Activation of Peroxydisulfate by Bimetallic Nano Zero-Valent Iron for Waste-Activated Sludge Disintegration

Abstract

Waste-activated sludge (WAS) disintegration using peroxydisulfate (PDS) has attracted scientific attention over the past few years. Despite several advantages offered by a sulfate radical-advanced oxidation process, there are still too many downsides of this treatment that limit its facile large-scale application. This study investigated whether modifying nano zero-valent iron (nZVI) with a second metal such as Ag and Cu enhanced the disruption of WAS. The disintegration efficiency was assessed using standard techniques, i.e., soluble chemical oxygen demand, Fourier-transform infrared spectroscopy and a scanning electron microscope. The bimetallics were shown to have an improved disintegration efficiency of > 2.5-fold compared with the untreated sample. Furthermore, nZVI/Ag was found to be more efficient than nZVI/Cu for PDS activation, which was validated by the higher ratio (3 and 2.5 for nZVI/Ag and nZVI/Cu, respectively) between the soluble extracellular polymeric substances and the bound extracellular polymeric substances (S-EPS/B-EPS). Similar conclusions were derived from a SEM analysis. The improved disintegration efficiency could be related to the enhanced electron transfer from nZVI to PDS or the intrinsic properties of silver, which was found to be one of the best activators for PDS under homogeneous conditions. We believe that this study deepens the understanding of PDS heterogeneous activation processes.

1. Introduction

Over the course of several centuries, we have observed an increase in the level of pollution caused by human activity. The population continues to grow, with increased pollution and wastewater production [1]. This problem concerns, among others, high water contamination worldwide. As a result, increasing demand and interest in water purification has led to the development of novel water purification techniques that degrade contaminants more effectively, in less time and consuming less energy.

A wastewater treatment system should operate at maximum efficiency to disinfect and purify the water. Thus, oxidizers such as ozone [2], hydrogen peroxide [3], peracetic acid [4] or potassium permanganate have been extensively studied for the removal of contaminants [5]. Wastewater treatment plants (WWTPs) are usually based on an activated sludge (AS) process to meet the current restrictive requirements [6]. AS is an essential wastewater treatment for converting and neutralizing dangerous contaminants. AS possesses microorganisms used to remove organic matter from wastewater, i.e., in nitrification, denitrification and phosphorus removal processes as well as the oxidation/reduction process for certain compounds [7]. One of the main problems of conducting this process is the increase in biomass. Therefore, the production of waste-activated sludge (WAS) increases as the amount of wastewater discharged intensifies [8]. Furthermore, the microorganisms involved in water purification processes are unable to degrade different types of pollutants [9]. Due to this, there is still a wide variety of harmful chemicals in WAS such as petroleum products, organic compounds, xenobiotics and others [10,11].

Currently, the surplus of WAS is mainly released into the environment via landfill, composting, combustion and methane production [12,13,14]. Therefore, the methods used today for WAS disposal are associated with additional costs. As WAS consists mainly of water, reducing its volume could facilitate the transport and disposal of the sludge. Nowadays, different methods can be used for WAS dehydration such as centrifugation [15], wet milling [16], ultrasound [17], advanced oxidation processes (e.g., the Fenton process [18]), filtering (or belt press), magnetic sedimentation, spinning and compression/filtration [19,20]. Nevertheless, existing disintegration methods remain far from ideal; the main problem is associated with retaining water in the bacteria cells [19,20].

One of the chemical disintegration methods of WAS is a chemical treatment by persulfates. The persulfates—i.e., peroxymonosulfate (PMS) or peroxydisulfate (PDS) anions are highly reactive by themselves because of their chemical structure, which includes a peroxide bond [21,22]. In addition to this reactivity, in combination with several types of activators (such as transition metals and UV radiation), they are able to generate reactive species such as sulfate radicals (SO₄^•−) and hydroxyl radicals (^•OH). The activation is based on the donation of an electron or energy to the peroxide bond with a further SO₄^•− formation [23]. An example of PDS activation by Fe²⁺ can be found in Equation (1).

$${\mathrm{Fe}}^{2+}+{\mathrm{S}}_2{\mathrm{O}}_8^{2-}\to {\mathrm{Fe}}^{3+}+{\mathrm{SO}}_4^{•-}+{\mathrm{SO}}_4^{2-}$$

(1)

Our research group was one of the first to report WAS disintegration by persulfates [24]. Additionally, it has been found that this treatment can contribute to removing harmful and persistent contaminants in sludge [23,25]. Our group also recently reported that nZVI can be used as a PDS activator in a relatively broad pH spectrum [26] and that modifications can enhance the electron transfer process [27], which (we hypothesized) could improve the formation of sulfate radicals.

To the best of our knowledge, the impact of bimetallic nanoparticles based on nZVI on PDS activation with a subsequent use for WAS disintegration has not been investigated before. Al-Shamsi et al. [28] reported the possibility of enhancing PMS and PDS activation using bimetallic nanoparticles; however, their study did not focus on WAS.

The present work investigates, for the first time, the potential use of nZVI bimetallic nanoparticles for the catalytic activation of PDS in a complex matrix such as sludge. The effect of a second metal (Ag and Cu) in WAS disintegration was also investigated. The efficiency of the process was evaluated by commonly used techniques: soluble chemical oxygen demand (SCOD); extracellular polymeric substance (EPS) determination; Fourier-transform infrared (FTIR) spectroscopy; and scanning electron microscope (SEM) observations. From this, it was found that decorating nZVI with a second metal could enhance the process of PDS catalytic activation. The authors believe this work could help scientists to optimize the heterogeneous activation process of persulfates.

2. Results and Discussion

The most common analytical method used to estimate WAS disintegration efficiency is soluble chemical oxygen demand (SCOD) [29]. The concentration of organic matter is determined by assessing the oxygen consumption for its oxidation [30]. During WAS disintegration by chemical pretreatments, flocs can be destroyed and microorganisms disrupted. As a result, intracellular organic material can be released from the solid particles into the liquid phase, raising the SCOD value. As Figure 1A shows, after 30 min of reaction, the SCOD value was significantly affected by the process. The SCOD value increased from 95 mg/L (R1, untreated sample) to 189, 206, 225, 242 and 241 mg/L for R2, R3, R4, R5 and R6, respectively. The addition of nZVI or bimetallic nanoparticles to PDS increased the SCOD value in all cases. From the reported data, PDS activated by nZVI or bimetallic nanoparticles could improve the breaking down of the cell walls of the bacteria and WAS flocs by the enhanced generation of radicals.

The protein concentration changes in the two fractions of EPS are shown in Figure 1B. The samples R1, R2 and R3 had a small fraction of soluble EPS (S-EPS), but a large fraction of the bound ones. The marked increase in the protein concentration of S-EPS in reactors R4, R5 and R6 suggested the partial or complete lysis of the bacterial cells and subsequent release into the liquid phase of the organic matter from inside the cells [31]. Thus, the decrease in the bound EPS (B-EPS) to 17.6 mg/L, 16.3 mg/L (the highest observed in this study) and 26 mg/L for R4, R5 and R6, respectively, confirmed the EPS disintegration. This phenomenon was not observed for R2 and R3. Furthermore, it was noted that in all the PDS-activated systems, the ratio between S-EPS and B-EPS significantly increased. This was an additional confirmation that the PDS was easily activated by the nZVI or bimetallic nanoparticles to form sulfate radicals. Analogous results were obtained by Guo et al. [32], where Fe(II)-activated PDS caused the conversion of B-EPS to S-EPS.

Decorating the nZVI surface with a second metal—in this case, silver—caused a decrease in the B-EPS and S-EPS amount (compared with the PDS/nZVI system). Nonetheless, the ratio between S-EPS/B-EPS was the highest of all the samples in reactor R5 and was equal to 3 (for R4, it was 2.8; for R6, it was 2.5). In the presence of nZVI/Ag, PDS generated the radicals with greater efficiency; this can help to break down cells (releasing organic matter from them), but eventually it also oxidizes organic substances into CO₂ + H₂O.

Based on the SCOD/EPS results, we decided to analyze samples R1, R4, R5 and R6 by FTIR spectroscopy (Figure 2). This showed the presence of five functional groups (primarily) representing hydrocarbons, proteins, polysaccharides, aromatic amino acids and nucleic acids [33]. The band region from 3650 to 3200 cm⁻¹ showed the O–H groups whereas the peaks at ~2925 and ~2856 cm⁻¹ could be attributed to the C–H, C–H₂ and C–H₃ groups of organic compounds. The peaks at ~1634, ~1528 and ~1402 cm⁻¹ were attributed to amide I (C–N and C=O stretching), amide II (C–N and N–H stretching) and amide II (C=O symmetric stretching) of the protein fraction, respectively. The peak at ~1026 cm⁻¹ was due to the presence of polysaccharides (C–O–C and C–OH). The peaks at ~906 cm⁻¹ were related to nucleic acids (O–P–O stretching). Of note was the presence of two additional peaks at wavenumbers ~777 and ~696 cm⁻¹. These peaks were ascribed to the presence of aromatic amino acids (C–C and C–OH ring vibrations), as reported by Ruan et al. [34]. As shown in Figure 2, the position of most of the peaks was unchanged or was slightly shifted, which was in agreement with the literature [35,36]. On the other hand, the intensities weakened for most of the peaks, implying that the SO₄^•− and ^•OH radicals generated by the nZVI/PDS and nZVI-bimetallic/PDS processes could oxidize the organic substances to CO₂ + H₂O. The exception was the peaks ascribed to amino acids (777 and 696 cm⁻¹) in which the intensity increased, indicating a release of substances from the disintegrated cells. This agreed with the EPS analysis, suggesting the release of soluble proteins. The largest increase, at ~777 cm⁻¹, was again reported for R5 (Figure 2, insert). The FTIR spectroscopy analysis confirmed the results of the SCOD/protein determination, which indicated an increase in proteins in this reactor; thus, this was the most efficient disintegration.

Figure 3A shows the images of the samples. As the figure shows, it was possible to observe a decrease in the turbidity of the supernatant in R5 and R6 along with an increase in the sedimentation rate. In R5, the supernatant appeared to be clearer compared with the other reactors.

A SEM analysis was conducted on the R1 and R5 samples (Figure 3B,C). R1 mostly showed an unaltered structure of the flakes and bacteria (Figure 3B). In this reactor, the shapes were regular, relatively round and smooth, but those treated with metal-activated PDS (Figure 3C) were much less regular and formed a structure with numerous pores and smaller particle sizes. This might have been connected to the improved sedimentation properties of WAS after an efficient disintegration, which our group has repeatedly reported before [37,38]. Changes in the WAS particle size suggested an increase in its liquid value, which aided in improved WAS sedimentation and, at the same time, better dehydration. Wacławek et al. [24] and Sponza [39] arrived at the same conclusion, finding that large flocs (particles) had high sludge volume index values, suggestive of poor settling characteristics.

Based on all the gathered results, the bimetallic nanoparticles were more prone to activating PDS than their monometallic counterparts. The second metal could enhance the electron transfer process needed for breaking the O–O bonds in persulfates [28]. This could be translated into a more efficient generation of sulfate radicals, which could interact faster (compared with other radical species [23]) with the sludge flocs and thus improve the disintegration efficiency. Furthermore, the soluble substances of the sludge were released into the liquid phase, which could be oxidized and mineralized further.

3. Materials and Methods

3.1. Solution and Reagents

The WAS was taken from a Liberec municipality wastewater treatment plant (WWTP) in the Czech Republic. The treatment plant was designed for nutrient removal. The concentration of suspended solids (SS) was, on average, 9.3 g/L and the amount of treated wastewater was about 54,000 m³/d. The solid retention time (SRT) was approximately 25 days and the hydraulic retention time (HRT) was approximately 15.5 h.

All chemicals used in the experiments were of an analytical reagent grade. Iron(III) chloride hexahydrate (≥98%), copper(II) chloride (97%), silver nitrate (≥99%) and sodium borohydride (≥98%) were purchased from Sigma-Aldrich, Czech Republic. Sodium peroxydisulfate–sodium persulfate (99%) was purchased from Lach-Ner, Czech Republic.

In the experiments, deionized water (18.2 MΩ·cm⁻¹, ELGA, Veolia Water, Marlow, UK) was used to prepare all of the reagents.

3.2. Nanoparticle Synthesis and Assessment

Both nZVI and bimetallic nZVI were synthesized following our previous work [27]. Briefly, 0.2 M of Iron(III) was reduced to Fe⁰ by sodium borohydride (0.5 M) dosed to the reactor at a rate of 1.5 mL/min under a nitrogen atmosphere. After adding the last drop of sodium borohydride, the reaction was carried out for 10 min to ensure the complete reduction of the iron precursor. The manufactured nanoparticles were rinsed three times with ethanol before freeze-drying (for future use). Regarding the bimetallic synthesis, nZVI was activated with 1 M hydrochloric acid for 20 s then 1% (wt% of Me/wt% of nZVI) Ag or Cu metal precursors were added sequentially and mixed for 10 min. The bimetallic nanoparticles were rinsed three times with water.

A full characterization of the monometallic and bimetallic nanoparticles used in this research is available in our previous work [27]. The content of Ag and Cu (in nZVI) was found to be in the range of ~0.5–0.2%.

A leaching test was performed to assess the stability and potential release of the metals during the process. The tests were performed for 24 h at room temperature (25 ± 1 °C) and the concentration of monometallic and bimetallic nanoparticles was 1 g/L. Under these conditions, nZVI released 26.6 mg/L of total iron whereas nZVI/Ag and nZVI/Cu released 25.6 and 15.3 mg/L of total iron, respectively. Regarding the leaching of the second metal, nZVI/Ag released 0.06 mg/L of silver whereas in the case of nZVI/Cu, the Cu concentration in the solution was lower than the detection limit; i.e., < 0.05 mg/L.

3.3. Method of Sludge Disintegration

The WAS disintegration was performed by a metallic/bimetallic chemical disintegration. For the chemical disintegration, PDS was used at a concentration of 8.4 mM in accordance with our previous report [40]; in nZVI and bimetallic nZVI (nZVI/Cu, nZVI/Ag), the concentration was 0.2 g/L [41]. The experiments were performed in six glass 250 mL reactors and continuously mixed: R1: control (no PDS, no nZVI); R2: nZVI (0.2 g/L); R3: PDS (8.4 mM); R4: PDS + nZVI (PDS: 8.4 mM, nZVI: 0.2 g/L); R5: PDS + nZVI/Ag (PDS: 8.4 mM, nZVI/Ag: 0.2 g/L); and R6: PDS + nZVI/Cu (PDS: 8.4 mM, nZVI/Cu: 0.2 g/L). The duration of the experiment was 30 min. All tests were performed in duplicate at room temperature. The average relative standard error for the SCOD and EPS analyses was <5% and <15%, respectively.

3.4. Characterization and Analysis

The SCOD was determined following the standard methods for examining water and wastewater, procedure 5220D [38].

The EPS extraction was performed according to the method developed by Wang et al. [42]. The experimental samples were centrifuged (8000 rpm for 10 min) and the supernatant was collected as S-EPS. A NaCl solution (0.85 wt%, warmed at 70 °C) was used to suspend the bottom sediment. After that, the suspension was stirred for 1 min using a vortex mixer and then centrifuged (8000 rpm for 10 min). The resulting supernatant was kept as B-EPS. A PTFE membrane syringe filter (0.22 μm) was used to filter the S-EPS and B-EPS for further analyses. The determination of protein (as one of the main EPS components) was carried out using the Lowry method [43,44].

The FTIR spectra (resolution of 4 cm⁻¹ at a range of 4000–700 cm⁻¹) were acquired with a germanium ATR crystal (NICOLET IZ10, Thermo Scientific, Waltham, MA, USA) equipped with a single reflection angle 45° horizontal ATR accessory.

A Tescan Vega XMU SEM (TESCAN s.r.o., Brno, Czech Republic) was used to observe the disintegration of the WAS particles.

The dissolved iron, silver and copper concentrations were analyzed by ICP-MS (Elan 6000, Perkin Elmer, Akron, OH, USA).

4. Conclusions

We investigated for the first time the influence of a second metal (Ag and Cu) on an nZVI surface for the catalytic activation of PDS in a complex matrix such as WAS. The bimetallic nanoparticles exhibited superior behavior for the disintegration of WAS by increasing the SCOD from 95 mg/L (untreated WAS) to 242 mg/L (WAS treated by PDS + nZVI/Ag). The quality of the treatment was also confirmed by FTIR spectroscopy, EPS determination and a SEM, which showed an improved decomposition of the WAS.

Such an enhanced catalytic efficiency of bimetallic nanoparticles, especially for nZVI/Ag, might be due to an improved electron transfer from the nZVI core to the second metal by forming a galvanic couple with nZVI, as reported previously by He et al. [45]; silver reacted with the PDS in a different way by which sulfate radicals were produced. The better efficiency of Ag (compared with other transition metals) as an activator for persulfates in homogeneous systems has already been reported by Edwards et al. [46] and Anipsitakis and Dionysiou [47]. We concluded that the intrinsic properties of Ag could also be beneficial in heterogeneous systems (such as presented in this communication), causing an increased formation of free radicals.

To sum up, the experimental results presented an improvement in the disintegration of WAS by bimetallic nZVI (with a slight increase by Ag over Cu). We believe that this can help provide a better comprehensiveness of the heterogeneous activation of PDS.