Stabilization and Utilization of Pyrite under Light Irradiation: Discussion of Photocorrosion Resistance.

The control of pyrite (FeS2) oxidation from a source is a problem of great concern on treatment of acid mine drainage (AMD). Compared with air and water, the effect of light on pyrite oxidation has not attracted enough attention. However, we found that pyrite photocorrosion in the light promoted the oxidation of pyrite. Herein, we introduce a method of coating pyrite with graphene oxide (GO), which can inhibit the oxidation and photocorrosion of pyrite while it can also degrade organic pollutants. The characterization results show that a covalent bond forms between the GO and pyrite. The stable and uniform GO coating prevents the permeation of O2 and H2O and promotes the transfer of photogenerated electrons. Moreover, it changes the conduction band (CB) and valence band (VB) levels of GO-pyrite. All of these are vital for preventing the corrosion of pyrite and promoting its photocatalytic ability. More importantly, the effect of CB and VB levels on the oxidized species was discussed. The inhibition of photocorrosion is achieved by the reaction of GO with the photoinduced h+, •OH, and •O2 -. The study provides insights for source treatment of AMD under light and the reuse of massive abandoned pyrite.

Introduction

Pyrite (FeS2) is one of the most abundant sulfide minerals.[1] The oxidation of pyrite in the tailings pond is an important source of acid mine drainage (AMD). Air and water are the important factors in the oxidation of pyrite to weaken the stability of pyrite.[2] In addition to oxygen, light irradiation may also influence the stability of pyrite. Recently, the photocorrosion of sulfide minerals has received a great deal of attention. Photocorrosion results from the photogenerated electrons/holes and the interaction among photocatalysts, photogenerated electrons/holes, and the surrounding media such as O2 and H2O.[3] As one of the most abundant sulfide minerals, the photocorrosion of pyrite may increase the leaching of iron and the production of acid mine wastewater. Lu et al. conclude that the solar light response and photocurrent production of widespread semiconducting mineral coatings are capable of playing important roles in biogeochemical processes on Earth’s surface.[4] As the most abundant semiconducting mineral, the effect of light irradiation on the stability of pyrite deserves more attention in the treatment of acid mine wastewater by inhibiting pyrite oxidation.

However, the effect of light irradiation on pyrite is always discussed in the reduction of environmental contaminants by pyrite. Due to its surface chemical properties[5] and suitable band gap (Eg = 0.95 eV),[6] pyrite is a promising photocatalysis material in the degradation of organic contaminants.[5,7] The basic mechanism of photocatalytic degradation of organic pollutants by pyrite is the separation of photogenerated electrons and holes, which further generates various radicals to oxidize organic pollutants. From another point of view, the separated photogenerated electrons or holes may oxidize pyrite itself in the absence of organic pollutants. This is actually the photocorrosion process of pyrite.

In this context, photocorrosion resistance approaches of pyrite can be accomplished to mix pyrite with another material. The added material accelerates the transfer of photoinduced electrons/holes[8] and reacts with the charge carriers instead of pyrite to inhibit photocorrosion.[9,10]

Consequently, it is important to discover a coating that can prevent O2 and H2O and promote the transfer of charge carriers to overcome the above corrosion shortcomings of pyrite under ultraviolet visible light irradiation. Graphene is an appropriate protective coating due to its high electric potential density, electron mobility, optical absorption, and high impermeability.[11,12] In order to ensure the stability of the coating, graphene oxide (GO) was used as the protective layer on pyrite. The part of oxygen-containing functional groups in the graphene layer provide more surface modification active points and a larger specific surface area.

Therefore, a physical grinding method was used to load the GO onto the natural pyrite surface (the compound was marked as GO–pyrite). After the bonding of GO–pyrite was analyzed, the effect of photocorrosion resistance was investigated. The GO is proved to inhibit the photocorrosion and enhance the photocatalytic ability. Then, various characterization and photoelectrochemical methods were used to investigate the photocorrosion resistance and photodegradation mechanism of GO–pyrite, especially discussions for the effect of conduction band (CB) and valence band (VB) levels on the oxidized species. Our research helps to deeply understand AMD under light. The improvement of photocatalytic performance makes it possible to put abandoned pyrite waste to reuse, which achieves the dual benefits of being environmental and economical.

Results and Discussion

Characterization and Formation Mechanism of GO–Pyrite

The pyrite (Figure S1a) was a regular crystalline particle with smooth surfaces and some fine convex particles. After coating with the GO, there were obvious layer of wrinkles on the surface of the GO–pyrite (Figure S1b). Elements of O, C, S, and Fe were evenly distributed on the surface of the GO–pyrite (Figure S1e–h). The O and C are response for the GO. This implies that the GO is evenly distributed on the pyrite and the graphene oxide film forms.

The X-ray diffraction (XRD) peaks of GO–pyrite and pyrite (Figure S2) at 28.49, 33.04, 37.05, 40.72, 47.37, 56.19, and 58.93 correspond to the (111), (200), (210), (211), (220), (311), and (222) faces of the pyrite, respectively.[13,14] However, the peak intensity of the GO–pyrite was weakened, and there were no characteristic peaks for GO compared to those of pyrite. This reveals that the GO coating does not change the pyrite’s crystalline structure. The coating of GO on the surface of the pyrite may weaken the diffraction intensity of pyrite. The diffraction intensity of GO was so weak that the diffraction peak of GO could not be observed.[15,16]

High-resolution transmission electron microscopy (HRTEM) in Figure shows that there were clear lattice fringes on both the pyrite and the GO–pyrite, with a distance of 0.16 nm, corresponding to the (311) crystal plane of pyrite.[17] However, obvious 2.67 nm thick wrinkles were observed on the GO–pyrite (Figure b). The observed obvious wrinkles may result from the GO coating that are consistent with the scanning electron microscopy (SEM) results.

Figure 1

HRTEM images of pyrite (a) and GO–pyrite (b).

For the Fourier transform infrared (FTIR) peaks of GO, the 1726, 1627, 1400, and 1054 cm–1 bands were assigned to the stretching vibrations of C=O in −COOH, C=C, C–OH, and C–O, respectively (Figure a).[18−20] Peaks at 1629 cm–1 (C=C) and 1400 cm–1 (C–OH) were observed in GO–pyrite too (Figure a). Thus, the GO is successfully coated onto the surface of the pyrite. However, the stretching vibration of C=O and C–O with peaks at 1726 and 1054 cm–1 disappeared in GO–pyrite. The C=O in the −COOH may bond with the functional group of pyrite.

Figure 2

(a) FTIR spectra of GO and pyrite and GO–pyrite, (b) Raman spectra of GO and pyrite and GO–pyrite, (c) XPS spectra of C 1s for GO, (d) XPS spectra of C 1s for GO–pyrite, (e) XPS spectra of Fe 2p for pyrite, and (f) XPS spectra of Fe 2p for GO–pyrite.

The obvious peaks at 1355 and 1605 cm–1 in the Raman spectra (Figure b) can be attributed to the disordered structure (D band) and the graphite structure (G band) of GO, respectively.[21] The integrated peak intensity ratio of the D-band and G-band (ID/IG) value of GO–pyrite (1.33) was slightly larger than that of GO (0.92), suggesting a higher level of disorder for the graphene layers during the functionalization process.[22,23] This may indicate the reduction of oxygen-containing functional groups on the surface of the GO–pyrite,[19] which is consistent with the FTIR results. The enhanced ratio for GO–pyrite suggests that the GO is successfully coated onto the pyrite.

Based on the above analysis, GO is successfully loaded onto the surface of the pyrite. In addition, the interaction of GO and pyrite is not a simple physical adsorption. The carboxyl or carbonyl group on the surface of the GO may have bonded with the functional group on the surface of the pyrite.

Figure c shows the X-ray photoelectron spectroscopy (XPS) spectra of the C 1s for GO. The peak around 285.1 eV was assigned to C=C or C–C. The peak around 287.1 eV corresponds to C=O. The peak around 288.9 eV can be attributed to O–C=O.[24] This confirms the existence of carbonyl and carboxyl groups. As shown in Figure d, the XPS spectra of GO–pyrite contained three peaks at 284.8 eV (C=C), 286.7 eV (C–O), and 288.9 eV (O–C=O).[25] Compared with GO, the C=O bond at 287.1 eV disappeared for GO–pyrite, and the C–O bond at 286.7 eV appeared. This is consistent with the FTIR spectra results for GO–pyrite.

The energy values of the Fe 2p spectrum (Figure e) for pyrite were 707.1, 709.5, and 720.0 eV. The peaks around 707.1 and 720.0 eV corresponded to Fe(II)–S.[26−28] The peak around 709.5 eV corresponded to Fe(II)–O.[29] Compared with pyrite, a new peak (Figure f) attributed to Fe(III)–O around 711.8 eV appeared in the GO–pyrite.[30] This implies that some of the Fe(II)–O are oxidized to Fe(III)–O on the surface of the GO–pyrite.

Combined with the C 1s and the Fe 2p spectrum results, the C=O bond in the −COOH in the GO and the Fe(III)–O in the pyrite may form covalent bonds, which allow the GO to be stably coated onto the pyrite surface. We speculate that a stable and uniform coating of GO on pyrite via covalent bonding may act as a protective film.

Effect of GO on the Oxidation Corrosion and Photocorrosion of Pyrite

The quantity of iron ion leaching at different times was used to characterize the corrosion of the pyrite. The dark and light irradiation groups (marked with illumination in figures) were measured to evaluate the oxidation corrosion and the photocorrosion resistance effects, respectively.

The amount of total ion leaching in the illuminated case for pyrite was higher than that for the dark case (Figure a). It indicates that illumination can accelerate the oxidation process of pyrite. This is an important result in efforts to verify the effect of illumination on the oxidation of pyrite, which has received little attention in previous studies. However, the number of total ions dissolved in the GO–pyrite under the condition of light irradiation and dark was almost the same with increasing time. So, the GO coating can effectively inhibit photocorrosion for 10 h (maximum experimental time). The number of total ions dissolved from the GO–pyrite was significantly less than that from pyrite under both light irradiation and dark conditions. These results demonstrate that the GO coating can effectively inhibit the corrosion of pyrite caused by oxidation and light irradiation.

Figure 3

Concentration of total Fe ions (a) and Fe(II) (b)in the solution (pH = 3) of pyrite and GO–pyrite as a function of time under light irradiation and dark treatment, and the cyclic voltammetry (CV) curves of pyrite (c) and GO–pyrite (d) for different treatments (0 h: primary sample; illumination 10 h:10 h light irradiation; and dark 10 h:10 h dark treatment in air).

The trends in the amount of divalent ion leaching for pyrite and GO–pyrite under light irradiation and dark conditions (Figure b) were similar to those of total iron leaching. The inhibition rate of GO for total iron and divalent iron was about 65 and 70% under the dark condition and was about 65 and 75% under the light irradiation condition, respectively. Therefore, the ions dissolved from pyrite and GO–pyrite under acid conditions were mainly divalent ions. As a coating on the surface of the pyrite, GO mainly inhibits the leaching of divalent ions to protect the pyrite from corrosion.

The morphologies of the pyrite under illumination for 10 h (Figure S1c) show that the pyrite still contained crystalline particles with regular shapes. However, the particles on the surface of the pyrite were no longer smooth, and there were some micropores which were in contrast with primary pyrite (Figure S1a). It is obvious that illumination influences the stability of pyrite. The surface of the GO–pyrite was still smooth without damage or peeling after 10 h of light irradiation (Figure S1d). Thus, the GO coating on the pyrite inhibits photocorrosion.

The CV curves for primary pyrite (pyrite-0 h) and pyrite exposed to air for 10 h under light irradiation (pyrite-illumination 10 h) and dark conditions (pyrite-dark 10 h) are shown in Figure c. There were three oxidation peaks (A1, A2, and A3) and two reduction peaks (C1 and C2) on the curves of all three samples. These peaks result from the oxidation of pyrite. The intensity order of the peaks from highest to lowest is as follows: pyrite-illumination 10 h > pyrite-dark 10 h > pyrite-0 h. This illustrates that pyrite is easily oxidized, and light irradiation can accelerate this process. However, there were no oxidation or reduction peaks on the curve of GO–pyrite under the same conditions (Figure d). It means GO can inhibit the oxidation corrosion and photocorrosion of pyrite.

Figure S3 shows the XPS results of the pyrite and GO–pyrite exposed for 10 h under dark and light irradiation conditions. All of the samples exhibited four peaks around 707, 709, 712, and 720 eV, which were attributed to Fe(II)–S, Fe(II)–O, Fe(III)–O, and Fe(II)–S, respectively.[26−30] The intensity of the Fe(III)–O peak was used to indicate the oxidation degree of the pyrite. The integration results of the peak area for Fe(III)–O are displayed in Table . There were no Fe(III)–O bonds in the primary pyrite (Figure e). However, the appearance of Fe(III)–O in the pyrite after exposure to dark (Figure S3a) and 10 h of light irradiation (Figure S3c) shows that the pyrite was oxidized. In addition, the peak area of the Fe(III)–O bond for pyrite-illumination-10 h was obviously greater than that of pyrite-dark-10 h. This further demonstrates that light promotes pyrite oxidation.

Table 1

Peak Area Integral of Fe(III)–O for Pyrite and GO–Pyrite at Different Conditions

sample	0 h	illumination 10 h	dark 10 h
pyrite	0	8473.96	6795.62
GO–pyrite	4029.99	3047.11	3189.34

The order of the Fe(III)–O peak area for all of the samples (Table ) from highest to lowest is as follows: pyrite-illumination-10 h (8473.96) > pyrite-dark-10 h (6795.62) > GO–pyrite (4029.99) > GO–pyrite-dark-10 h (3189.34) > GO–pyrite-illumination-10 h (3047.11). The oxidation of pyrite was higher than that of GO–pyrite under the same conditions, whether in illumination or in the dark. Thus, it is concluded that GO can inhibit the corrosion of pyrite under both illumination and dark conditions. The Fe(III)–O in the GO–pyrite was caused by the oxidation of Fe(II)–O in the process of sample preparation. Thus, the peak strength of Fe(III)–O for GO–pyrite-0 h was stronger than that for pyrite-0 h. Once the pyrite is coated with GO, the coating prevents the oxidation of pyrite in theory. The Fe(III)–O peak areas of GO–pyrite-illumination-10 h and GO–pyrite-dark 10 h (3189.34 and 3047.11, respectively) were almost the same. This demonstrates that the GO coating is stable. It inhibits the pyrite from coming in contact with water and air. The GO coating effectively inhibits the corrosion of pyrite by oxygen and light irradiation.

The corrosion resistance experiments on the GO–pyrite indicate that the GO coating prevents the pyrite from photocorrosion. However, the photocorrosion mechanism of GO–pyrite needs further study.

Recombination and Transfer of Photoexcited Carriers

Figure a compares the photoluminescence (PL) spectra of pyrite and GO–pyrite. The GO–pyrite displayed much lower PL intensities than the pyrite. It indicates that the GO coating can effectively transfer photoexcited electrons from the pyrite to the GO and restrain the recombination of photogenerated electrons and holes. This may be due to the formation of a barrier at the GO–pyrite interface.[31,32]

Figure 4

(a) PL spectra of pyrite and GO–pyrite with an excitation wavelength of 320 nm, (b) TRPL spectra of pyrite and GO–pyrite with an excitation wavelength of 320 nm, (c) electrochemical impedance spectroscopy (EIS) of pyrite and GO–pyrite, and (d) transient photocurrent responses of pyrite and GO–pyrite.

The time-resolved transient PL (TRPL) decay spectra of pristine pyrite and GO–pyrite were used to measure the interfacial charge separation and transfer efficiency (Figure b). The average fluorescence lifetime (τa) of pristine pyrite and GO–pyrite was 0.78 and 0.94 ns (Table S1), respectively. GO–pyrite had longer carrier lifetime than those of pyrite. This may indicate that face-to-face contact interfaces and numerous high-speed charge-transfer channels exist on the GO–pyrite. They help to prolong the lifetime of charge carriers and efficiently separate carriers.[33−35]

The pyrite and GO–pyrite samples were studied using EIS to understand the transport behaviors of interfacial carriers.[36,37] The arc radius of the EIS for the GO–pyrite electrode (Figure c) was much smaller than that for the pyrite electrode, which reflected a relatively small charge-transfer resistance at the GO–pyrite electrode interface. The GO–pyrite electrode under light irradiation is more effective to separate photoinduced electron–hole pairs as well as transfer the interfacial charge than the pyrite electrode.

In addition, the transient photocurrent of pyrite and GO–pyrite are shown in Figure d. Both of the samples showed a rapidly increasing photocurrent response under light irradiation. In contrast, the photocurrent response intensity was still low in the dark. Compared to pyrite, the photocurrent response of GO–pyrite was significantly greater. This confirms the faster charge-transfer ability of the GO–pyrite. The photocurrent produced by the GO–pyrite electrode under light irradiation had a good reproducibility. The photocurrent value of the GO–pyrite electrode was 4 times than that of the pyrite. Thus, GO–pyrite can not only stably separate electrons and holes but also reduce the recombination rate of photogenerated electrons and holes.[38] The GO coating on the pyrite has an efficient interfacial electron transfer ability. The smaller charge-transfer resistance and higher photocurrent of the GO–pyrite are consistent with the longer carrier lifetime and highly efficient carrier separations of the GO–pyrite.

In the entire UV visible near-infrared region of 250–2500 nm, the pyrite and GO–pyrite absorbed light in the full band, and the absorbance of GO–pyrite was higher than that of pyrite (Figure S4). The GO promotes the light absorption ability of pyrite. The band gaps (Eg) of pyrite and GO–pyrite were 0.84 and 0.71 eV (Figure S5), respectively. The lower band gap of the GO–pyrite would capture more light. It means that the light prevention is not the reason of photocorrosion resistant for GO–pyrite. Instead, the coating of GO rapidly transfers charge carriers and consumes them to prevent pyrite from reacting with photoinduced electrons/holes.

Energy Level Changes in the CB and the VB

The synchrotron radiation photoemission spectra (SRPES) were used to study the band structure of pyrite and GO–pyrite, as shown in Figure S6. The calculation method[39] and the value of the Fermi level, CB, VB, and band gap are displayed in Table S2. The energy level diagrams of pyrite and GO–pyrite are shown in Figure . The CB and VB of pyrite were −3.7 and −4.65 eV versus the vacuum level [−0.73 and 0.15 V vs reversible hydrogen electrode (RHE)], respectively. The CB and VB of GO–pyrite were −4.18 and −4.95 eV versus the vacuum level (−0.32 and 0.35 V vs RHE), respectively. Compared to the pyrite with CB/VB levels of −0.73/0.15 V versus RHE, both the CB and VB levels of GO–pyrite shifted in the positive direction. In particular, the more positive VB level of the GO–pyrite facilitates oxidation by the holes.[39]

Figure 5

Experimentally determined energy levels of pyrite and GO–pyrite corresponding to the SRPES.

Although the slightly more positive CB position of the GO–pyrite (−0.32 V vs RHE), the reaction of dissolved oxygen and the conduction electrons in the GO–pyrite could be easily generated abundantly •O2– due to its high surface area [EΘ(O2/•O2–) = −0.33 V vs NHE, pH = 7].[40,41] The VB potential of the pyrite and GO–pyrite was 0.15 and 0.45 V versus RHE at pH 7, respectively. Although the holes are rich, they are not possible for the direct generation of •OH with the reaction of H2O or OH– because of the lower VB potential [EΘ(H2O/•OH) = 2.27 eV; EΘ(OH–/•OH) = 1.99 eV vs NHE, pH = 7.[42] Thus, the interaction of the photoinduced electrons, •O2– and H2O2, may be a potential pathway for •OH generation. The photochemical processes are listed as follows[43]

Detection of Active Oxidation Species during Light Irradiation

The electron spin resonance (ESR) results are shown in Figure a–c. There were no ESR signals of h+, •OH, and •O2– for the pyrite and GO–pyrite in the dark. However, they were observed in the pyrite and GO–pyrite under light irradiation. It is consistent with the SRPES results. The photoelectrochemical activity results of GO–pyrite manifest that GO leads to the transfer of photogenerated electrons from the pyrite to the GO, which inhibits the recombination of the charge carriers. Moreover, the enhanced light absorption capacity of pyrite, the decreased band gap, and the positive shift in the CB and VB levels of GO–pyrite hybrids indicate that the GO coating reacts with the photoinduced h+, •OH, and •O2– to prevent the photocorrosion of pyrite. However, all of the photoelectrochemical activities also hint a better photocatalytic performance of GO–pyrite with faster electron transfer ability, higher light absorption capacity, and a positive shift in the CB and VB levels. Therefore, rhodamine B (RhB) was used to assess the photocatalytic activity of GO–pyrite.

Figure 6

Electron paramagnetic resonance spectra of pyrite and GO–pyrite in the presence of DMPO as a radical scavenger. The signals were collected under light illumination and dark treatment (a) h+, (b) •OH, (c) •O2–, and (d) photodegradation efficiency of GO–pyrite to RhB in solution after adding different scavengers under light conditions. GO–pyrite: without scavenger; isopropyl alcohol (IPA):•OH scavenger; AgNO3: e– scavenger; ethanol (EA): h+ scavenger; and p-benzoquinone (p-BQ): •O2–scavenger.

Photocatalytic Activity of GO–Pyrite

First, the adsorption equilibrium of RhB (30 mg L–1) by the pyrite and GO–pyrite was about 2 h (Figure S7a). The adsorption capacity of RhB on the GO–pyrite was 3.65 times than that on the pyrite after 2 h (Table ). This may be due to the specific surface area of GO–pyrite was 11.45 times than that of the pyrite (Table S3).

Table 2

Degradation Ratio of RhB for Pyrite and GO–Pyrite

catalyst	adsorption ratio (%)	photocatalytic degradation ratio (%)	degradation ratio (%)
pyrite	11.80	8.47	19.27
GO–pyrite	43.03	40.47	66.08

The curves for 0–2 h illustrated the adsorption of RhB under dark conditions, and the curves for 2–7 h illustrated the photocatalytic degradation of RhB under light irradiation conditions (Figure S7b). The adsorption efficiency was 11.80% for the pyrite and 43.03% for the GO–pyrite at 2 h (3.65 times greater). The total degradation efficiency was 3.43 times greater for the pyrite coated with GO at 7 h (19.27% for pyrite and 66.08% for GO–pyrite, as listed in Table ). The photocatalytic degradation efficiency for GO–pyrite (40.47%) was 4.78 times greater than that for pyrite (8.47%). The results show that the GO coating improved the photocatalytic activity of pyrite. It should be noted that the RhB degradation efficiency of GO–pyrite was not high, which may be related to the concentration of RhB (30 mg L–1 in the experiment) and the quantity of GO–pyrite (30 mg in the experiment) involved in the degradation process.

In order to measure the active oxidation species in the process of RhB degradation under simulated solar light irradiation, IPA, EA, p-BQ, and AgNO3 were added as scavengers of •OH, h+, •O2–, and e–, respectively. As shown in Figure d, the removal rate of RhB decreased after all of the scavengers were added to the solutions. The order of the degradation rate of RhB for the scavengers was •OH > h+ > •O2–. The results indicate that •OH is the main active oxygen species in the photocatalytic process. When e– was captured, the photocatalytic efficiency was reduced. The reason is that the e– may affect the production of •O2– and some of the •OH. The h+ can directly oxidize RhB. However, the contribution of h+ was smaller than •OH. The h+ should be consumed except for RhB. The smaller contribution of h+ may result from part of the Fe2+ existed in the system. The following processes may have occurred

When the organic pollution is absent in the solution, the •OH, h+ and •O2– may react with the GO coating. The photocatalytic activity of GO–pyrite gives us a new insight for the source control of AMD. Except for preventing the corrosion of pyrite, reasonable utilization of pyrite from the tailings pond may deserve to be considered.

Conclusions

We discuss the effect of light on the stability of nature pyrite and report an easy, effective, environmentally friendly method of coating natural pyrite surfaces with GO. The stable and uniform GO coating prevents the permeation of O2 and H2O and promotes the transfer of photogenerated electrons. Moreover, it changes the CB and VB levels of GO–pyrite. All of these are vital for preventing the corrosion of pyrite and promoting its photocatalytic ability. More importantly, the effect of CB and VB levels on the oxidized species was discussed. The inhibition of photocorrosion is achieved by the reaction of GO with the photoinduced h+, •OH, and •O2–. This study provides a new insight for the source treatment of AMD under light and the reuse of massive abandoned pyrite.

Experimental Section

Chemicals

Natural pyrite was obtained from Wanbao Mining Ltd., China. Commercial GO (500 nm to 5 μm) was purchased from Xianfeng Nanomaterial Corporation, Ltd., China. The other chemical reagents used in the experiments were all of the analytical grade and were purchased from National Medicines Corporation, Ltd., China.

Preparation of GO–Pyrite

One milliliter of 12 mg mL–1 GO dispersion solution and 0.2 g of 100 mesh natural pyrite were ground for 5 min in the agate mortar. The mixture (GO–pyrite) was dried, sealed, and stored in the dark until it was used.

Characterization

The Brunauer–Emmett–Teller (BET), SEM, XRD, XPS, FTIR, Raman, UV–vis near-infrared spectroscopy, CV curves, fluorescence spectrometer, SRPES, EIS, and current density–time (I–T) curves were used to analyze the samples.

Stability Measurement

Ten milliliters of hydrochloric acid with pH = 3 and 0.2 g of GO–pyrite were placed in a 50 mL glass beaker. The sample was allowed to react for 0–10 h in the dark and in visible light irradiation conditions (a 500 W high-pressure xenon long-arc lamp).[44,45] The dissolved ferrous ions and the total dissolved iron ions of supernatant were detected at regular time intervals.

Photocatalytic Measurement

The initial pH value of RhB solution was 6.2. The batch reaction was conducted by adding 0.03 g of GO–pyrite to 20 mL of RhB (30 mg L–1). After a typical photocatalytic degradation procedure, the concentration of the RhB at different times was determined.

Experiment of Active Species Capture

The ESR spin-trap technique was used to measure the active oxidation species in the pyrite or GO–pyrite solution (2 mg mL–1) for 10 min of light irradiation with 5,5-dimethyl-1-pyrroline N-oxide (DMPO, 50 mM) as the scavenger at room temperature. In order to examine the active species in photodegradation, IPA (5mL L–1),[46] EA (5mL L–1),[47] AgNO3 (0.1 mmol L–1),[48] and p-BQ (2.5 mmol L–1)[49] were used as the scavengers for •OH, h+, e–, and •O2–, respectively.

The more detailed experiment section information is shown in the Supporting Information.