanometer (nm). All other factors equalized such as composition, strength, and a thicker passive film will have better and longer-lasting corrosion resistance versus a thinner counterpart.
This article provides a comparison of the passive layer thickness which forms on three stainless steels, namely 304L, 316L, and Duplex 2205, in thiosulfate solutions. Firstly, the passive layer thicknesses are estimated with models that take the electrochemical parameter outputs of electrochemical impedance spectroscopy (EIS) tests as inputs. These estimates are then compared to physical tests of passive layer thickness through Time-of-Flight Secondary Ion Mass Spectroscopy (Tof-SIMS)1, providing a direct evaluation of the accuracy of EIS-based passive layer thickness models.
The three stainless steels (SS) studied were SS 304L, SS 316L, and duplex SS (DSS) 2205. In preparation for the EIS experiments, samples from each alloy were wet-ground with abrasive paper of sequentially increasing grit and then polished down to 1-micrometer diamond suspension abrasive. Samples were etched after degreasing in acetone for 10 minutes using sonication, washing with ultra-pure deionized water, and drying in a stream of hot air. A solution of 40 vol.% nitric acid (HNO3) in deionized water was used for etching SS 316L and DSS 2205, whereas SS 304L was etched with a solution of 10 wt.% oxalic acid in deionized water. All samples required an electrolytic etching procedure to reveal microstructural features. In this procedure, a constant potential of 0.6 V was applied using an auxiliary electrode for approximately 10-30 seconds.
Of the varying thiosulfate solutions used in the gold leaching process, both ammoniacal thiosulfate (AmTS) and calcium thiosulfate (CaTS) solutions were studied. The following points detail the composition and pH of the solutions used in the electrochemical tests:
- CaTS solution:
– 0.1 M calcium thiosulfate, 2.8 x 10^-3 M sodium chloride, and 3.2 x 10^-4 M cupric sulfate pentahydrate.
– pH adjusted to 8 with a saturated lime solution.
- AmTS solution:
– 0.1 M sodium thiosulfate, 0.3 M ammonia, and 0.05 M sodium chloride.
– pH of 11.6, unadjusted.
Electrochemical experiments were conducted using a conventional three-electrode cell, where the SS samples were the working electrodes, a saturated calomel electrode (SCE) was the reference electrode, and a graphite rod was the counter electrode. All the experiments were performed at 20°C, in open-to-air conditions, and on a Reference 600 potentiostat. The parameters of the EIS experiments were as follows:
- Amplitude of potential: 10 mV.
- Frequencies: 100 kHz to 1 mHz.
- Impedance analysis software: ZSimp-Win.
A Trift V nano ToF-SIMS instrument was used for ToF-SIMS depth profiles and analysis. The following were the parameters associated with the ToF-SIMS experiments:
- Ion gun accelerating voltage: 3kV
- Ion gun raster area of 800 µm × 800 µm
- DC current of Ar+ primary ions: 400 nA
- Sputter cycle: 5 seconds
- Etching rate: 0.2 nm/second
Passive Film Thickness Models
A detailed discussion of passive fi lm thickness models used in this study is beyond the scope of this article. More information can be found in relevant references such as 2, 3, 4, 5, and 6. The following are the models used in the present study, with the supporting background reference for each:
- Simple distribution of constant-phase element (Q)2
- Power-Law5 and 6
Calculated Film Thickness (EIS)
Using the values for the CPE parameters extracted from the EIS tests (see data in reference 2), the effective capacitances were calculated following the approaches associated with the models used to calculate film thickness. Results reveal that effective capacitance values calculated by the simple distribution and Brug’s model were comparable. Conversely, the effective capacitance values for the P-L model were considerably lower.
Using these effective capacitance values, the thicknesses of the passive layers formed at different potentials were calculated. Figure 1 shows the calculated passive layer thicknesses on each of the three SS grades studied, with each of the four models tested. Overlaid on this figure are the measured passive layer thickness results from ToF-SIMS. According to these results, the simple distribution model, Brug’s model, and H-M model predicted the passive layers of thicknesses less than 1 nm. Again, the P-L model yielded the largest value for the passive film thickness among all the studied models and was most in line with the measured results from ToF-SIMS.
Measured Film Thickness (ToF-SIMS)
The ToF-SIMS depth profiles of Fe and Cr for the films formed on the SS 304L and SS 316L alloys after 120 hours of immersion (open-circuit) in the CaTS solution are shown in Figure 2. Similarly, the ToF-SIMS depth profiles of Fe and Cr for the films formed on the DSS 2205 alloy after 6 hours of potentiostatic polarization (0.1 VSCE and 0.5 VSCE) in the AmTS solution are shown in Figure 3. Depth profiles in Figure 3 were similar in shape at both the 0.1 VSCE and 0.5 VSCE potentials. All passive films clearly show dual-layer characteristics. The thicknesses of the passive layers were determined from the ToF-SIMS results using the methodology described in reference 2. According to these results, the thicknesses of the passive films formed on the SS grades tests and the potentials they were polarized to did not exceed 4 nm and the maximum variance of the results was ±0.2 nm.
Comparing the calculated fi lm thickness results with the measured ToF-SIMS results reveals that using the simple distribution, H-M model, and Brug’s model to find effective capacitance eventually yields a gross underestimation of film thickness. This is supported by results for all alloys and solution combinations tested in this study, i.e., SS 304L, SS 316L, and DSS 2205 in CaTS and AmTS solutions. Conversely, comparing the calculated film thickness results with the measured ToF-SIMS results reveals that uses using the P-L model results in a much more accurate estimation of the film thickness, supported by the ToF-SIMS measurements as a benchmark.
The passive layer thicknesses of 304L, 316L, and Duplex 2205 stainless steel alloys were evaluated in thiosulfate leaching solutions using Electrochemical Impedance Spectroscopy (EIS) and Time-of-Flight Secondary Ion Mass Spectroscopy (ToF-SIMS). None of the examined stainless steel alloys (304L, 316L, and 2205) were susceptible to pitting corrosion in thiosulfate solutions at ambient temperature. Performance-wise, the DSS 2205 exhibited higher polarization resistance compared to the SS 304L and SS 316L alloys at any time of the immersion period up to 120 hours in the CaTS solution, indicating better corrosion resistance in these thiosulfate environments.
More importantly, it was shown here that the several models for passive film thickness calculation from EIS results, namely simple distribution model, Brug’s model, and Hsu-Mansfeld model, underestimate film thickness values. In contrast, the more recent power-law model much more accurately predicts the thickness of the passive film formed on the stainless steels studied. This improved accuracy was not adversely affected by solution type or potentiostatic polarization potential.