A review on heterogeneous oxidation of acetaminophen based on micro and nanoparticles catalyzed by different activators

Emerging contaminants are the contaminants that newly identified their adverse effects on the environment. Pharmaceutical compounds have gained researchers’ attention among developing organic pollutants as the demand for pharmaceutical compounds has increased, implying their continuing release into the environment. Acetaminophen (ACT) is a popular drug that is wildly used without prescription for the relief of headaches and rheumatic pains. In some places, the detected values of ACT are more than the natural values, which may seriously threaten the environment. Many methods have been applied to remove ACT from water. The advanced oxidation process (AOP) based on micro and nanoparticles has shown promising results to remove ACT from an aqueous medium. This review provides a summary and an organization of the scattered available information related to studies that investigated the removal of ACT from water by an AOP based on micro and nanoparticles. Many topics investigated in this review include the influence of temperature, pH, catalyst concentration, pollutant concentration, the effects of scavengers and oxidants, the stability of the catalyst, and doping ratio. The main results obtained for the removal of ACT by using micro and nanoparticles have been discussed in this review.


Introduction
Emerging organic contaminants (EOCs) have recently gained attention due to their resistance to oxidation and negative impact on the environment [1]. EOCs groups including pharmaceuticals and personal care products, pesticide, disinfection byproducts, wood preservation, endocrine disrupting compounds, bacteria, cyanotoxins, and industrial chemicals [2]. The continuous introduction of these bioactivity compounds into the environment in many ways. Even in low concentrations, they attracted the regulatory organization and governments [3]. Among EOCs compounds, pharmaceutical compounds have attracted real attention because of their negative impacts on public health and the environment [4,5]. Many pharmaceutical products are widely used as anti-inflammatories, analgesic, lipid regulators, antibiotics, antiepileptics, antiseptics, and disinfections. Nowadays, a large number of prescription and nonprescription cure have been used around the world [6][7][8].
Recently, the consumption of pharmaceutical compounds has increased, which means the continuous release of them into the environment. In the European Union, the use of pharmaceutical compounds could reach thousands of tons per year [9]. With passing time, they may reach a specific concentration causing chronic toxicity effects for humans and the organisms. Moreover, pharmaceutical compounds may enter into the human body through inhalation, ingestion, or transdermal delivery leading to accumulation in tissues, reproductive damage, inhibition of cell proliferation, and behavioral changes [10][11][12]. It should be noted that there is no standard set for the discharge limits of the pharmaceuticals in an aqueous medium, and the researches that examined pharmaceutical concentrations in water are limited [13]. The sources of pharmaceutical compounds are varied: they could be released from wastewater treatment plants (WWTPs), hospitals, medical care centers, landfills, domestic sewage systems (the drugs could discharge with the urine and manure of the human body through the sewer system), and industries (unused, expired, and residual); as a result, pharmaceutical compounds have been found in the surface water and groundwater [14,15]. Acetaminophen (ACT) or paracetamol (C 8 H 9 NO 2 , M w = 151.163, N-(4 hydroxyphenyl)ethanamide) is a popular drug that is widely used without prescription for the relief of headache, migraine, neuralgia, backache, and rheumatic pains [16,17]. The natural value of ACT in the surface water and municipal WWTPs has been detected to be less than 100 ng/L. However, the detected values of ACT in the industrial effluent from a few to tens of milligrams per liter, which may seriously threaten the aquatic organisms and the environment [18][19][20]. In addition, ACT has some toxic byproducts, such as 1,4-benzoquione and N-acetyl-P-benzoquinone, which can harm the kidney and liver in humans [21]. Freshwater scarcity, and the rising drinking water demand, is considered one of the environmental problems in the twenty-first century according to its effect on environment, economic, and society [22]. The increase in drinking water demand may attribute to many reasons, such as industrialization, the increase in population, environmental change, and environmental demand. To fulfill the increasing demand for drinking water and to skip any further accumulation of pollutants in the environment, it has become important to treat contaminated water, and also it helps to remove the pollutants mixing with clean water sources [23]. Many approaches have been applied to remove ACT from water [24][25][26][27]. Thus, approaches are classified into three major processes: physical, chemical, and biological process. Among these approaches, chemical oxidation proved its ability to degrade recalcitrant compounds, such as ACT, that resist the bioremediation [18]. The researchers' concern for the ACT degradation is evident from the number of research publications on ACT removal per year, as shown in Figure 1. Last 50 year survey on oxidation of ACT showed 1,620 documents in scopus.com, out of which 56% (900) publications were reported in the last 10 years. Data analysis on the literature survey showed that 92.2% research articles, 4.3% review articles, and 1.6% conference articles were published during this period. This current study, aims to review the recent studies that deal with advanced oxidation process (AOP) based on micro and nanoparticles to remove ACT from an aqueous medium. Also, this study provides a summary and an organization of the scattered available information related to this subject. Many topics investigated in this review include the influence of temperature, pH, catalyst concentration, pollutant concentration, the effects of additives (scavengers and oxidants), the reusability and durability of the catalyst, and doping ratio. The main results obtained for the removal of ACT by using micro and nanoparticles have been discussed in this review.

AOP
The mechanism of AOP relies on the activation of some molecules to create reactive species called radicals. The examples of AOP techniques are varied such as Fenton, metal/PS, electrochemical, ultrasound/oxidant, ultrasonic irradiation, ozonation in the presence of particles, and nanoparticles/ultraviolet light (UV). Application of these processes into the water medium generates radicals with a high oxidation-reduction potential that can oxidize different organic compounds [28][29][30][31]. There are two mechanisms for the AOP: heterogeneous and homogeneous. In the microcomposite-based and nanocomposite-based catalyst systems, oxidation mainly occurs on the catalyst surface; therefore, the heterogeneous reaction pathway is dominant. Many heterogeneous mechanisms were proposed, for example, singlet oxygen, surface-activated complex, surface-confined sulfate radicals, and surface electron transfer (catalysts as an electron conductor) [32]. Chemical Eqs. (1)-(3) represent iron oxide-based catalyst for heterogeneous nonradical-based reaction and Eqs. (4)-(12) represent iron oxide-based catalyst and persulfate (PS) as an oxidant for homogeneous radical-based reaction [5].
A wide variety of AOP systems, including homogeneous and heterogeneous mechanisms, have been applied to degrade ACT from an aqueous medium. AOP systems that are based on homogeneous mechanisms have many drawbacks detected, such as the difficulty of catalyst recovery and metal ions leaching in the reaction media, which caused secondary water pollution [33]. As revealed, heterogeneous catalytic systems such as AOP based on micro and nanomaterials have shown promising results to remove ACT from water [34]. Many semiconductors, such as WO 3 , ZnO, SnO 2 , TiO 2 , CeO 2 , Bi 2 O 3 , and C 3 N 4 , and metals, such as Fe 3 O 2 , Fe 2 O 3 , Al 2 O 3 , zero-valent aluminum, and Cu 2 O, have been used as heterogeneous catalysts. Also, the doping between semiconductors and metals has attracted the researcher's attention according to their advantages, such as reduction in the bandgap of the semiconductors, increase in the bandwidth of absorbance light of the semiconductor surface, and reusability of the composites many times after regeneration.

The degradation mechanism of ACT by nanoparticles
The oxidation of ACT based on semiconductors and metal nanomaterials has been gaining the attention because of its high degradation performance, low toxicity, low cost, and ability to function under different conditions. A wide range of activators have been used to catalyze the nanoparticles, such as ultrasound, irradiation (UV and visible light), plasma, and oxidants. The oxidation mechanisms for ACT degradation were varied. Most of the studies observed that the main degradation mechanism was based on the formation of superoxide radicals.

Mechanism by nanoparticles catalyzed by UV and visible light
Semiconductor nanoparticles capable to generate a hole (h + ) and electrons (e − ) after illuminated with UV or visible light make it the most promising oxidation process as electrons act as reduction agents, whereas holes act as oxidation sites [2]. The mechanism of semiconductors that catalyzed by irradiation, based on photoexcited of the electrons that exist on the catalyst surface, leading to the movement of electrons (e − ) from valance band to conduction band leaving positives holes (h + ). Both electrons and holes can start the redox reactions and oxidize ACT. The mechanism of nanoparticles catalyzed by UV and visible light is depicted in Figure 2.
The equations (13)-(18) represent the possible chemical reactions of TiO 2 catalyzed by UV or visible light system [35].
Some disadvantages have been observed related to photodegradation in the presence of semiconductor TiO 2 . For example, the high rate of recombination of electronsholes leading to minimize the degradation of ACT, a wide bandgap of TiO 2 , low ionic and electrical conductivity, slow charge transfer rate limits the quantum efficiency of TiO 2 in the photocatalytic reactions, limited adsorption capacity and porosity, and lower efficiency under solar irradiation restricts the application of this system. To decrease the bandgap and enhance the absorption of irradiation of surface catalyst, the researchers have doped semiconductors with transition metals (Co, Fe, Ni, Ag, Au, Cu, Mg, Pt, Zn, Si, and Al) and nonmetals (N and Cl) [17,[36][37][38][39][40][41][42][43]. Kohantorabi et al. [38] revealed that ACT was completely degraded after 15 min of reaction, and the mineralization was 63% within 1 h when 1.0% w w Ag ZnO@NiFe O PMS UVA 2 4 / / / / was applied. In addition, Yang et al. [44] examined TiO 2 nanoparticles activated by UVC to degrade ACT from a liquid medium. In this system, the initial concentration of ACT was 2.0 mM, and the catalyst concentration was 0.4 g/L. The degradation of ACT was 96% after 80 min of reaction at a pH range of 5.1-3.2. Moreover, Montenegro-Ayo et al. [45] applied TiO 2 nanoparticles/UV/0.02 M Na SO 2 4 system to oxidize ACT. Around 72% of ACT was degraded after 300 min. Sun et al. [46] applied 0.5 mM N/S codoped ordered mesoporous carbon to catalyze peroxymonosulfate (PMS) for ACT degradation. Around 50 mg/L of ACT was totally oxidized within 30 min, and the mineralization was 27%. The kinetic reaction of ACT in this system was 2.4 10 1 × − /min. Wang et al. [47] examined BiOCl UV Persulfate / / system to oxidize ACT from an aqueous medium. The results showed that 50 µM of ACT was completely degraded within 150 min, and the mineralization was 83% after 180 min at pH = 5.4. The kinetic reaction was 7.13 10 4 × − /min. The studies that applied semiconductor as a catalyst and activated by UVA or UVC to degrade ACT from an aqueous medium have summarized in Table 1.
As the catalyzing of nanoparticles by using UV is costly, to avoid that, the researchers intensified on an alternative photocatalytic method such as visible light. As mentioned, TiO 2 powder has a limited absorption capacity of solar light: just around 5% could be absorbed by TiO 2 powder. To enhance the optical absorption of the catalyst, the researchers examined different approaches to improve the TiO 2 performance. Gómez-Avilés et al. [2] used C-modified TiO 2 nanoparticles/solar irradiation system for ACT degradation. In this system, ACT was completely removed within 60 min, and the mineralization was 20.4% after 120 min. Da Silva et al. [40] studied the degradation of ACT by using 2 g/L of 25% MgO doped with TiO 2 catalyzed by the solar light system. About 48.3% of ACT was degraded within 1 h at pH = 7. Furthermore, Aziz et al. [43] applied 5 wt% of TiO 2 doping with SiO 2 -ZSM-5/visible light system to remove ACT from an aqueous medium. The results showed around 96% of ACT was oxidized after 180 min, and the mineralization was 77.8%. In addition, 0.1 wt% Cu doped TiO 2 -/visible light system was applied by Lin and Yang [48]. The results indicated that 50 mg/L ACT was completely decomposed after 3 h of reaction at pH = 6, and the degradation rate was 0.0243/min. Also, Feng et al. [49] examined oxygen vacancies and phosphorus codoped black titania-coated carbon nanotube composites (OVPTCN) activated by a visible light system to degrade and mineralize ACT from a liquid medium. The results were 96 and 20.4% of degradation and mineralization, respectively. Table 2 includes the studies that investigated the semiconductor particles/ visible light systems to remove ACT from an aqueous medium.

Mechanism by nanoparticles catalyzed by oxidants
Many studies investigated the oxidation of ACT by using synthesized particles in the presence of oxidants. For example, Ikhlaq et al. [50] examined the oxidation of ACT by zeolite/O 3 system. They proposed that both ACT and O 3 were adsorbed on the surface of zeolite then react with each other, which supports that the nonradical mechanism was dominant. Figure 3 illustrated the adsorption of oxidant and ACT on the catalyst surface, then the oxidant attacked ACT, which resulted in the degradation of the pollutant via the heterogeneous or nonradical mechanism. In addition, heterogeneous and homogeneous could happen together. Mashayekh-Salehi et al. [3] applied MgO/O 3 system to oxidize ACT in an aqueous solution. They proposed the following chemical equations (19)- (25) that might happen while the degradation reaction is running.   *Means data not available; BET: Brunauer, Emmett and Teller.
Oxidation of acetaminophen based on micro and nanoparticles  503  Photo-Fenton using FeO • Radical type catalytic oxidation on MgO surface: • Direct oxidation with O 3 molecules in the bulk solution: • Radical type catalytic oxidation in the bulk solution: The symbol S in the Eqs.
The possible reactions Eqs. (37)- (39) in the presence of PMS are as follows [53]: Furthermore, in the systems that relay on PS and PMS as a catalyst, the reaction may go further to produce hydrogen peroxide (OH˙) as in the following Eqs. (40)-(44) [5].  Table 3 lists the studies that applied metals and semiconductor particles catalyzed by oxidants to remove ACT from the aqueous medium.

The influence of different parameters on ACT degradation
Many different parameters that influence the degradation of ACT such as temperature, pH, catalyst concentration, pollutant concentration, effects of additives (scavengers Oxidation of acetaminophen based on micro and nanoparticles  505

The influence of ACT concentration
Major studies that have applied AOP systems in the presence of synthesized particles for ACT oxidation pointed out that when the ACT concentration increases, the degradation efficiency decreases. For systems based on oxidants such as hydrogen peroxide, PS, and PMS, high ACT concentrations may adsorb and cover a wide number of the active sites on the catalyst's surface, consequently, suppressing the production of super oxidant radicals. Also, for the systems that depend on UV or visible lights as a catalyst, a high ACT concentration may accumulate on the catalyst surface and prevent the penetration of the irradiation, which may reduce the photocatalytic efficiency. For example, Yang et al. [44] applied TiO 2 catalyzed by UVA and UVC. They carried out different ACT initial concentrations from 2 to 10 mg/L. The removal decreased from 95% to less than 20%. Montenegro-Ayo et al. [45] studied the changes in the degradation rate when the initial concentration of ACT increased from 5 to 50 mg/L. The results showed a decrease in the oxidation rate from 2.05 10 4 × − to 2. 86 10 5 × − , respectively. Also, Tan et al. [54] examined TiO 2 activated by UV system to eliminate ACT from liquid medium. The oxidation of ACT was declined from 72.7 to 40.2% when the ACT initial concentration increased from 0.017 to 0.067 mM, respectively. Furthermore, Kurniawan et al. [55] applied photocatalytic of BaTiO 3 /TiO 2 composites to remove ACT; 5 and 25 mg/L of initial ACT concentration were implemented, and the results were a decline in the degradation efficiency from 81 to 19%. The same results were observed when Yaghmaeian et al. [56] carried out modified MgO nanoparticles catalyzed by ozone. When the initial concentration of ACT was 10, 50, 100, and 200 mg/L, the removal was 99.5, 99.4, 77, and 45%, respectively. However, Fan et al. [57] observed that when the initial ACT concentrations were 0.5, 1.0, and 1.5 mg/L at Ag/AgCl@ZIF-8/visible light system, the removal efficiency was stable at 99% after 60 min. They reported that when the concentration was between 0.5 and 1.5, the reaction rate reached the fastest at 1.5 mg/L. However, when the initial concentration exceeds 2.0 mg/L, the reaction rates decrease because the permeability of the photon would reduce when the substrate concentration was too high.

The influence of semiconductor and metal dosages
Most studies agreed that when the catalyst concentration increases to a certain level, it may benefit and increase the degradation reaction. However, if an excessive amount of catalyst adds, that maybe affected adversely on the degradation performance, or at least the degradation performance stays similar. For systems based on photocatalytic, that might interpret because, at high catalyst concentrations, the agglomeration and the shielding effect of the suspended catalyst is due to increased turbidity and impedes the light penetration, which reduces the accessible light to the catalyst surface resulted decreasing in the photocatalytic. For AOP-nanocomposite systems that based on irradiation activation such as hydrogen peroxide, PMS, and PS, high catalyst concentration leading to an increase in the number of activated radicals, which leads to the self-consumption of generated radicals.  [59] mentioned the effect of catalyst dosage when the LED/ titanium dioxide doped with graphene oxide (TiO 2 @rGO) system was applied to remove ACT. They implemented TiO 2 @rGO concentrations from 0.4 to 4 g/L. It observed that when the catalyst concentrations were 0.4, 1, and 2 g/L, the removal was 53, 57, and 81%, respectively. Meanwhile, when the concentration of TiO 2 @rGO increased from 2 to 4 g/L, there was no improvement in the degradation of ACT. Moreover, Abdel-Wahab et al. [60] examined the TiO 2 /Fe 2 O 3 /UV system that when the catalyst concentration increased from 0.1 to 2 g/L, the reaction rate was strongly affected, while concentrations between 0.1 and 1.2 g/L, the removal rate increased because the number of active sites and activated radicals increased. Hence, at 2 g/L of TiO 2 /Fe 2 O 3 was applied, the degradation rate was declined. Also, different K 2 S 2 O 8 -doped TiO 2 dosages have been applied. From 0.25 to 0.5 g/L concentration, the oxidation of ACT increased from 90 to 100% after 9 h. When the dosage increases from 0.5 to 1.5 g/L, the removal was kept around 100% and reached the fastest reaction rate at 1 g/L. While at 2 g/L was applied, the degradation dropped from 100 to 97% [61].

The influence of pH
pH is a significant factor in the AOP based on semiconductors and metal systems. The effect of pH in (AOP/ composites) system for the degradation of ACT is widely investigated. It has been observed that ACT has two chemical forms depending on the pH: (i) nonionic form when the pK a is under 9.4 and (ii) ionic form when pK a is more than pH 9.4. In addition, pH zpc and the type of composite were the main variables to define the oxidation performance of ACT [62]. For example, Ziylan-Yavaş and Ince [17] studied the degradation of ACT by using Pt-supported nanocomposites of the Al 2 O 3 /O 3 system. The results showed that at the base and neutral pH conditions, ACT was eliminated in 7 and 10 min, respectively. The carbon mineralization increased when increasing the pH because the consumption of ozone increased with increasing pH. That might be attributed to the increase in O 3 that converted to OH˙at high pH atmosphere. The impact of pH between 4 and 10 on the oxidation of ACT by using 1.0% w/w Ag/ZnO0.4@ NiFe 2 O 4 /PMS/UVA system was investigated. The best result was obtained at pH between 6 and 7. Because at this pH, the HSO 5 − ions can be attracted to the positive surface of the catalyst, which improve the oxidation efficiency by the production of OH SȮ 4 / − . At acidic conditions, the activation of PMS was decreased due to the H-bond formation between H + and O-O group of HSO 5 , which declines the ACT removal [38]. Hassani et al. [39] studied the impact of pH from 3 to 11 on the oxidation of ACT by using the CoFe 2 O 4 /mpg-C 3 N 4 system. They revealed that the best result was obtained when the pH was 7. Both acidic and base conditions were not favorable for degradation in this system. In the acid atmosphere, the oxidation of ACT was not good for two reasons because the activation of PMS was decreased due to the H-bond formation between H + and O-O group of HSO 5 − , which decreased the ACT removal, and both SO 4 − and OH˙react with H + resulted in reducing in the degradation of ACT. In alkaline conditions, there were some possible reasons responsible for decreasing ACT degradation: (1) converting SO 4 − radicals to OH˙species with relatively lower redox potential by OH − ions, (2) formation of metal hydroxide complexes of CFNPs and subsequently decrease of PMS decomposition reactions, and (3) self-decomposition of PMS to water and sulfate ions. Ling et al. [41] examined the influence of pH on the mineralization of ACT when (solar/4% Ag-g-C 3 N 4 /O 3 ) system was implemented. They applied the following pH: 3, 5, 7, 9, and 11. From pH 3-7, the degradation of ACT was enhanced because the increase of pH value increases the conversion of O 3 to OH˙, which degrades more ACT effectively when the pH increased from 7 to 11, the removal of ACT kept constant because at these pH values, the Ag-g-C 3 N 4 and ACT have the same charge, and the repulsion force was dominant. Sun et al. [46] applied pH from 3.5 to 9 on N/S codoped/ PMS system. The results showed that when the pH increased, the degradation was improving because the high concentration of OH − caused the decomposition of PMS to produce SO 4 − . Ikhlaq et al. [50] studied the oxidation of ACT by using a zeolite/O 3 system at pH 3, 7, and 10. In this study, the optimum pH for zeolite/O 3 was 7.12 because, at this pH value, the ACT and hydroxyl groups on the zeolite surface were protonated. However, at pH 3 and 10, the oxidation of ACT decreased because ACT and zeolite at this pH have the same charge. Zhang et al. [52] studied the pH in the ironcopper bimetallic system activated by PS to remove ACT. The best pH values were between 5 and 7. The strong acid and alkaline conditions were not favorable for this system. Moreover, the effect of pH in Fe 3 O 4 magnetic nanoparticles investigated by Tan et al. [53]. They reported that there were two effects of pH on the experiment: Kurniawan et al. [55] examined a wide range of pH values from 3 to 11 for ACT degradation using BaTiO 3 /TiO 2 /UVA. From pH 3 to 7, the degradation enhanced from 7 to 95%, and the optimum value for ACT oxidation was 7; meanwhile, in alkaline conditions, the degradation decreased from 95 to 54% because both the catalyst and the ACT molecules had negative charges in alkaline conditions. As a result, the catalyst's surface was repelled to the negatively charged ACT molecules, leading to a low ACT removal. Yaghmaeian et al. [56] investigated the influence of different pH to oxidize ACT by using modified MgO nanoparticles catalyzed by ozone. They observed that the consumption of ozone was related to the increasing of pH: when the pH increased from 2 to 8, the consumption of ozone increased from 17 to 41.5%, and from 75 to 90% when the pH increased from 9 to 10. Also, more O 3 consumption means more OH production. OH˙, has a high oxidation potential, which is more than O 3 . The best pH value in this system was the natural pH solution close to 5.4; at this pH, the ACT molecule was mostly in its molecular form and could better interact with OH. In acidic conditions, there was no O 3 converted to OHė nough. In the alkaline conditions, the isoelectric point of modified MgO and ACT were 10.4 and 9.4, respectively, which means the catalyst m-MgO significantly promoted the decomposition of O 3 . In addition, Fan et al. [57] mentioned the influence of pH on the Ag/AgCl@ZIF8 system for oxidation of ACT. In this system, the optimum pH value was 5. The pH values between 7 and 9.4 were not desired because of weaker electrostatic integration between ACT and Ag/AgCl@ZIF8. For pH, more than 9.4 any pH values less than 7 were favorable for degradation ACT in this system. However, they noted that Ag/AgCl@ZIF8 dissolved in strong acid, which decreased the efficiency of this system for ACT removal. The influence of pH for decomposition of ACT in TiO 2 @rGO nanoparticle system was studied by Cheshme Khavar et al. [59]. They revealed that when the pH increased from 4 to 9, the degradation of ACT promoted from 68 to 93%, respectively. Thus, it can be explained that at the natural pH solution, the surface catalyst has a negative charge, the anion species could not adsorb at the catalyst surface, allowing higher functional group interaction of OH − , which resulted in a high amount of OH − converting to OH˙and finally enhanced the degradation of ACT. Zhang et al. [63] observed that when the pH increased from pH 3 to 6.5, the degradation efficiency increased from 80 to 91%, respectively. In the acidic atmosphere, the oxidation of Fe 2+ to Fe 3+ was slower, according to the increasing of H + , which inhabited the catalyst, resulting in a decrease in the degradation of ACT. When the initial pH value was 11, it has been observed that the pH decreased to be 3. They explained that because PS produced a large amount of H + , the pH was drastically reduced. At pH 11, the OH − molecules combined with Fe 2+ caused a rapid reduction of Fe 2+ , resulting in an insufficient coupling effect of Fe 2+ and CuO, thus limiting the oxidation of ACT. Peng et al. [64] carried out different pH conditions. They revealed that pyrite/PDS could apply for a wide pH range, whereas pyrite/H 2 O 2 in a narrow range. When pH 6 was implemented, the degradation of ACT in the pyrite/PDS system was 50%, whereas 0% when pyrite/H 2 O 2 system was applied. Also, at pH 8, the results were 10 and 0%, respectively. It is believed that pH playing a central role in the determination of radical species of PDS, from pH 2 to 7 SO 4 − , from 8 to 10 OH SȮ 4 / − , and from 10 to 12 OH˙was dominant. Also, Dong et al. [65] applied pH 3, 9, and 12. The results were 98.5, 79.5, and 13.5%, respectively. They observed, when the initial pH was 12, the pH value was decreasing to 3. This, because SO 4 − would react with OH − and H 2 O resulting in the removal of H + and consumption of OH − . In the base conditions, OH − combined with Fe 2+ to form oxyhydroxides and leads to precipitation. The absence of OH − leading to insufficient activation of PS, which was due to a decrease in the degradation of ACT. Zhang et al. [66] also mentioned the influence of pH by using a S-doped graphene/Pt/TiO 2 system. When pH value increased from 4, 8, and 10, the removal was 99.9, 95.3, and 95.1%, respectively. In acidic conditions, hydroxyl radical behaved like a weak acid and reacted with hydroxyl ions under neutral and alkaline conditions, which was due to decreasing degradation reaction. The effect of pH on green rust coupled with Cu(II) has been reported by Zhao et al. [67]. They investigated three systems GR SO4 /Cu(II), GR CO3 /Cu(II), and GR Cl / Cu(II), and it was observed that when pH 6 was applied on GR SO4 /Cu(II), and GR Cl /Cu(II), the pH declined to 4, and 5.4, respectively. However, when pH 6 was applied to GR CO3 / Cu(II), the pH increased to 6.4 and then decreased to 4.2. This pattern may be due to the buffering effect of CO 3 in this H + system. Hydrolysis due to the rise in the pH, but as more accumulated, the buffering ability was exceeded, and then the pH decreased. When the pH decreased in the efficiency of GR SO4 /Cu(II) and GR CO3 /Cu(II) decreased from 100 to 82% and 84 to 28%, respectively.

The influence of scavengers
There are two mechanisms for the AOP: heterogeneous and homogeneous. In the mineral-based catalyst systems, oxidation mainly occurs on the catalyst surface; therefore, the heterogeneous reaction pathway is dominant. The mechanism of AOP is complex because some radical species are generating in parallel or series. To identify the radicals that are responsible for the degradation of ACT, the researchers added some substance called scavengers or quenching agents acting to trap the activated radicals: after adding these scavengers, the oxidation significantly declines, which means the radical trapped is responsible for the oxidation process. Table 4 represents the radicals generated in the AOPsynthesized particle-based system for degradation of ACT. Many scavengers such as isopropanol (IPA), tert butyl alcohol (TBA), methanol, salicylic acid, benzoquinone (BQ), KI (potassium iodide), triethanolamine (TEOA), ammonium oxalate, ethylenediaminetetracetic acid disodium (EDTA-2Na), ethanol (EtOH), sodium oxalate, N 2 , Oxidation of acetaminophen based on micro and nanoparticles  511 ) was responsible for ACT degradation [102] (Continued) 4-hydroxy-2,2,6,6 tertamethylpiperidinyloxy (TEMPOL), and L-histidine (L-his) have been applied.

The influence of doping ratio
The impact of the doping ratio in the synthesized particles, such as TiO 2 @rGO, Ag-g-C 3 N 4 , ZSM-5/TiO 2 , BaTiO 3 / TiO 2 , La-doped ZnO, TiO 2 /Fe 2 O 3 , Fe 3 O 4 @SiO 2 , Mg/SiO 2 , and iron-copper bimetallic doped with silica (Cu/ Fe 3 O 4 @SiO 2 ), for the ACT removal has been investigated. Most of these studies agreed that when the doping ratio increases, the degradation of ACT increases. But if the doping ratio increased above the threshold, it might negatively impact the degradation performance. For example, Ling et al. [41] applied the Ag-g-C 3 N 4 system to degrade ACT. The excessive amount of Ag might be accumulated on the g-C 3 N 4 nanoparticle surface and cover the active sites, which increase the recombination of photogenerated charges. Lin and Yang [48] examined Cu-doped TiO 2 to eliminate ACT, many doping ratios were applied, 0.1, 1, and 10 wt%. The best result was 100% of ACT degradation within 3 h by using 0.1% of Cu. They revealed that when the Cu ratio increased, it may generate isolated CuO aggregates expelled from the Cu-TiO 2 framework to the pore channels, which may act as a center of charge recombination and decline the mass transport between the pore channels. Also, Kurniawan et al. [55] prepared BaTiO 3 / TiO 2 composites catalyzed by solar irradiation. Three different ratios of BaTiO 3 /TiO 2 were applied, composite-A (1:3), composite-B (1:1), and composite-C (3:1). After 4 h of reaction, the results were 76, 39, and 26%, respectively. Cheshme Khavar et al. [59] applied TiO 2 @rGO nanocomposites catalyzed by UVA to oxidize ACT. Different doping ratios starting from 0, 1, 3, 5, 7.5, and 10 wt% of rGO were applied, and the degradation of ACT was 53, 83, 100, 87, 76, and 70%, respectively. These clearly showed that 3 wt% was the best doping ratio in the TiO 2 @rGO system. If the doping ratio exceeds 3 wt%, the degradation efficiencies begin to decrease because a large amount of ACT adsorbs on the catalyst surfaces, which due to some occupied active sites of TiO 2 resulted in decreasing the UV light that reaches to TiO 2 surface and reduces the photocatalytic activity. In addition, when the UV transmission decreased by TiO 2 , it might increase the recombination rate. Thi and Lee [68] observed that the presence of La doped on ZnO nanoparticles enhanced the photocatalytic activity and reduced the bandgap energy when applied 0.5 and 1.0 wt% La-doped ZnO. Meanwhile, too much adding like 1.5 wt% of La may adversely affect the system performance and increase the bandgap energy because, in each semiconductor, there is an energy level named Fermi, which is the highest energy level occupied by electrons in a particular site. In ZnO, the Fermi level is between the conduction band and valance band. When more than 1.0 wt% of La doped onto ZnO, the bandgap was increased because Burstein-Moss effect. These states could push the Fermi level to a higher energy position, and then the Fermi level would lie in the conduction band the process depicted in Figure 4. Aziz et al. [69] studied TiO 2 doped onto fibrous silica ZSM-5 system catalyzed by solar light to oxidize ACT. Different doping ratios starting from 1, 3, and 5 wt%, of TiO 2 were carried out, and the results were 65, 90, and 71%, respectively. The results indicated that 3 wt% was the best doping ratio in the ZSM-5/TiO 2 system. They explained that TiO 2 might agglomerate on the surface of fibrous silica and cover the active sites, which caused low penetration of the visible irradiation. This effect was detected when electron carrier concentration exceeded the conduction band edge density of the state. However, some studies did not observe any decreases with an increase in the doping ratio. For example, Da Silva et al. [40] applied Mg/SiO 2 /UV system. Many Mg concentrations experimented with to oxidize ACT, 1, 2, 10, and 25 wt%. The best result was 60% of ACT was removed within 60 min obtained when 25 wt% Mg concentration was applied. Moreover, Abdel-Wahab et al. [60] applied TiO 2 / Fe 2 O 3 with different TiO 2 ratios, 15, 33, and 50 wt%. They observed that when the concentration of TiO 2 increased from 15 to 50%, the degradation increased from 52.5 to 98%, respectively. The improvement of ACT degradation could be correlated to the effective separation of photogenerated electron-hole pairs accomplished by a combination of narrow bandgap Fe 2 O 3 with wide bandgap TiO 2 . In addition, the oxidation rate of ACT was accelerated proportionally when increasing copper concentration from 0 to 1% in Cu/Fe 3 O 4 @SiO 2 system, and the degradation increases from 59.2 to 100%, respectively. That was because ACT could quickly and efficiently adsorb on the catalyst and thus increase the catalyst activity. Do et al. [51] observed that when the molar ratio of Cu increased from 2.1 to 2.94, the reaction rate begins to slow down because of the dispersion morphology of Cu nanocomposites of the surface of iron doped with silica (Fe 3 O 4 @SiO 2 ).

The influence of oxidants dosage
The influence of addition and the concentrations of PMS, PS, oxygen (O 2 ), ozone (O 3 ), and hydrogen peroxide (H 2 O 2 ) have been widely investigated. All the studies agreed that the addition of oxidants enhances the degradation and the reaction rates of ACT. However, when an excessive amount of oxidant is added, it may impact adversely on the degradation performance. This attributed to many reasons: (1) the excessive amount of oxidants generate more radicals, these radicals may consume each other as shown in the Eqs. (46)(47)(48)(49)(50): (2) The limitation of active sites on the catalyst surface according to the presence of a high concentration of oxidant, and (3) if the excessive concentration of H 2 O 2 was added, the generated hydroxyl radicals may react with H 2 O 2 to produce HO 2 HO 2˙, which contributed less oxidation potential than OH˙. Several studies investigated the addition of oxidant in AOP based on composites systems. Ziylan-Yavaş and Ince [17] observed that when the ozone flow increased from 3, 6, and 9 mg/min on Pt/Al 2 O 3 /O 3 system, the oxidation was enhanced, and 9 mg/min contributed the best ozone flow in this system. That may refer to the excessive amount of H 2 O 2 produced from the oxidation of ozone, which in turn increases the OH˙that attacks ACT resulting in an increased ACT degradation. Hassani et al. [39] also examined different PMS dosages in CoFe 2 O 4 / mpg-C 3 N 4 catalyzed by PMS to degrade ACT. At 0.5 mM of PMS, the degradation efficiency was 60.9% after 25 min reaction. At 1.5 mM of PMS, the efficiency increased to 92%. They mentioned that higher PMS concentration was not favorable in this system because of the reasons mentioned above. Moreover, Sun et al. [46] examined many concentrations of PMS on N/S codoped ordered mesoporous carbon system. When 0.25, 0.5, and 1.0 mM of PMS were applied, the k values increased from 2.0 ± 0.04 × 10 2 − to 2.4 ± 0.06 × 10 1 − and 3.7 ± 0.2 × 10 1 − , respectively. Also, Wang et al. [47] pointed out the addition of Na 2 S 2 O 8 and H 2 O 2 on the BiOCl/UVA system. When Na 2 S 2 O 8 was added, the degradation rate and the mineralization were accelerated and enhanced. They attributed the improvement to three reasons: (1) the direct reaction between the photon and PS molecules, which results in generating sulfate radicals, (2) also, PS may react with conduction band electrons yielding the formation of sulfate radicals, and (3)  [53] studied the influence of adding different PMS concentrations on the degradation rate in Fe 3 O 4 magnetic nanoparticles/PMS system. They observed that when the initial concentration of PMS increased from 0.0 to 0.2 mM, the reaction rate was promoted from 0.23 10 2 × − to 1.22 10 2 × − /min. However, when the initial concentration of PMS increased from 0.2 to 0.5 mM, the reaction rate slightly decreased from 1.45 10 2 × − to be 1.13 10 2 × − /min. The increase of degradation rate in the initial concentration 0.2 mM was attributed to the availability of PMS. At this concentration, PMS acting as a limiting factor controlling the yield of radicals. Furthermore, Tan et al. [58] applied different PMS dosages on MnFe 2 O 4 and CoFe 2 O 4 to eliminate ACT. The dosages were 0.05, 0.1, 0.15, and 0.2 g/L, and the removal of ACT when MnFe 2 O 4 and CoFe 2 O 4 were used were 89, 100, 100, 100% and 55.6, 85.7, 94, 100%, respectively. They noted that when the initial concentration of PMS increased to 0.4 mM, the degradation rate started to decline. In this study, there was no adverse effect observed because the initial concentration of PMS did not reach the threshold level. Dong et al. [65] mentioned that when higher PS concentration applied, PS got in the micropores in the catalyst and would react with SO 4˙− and led to PS consumption, thereby causing the undesired inhibiting effect. In addition, Velichkova et al. [70] As mentioned above, HO 2 contributed less oxidation potential than OH˙, which adversely affect the degradation performance.

The influence of oxygen
The impact of the oxygen for ACT degradation by using AOP based on composites has been investigated. Moctezuma et al. [8] revealed that oxygen has a strong effect on photocatalytic degradation. Bubbling O 2 acted to trap the free electrons to inhabit the recombination of (e − /h + ), which affect positively on the degradation performance. Also, Yang et al. [44] reported that O 2 increased the degradation of ACT more than six times. O 2 could inhabit electron-hole recombination as O 2 consumes conduction band electrons allowing valance band holes too, directly and indirectly as shown in Eqs. (52)- (55).
Zhang et al. [71] pointed out, that DO plays an important role in the oxidation of ACT and for radical's generation when Fe 2+ /CuO was applied. When Fe 2+ /CuO was added, the concentration of DO decreased from 9.48 to 4.85 mg/L in the first 10 min of reaction and then increased to 7.64 mg/L after 6 h reaction. In addition, they observed that the removal of DO completely inhibits the degradation of ACT.

The influence of temperature
According to the literature, the temperature is directly proportional to the removal of pollutant because the pollutant migrates from the bulk solution to the gas-liquid interface region where temperature and OH˙are high. Moreover, it has been proposed that the optimum temperature was based on the characteristic of organic matter and the kinetics of the reaction between OH˙and pollutant. Velichkova et al. [70] studied the effect of temperature on the degradation of ACT by three types of nanoparticles of iron oxide/H 2 O 2 system. They revealed that the increase in the temperature from 30 to 60°C had a beneficial effect for all examined conditions. Also, a higher temperature increases the rate of OH˙formation. In contrast, high temperature increases the decomposition of H 2 O 2 into O 2 and water, which reduces the removal efficiency of ACT. In addition, Im et al. [72] studied the effect of temperature on the removal of ACT by ultrasound. Two ultrasound waves were applied at 28 and 1,000 kHz, and the best temperature was 25 and 35°C, respectively. They mentioned that beyond the optimum temperature leads to an increase in bubble vapor and then bubble collapse due to the reaction of net energy and free radicles. Tan et al. [53] applied different temperatures from 30 to 70°C in Fe 3 O 4 magnetic nanoparticles/H 2 O 2 and PMS system. The results showed that temperature affects positively on the elimination of ACT and the kinetic increased from 1.5 × 10 −2 to 10 × 10 −2 /min, for 30 and 70°C, respectively. In addition, Sun et al. [46] reported that when N/S codoped ordered mesoporous carbon was applied for ACT degradation at 25°C, the removal was 100% after 30 min, whereas at 45°C, ACT was completely degraded after 20 min, and the kinetic at 25 and 45°C were 2.4 × 10 −1 and 3.5 × 10 −1 / min, respectively. The activation energy, E a , of the oxidation system was calculated as 13.8 kJ/mol, which suggested the reaction temperature does not have a significant effect on the oxidation reaction.

Stability and reusability of the catalysts
One of the main advantages of synthesized particles, which use in AOP systems, is their durability and reusability without any considerable change in the degradation performance. According to the literature, the reduction in the degradation efficiency attributes to the following reasons: (1) metals such as Ag, Zn, Fe, Cu, and Ni could be  Oxalic acid, acetic acid, and formic acid [116] Direct electron transfer by reactive Mn 2 O 3 Acetic acid and a-nitrosophenol [117] leaching during the reaction, or/and (2) loss of the catalyst weight during the regeneration, or/and (3) agglomeration during the reaction. Most applied experiments using the following procedure to recover the catalyst: washing the catalyst three times with deionized water and dry them at a temperature between 80 and 100°C for 24 h. Sun et al. [46] used NS-CMK-3 catalyst to remove ACT. After five times of running, the catalyst showed high durability to oxidize ACT from an aqueous medium. Soltani et al. [73] applied ZnO/PSW nanopowder for four times to remove ACT. The authors revealed the first three times of running, the catalyst showed good stability, whereas at the end of the fourth cycle, the efficiency decreased to 20%. Moreover, Pham et al. [74] examined the durability of Fe/N-CNT particles to degrade ACT. Palas recovered and reused the catalyst for ten times. The catalyst showed superior stability even after ten times of degradation cycle, the oxidation efficiency was kept as 99.8%. In the Table 5, some studies have examined the catalyst for many cycles to oxidize ACT.

Byproducts formation
As mentioned in the introduction, some ACT byproducts such as 1,4-benzoquinone and N-acetyl-P-benzoquinone have negative effects on human health. Most of the byproducts mainly consist of hydroquinone and carboxylic acid derivatives. Ling et al. [41] applied Ag G C N O 3 4 3 − − / catalyzed by Vis-UV light to oxidize ACT from a liquid medium. After the degradation process, the main byproducts were hydroquinone, di-hydroxyphenyl, and tri-hydroxyphenyl. Moreover, Montenegro-Ayo et al. [45] observed small byproducts such as formic acid, oxamic acid, and oxalic acid after ACT oxidation by using TiO UV 2 / system. In addition, Zhang et al. [71] detected oxalic acid, hydroquinone, formic acid, acetic acid, and ammonium as byproducts after ACT degradation by using ferrous ion and copper oxide/O 2 system. Table 6 lists the studies, which monitored the byproducts of ACT after treatment processes.

Conclusions and future prospective
This review article has attempted to cover a wide range of state-of-the-art studies related to the oxidation of ACT by using semiconductor and metal catalysts. Hydroxyl radical was the most dominant superoxide responsible for ACT degradation because hydroxyl radical could be generating in all AOP systems, such as ultrasound systems, photocatalytic systems, and AOP-based oxidants systems, that were investigated for ACT degradation. Also, the stability and reusability of the catalysts have been studied. Most of the semiconductor catalysts have shown good stability, but Fe and N codoped carbon nanotube/PS system have shown superior stability, and the degradation efficiency is still 99.8% after the tenth cycle. pH has played a central role in ACT degradation by control of zero of point charge of the catalyst, pK a of ACT, and the formation of radicals. In addition, the influence of catalyst, ACT, and oxidant concentrations has been reported. The increase of catalyst concentration was beneficial, but if the catalyst concentration exceeds the threshold point, it adversely impacted the degradation efficiency. Because of high catalyst concentrations, the agglomeration and the shielding effect of the suspended catalyst are due to increased turbidity and low light penetration, which reduces the accessible light to the catalyst surface resulting in decreased photocatalytic activity. High catalyst concentration leads to an increase in the number of activated radicals, which results in the self-consumption of generated radicals. If excessive amount of ACT was added that is due to, for systems based on oxidants such as hydrogen peroxide, PS, and PMS, high ACT concentrations may adsorb and cover a wide number of the active sites on the catalyst's surface consequently, suppresses the production of super oxidant radicals. Moreover, for the systems that depend on UV or visible lights as a catalyst, a high ACT concentration may accumulate on the catalyst surface and prevent the penetration of the irradiation, which may reduce the photocatalytic efficiency. For oxidants, the excessive amount of oxidants (1) generate more radicals, which may consume each other, (2) the limitation of active sites on the surface catalyst according to the presence of a high concentration of oxidant, and (3) if the excessive concentration of H 2 O 2 was added, the generated hydroxyl radicals might react with H 2 O 2 to produce HO 2 , which contributed less oxidation potential than OH˙. It has been noted that the increase in doping ratio was not beneficial because the agglomerate on the surface covers the active sites, which caused low penetration of the irradiation. The studies agreed that DO improved ACT degradation, which attributed to the reaction between O 2 and generated radical yielding to the formation of superoxide radicals. A high degree of temperature was not good for ACT degradation because high temperature increases the decomposition of H 2 O 2 into O 2 and water, which reduces the removal efficiency of ACT. Finally, most of the byproducts mainly consist of hydroquinone and carboxylic acid derivatives. AOP systems based on micro and nanoparticles are considered a promising method for ACT degradation. There is a deficiency in the literature about the prediction of the oxidation mechanism of ACT in the presence of nanomaterials. This topic needs further investigation. In addition, the threshold of concentrations and the ratio of oxidants, pollutants, and catalysts need further investigation to expect the optimum ratio between them. Author contributions: All authors have accepted responsibility for the entire content of this manuscript and approved its submission.