Sulfate Radical Technologies as Tertiary Treatment for the Removal of Emerging Contaminants from Wastewater

Water scarcity and water pollution is a worldwide problem and has driven research into eco-friendly and low-energy cost efficient remediation. The reuse of wastewater for non-potable purpose after proper treatment is the only sustainable solution to the problem. Advanced oxidation processes (AOP) based on the in-situ generation of hydroxyl radicals have been intensively investigated for this purpose as a treatment step to achieve wastewater reuse. The main degradation mechanism of AOPs is based on the reaction of hydroxyl radicals with dissolved organic matter. However, hydroxyl radicals follow unselective multi-step pathways, limiting their efficiency in complex environmental matrices. To overcome such limitations, AOP treatment, based on generation of sulfate radicals, has been developed and widely investigated. This current mini-review will cover the most recent developments regarding emerging contaminant removal, i.e., organic micropollutants, using sulfate radicals generated by active persulfate or peroxymonosulfate, with a focus on an application to wastewater effluents for possible wastewater reuse.


Introduction
Water pollution by recalcitrant organic compounds is an increasingly important problem due to the continuous introduction of new chemicals into the environment. The recalcitrant organic compounds are well known as emerging contaminants-such as pharmaceuticals, pesticides, endocrine disruptors, and personal care products-and recent studies have showed that they are widely detected in water bodies [1][2][3]. Many emerging contaminants are toxic, biorefractory and difficult to be removed from wastewaters by the conventional treatment methods [4]. For this reason, it is necessary to improve the conventional wastewater treatment to enable the efficient removal of recalcitrant organics. Therefore, advanced oxidation processes are considered as an alternative or an additional step to conventional wastewater treatment plants, in order to enable the degradation of resistant organic compounds. The concept of "advanced oxidation processes" was defined by Glaze and co-workers in 1987 as the processes involving the in-situ generation of highly reactive species, able to attack and degrade organic substances [5].
Hydroxyl radicals are powerful, non-selective chemical oxidants that act rapidly with most organic compounds. Advanced oxidation processes give the possibility to produce hydroxyl radicals in situ by different possible processes. In water treatment, the most used are (i) Fenton oxidation; (ii) ozonation; (iii) photocatalysis and (iv) other methods. Researchers' interest in AOPs began only around 1995 and continues today; in fact, there is an increase of the number of scientific papers in this topic, as is illustrated in the graph (Figure 1   Briefly, the main processes used in wastewater treatment are the following: (i) The Fenton and/or photo Fenton processes: Fenton H.J.H. in the 1890s discovered the reaction of polycarboxylic acid with hydrogen peroxide (H 2 O 2 ), and observed that a strong promotion took place in the presence of ferrous ions (Fe 2+ ). The Fenton process, or dark Fenton, invo0lves the use of an oxidant, generally H 2 O 2 , and a catalyst, usually iron in the form of ferrous ions. The generation of hydroxyl radicals are formed by the oxidation of Fe 2+ . The main applicative limits are that this process occurs in the acid medium, and that iron removal is required.
The photo Fenton process involves irradiation with sunlight or an artificial light source. The presence of light improves the efficiency of the treatment thanks to the photoreduction of the Fe 3+ to Fe 2+ , which results in additional generation of hydroxyl radicals. This process requires additional cost for the UV irradiation.
(ii) Ozonation: Ozone is a powerful oxidizing agent, which is able to participate in a great number of reactions with organic and inorganic compounds. Since the beginning of the century, it has been used as disinfectant, but only in the last decades has it acquired notoriety in wastewater treatment. In the absence of light, ozone can react directly with organic compounds through a slow and selective reaction, or through a fast and non-selective reaction through the formation of hydroxyl radicals. The presence of UV light can generate additional hydroxyl radicals due to the photolysis of O 3 , increasing the efficiency.
The main disadvantages are the low solubility of O 3 in water, the formation of hazardous by-products, in particular, bromate formation, and the elevated energy costs [6].
Unfortunately, there are many literature reports concerning industrial wastewater treatment by AOPs, especially about textile wastewater, but most of them are not oriented on the industrial application. In this scenario, [7] demonstrated that some of those oxidative processes can be suitable for industrial application. The research group reported that, based on the experimental result and cost evaluation, the use of ozone or ozone with a very low concentration of hydrogen peroxide (0.005 M) can be recommended to the textile industry. In contrast to the O 3 /UV treatment, the investment and operating cost are much higher than ozonation itself. Therefore, the main application of ozonation is for disinfection in wastewater treatment plants for drinking water.
(iii) Photocatalysis: This AOP is based on the use of a semiconductor as catalyst and a light source. Several semiconductors have been used in this field, but titanium dioxide (TiO 2 ) proved to be the most appropriate due its characteristics such as high photoactivity, chemical inertness, non-toxicity, and low cost [8]. The initiating procedure of the photocatalytic reaction is the adsorption of the radiation with the formation of holes (h + ) in the valence band and electrons (e − ) in the conduction band.
(iv) There are other methods among AOPs; however these are non-photochemical, such as electro-Fenton and wet air oxidation. There are two main types of electro-Fenton. In cathodic processes, iron is added as Fe 2+ or Fe 3+ salt. The source of H 2 O 2 may be either via direct H 2 O 2 addition or produced by reduction of oxygen at the cathode. In the anodic process, the source of iron is a sacrificial iron anode. This treatment requires a high iron concentration and elevated costs.
The wet air oxidation is defined as the oxidation of substances in aqueous solution by means of oxygen or air at elevated temperature and pressure; for this reason, it has a high operational cost.
Hydroxyl radicals proceed through unselective multi-step pathways which limit its efficiency in complex environmental matrices; for example, waters which contain high concentrations of the main hydroxyl radical scavengers in wastewater, i.e., dissolved organic matter (DOM) and carbonate/bicarbonate anions [9]. This increases the overall scavenging rate of waters to be treated [6]. To overcome this limitation, novel more selective methods, producing reactive radicals, were developed. Use of sulfate radical (SO 4 − ) technologies is a recent example. In fact, Ahmed et al. [10] reported that sulfate radicals attack through the one electron oxidation mechanism, limiting the scavenging effect of DOM and inorganic ions [10,11]. Few works have reported the comparison between sulphate radical (SR) and hydroxyl radical (HR) technologies. Table 1 compares the advantages and disadvantages of various AOPs, while Table 2 compares the kinetic rates of HR and SR. [12] investigated the kinetic rate of six organic compounds under SR and HR oxidation. The SR technology was always 10 times faster than HR systems; except for carbamazepine, for which the kinetic constant was less than seven-fold higher than with HO. Other works reported that HR systems were more efficient than sulphate radicals on ibuprofen or levofloxacin abetment [13][14][15].
Sulfate radicals (SO •− 4 ) serve as powerful oxidants with a standard redox potential of 2.6 V, comparable to the standard potential of HO • (2.8 V) [16] that can be generated in situ by the activation of the most common precursors, namely potassium persulfate (PS) or peroxymonosulfate salt (PMS), by a variety of approaches such as photolysis, radiation and thermal activation (Equations (1)-(8)). Between these methods, the addition of transition metals is recognized to be a viable way to enact the homogenous activation of (persulfate/peroxymonosulfate) PS/PMS [17].
Therefore, this technology is considered as a promising method for tertiary wastewater treatment because it is able to produce treated effluent that can be reused for agriculture or industrial activities. Table 1 summarizes the advantages and disadvantages of the AOP methods used in the wastewater treatment. Based on this information, the aim of this review study is to report the fundamentals and the recent advances of the application of sulphate radicals in wastewater treatment technologies, with the purpose of providing insight towards the application of such technologies for efficient water reuse for applications in water-scarce regions and vulnerable communities.

Sulfate Radicals for the Treatment of Wastewater
Different advanced oxidation technologies have been studied and applied for the treatment of domestic and various industrial (e.g., winery, textile, paper, etc.) wastewaters [22,23].The efficiency of the treatment depends on many parameters, such as the type of wastewater, chemical proprieties of contaminants, pH of water, concentrations of scavengers, operating conditions. In this article, we summarized the advanced oxidation processes based on sulfate radicals for the treatment of a domestic wastewater effluent. Wastewater effluent is considered the most promising source for production of recycled water [24]. Reused wastewater should be free of toxic or xenobiotic substances such as pesticides, pharmaceuticals, and endocrine disrupting compounds [25].

Ferrous Activation Systems
One of the most popular advanced oxidation processes are Fenton-like processes. The Fenton-like reaction can be performed through the activation of PS or PMS, as an oxidant, and the pair of Fe 2+ /Fe 3+ , as a catalyst, according to Equations (2), (3), (7) and (8). The generation of sulfate radicals through Fe 2+ has a great potential for the degradation of various recalcitrant compounds such as aniline, pesticides, chlorophenols and azo dyes [26,27] or for the treatment of stabilized leachates [28]. Rastogi and co-workers [29] have shown that sulfate radicals generated from activation of peroxymonosulfate by Fe(II) were able to destroy polychlorinated biphenyls (PCBs) in aqueous and sediment systems. This system had been used in situ for chemical oxidation in wastewater treatment as reported in different studies [30,31]. This treatment has some limitations such as the fast conversion of Fe(II) in Fe(III) and acidic pH of the solution [32]. To overcome this applicative limit, various studies were developed using different sources of iron. Zero valent iron (ZVI) was deeply researched as an alternative source of Fe(II) which is able to activate persulfate. In the system using ZVI/PS, Fe(II) is produced by corrosion under oxic/anoxic condition and/or oxidation by PS, resulting in the release of Fe 2+ , which can activate persulfate to generate sulphate radicals (Equations (9)-(12)) [33]: SR generation by the reduction of zero valent iron (ZVI) has been recently investigated and considered as good strategy for the removal of recalcitrant compounds such as polyvinyl alcohol (PAV) [34], pesticides such as alaclhor [35], herbicides such as bentazon [36], volatile gas as methyl mercaptan [37] and 1-4 dioxane [38]. The use of this source of iron has the advantages of low cost and easy operation; but it has a tendency to agglomerate due to the high surface energy and intrinsic magnetic interaction [39]. To improve the efficiency and decrease the aggregation of ZVI particles, recent research has investigated the encapsulation on a catalyst. Different kinds of materials have been investigated such as mesopourous silica microsphere, kaolinite, carbon [40][41][42], all of which have shown a great performance in transforming persistent pollutants in less toxic products.
These results were also confirmed by studies that have showed that modified zero valent iron showed a greater performance in the removal of various pollutants than non-modified ZVI [39,43]. In more recent studies, Diao and co-workers [44] reported the efficiency of bentonite-supported nanoscale zero-valent iron as a catalyst able to activate persulfate, and produce SR, which was confirmed from a positive correlation between persulfate decomposition and dissolved Fe(II).
This heterogeneous treatment was able to remove, simultaneously, heavy metals such as chromate (Cr(VI) and organic pollutants such as phenol from contaminated aqueous solutions.
Additionally, several studies have been reported in literature to use different organic chelating agents (e.g., citric acid, ethylenediaminetetraacetic acid (EDTA)) which were able to activate the persulfate system and to increase the contaminant degradation effectiveness. Liang and co-workers reported that the addition of citric acid in Fe(II) [45] in persulfate system improved the trichloroethylene (TCE) degradation by stabilizing Fe(III) and preventing its precipitation as iron hydroxide/iron oxide. Other studies investigated the development of SR using EDTA to remove TCE from aqueous solution [46]. This study showed that that the presence of Fe(III) and EDTA alone and in combination did not cause any significant increase in persulfate activation. The presence of TCE in the EDTA/Fe(III) activated persulfate system significantly enhanced persulfate activation, and the application of an EDTA/Fe(III) mixture was successfully applied for the degradation of TCE in solution at various pH conditions. Furthermore, Dulova and co-workers [47] confirmed that the addition of citric acid as a chelating iron agent improved the naproxen destruction by the oxidation process. The highest performance was obtained when the oxidant agent used was persulfate instead of hydrogen peroxide.
In Fenton or Fenton-like processes, the slow regeneration of Fe(II) and production of ferric hydroxide sludge restrict its usage, because of passivation and loss of activity [48]. Therefore, an improvement of this process implies the application of irradiation, which greatly improves the photo-reduction of Fe(III)-complex in Fe(II) (Equation (13)).
The efficiency of the Fenton process can be improved through the addition of radiation (sunlight or artificial sources of UV radiation). The photochemical reduction of Fe(III) is able to generate additional hydroxyl radicals, so the efficiency of the processes increases [49,50]. The artificial generation of photons required for the detoxification of polluted water is the most important source of costs during the operating of photocatalytic wastewater treatment plants. In fact, the photo-Fenton treatment developed under solar irradiation as a UV-Vis photon source is a potentially low-cost technique [51,52]. Recent findings demonstrated the ability of photo-Fenton treatment to remove micro-pollutants at relatively low concentrations (ng/L-µg/L) [53,54]. Recently, Malato's group [55] studied the solar photo-Fenton optimization for the treatment of municipal wastewater effluents. The important outcome of this study was that iron concentration, temperature and their interaction are the most important parameters necessary to remove over 95% of the organic contaminant targets. Another important observation was that a high concentration of hydrogen peroxide (up to 70 mg/L) decreased the processes efficiency [55].
Solar photo-Fenton is not the only treatment investigated in this topic; solar photocatalysis using titanium dioxide (TiO 2 ) have also been investigated. It has also reported [56] that solar TiO 2 technologies were able to eliminate toxicity and completely mineralize recalcitrant compounds present in raw oil sand process-affected water (OSPW). Malato et co-workers [57] developed an interesting review on the use of sunlight to produce the hydroxyl radicals by TiO 2 and encourage further relevant research with possible applications in industrial processes.
Titanium dioxide is not only the catalyst used in solar photocatalysis technologies. Dhatshanamurthi and co-workers [58] developed an interesting study and dealt with an efficient pilot scale solar for the treatment of dye industry effluent based on the generation in situ of hydroxyl radicals. The solar treatment was carried out using prepared nano-ZnO, commercial ZnO and titanium dioxide P25 (TiO 2 -P25). The catalyst made on thin film was stable and reusable for several cycles, so this make the solar treatment processes suitable for industrial effluent.
Despite the growing interest in AOPs, there are very few large-scale applications of solar-activated AOPs based on the generation of sulfate radicals. Among them, Brienza and co-workers [19] demonstrated that sulfate radical-based homogenous advanced oxidation technologies involving peroxymonosulfate as an oxidant and ferrous iron (Fe(II)) as a catalyst had better kinetic results when compared to solar photo-Fenton (UV-Vis/H 2 O 2 /Fe(II)) and heterogeneous photocatalysis using titanium dioxide (UV-Vis/TiO 2 ) for the removal of 17β-estradiol in wastewater. Little attention has been paid to advanced oxidation processes applied directly to wastewater effluents. Brienza and co-workers [1] reported that solar advanced oxidation processes showed that emerging contaminants at low concentrations can be successfully degraded. The aim of this work was to apply the solar AOP based on sulfate (SR) or hydroxyl radical (HR) on wastewater effluents. Both technologies (SR or HR) were able to transform part of the 53 micro-pollutants detected in the effluents. The results were very promising and further research is under way.
The full-scale treatment developed by solar-enhanced Fenton-like processes involving persulfate and ferrous ion was able to cause a complete degradation of atrazine and a 20% reduction in total organic carbon (TOC) content observed after 30 min of the treatment, as reported by Khandarkhaeva et co-workers [59]. The relevance of this study was the demonstration that the high efficiency of the solar processes was due to the additional formation of hydroxyl and sulfate radicals generating from the photo-reduction of Fe aq 3+ , FeOH 2+ and Fe 2 (OH) 2 4+ and decomposition of S 2 O 8 2− . Therefore, in this scenario, iron was acting not only as catalyst but also as photo-sensitizer [59]. Lin and Wu [60] investigated at a large scale the degradation of ciprofloxacin by UV/S 2 O 8 2− .
The relevance of this study was that the degradation of the target compound increased with persulfate concentrations until the excess of it started to inhibit the processes.

Non-Ferrous Activation Systems
Iron is not the only transition metal investigated in this field. In fact, the cobalt used for activation of peroxymonosulfate (Co 2+ /PMS) was studied [61]. Later studies showed that Co 2+ /PMS systems were found to show the best performance at neutral pH and with a lower dosage of PMS [62,63]. Moreover, Zhang et al. [63] reported the activation of persulfate by nano-Co 3 O 4 for orange G degradation, and it was reported that the nanoparticles were stable, with a slight decline in the catalytic ability due to the oxidation of surface Co(II) to Co(III). The nanoparticles were able to activate potassium persulfate (PS), peroxymonosulfate (PMS) and sodium persulfate (NaPs). The efficiency in terms of time need to remove orange G followed the order nano-Co 3 O 4 /PMS > nano-Co 3 O 4 /PS > nano-Co 3 O 4 /NaPS.
The main disadvantage observed was the Co leaching during the catalytic processes, which limits its use in environmental applications. In order to reduce this undesirable effects, Co 3 O 4 has been decorated on various supports including cobalt ferrite [64], silica support [65], graphene oxide [66]. The supported catalyst of cobalt, generally, showed better catalytic activity, reducing Co leaching, and better separation of the catalyst from the treated water. However, this did not eliminate the Co leaching problem.
Other transition metals used for the activation of persulfate or peroxymonosulfate were investigated. Saputra et al. [67] used various types of manganese oxides (MnO, MnO 2 , Mn 2 O 3 and Mn 3 O 4 ) for activation of PMS and studied phenol removal. They found that the catalytic activity of Mn oxides was dependent on the chemical states of Mn. Mn 2 O 3 was the most effective among the Mn oxide tested.
Liu et al. [68] reported a comparison study of propachlor degradation by persulfate activation using two metal ions such as Cu(II) and Fe(II), respectively. Both metals ions were able to activate persulfate and degradate the target molecules. The kinetic behaviour was different, in fact, when Fe(II) was used; the degradation was initially quite fast and afterwards slower, due to the rapid depletion of Fe(II) by the generated sulfate radicals. Copper ion activated persulfate was proven to achieve steady and efficient degradation of propachlor.
Recently, for the first time, Fang and co-workers [69] demonstrated that vanadium species are able to activate persulfate for the degradation of contaminates. The most efficient system for the degradation of target molecules (2,4,) was V 2 O 3 /PS; VO 2 and V 2 O 5 also can activate PS, but the degradation of PCB28 was significantly lower. Vanadium is a natural element, widely-distributed particularly in soil, so the natural occurrence is able to activate persulfate and can provide the possibility to apply this technique for the purification of wastewater.

Non-Metal Carbon Catalysts
Recently, several studies investigated the ability of non-metal carbon catalysts as a potential solution to the metal leaching problem, and to constitute metal-free environmental catalysts [70,71]. The most studied was activated carbon (AC) due to its relatively low cost and the wide applications in the water/wastewater treatment industry. Activated carbon shows catalytic activity due to its oxygen functional group on the surface, which favors electron-transfer. Kimura et al. [72] and Liang et al. [46] reported the ability of activated carbon to activate hydrogen peroxide and persulfate, respectively.
Different forms of activated carbon have been intensely investigated as material able to show synergistic effects both as adsorbent and catalyst. The particle size plays an important role on activated carbon activity, because its performance as a PMS activator depends of the density of the surface. Saputra et al. (2013), [67] demonstrated that powdered activated carbon (PAC) was much more effective than granular activated carbon (GAC) and proposed the following reaction mechanisms: Several studies showed the potential of activated carbon to activate oxidant agents able to destroy several recalcitrant organic compounds. Muhammed and co-workers [73] reported the comparison of catalyst efficiency between RuO 2 /AC and RuO 2 /ZSM5. Both catalysts were able to activate PMS and able to destroy phenol in aqueous solutions, but activated carbon had better performance. Yang et al. [74] proved that, in a carbon filter/PMS system, the catalytic oxidation made a major contribution to total acid orange 7 removal rather than adsorption. Similar results were obtained by Zhang et al. [75], where the decolorization of acid orange 7 solution was accomplished using granular activated carbon/peroxymonosulfate system. Unfortunately, the catalyst capacity of AC is not durable, as its adsorption capacity decreases with repeated use [75]. To overcome these limitations, post-treatment processes such as heating or chemical activation are required, increasing the cost of the treatment.

Graphenes as Activation Medium for PMS
Activated carbon is not the only carbon catalyst investigated in water/wastewater treatment. Recently, great attention has been given to the use of graphene and its derivatives. Wang's group investigated intensively on this topic, developing research on a wide variety of carbon allotropes in different molecular dimensions, such as reduced graphene oxide (rGO), single walled carbon nanotubes (SWCNTs), mesoporous carbon (CMK-8), nanodiamonds, graphitic carbon nitride (g-C 3 N 4 ), and fullerene (C 60 ) [70]. At first, they discovered that rGO developed by chemically synthesized graphene oxide by the one-pot thermal method was found to be effective in the metal-free activation of PMS to generate reactive radicals for the oxidation of methylene blue [76]. Successively, they showed that the adsorption capacity of nanocarbons was related to the structure; in fact, the 3D structure had a bigger porosity and larger specific area, thus giving higher phenol adsorption. Additionally, its catalytic activity for the activation of PS was tested on phenol degradation, and it was found that SWCNTs, CMK-8, and rGO-900 presented outstanding performances for PS activation, providing complete phenol degradation in 90, 45 and 30 min, accordingly [70].
The activation of persulfate or peroxymonosulfate on carbon based catalysts is not fully understood. It has been reported that the activation mechanism is not only by radicals but also a non-radical pathway bridging the formation of both OH • and SO •− 4 radicals using N-doped single walled carbon nanotubes for PMS activation [70]. The occurrence of radical and non-radical pathways depends on the type of carbon structure [77]. The non-metal carbon catalyst/oxidant agent system can be an alternative treatment for wastewater effluents. More investigations should be performed because this is a very promising idea, and to date, no large-scale applications or pilot plants have been carried out. It should be observed whether this system can also be efficient in real effluent, rich in normal organic matter, which normally are able to scavenge the radical species.

Hybrid Advanced Oxidation Processes Based on Sulfate Radicals
Hybrid advanced oxidation processes based on coupling of either UV-Vis or microwave irradiation or ultrasonic (US) with heterogeneous catalyst/PMS have been investigated to improve the oxidation efficiency. Under UV-Vis irradiation, one of the catalysts studied is titanium dioxide (TiO 2 ). Chen et al. [78] reported that peroxymonosulfate was able to accelerate the TiO 2 photocatalyst of acid orange 7 under visible light. The performance of this system was better than UV-Vis/TiO 2 /PS and UV-Vis/TiO 2 /H 2 O 2 . Furthermore, the presence of humid acid has positive effect on degradation of acid orange 7 under UV-Vis/TiO 2 /PMS. Similar results were reported by Kuriechen et co-workers [79], where the photodegration rate of textile dyes such as reactive red 180 (RR180) increased by 10-fold adding peroxymonosulfate (e.g., oxone) in the UV-Vis/TiO 2 system. Other hybrid systems coupling microwave irradiation with catalyst/PMS have been reported to have a positive effect on P-nitrophenol removal (PNP) [80]. In fact, the microwave irradiation accelerated the processes and achieved almost 98% degradation of PNP in 2 min using PMS/MnFe 2 O 4 /MW. The main advantage of microwave irradiation is that the temperature of aqueous solution increases and can create hotspots on the catalyst surface. Therefore, in such cases, it is really important to know the stability of the used catalysts. Many parameters influence the efficiency of this treatment such as (i) the right quantity of the catalyst; (ii) the quantity of the oxidant agent; (iii) pH of the solution; (iv) the photon flux. Table 3 summarizes the main characteristics of metal catalysts used in the production of sulphate radicals.

Conclusions and Implications for Future Applications
AOPs based on sulfate radical generation have attracted attention for applications in wastewater treatment and have been considered as a promising alternative treatment to remove recalcitrant organic compounds and traceable emerging contaminants. The advantages of sulfate radicals over traditional hydroxyl radicals are that they are more selective in the oxidation of target compounds and that they have a longer lifetime. Persulfate or peroxymonosulfate systems are often activated by alkali, heat, radiation, carbon based catalysts, and transition metals. The most used activation method is the use of transition metals; precisely, by Fe(II) or any other source of iron(II) such as zero valent iron. This activation is used more often because its performance can be enhanced by combining irradiation. Non-ferrous systems are also used, such as the use of Co, but the application is limited due to leaching of the metal into the environment. To overcome this problem, recent developments used persulfate or peroxymonosulfate activation by several carbon-based catalysts.
In summary, this work intended to highlight the major findings with regard to the application of sulphate radicals in the wastewater treatment for recovery and reuse. The results to date are very promising, and possible further applications could be in the treatment of leachates from irrigation waters, in order to be reused again for irrigation, which is an extremely important issue for arid areas, which rely heavily on agricultural activities.