Effect of pH on the photocatalytic removal of silver ions by β-MnO₂ particles

Silver (Ag) is a valuable metal that has been widely used in the photovoltaic, [1], medical [2], photography [3], electroplating [4], catalysis [5], and jewelry [6] industries. In particular, Ag is mainly used as a conductive ink to provide electrical pathways in the fabrication of photovoltaic cells, whereas its powerful antibacterial property with low toxicity to human cells is useful in the medical industry. The worldwide demands for Ag have surpassed more than 1 million ounces in 2018. This amount is expected to increase annually [7].

Thousands of tons of Ag have been released annually to land and water sources since 2007 [8]. The disposal of silver, which is a precious natural resource, into water and land is a wastage. Concerns have also been raised about the potential hazardous effect of Ag(I) ions on the health of human and ecosystem [9–11]. For instance, overexposure to Ag can cause permanent bluish-grey discoloration of the skin (argyria) or eyes (argyrosis) [12]. This adverse effect is worrying because about 144 billion people are using untreated surface water from the lakes, ponds, rivers, and streams daily according to the statistic from World Health Organization (WHO) [13]. Thus, the National Institute for Occupational Safety and Health recommended an exposure limit of 0.01 mg/m³ for all forms of Ag.

The recovery/removal of silver from wastewater produced by various industries is essential [14–17]. Several metal removal techniques, such as electro-deposition [18], metallic replacement [19], precipitation [20], and ion exchange [21], are implemented to recover/remove silver from wastewater. However, these techniques have their inherent drawbacks. Electro-deposition is costly and cannot recover Ag(I) ions at low concentrations. Metallic replacement is relatively economic but can only recover around 95% of silver from the reacting solution. The remaining Ag(I) ions may still be above the safety limit [22]. Different from the above-mentioned techniques, precipitation can effectively reduce silver concentration to safety levels. However, the utilization of additional chemicals during precipitation and the final treatment of the sludge are the considerable drawbacks of this technique [23]. To date, ion exchange remains the most effective technique to remove silver ions at various concentrations. Nonetheless, the high installation and maintenance cost limit its prevalent application [24–25].

Advanced oxidation is a relatively new technique for the removal of metal ions in wastewater [26–27]. This process utilizes solar energy and semiconductor photocatalysts. In general, metal ions are reduced by the photogenerated electrons upon optical excitation on semiconductor photocatalysts [28]. This environment-friendly approach is a potential alternative to conventional techniques because of its process simplicity, low installation, and low operational cost. In this process, metal ions are deposited onto the semiconductor surface during photocatalysis and can subsequently be extracted from the slurry by mechanical and/or chemical means [29–31]. The photocatalyst may be reused for the next reaction [32–33], or metals such as silver can be recovered from wastewater.

The present study aimed to investigate the potential/feasibility of β-MnO₂ particles as photocatalysts to remove Ag(I) ions from wastewater effluents. To date, the most commonly studied photocatalysts are ultraviolet (UV) light-driven titanium dioxide (TiO₂) [34] and zinc oxide (ZnO) [35] semiconductors. In the present study, β-MnO₂ particles were selected as semiconductor because of their low cost, good acid resistance, and non-toxicity. In addition, β-MnO₂ can be activated by visible light because of its narrow-bandgap energy (1–2 eV) [36–37]. Considering that wastewater effluents can be in acidic or alkaline medium, this study analyzed the effect of pH on the removal efficiency of Ag(I) ions by β-MnO₂ particles.

In each test, 50 mg/L of Ag (I) in silver nitrate solution (AgNO₃, 55152 Qrec) was prepared. Subsequently, the pH of this solution was adjusted to pH 2, 4, 8, and 10 by adding either hydrochloric acid (HCl, 37wt%, H8040 Qrec) or sodium hydroxide (NaOH, 221465 Sigma Aldrich). Manganese(IV) oxide (β-MnO₂, 243442 Sigma Aldrich) particles (1 g/L) were dispersed into the silver nitrate solution and then stirred in the dark for 30 min to achieve adsorption/desorption equilibrium. Visible light (λ = 445 nm to 665 nm) with a light intensity of 2.0 W/m² was used for the photocatalytic reaction in the removal of Ag(I) ions by β-MnO₂ particles. Ag(I) ions were removed from the solution by depositing onto the surface of β-MnO₂ particles, which triggered the formation of metal/β-MnO₂ hybrid particles.

The UV-Vis spectra of β-MnO₂ were obtained using UV-Visible spectroscopy (Varian Cary 50). Then, the optical band gap was estimated from Tauc’s relationship as shown in Eq. (1) to prove its absorption ability in visible light.

where α₀ is the proportionality constant, E_g is the optical band gap energy, and n is the power factor of the transition mode that is dependent on the nature of the material. In consideration that MnO₂ nanoparticles are indirect transition materials, the value of n = 2 was taken to draw the graph of (αhυ)^1/2 versus energy. The E_g value was estimated from the intercept of the straight line extended to the X-axis at (αhυ)^1/n = 0 in Tauc’s plot.

The concentration of Ag(I) ions in the solution was measured every 30 min by using inductively coupled plasma mass spectrometry (ICP-MS, Perkin Elmer, Nexion 300). The Ag(I) ion removal efficiency by β-MnO₂ particles was calculated using Eq. (2). These hybrid particles were collected through filtration and then dried at room temperature for further characterization.

where C₀ is the initial concentration of Ag(I) ions and C_t is the concentration of Ag(I) ions at time t (min).

The crystal structure of the particles was characterized by X-ray diffraction (XRD, Bruker D8- Discovery) operated at 40 kV and 40 mA with a copper-monochrome Cu K_α radiation source. The 2θ range was scanned from 10° to 90° at a scanning step of 0.01°. The Joint Committee on Powder Diffraction Standards (JCPDS) database was used as a reference to match the XRD spectrum. The surface morphologies of the particles were analyzed using scanning electron microscopy (SEM, Zeiss Supra 35 VP) with an accelerating voltage of 15 kV and transmission electron microscopy (TEM, TECNAI G2 20S-TWIN, FEI) with a working voltage of 200 kV. TEM samples were prepared by dispersing the particles in ethanol under sonication, followed by dispensing a few droplets of mixture on the holey carbon TEM grids. Zetasizer (Malvern Zetasizer Nano ZS) was used to analyze the change in surface charge of the particles in acidic and alkaline solutions.

The optical band gap of a photocatalyst is important because it determines the photocatalytic behavior of a catalyst in response to appropriate light sources, either in the ultraviolet, visible, or infrared regime of the light spectrum. Fig. 1(a) shows the UV-vis spectra of the β-MnO₂ particles used to estimate the optical band gap. The estimated E_g by Eq. (1) is 1.97 eV. Thus, visible light is adequate to optically excite β-MnO₂ particles for photocatalytic activity in this work.

Figure 1.  UV-vis absorption spectra of β-MnO₂ particles (a) and Tauc plot with n=2 for indirect transition of β-MnO₂ particles (b).

Fig. 2(a) shows the selective removal of Ag(I) ions from the solution at pH 6 by β-MnO₂ particles under visible light. Poor removal efficiencies (i.e., <20%) of Pb(II), Cd(II), Ni(II), and Mn(II) ions by β-MnO ₂ particles were recorded. A series of blank experiments was conducted without β-MnO₂ particles under different pH levels in dark and visible light irradiation. As shown in Fig. 2(b), the removal efficiency of Ag(I) ions was less than 2% without adding β-MnO₂ particles, indicating the role of β-MnO₂ particles as a photocatalyst in the removal of Ag(I) ions.

Figure 2.  Removal efficiency of Cr(II), Ni(II), Cu(II), Ag(I), and Cd(II) metal ions at pH 6 (the original pH of the reacting solution without additional of HCl/ NaOH) (a), removal efficiency of Ag(I) ions in dark and visible light without using β-MnO₂ particles (b), change in the concentration of Ag(I) ions with time in various pH (c), removal efficiency of MnO₂ particles in dark (d), total removal efficiency (in dark and visible light) (e), and removal efficiency in visible light (f).

The removal of Ag(I) ions by MnO₂ particles under various pH conditions was tested as shown in Fig. 2(c). In general, the concentration of Ag(I) ions reduces with time. The difference in removal rates under dark and visible light suggests MnO₂ particles have two Ag(I) ion removal mechanisms. The removal of Ag(I) ions by MnO₂ particles under dark condition, where the mixture was stirred to achieve adsorption/desorption equilibrium, is depicted in Fig. 2(d). Under strong acid medium (pH 2), only 10% of Ag(I) ions was removed by MnO₂ particles. The removal efficiency increased when the medium was changed from acidic to alkaline solution. The highest Ag(I) ion removal efficiency was recorded as 22% at pH 8–pH 10, which was twofold higher than that at pH 2. In consideration that no optical excitation source was used to produce electrons and holes, the removal of Ag(I) ions by MnO₂ particles under dark condition was through adsorption.

Fig. 2(e) shows the total Ag(I) ion removal efficiency by MnO₂ particles when the visible light was switched on. Almost 99% of the Ag(I) ions were removed under pH 4, followed by 80% at pH 2, 51% at pH 6, 24% at pH 8, and 23% at pH 10. Fig. 2(f) shows the actual removal efficiency of the Ag(I) ions by MnO₂ particles under visible light. It was calculated by subtracting the total removal efficiency (under dark and visible conditions) from the removal efficiency under dark condition. The MnO₂ particles achieved a high removal efficiency of Ag(I) ions under strong acid solution, i.e., 70% at pH 2 and 81% at pH 4. However, the removal efficiencies deteriorated to less than 5% in alkaline solution. This observation was different from that in the removal of Ag(I) ions under dark condition (Fig. 2(d)).

The best removal of Ag(I) by MnO₂ particles was achieved in the slightly acidic condition of pH 4 after 1 h of visible light irradiation. If adsorption was dominant in this study, the removal of Ag(I) would be observed in an increasing trend with the increase in pH. However, pH 4 showed the best removal efficiency. Thus, the dominant process in the removal of Ag(I) ions by MnO₂ particles under visible light irradiation was through the photocatalytic reduction of Ag(I) ions to Ag metal because the removal efficiency significantly increased upon the irradiation of visible light in acidic environment (Fig. 2(f)). This result indicates that pH played a great impact on the removal of Ag(I) ions by MnO₂ particles under dark condition and visible light irradiation. The Ag(I) ion removal mechanisms by MnO₂ particles will be discussed in Section 3.2.

Zeta potential measurement was carried out to analyze the alteration in surface charge of MnO₂ particles in acidic and alkaline solutions. As shown in Fig. 3, the original pH of MnO₂ particles in the AgNO₃ solution was −23.9 mV. The surface charge became more positive after adding acid (pH 2 and 4) and became negative in the alkaline environment (pH 6 and 8). This observation can be ascribed to the fact that transition metal oxides such as MnO₂ are basic in nature. Hence, the negative surface charge was detected in the MnO₂ at its original pH of 6, as shown in Eq. (3). After adding acid/alkaline, the ionization state of the MnO₂ surface was altered in accordance with Eqs. (4) and (5) [38]:

Figure 3.  Change in surface charges of MnO₂ particles under various pH solutions.

Hence, the relatively high removal of Ag(I) ions at pH 8 and 10 in dark condition was due to adsorption as a result of the interaction between the negatively charged surface of MnO₂ particles and positively charged Ag(I) ions in alkaline medium. Under visible light irradiation, photoreduction was more dominant than adsorption as the effect of interaction force becoming less conspicuous.

Fig. 4 shows the XRD profile of MnO₂ particles and Ag/MnO₂ hybrid particles collected after 1 h of visible light irradiation. In general, the diffraction peaks at 29.5°, 38.4°, 41.5°, 43.1°, 46.7°, 57.1°, 60.6°, 68.2°, and 72.6° corresponded to the (110), (101), (200), (111), (211), (220), (002), and (310) crystal planes of β-MnO₂ structures, respectively (JCPDS no. 24-735). At pH 2, pH 4, and pH 6, two additional diffraction peaks at 35.2° and 46.8° corresponded to the (111) and (200) cubic structures of silver (JCPDS no. 98-008-3900). The intensity of silver diffraction peaks was the highest for Ag/MnO₂ hybrid particles collected at pH 4, followed by pH 2 and pH 6. This result suggests that the highest deposition of Ag was at pH 4, followed by pH 2 and pH 6. This result is in accordance with the ICP-MS analysis (Fig. 2(b)) where the optimum pH for the removal of Ag(I) ions was at pH 4. Meanwhile, no additional peaks were observed at pH 8 and pH 10. This result was due to the poor removal (deposition) of Ag(I) ions via adsorption on β-MnO₂ particles and thus beyond the detection limit of XRD.

Figure 4.  XRD spectra of β-MnO₂ particles and Ag/β-MnO₂ hybrid particles after deposition for 1 h under visible light irradiation.

As shown in Fig. 5, the β-MnO₂ particles consist of flake-like structures in different diameters and lengths. After metal ion removal, deposition of tiny Ag particles (greyish white, as highlighted in yellow dotted circles) was observed on the surface of β-MnO₂ particles (Fig. 5). The amount of deposition of Ag on the β-MnO₂ particles differed under the different pH environments. Energy dispersive X-ray (EDXS) analysis was performed to measure the atomic percentage of Ag deposited at each selected pH level. As depicted in Fig. 6, the highest Ag deposition (3.24at%) was at pH 4. The result agrees with the XRD and ICP-MS analyses.

Figure 5.  SEM images of β-MnO₂ (a) and Ag/β-MnO₂ hybrid particles after removal of heavy metal ions under visible light irradiation at pH 2 (b), pH 4 (c), pH 6 (d), pH 8 (e), and pH 10 (f). The dotted circles in yellow indicate the deposition of Ag on β-MnO₂.

Figure 6.  Atomic percentage of Ag deposited on β-MnO₂ hybrid particles under various pH solutions as measured using EDXS.

TEM analysis was carried out to explore the structure of Ag/β-MnO₂ hybrid particles produced at pH 4 after visible light irradiation. As shown in Fig. 7(a), tiny particles appeared on the flake-like surface of β-MnO₂ particles. The HRTEM in Fig. 7(b) clearly reveals that the tiny particles are Ag particles with an inter-planar spacing of 2.3 nm and corresponding to the (111) phase of Ag. The inter-planar spacing of β-MnO₂ was 3.1 nm and corresponded to the (110) phase of β-MnO₂.

Figure 7.  TEM (a) and HRTEM (b) of Ag/β-MnO₂ hybrid structure produced at pH 4.

The wide-scan survey results of XPS on pure β-MnO₂ particles and used β-MnO₂ particles (under pH 4) are shown in Figs. 8(a) and 8(b). Comparison of the wide scan survey revealed an additional Ag peak for the used β-MnO₂ particles, indicating the deposition of Ag on the surface of β-MnO₂ particles. Fig. 8(c) shows the high-resolution scan of Ag on the used β-MnO₂ particles. The binding energy at 368.1 and 374.1 eV fitted well with the Ag⁰ oxidation state of Ag 3d_5/2 and Ag 3d_3/2, respectively. The result indicates clearly the photoreduction of Ag(I) ions into Ag. The deposition of Ag on the surface of β-MnO₂ particles was in particle form as shown in Fig. 6. In summary, Ag particles were deposited on the surface of β-MnO₂ particles via the photoreduction of Ag(I) ions as supported by the evidence from X-ray diffraction (XRD), Field emission electron microscope (FESEM), Energy dispersive X-ray (EDXS), transmission electron microscopy (TEM), and X-ray photoelectron microscopy (XPS) analyses.

Figure 8.  Wide survey scan of β-MnO₂ particles (a), Ag/β-MnO₂ particles hybrid particles (b) and high-resolution Ag scan of Ag/β-MnO₂ hybrid particles (c).

Fig. 9 shows the reusability study of β-MnO₂ particles in Ag(I) ion removal performed under pH 4. The removal efficiency of Ag(I) ions by β-MnO₂ particles deteriorated significantly from the first (99%) to third run (38%). This phenomenon was mainly due to the saturation of the surface area of β-MnO₂ particles available for Ag deposition. The result suggests that β-MnO₂ particles are an effective photocatalyst to remove Ag(I) ions from water via reduction. However, regeneration of β-MnO₂ particles is needed to pro-long their lifespan in practical applications. The reusability of the β-MnO₂ particles could be improved after the recovery of Ag from their surface through biphasis catalysis [39–40] or ion exchange [41–42].

Figure 9.  Reusability study of β-MnO₂ particles in Ag(I) ions removal under 90 min of visible light irradiation at pH 4.

On the basis of the above findings, the Ag(I) ion removal mechanisms by β-MnO₂ particles under dark condition and under visible light irradiation are proposed as follows:

Under dark condition, no optical excitation source was available for β-MnO₂ particles to produce electrons/holes for the subsequent photoreduction/photooxidation of Ag(I) ions. Therefore, the removal of Ag(I) ions by β-MnO₂ particles was through adsorption. In strong acid medium, the positively charged surface of β-MnO₂ particles repelled the positively charged Ag(I) ions. Therefore, less Ag(I) ions were adsorbed on the surface of β-MnO₂ particles, resulting in the poor removal efficiency of Ag(I) ions. By contrast, the surface charge of β-MnO₂ particles was negatively charged under alkaline solution. This phenomenon was favorable for the attraction between the negatively charged β-MnO₂ particles and positively charged Ag(I) ions. Thus, high removal of Ag(I) ions by β-MnO₂ particles was observed at pH 8 (22%) and pH 10 (21%) in dark condition.

The reduction potential of Ag⁺/Ag (0.79 V) must be more positive than the e^–_CB level of β-MnO₂ for the photoreduction of metal ions to occur [28]. The band gap energy of β-MnO₂ is around 1.0 V with e^–_CB of 0.2 V. The band edge of e^–_CB of β-MnO₂ in NHE diagram is feasible for photoreduction to occur [43]. However, the relatively low band gap energy of β-MnO₂ promoted the electron–hole pair recombination, which is a considerable drawback for photocatalytic reaction. The presence of H⁺ in the acidic environment affectively mitigated this problem. On the basis of the aforesaid observation, the removal mechanism of Ag(I) ions at pH 4 is proposed in Fig. 10. The process started with the adsorption of Ag(I) ions on the surface of MnO₂ particles followed by the photoreduction of Ag(I) ions.

In the acidic solution of pH 4, the slightly negative charged surface of β-MnO₂ particles favored the adsorption of positively charged Ag(I) ions (Eq. (6)). The excess H⁺ ions reacted with β-MnO₂ particles, forming Mn²⁺ ions (surface species) ((Eq. (7)). In consideration that Mn²⁺ ions (surface species) were in highly unstable state, Mn²⁺ ions (surface species) reacted with the photo-generated holes on the surface of β-MnO₂ particles, forming Mn³⁺ ions (surface species) upon the irradiation of visible light (Eq. (8)) [44]. The trapping of holes by Mn²⁺ ions (surface species) suppressed the recombination between electrons and holes. The photo-excited electrons could then freely react with the Ag⁺ ions in aqueous solution to form Ag particles, as depicted in Eq. (9) (photocatalytic reduction). Hence, the slightly acidic condition of pH 4 possessed the best Ag(I) ion removal ability.

In pH 2, the acidic condition promoted the trapping of electrons by β-MnO₂ particles, which was greatly advantageous for the photocatalytic reduction of Ag⁺. However, the acid solution also significantly changed the surface of β-MnO₂ particles to positively charge. Thus, the positively charged β-MnO₂ particles repelled positively charged Ag(I) ions, which made the absorption of Ag⁺ on the catalyst surface difficult and deteriorated the removal of Ag(I) ions.

In alkaline solution (pH 8 and pH 10), the absorption of Ag(I) ions was favored because it showed the highest absorption ability at the first 30 min of reaction. However, the reaction of β-MnO₂ particles with H⁺ was inhibited due to the excess OH^–. Thus, the recombination of electron–hole pairs possibly occurred and affected the photo-reduction ability of β-MnO₂. Therefore, the maximal removal of Ag(I) ions at pH 4 contributed to the synergetic effect between the electrostatic attraction and the suppression of electron–hole recombination.

This study demonstrated that β-MnO₂ particles can remove Ag(I) ions from aqueous solution selectively. The β-MnO₂ particles showed optimum removal efficiency at slight acidic solution (pH 4), and it was twofold better than that at its original pH of 6 under visible light irradiation. Removal of Ag(I) ions by β-MnO₂ particles from the solution via deposition was verified by XRD, SEM, EDXS, and HRTEM analyses. Tiny Ag particles appeared on the surface of β-MnO₂ particles. In addition, the zeta potential measurement revealed that the surface charge became more positive in the acidic environment (pH 2 and pH 4) and more negative in the alkaline environment (pH 8 and pH 10). The removal mechanisms of Ag(I) ions are proposed as follows: (i) adsorption under dark condition and (ii) photocatalytic reduction of Ag(I) ions to Ag particles under visible light irradiation. This study demonstrates the potential of β-MnO₂ photocatalyst for the selective removal of Ag(I) ions from wastewater.

This study was funded by Universiti Sains Malaysia, Research University Grant (1001.PBAHAN.8014095). The authors also acknowledge the support from USM Fellowship.