*3.1. Characteristics of Refractory Gold Ore and Concentrate*

The refractory gold mineral sample was observed using an optical microscope. As described in Figure 2, the internal textures of the pyrite samples were investigated using nitric acid etching. Minerals containing arsenic are deeply etched by nitric acid; therefore, this procedure can greatly assist in highlighting pyrite growth zones [18]. However, arsenopyrite was not found. Optical microscopy images of pyrite with core and rim zones were mainly targeted for this study. The main features observed were pyrite and chalcopyrite; native gold and electrum were not found. The general crystals were angular and irregularly shaped, with a grain size distribution. Some typical textures, such as pores and cracks, were found in the pyrite. Chalcopyrite crystals contained in the pyrite were also observed.

**Figure 2.** Microphotographs of gold ore samples. Reflected light microscopy of the etching of pyrite grains (**a**) before and (**b**) after etching in HNO3 solution. Ccp is chalcopyrite and Py is pyrite.

Figure 3 presents the SEM-EDS analysis for the refractory gold concentrate. Idiomorphic cubic pyrite was observed. Generally, the Fe:S ratio is approximately 1:2 in pyrite; however, this phase has an S/Fe ratio of > 3, identical to that of pyrite. EDS analysis of the surface of pyrite showed the presence of arsenic, iron, sulfur, and gold, suggesting the formation of sulfur-containing arsenic and iron compounds. XRD analysis of the refractory gold concentrate revealed that consists of pyrite and quartz. The main chemical composition of the concentrate is listed in Table 1.

**Figure 3.** SEM-EDS (Scanning Electron Microscopy-Energy-dispersive X-ray spectroscopy) of pyrite in the refractory gold concentrate.

**Table 1.** Chemical composition of the complex sulfide concentrate.


#### *3.2. Leaching Experiments*

#### 3.2.1. Effect of Nitric Acid Concentration

As shown in Figure 4, As, Cu, Fe, and Zn show increasing leaching efficiency with increasing nitric acid concentration from 0.1 to 1.0 M. Au was not detected under most of the experimental conditions. The increase in the concentration of nitric acid should accelerate the reaction rate of the refractory gold concentrate. As this effect is evident only at the highest nitric acid concentrations, it suggests that oxidation is the most likely factor. Based on the experimental results, the leaching parameters of the leaching experiment conditions estimated using Equation (2) are presented in Table S1. Initially, the metal ion leaching rate was slow, but steadily increased with increased nitric acid concentration. The sample mass decreased from 19.0% at 0.1 M, 27.0% at 0.5 M and 51.5% at 1.0 M. Nitric acid concentration was shown to be effective for the decomposition of the dominant sulfide minerals, due to the oxidation of insoluble sulfides to water-soluble sulfate phases during leaching.

**Figure 4.** Effect of HNO3 concentration on the leaching efficiencies of As, Cu, Zn, and Fe from the refractory gold concentrate. The leaching conditions were a reaction time of 15 min, reaction temperature of 80 ◦C, and HNO3 concentration of 0.1, 0.5, or 1.0 M. According to Equation (2) in Table 2.


**Table 2.** Refractory gold concentrate and leach residue surface atomic composition with different treatments. The leaching conditions were a reaction temperature of 80 ◦C reaction time of 15 min, and HNO3 concentration of 0.1 and 0.5 M.

#### 3.2.2. Effect of Temperature

The results are presented in Figure 5. The As leaching increased from 84.0% to 92.5%, as the temperature increased from 80 to 120 ◦C. Initially, the As leaching rate was slow, but steadily increased with increased temperature. However, Cu leaching decreased from 85.8% at 80 ◦C to 75.8% at 120 ◦C and Zn leaching decreased from 14.6% at 80 ◦C to 10.5% at 120 ◦C. The Fe leach ability was 98.7% at 80 ◦C and slightly decreased to 94.6% at 120 ◦C. As the temperature increased from 80 to 120 ◦C, the Cu leaching rate constant increased from 0.18 to 0.36 min−<sup>1</sup> and the Fe leaching rate constant increased from 0.21 to 0.66 min-1. Based on the experimental results, the leaching parameters of the leaching experiment conditions estimated using Equation (2) are presented in Table S2. Due to the effect of temperature, it is necessary for sulfide to be oxidized into sulfate, with a minimized formation of elemental sulfur that impedes the subsequent recovery of gold.

**Figure 5.** Effect of temperature on the leaching efficiencies of As, Cu, Zn, and Fe from the refractory gold concentrate. The leaching conditions were a reaction temperature between 80 and 120 ◦C, HNO3 concentration of 1.0 M, and reaction time of 15 min. According to Equation (2) in Table S2.

*Metals* **2020**, *10*, 571

Figure 6 shows the effect of temperature on the Pb leach ability from the refractory gold concentrate at nitric acid 1.0 M. Pb leaching was increased for the first 5 min; the rate decreased after that, and leaching efficiency decreased from 57.8% at 40 ◦C, 15.3% at 80 ◦C, and was not detected at 120 ◦C. The ORP (mV) was reduced by leaching, particularly at the end of the leaching time. The ORP slowly decreased to 686 mV at 40 ◦C, 638 mV at 80 ◦C, and 623 mV at 120 ◦C (Figure 6b). These results indicate that lead complexes, such as PbNO3<sup>+</sup>, Pb(NO3)2, Pb(NO3)3 <sup>−</sup> and Pb(NO3)4 <sup>2</sup>−, are formed by oxidation at high-temperatures [10]. The respective reactions may involve the formation of a passive layer of elemental sulfur, which is then leached, and passivation by lead sulfate or basic sulfate [10]. The lower Pb leaching at > 80 ◦C is likely due to passivation by increased elemental sulfur formation. This shows that it is possible to obtain lead in solution, although lead may not precipitate at lower potentials. XRD analysis of the leach residues mainly show untreated pyrite, with a small quantity of quartz and sulfur. The reaction produced elemental sulfur (Figure 7) and there is a clear indication that surface passivation prevented further leaching. Therefore, the decrease in Pb leaching efficiency may be due to the consumption of oxidants and acids.

**Figure 6.** Effect of temperature on the leaching efficiencies of Pb (**a**) and ORP (mV) (**b**) from the refractory gold concentrate. The leaching conditions were an HNO3 concentration of 1.0 M, reaction time of 15 min, and reaction temperatures of 40, 80, and 120 ◦C.

**Figure 7.** XRD patterns of raw and leach residue from nitric acid leaching at 40, 80, and 120 ◦C. The leaching conditions were an HNO3 concentration of 1.0 M and a reaction time of 15 min.

As the temperature increased, Pb leaching decreased, which mainly occurred due to the formation of a passive layer. The leaching efficiencies of Fe in the leaching experiment were not significant (Figure 8). The Fe leaching rate was 90.5% at 40 ◦C, 93.1% at 80 ◦C, and 94.61% at 120 ◦C after 15 min. Sulfide oxidation [21] is strongly affected by temperature (> 100 ◦C) and a temperature increase has a negative effect on Pb leaching. It is noteworthy that the Pb leaching efficiency was lower than the Fe leaching efficiency. Therefore, it is difficult to simultaneously leach Pb and other base metals, such as As, Cu, and Zn, from sulfide minerals in nitric acid at elevated temperatures. In addition, Pb shows a different leaching behavior compared with the other metal ions, due to the leaching of elemental sulfur to sulfate that occurs after the complete oxidation of sulfide to sulfur [13].

**Figure 8.** Effect of temperature on the leaching efficiencies of Pb and Fe from the refractory gold concentrate. The leaching conditions were an HNO3 concentration of 1.0 M and a reaction temperature of 40, 80 and 120 ◦C.

#### *3.3. Characterization of the Passive Layer in the Refractory Gold Minerals*

XPS analysis was conducted to detect the dissolution changes of metal ions (S, Fe, and As) on the surface of minerals. Table 2 shows the changes in the chemical environment of S, Fe, and As, before and after leaching. These results indicate a change in the surface of the metal ions during nitric acid leaching. The Fe atom percent was 39.64% in the raw sample and 17.93% at 0.1 M, 18.69% at 0.5 M in the nitric acid leaching residue, indicating that the superficial pyrite in the minerals did not significantly dissolve during the reaction time of 10 min. Some Fe existed in the crystal lattice of the sulfide minerals in the refractory gold concentrate, due to the superficial pyrite on the minerals dissolving. The XPS analysis indicates that the pyrite in the refractory gold concentrate did not significantly dissolve during the nitric acid leaching at < 0.5 M. The S/Fe ratio determines the surface of sulfide mineral decomposition and affects the efficiency of passive layer removal [22]. S/Fe ratio in nitric acid leaching (reaction time of 10 min) reaches 4.34 at 0.1 M and 4.20 at 0.5 M. This indicates that a relatively lower proportion of iron was consumed. However, the atomic percentages of arsenic on the surface of the raw concentrate decreased from 6.39% to 0.11% at 0.1 M and was not detected at 0.5 M. This indicates that a relatively large proportion of arsenic was consumed.

According to the XPS reference [23], the Fe2p peak with a binding energy of 708.17 eV and the S 2p peak with a binding energy of 160.93 eV belong to pyrite. In the nitric acid leaching residue from leaching at < 0.5 M, the Fe2p peak with a binding energy of 708.54 eV and the S2p peak with a binding energy of 161.02 eV were assigned to pyrite, both having small changes compared with those in the raw sample. The S2p spectrum demonstrates that S exists in multiple oxidation states, including monosulfide (S2−), disulfide (S2 <sup>2</sup>−), polysulfide (Sn <sup>2</sup>−), elemental sulfur (S0), thiosulfate (S2O3 <sup>2</sup><sup>−</sup>), and sulfate (SO4 <sup>2</sup>−). Raw sample disulfide species of sulfide minerals are oxidized to sulfate via several intermediate steps [23,24]. The leaching of sulfide is complex, involving the dissolution of various sulfur species. For the leaching of sulfide minerals, the preferential release of metal ions such as Fe, Cu, Zn, etc., into the leaching solution results in the passive layer of surface S2<sup>−</sup>, accounting for the formation of S2O3 <sup>2</sup><sup>−</sup> and SO4 <sup>2</sup>−. This was possibly due to the S-S bonding being weaker than the Fe-S bonding, meaning that the S-S bonds were more easily broken in the lattice pyrite. The passivation layer on the residue surface mainly consisted of iron and sulfur species, as seen in Figures 9 and 10.

**Figure 9.** XPS spectra of Fe2p for (**a**) raw concentrate, (**b**) leach residue (0.1 M), (**c**) leach residue (0.5 M), and (**d**) distribution of the surface Fe species. The leaching conditions were 80 ◦C and 15 min.

Under different conditions, refractory gold concentrate leaching results in the formation of a passive layer on its surface, consisting of sulfur- and iron-rich layers. XRD analyses were conducted to confirm the changes in refractory sulfide minerals during leaching; the results of these analyses are presented in Figure 11. The intensities of the pyrite peak (111) decreased, while those of the gangue peak increased in the leach residues. Sulfur was not found in the leach residues, which means that the sulfur in the sulfides was transformed into thiosulfate or sulfate, as opposed to elemental sulfur. The (111) plane shows a higher oxidation rate than the (100) and (110) planes [24]. This is likely due to the S-S bond in the pyrite, which is weaker than that in Fe-S. This indicates that the fraction of dissolved metal ions increased as the concentration of nitric acid increased. The SEM image (Figure 12) shows the morphology of the residue after leaching for 10 min, which is not noticeably different from that of the refractory gold concentrate. However, the surface of the leach residue became uneven; it is obvious that multiple holes appeared on the surface of the leach residue particles. It can be concluded that pyrite in the sulfide minerals is not decomposed by nitric acid at < 0.5 M (Figure 11). Therefore, we investigated the increase in nitric acid concentration to enhance the recovery of gold.

**Figure 10.** XPS spectra of S2p for (**a**) raw concentrate, (**b**) leach residue (0.1 M), (**c**) leach residue (0.5 M), and (**d**) distribution of the surface S species. The leaching conditions were 80 ◦C and 15 min.

**Figure 11.** XRD patterns of raw and leach residue from nitric acid leaching at 0.1 and 0.5 M. The leaching conditions were 80 ◦C and 15 min.

**Figure 12.** SEM images of leach residues. The leaching conditions were 80 ◦C, 15 min, and HNO3 concentration of (**a**) 0.1 M and (**b**) 0.5 M.

#### *3.4. Recovery of Gold by Microwave-Assisted Leaching*

The gold content of the leach residue was analyzed using a fire assay. The effect of nitric acid concentration on gold recovery was studied, with a leaching time of 15 min (Figure 13). The original sample showed a gold content of 94.37 g/t according to the fire assay. After the leaching process with 3.0 M nitric acid for 10 min, a gold content of 126.39 g/t was obtained. Compared to the untreated refractory gold concentrate, the higher gold recovery confirms that a large portion of the gold was refractory in nature, with the gold occurring as either solid solution components in sulfide minerals or encapsulated in the sulfide minerals. The weight of the leach residue decreased when the nitric acid concentration increased. The sample mass decreased by a maximum of 69.60% after leaching at 5.0 M, indicating that the sulfide minerals were decomposed or dissolved by microwave-assisted leaching.

**Figure 13.** Effect of HNO3 concentration on the (**a**) recovery of gold and (**b**) mass loss and reaction temperature from the refractory gold concentrate. The leaching conditions were a reaction time of 15 min and an HNO3 concentration of 1.0, 3.0 and 5.0 M.

Microwave-assisted leaching experiments for gold recovery were conducted for the refractory gold concentrate for different reaction times. The results showed that the reaction time enhanced the recovery of gold. The gold recovery was approximately 132.55 g/t after 20 min of leaching with 2.0 M nitric acid, whereas the mass loss in leach residue similarly decreased with reaction time (Figure 14). The sample mass decreased by a maximum of 40.5% after 20 min. The microwave-assisted leaching experiment resulted in the loss of the constituents in the refractory concentrate.

**Figure 14.** Effect of leaching time on (**a**) recovery of gold and (**b**) mass loss and reaction temperature from the refractory gold concentrate. The leaching conditions were an HNO3 concentration of 2.0 M.

Leach residues were collected by performing microwave-assisted leaching experiments for each nitric acid concentration. When XRD analysis was conducted on these leach residues, quartz, pyrite, and sulfur were detected (Figure 15). While pyrite was detected with 3 M nitric acid, it disappeared with 5 M nitric acid. This means that the intensities of the pyrite peak decreased, while those of the gangue peak increased in the leach residues, thus indicating the dissolution of pyrite during microwave-assisted leaching. However, sulfur appeared in the leach residue at 3.0 M and 5.0 M nitric acid, even though it was not detected in the concentrate. This appears to be due to the reactions of the pyrite included in the refractory gold concentrate with nitric acid, as shown in reactions (3) and (4) [25].

$$\text{FeS}\_2 + 8\text{HNO}\_3 = \text{Fe(NO}\_3\text{)}\_3 + 2\text{H}\_2\text{SO}\_4 + 2\text{H}\_2\text{O} + 5\text{NO} \tag{3}$$

$$\text{H}\_2\text{FeS}\_2 + 8\text{HNO}\_3 = \text{Fe}\_2(\text{SO}\_4)\_3 + \text{S}^0 + 8\text{NO} + 4\text{H}\_2\text{O} \tag{4}$$

**Figure 15.** XRD patterns of raw and leach residue from nitric acid leaching at 1.0, 3.0 and 5.0 M. The leaching time was 15 min.

From the redox reactions above, it can be inferred that sulfur was generated from the leaching decomposition. However, most studies [13,19] have reported that S0 is responsible for the blocking of the surface. The transformation of S species, such as S0, on the surface is the most important intermediate during the leaching of sulfide minerals, and it is considered to be the main component hindering the dissolution of sulfide minerals. The results also revealed that a larger proportion of sulfur was transformed and the generated hydroxyl precipitated (Figure 16), which could be confirmed by SEM-EDS. More extreme reaction conditions, such as the increase in nitric acid concentration from 1.0 M to 5.0 M, facilitated the decomposition of passivation species derived from metal ion dissolution and the liberation of gangue minerals from the sulfide surface. After leaching, SEM-EDS analysis was conducted on the leach residues (Figure 16). Based on the EDS analysis, the particles contain sulfur. Leaching increased with time, due to the oxidation of insoluble sulfides to soluble sulfate phases. A larger proportion of sulfur was transformed from the pyrite lattice in the refractory sulfide, thus leaving many vacant areas and microstructures, which can effectively liberate encapsulated gold and improve the recovery of gold.

**Figure 16.** SEM images of the leach residues from nitric acid leaching at (a) 1.0 M, (b) 3.0 M and (c) 5.0 M. The leaching time was 15 min.
