*2.1. Chalcopyrite Concentrate and Reagents*

The chalcopyrite concentrate sample was supplied by a Chilean mining company. Two particle sizes were used with P80 of 29.8 and 60.66 μm, respectively. The particle sizes distributions were determined using a Microtrac model S3500 laser diffraction Particle Size Analyser (PSA) (Verder Scientific, Newtown, PA, USA). The chemical composition was determined by digestion which was then analyzed using inductively coupled plasma atomic emission spectroscopy (ICP-AES, ICPE-9000, Shimadzu, Tokyo, Japan). The mineralogical characterization was determined by quantitative evaluation of the minerals by scanning electron microscopy (QEMSCAN) and morphology of the samples were determined by Scanning Electron Microscopy (SEM), equipped with a microscope-coupled X-ray dispersive energy analyzer, SEM-EDX (model JSM 6360 LV, JEOL Ltd., Tokyo, Japan). Table 3 shows the chemical analysis and mineralogical composition of the concentrate sample. Chemical analysis showed that the concentrate included mostly Cu, Fe and S. The XRD pattern (Siemens D5000 X-ray diffractometer, Bruker, Billerica, MA, USA) shows that the sample was mainly composed of chalcopyrite and pyrite, with small amounts of covellite and chalcanthite. Based on this data, the sample is composed of 63% of copper sulfides, following by 24.3% of other sulphides.


**Table 3.** Chemical and mineralogical analysis of the chalcopyrite concentrate.

Sodium nitrate (99.5% absolute, Merck, Darmstadt, Germany) and sulfuric acid (95–97%, Merck, Darmstadt, Germany) were used in the leaching tests. The scrubbing solution was prepared with NaOH (99.0% absolute, Merck, Darmstadt, Germany) and distilled water.

Seawater was used as source of water and chloride ions in all experiments. It was collected from the coast in San Jorge Bay (Antofagasta, Chile). The sample of seawater was passed through a quartz sand filter (50 μm) and a mechanical polyethylene filter (1 μm) to remove insoluble particulate matter. Table 4 shows the composition of the seawater, which was obtained by different analytical techniques (argentometric method, atomic absorption spectrometry-AAS and volumetric analysis).


**Table 4.** Major composition of seawater from San Jorge Bay, Chile.

#### *2.2. Experimental Procedure*

The experimental test work was carried out in three stages. The first stage was the leaching of chalcopyrite concentrate. A Taguchi L6(23) orthogonal array design experimental setup was employed in Series I to clarify the effects of sulfuric acid and nitrate concentration and a Taguchi L4(22) orthogonal array in Series II for temperature and particle size on copper extraction. Leaching experiments were performed in 2 L jacketed glass reactors. Each reactor was loaded with 1 L of leaching solution (0.1, 0.5 and 1.0 M of sulfuric acid 0.1 and 0.5 M of sodium nitrate and seawater as the source of water and chloride ions). Once the solution reached the desired temperature (room and 45 ◦C), 50 g of solid sample (P80 of 29.8 and 60.66 μm) was added to the reactor.

The pulp was stirred to a homogenous mix with a propeller at a rotation speed of 450 rpm. A 10-mL aliquot of the leached solution was withdrawn periodically during the test and analyzed for Cu and Fe using the atomic absorption spectrometry (AAS method, model 2380, Perkin Elmer, Wellesley, MA, USA). Redox potential (ORP) and pH were measured throughout the test with a portable meter (model HI991003, Hanna, St. Louis, MO, USA). All experiments were conducted in duplicate. The solid residues were carefully filtered, washed with distilled water, dried at 60 ◦C and samples were taken for mineralogical characterization and particle size determination.

The second stage was the alkaline gas scrubbing of NOx produced in the leaching stage at 45 ◦C (Test 9, see Table 5). At this temperature, enough NOx gases were produced. Scrubbing was conducted in duplicate using a 1.0 M sodium hydroxide solution. The initial pH of the scrubbing solution was 13. The gas was transported to the scrubber by pumping air to the reactor through a tube. The scrubber column was 25 cm high, the volume of the absorbing solution was 250 mL, and air was supplied at the rate of 3.0 L/min. The flue gases were dispersed into the solution using a fritted bubble disperser. The bubble size of the dispersed gas was generally >1 mm. All scrubbing tests were carried out at 2 ◦C. Figure 1 shows the schematic diagram of leaching and scrubbing tests. The amounts of NaNO2-NaNO3 and NaOH salt formation produced from the scrubbing solution were obtained by chemical analysis and ionic balance using atomic absorption spectrometry (AAS), volumetric analysis and colorimetric assays.

Last stage of the work was the recovery of nitrate-nitrite salts by evaporation. The scrubbing solution was dried at 100 ◦C to determine the amount of salts that crystallized with the evaporating solvent. Thermal stability was studied from crystallized solids. Nitrite-nitrate salts were characterized by differential scanning calorimetry using a Mettler Toledo TGA/DSC 1 StarSystem, (NETZSCH, Bavaria, Germany). A 10 mg sample was used in the TGA/DSC determinations. This equipment can

measure the heat capacity of a sample to determine the melting temperature and quantify the degree of crystallinity and characterization of the material.

**Figure 1.** Schematic diagram of leaching-scrubbing system.

#### **3. Results and Discussion**

Table 5 summarize the leaching test results of Series I and Series II. The tests are discussed in the following sections. The tests (1 to 10) were carried out in duplicate.


**Table 5.** Serie I (Leach time 94 h, and Serie II (Leach time 46 h). Leaching test results in seawater.

#### *3.1. Series I. Leaching Test Results at Room Temperature*

#### 3.1.1. Variation of Redox Potential (ORP)

Figure 2 shows the variations of the oxidation redox potential (ORP) according to leaching time. It is clear that redox potential increases when the dissolution of copper increases (see Tests 2 and 3, at 56 and 26 h, respectively).

The results of tests 2 and 3 indicate that ion oxidation increases with higher levels of aciditsiy in the solution due to NOx formation as an oxidant. NOx gas did not form immediately, but was observed with extending leaching time. Once the NOx is formed, this gas tends to oxidize rapidly due to the action of the injected oxygen, turning into a brown gas, which is characteristic of NO2(g). The potential remained above 700 mV vs. Ag/AgCl, facilitating the leaching of chalcopyrite.

The redox potential of Test 1 with 0.1 M H2SO4–0.1 M NaNO3 solution remained low, between 400 and 450 mV vs. Ag/AgCl, indicating a low level of chalcopyrite oxidation under these conditions. When 1.0 and 0.5 M of acid and 0.5 M of NaNO3 were added to the leaching solution the redox potential remained above 700 mV vs. Ag/AgCl throughout tests 4 and 5 enhancing the dissolution of copper.

**Figure 2.** Dissolution of copper (%) from chalcopyrite concentrate and averaged ORP (mV vs. Ag/AgCl) values over time (hours) for Series I at different concentrations of H2SO4 and NaNO3. (-) Test 1 (H2SO4 = 0.1 M and NaNO3 = 0.1 M); () Test 2 (H2SO4 = 0.1 M and NaNO3 = 0.5 M); () Test 3 (H2SO4 = 0.5 M and NaNO3 = 0.1 M); (•) Test 4 (H2SO4 = 0.5 M and NaNO3 = 0.5 M); (X) Test 5 (H2SO4 = 1.0 M and NaNO3 = 0.1 M); () Test 6 (H2SO4 = 0.5 M and NaNO3 = 0 M). (Test conditions: 94 h, P80 = 60.66 μm, 450 rpm and room temperature). The copper recoveries are showing with black dotted line in axis secondary and averaged ORP values are showing with red dotted line.

#### 3.1.2. Effect of the Nitrate Concentration

The influence of nitrate ion concentrations on copper extraction was investigated at room temperature. Figure 2 shows the effect of the addition of initial NaNO3 concentration. Copper extraction in solutions with 0.1 M of H2SO4 increased from 11.5% to 27.3% when the concentration of NaNO3 increased from 0.1 M (Test 1) to 0.5 M (Test 2). The dissolution of copper leached in a solution with 0.5 M of H2SO4 increased from around 47 to 77% when the concentration of NaNO3 increased from 0.1 M (Test 3) to 0.5 M (Test 4). This result agree with other found in literature [4] and shows that nitrates with a standard potential of 0.96 V are a powerful oxidants.

#### 3.1.3. Effect of Sulfuric Acid Concentration

Figure 3a shows the results of leaching copper concentrate at different concentration of acid and sodium nitrate. The dissolution of copper clearly is affected by the initial addition of both sulfuric acid and sodium nitrate concentrations to the leaching solutions. Copper extraction clearly increased with higher acid concentration (Tests 1, 3 and 5). Without enough H<sup>+</sup> ions in the system (Test 1), copper extraction was negligible (10–11.5%). The experiments also showed little or no effect of sodium nitrate on copper recovery under the conditions studied (Test 1).

**Figure 3.** (**A**) Effect of the acid concentration on the extend of the copper concentrate at room temperature and (**B**) Effect of concentrations of H2SO4 and NaNO3 on the leaching efficiency of copper concentrate.

The copper extraction rate in Test 5, with a high acid and low sodium nitrate content, was 55.7%. Although the low concentration of the oxidant the potential reminds high. This could be due to

the formation of a new oxidant in the NOx system (perhaps Fe3<sup>+</sup> according with Figure 7), but less powerful than NaNO3.

As shown in Figure 3A, the best copper extraction rate (76.9%) was obtained in Test 4, with a solution containing sulfuric acid and sodium nitrate both at a concentration of 0.5 M. These highlights, presented in Figure 3B, shown the importance of the role of the concentration of H<sup>+</sup> ions compared to that of the sodium nitrate (Tests 1–3 and Tests 2–4). However, increasing amount of nitrate ion concentration enhances the copper extraction rate for all acid concentration used in the leaching solution. These results are agreed with results from Hernández et al. [11].

Table 6 shows the effect of parameters (H2SO4 and NaNO3) using the analysis of variance (ANOVA) module of the Minitab 19 software (version 19, Pennsylvania State University, State College, PA, USA). According to Table 6, H2SO4 had the largest contribution (65.45%) for copper extraction compared to NaNO3 with 32.6% approximately.


**Table 6.** Analysis of variance (ANOVA) for Cu extraction.

The data of Tests 3–5, showed in Figure 2, also showed that dissolution rates slowed down in the final stage of the leaching. This is thought to be due to a formation of a layer of sulfur on the chalcopyrite particles surface (cf. Figure 4a,b). The morphology of the layer shows irregular grains with porosity that perhaps allows the solution to flow into the particles of chalcopyrite. This suggests that dissolution reaction is initially controlled by a surface mechanism, and later by diffusion reaction mechanism [16]. The extent of sulfur formation was influenced by reagent concentrations and the leaching conditions.

**Figure 4.** SEM micrographs and EDS spectrums of (**a**) residue (Test 4) and (**b**) Initial chalcopyrite concentrate sample.

#### *3.2. Series II. The E*ff*ect of Temperature and Particle Size on the Leaching Rate*

The effects of increase temperature and particle size on the dissolution of copper Tests 7 to 10 of Series II (Table 5). It is not surprising that the leaching rate increases as the temperature increases from room temperature (tests 8 and 10) to 45 ◦C (Tests 7 and 9) seen in Figure 5. Extraction was relatively rapid at both temperatures during the first 24 h (80% and 60% of copper extraction at 45 ◦C and room temperature respectively), but then slowed down, which was related to the formation of a thin sulfur layer on the chalcopyrite surface grains as shows in Figure 4a (see Section 3.2.1).

**Figure 5.** The effect of particle size and temperature on leaching efficiency of copper concentrate.

It seems that particle sizes between 29.8 and 60.66 μm have a negligible effect on the leaching of chalcopyrite under all conditions studied, according to Table 5. All ORP values (Series II) were over 700 mV vs. Ag/AgCl. This was due to the presence of relatively high concentrations (0.5 M) of NaNO3 and H2SO4. It also shows that the ORP of finer particle size ore takes longer to reach 700 mV when leaching is carried out at room temperature (Test 8).

#### 3.2.1. Characterization of the Residues

The solid leaching residue at 45 ◦C, with a P80 of 60.66 μm (Test 9) was characterized by XRD (Table 7) and SEM-EDS analyses. Results are shown in Figure 6.


**Table 7.** Mineral composition of the residue (Test 9).

Table 7 shows the mineral phases in the leaching residue identified by XRD, indicating the presence mainly of elemental sulfur (46%) and gangue minerals (46%), with 5.9% of unleached chalcopyrite. The presence of large quantities of sulfur confirms the expectation that elemental sulfur forms during the leaching. The presence of sulfur is also confirmed by SEM and EDS spectra, as shown in Figure 6. This residue was selected to analyze the products formed due to the highest extraction of copper. The spectrum and the micrograph corroborated the presence of a sulfur layer on the chalcopyrite surface. The grains of sulfur have intergranular porosity. This porosity decreases with the leaching progress. The data also shows that other major sulfide minerals were completely dissolved in the first

48 h. It is noted also the absence of pyrite, sphalerite and molybdenite in the residue (see Table 7), as compared to the feed material (chalcopyrite concentrate) shown in Table 4.

**Figure 6.** SEM micrograph and EDS spectrum of residue (Test 9).

In order to confirm that pyrite could leach under the studied conditions, a pure sample of pyrite (P80 = 60.66 μm) was leached with 0.5 M of H2SO4 and 0.5 M of NaNO3, the result showed that the entire pyrite sample was leached. According to thermodynamic calculations NO3<sup>−</sup> ions and the acidity allow Fe3<sup>+</sup> to remain in the solution and therefore acts as an oxidant. At higher ORP (600–800 mV vs. Ag/AgCl) according to the Eh-pH diagram (Figure 7) the Fe2<sup>+</sup> then could re-oxidize to Fe3+.

**Figure 7.** Diagram Eh-pH of Fe-S-N-Cu-Cl at 45 ◦C, during the interaction of chalcopyrite concentrate (pyrite included) (HSC Chemistry 9.0 version).

#### *3.3. Flue Gas Scrubbing Tests*

Gas scrubbing was tested with a solution of sodium hydroxide. Table 8 shows the results of the analysis of the scrubbing solution before the crystallization stage.


**Table 8.** Chemical analysis of nitrogen oxide scrubbing tests.

The amounts of NaNO2 and NaNO3 produced and of absorbed gas (NO2(g)), as shown in Table 8, were estimated with an ionic balance. The results are shown in Figures 8 and 9. The data shows a relationship between the scrubbing solution concentration and the percentage of crystallized salts. The alkaline concentration of the wash solution (1.0 M NaOH) allowed NaNO3 and NaNO2 salts to crystallize with a lower quantity of mass.

**Figure 8.** Amount of NaNO2 - NaNO3 and NaOH salt formation (obtained from chemical analysis and ionic balance).

**Figure 9.** Percentages of NaNO2 - NaNO3 and NaOH salt formed.

Differential scanning calorimetry (DSC) analyses were performed on the scrubbing systems (Tests A and B). Figures 10 and 11 shows the results of Tests A and Test B, in which the peaks occur at 228.48 ◦C and 186.65 ◦C, respectively. These values are different from those reported by Janz et al. [34] and Janz and Tomkins [35] for pure NaNO2 (281 ◦C) and for pure NaNO3 (307 ◦C). The difference between the crystallized products in this study and those in the literature are related to the combination of NaNO2 and NaNO3 in the samples of this study, showing the presence of a eutectic salt as a result of absorption and crystallization by evaporation. Raman spectroscopy measurements and differential scanning calorimetry data on solidified mixtures of different compositions have provided support for a simple eutectic diagram with a solid at 230 ◦C ranging from 0.25 to 0.80 in mole fraction of sodium nitrate [36].

The formation of sodium and nitrogen salts is also shown (qualitatively) by SEM-EDS patterns in sample from Test A (Figure 12). The particles crystallize, as shown by EDS analysis, in small agglomerates of nitrogen salts combined with sodium. NOx gas scrubbing is a relatively complex operation that requires optimizing several parameters like the NO to NO2 ratio, initial concentrations, bubble size, temperature and the type of alkaline solution, as these parameters influence overall scrubbing efficiency.

**Figure 10.** DSC curve for products crystallized from NOx gases in Test A.

**Figure 11.** DSC curve for products crystallized from NOx gases in Test B.

**Figure 12.** SEM micrograph and EDS spectrum of solids crystallized from Test A.

#### *3.4. Proposed Process for Leaching Chalcopyrite and Recovering Oxidants*

This paper proposes a process for leach chalcopyrite using nitrate-acid-seawater media, with subsequent recovery and recycling of oxidant salts. The recovery of sodium nitrate and sodium nitrite from nitrous gases is also shown in the process flowsheet in Figure 13.

**Figure 13.** Proposed process for chalcopyrite leaching.

The process includes leaching, absorption and crystallization. Leaching should be carried out in closed reactors with controlled agitation to ensure continuous solid-liquid contact. The pregnant leaching solution (PLS) will continue with solvent extraction (SX) and electrowinning (EW) stages. It should be noted that there are reagents that have been created and are capable of extracting copper from a matrix with a high concentration of nitrate and chloride ions [37–42]. The second nitrous gas absorption stage recovers contaminating gases (NOx) formed by dissolving the concentrate that would otherwise escape into the environment. The last stage involves recovering salts for reusing, which can be carried out by crystallization, or by returning the absorption solution in the case of an acid or neutral absorbent.

#### **4. Conclusions**

Leaching of a chalcopyrite concentrate using sulfuric acid and sodium nitrate in seawater was studied. The results of this study suggest that sodium nitrate in the presence of seawater is an effective oxidation agent that makes it possible to replace other lixiviants and recovery copper faster.

The highest copper extraction of 91% was obtained when the chalcopyrite concentrate with a P80 of 60.66 μm was leached with 0.5 M H2SO4 and 0.5 M NaNO3 at 45 ◦C for a period of 46 h.

This study shows that the rate of leaching of the finely ground ore concentrate is relatively fast for the first 24 h but then slows down. This is related to the formation of a passive layer of elemental sulfur on the chalcopyrite grains that was confirmed by XRD and SEM/EDS analysis. The analysis also showed that pyrite, sphalerite and molybdenite were completely dissolved within the first 48 h.

Leaching of chalcopyrite concentrate with sulfuric acid and sodium nitrate generated NO and NO2 gases. Scrubbing parts of these gases in a sodium hydroxide solution recovered sodium nitrate and sodium nitrite. The amount of sodium nitrite recovered tended to exceed the amount of sodium nitrate. The recovery of sodium nitrate can be beneficial to an industrial-scale leaching operation processing copper concentrate.

Leaching technologies with the absorption stage described above are technically feasible and environmentally friendly. Future studies should consider improving the environmental effects of leaching and making the process more economical. These might be the recyclability of recrystallized salts and the amenability of the PLS to solvent extraction for Cu recovery.

**Author Contributions:** C.I.C. contributed in research and wrote paper, P.C.H. contributed in project. L.V.-Y. and M.E.T. contributed in review and editing. All authors have read and agreed to the published version of the manuscript. All of the authors contributed to analyzing the results and writing the paper.

**Funding:** This research was funded by ANID-PFCHA/National Doctoral Program/2017-21171313, FONDECYT Project 11170179 and CODELCO Piensa Mineria scholarship.

**Acknowledgments:** César I. Castellón thanks ANID-PFCHA/National Doctoral Program/2017-21171313, Piensa Mineria scholarship from CODELCO and María E. Taboada thanks the Universidad de Antofagasta for the support provided.

**Conflicts of Interest:** The authors declare no conflict of interest.
