Lanthanum Nickel Oxide: An Effective Heterogeneous Activator of Sodium Persulfate for Antibiotics Elimination

In recent years, the presence of pharmaceutically active compounds (PhACs) in surface waters and wastewaters has b the effectiveness of conventional water treatment methods. Towards this direction, advanced oxidation processes (AOPs) for the complete elimination of micro pollutants in waters have become an emerging area of research. The present study reports the heterogeneous activation of sodium persulfate (SPS) by LaNiO3 (LNO) perovskite oxide for the degradation of sulfamethoxazole (SMX), an antibiotic agent. LNO was prepared according to a combustion method, and its physicochemical characteristics were identified by means of XRD, BET, TEM, and SEM/EDS. SMX degradation results showed the great efficiency of LNO for SPS activation. Increasing LNO and SPS dosage up to 250 mg/L enhanced the SMX degradation. In contrast, increasing SMX concentration resulted in longer time periods for its degradation. Considering the pH effect, SMX removal was obstructed under basic conditions, while the efficiency was enhanced at near-neutral conditions. The present system’s activity was also tested for piroxicam (PIR) and methylparaben (MeP) degradation, showing promising results. Unfortunately, experiments conducted in real water matrices such as bottled water (BW) and wastewater (WW), showed that SMX removal was limited to less than 25% in both cases. The hindering effects were mainly attributed to bicarbonate ions and organic matter present in aqueous media. The results obtained using suitable radical scavengers revealed the contribution of both hydroxyl and sulfate radicals in degradation reactions. Finally, LNO exhibited good stability under consecutive experimental runs.


Introduction
In recent years, advanced oxidation processes (AOPs), based mainly but not exclusively on the in-situ formation of hydroxyl radicals (•OH), have been considered as a promising family of technologies for the complete degradation of persistent micropollutants in aqueous media [1]. Some characteristic examples include phenolic compounds photocatalytic removal, [2], antibiotics degradation by Fenton and Fenton-like processes [3], and polyfluorinated alkyl substances (PFASs) electrochemical oxidation [4].
A subcategory of AOPs attracting great scientific interest includes the formation of sulfate radicals (SO •− 4 ) as the main oxidative species, known as sulfate radical-based AOPs (SR-AOPs) [5,6]. The SO •− In photocatalysis, there are few reports investigating LNO composite materials towards hydrogen evolution reaction (HER) [31].
In the present study, LNO was synthesized according to a combustion method followed by calcination at 700 • C for 2 h. Its physicochemical characteristics were investigated with the use of X-ray diffraction (XRD), scanning electron microscopy (SEM), and N 2 physical adsorption (BET). The performance of LNO towards SPS activation was examined for SMX degradation in ultrapure water (UPW). SMX, belonging to the group of antibiotics, has been detected in treated effluents at concentrations between ng/L and mg/L [32], thus proving the inadequacy of conventional water treatment methods for its degradation [33]. This phenomenon is closely related not only to adverse ecotoxicological effects but also to the development of antibacterial resistant genes posing serious threats to human health [34].
The effects of several experimental parameters, such as LNO, SMX, and SPS concentration, were investigated. Additional experiments altering initial pH were also carried out. Special emphasis was given to the water matrix effect, presenting results in real and synthetic water matrices. The reusability of LNO, as well as possible reaction mechanism, are also discussed.

Characterization of LNO
The X RD spectrum of LNO is presented in Figure 1A. All diffraction peaks can be indexed to rhombohedral LaNiO 3 (JCPDS No 341028), demonstrating the formation of the perovskite structure. The primary crystallite size of LNO, as calculated by means of Scherrer equation, is 14 nm. Figure 1B and 1C respectively show a representative SEM and TEM image of LNO. Figure 1C reveals a uniform texture of spherical particles. The particle sizes are in nanoscale which is in consistent with the crystallite sizes calculated by Scherrer equation.
Catalysts 2020, 10, x FOR PEER REVIEW 3 of 17 various reactions, as potential substitutes to the expensive and rare noble metals. This includes some widely studied reactions such as CO oxidation, methane oxidation/partial oxidation, methane dry reforming, and steam reforming of ethanol [30]. In photocatalysis, there are few reports investigating LNO composite materials towards hydrogen evolution reaction (HER) [31].
In the present study, LNO was synthesized according to a combustion method followed by calcination at 700 °C for 2 h. Its physicochemical characteristics were investigated with the use of Xray diffraction (XRD), scanning electron microscopy (SEM), and N2 physical adsorption (BET). The performance of LNO towards SPS activation was examined for SMX degradation in ultrapure water (UPW). SMX, belonging to the group of antibiotics, has been detected in treated effluents at concentrations between ng/L and mg/L [32], thus proving the inadequacy of conventional water treatment methods for its degradation [33]. This phenomenon is closely related not only to adverse ecotoxicological effects but also to the development of antibacterial resistant genes posing serious threats to human health [34].
The effects of several experimental parameters, such as LNO, SMX, and SPS concentration, were investigated. Additional experiments altering initial pH were also carried out. Special emphasis was given to the water matrix effect, presenting results in real and synthetic water matrices. The reusability of LNO, as well as possible reaction mechanism, are also discussed.

Characterization of LNO
The X RD spectrum of LNO is presented in Figure 1A. All diffraction peaks can be indexed to rhombohedral LaNiO3 (JCPDS No 341028), demonstrating the formation of the perovskite structure. The primary crystallite size of LNO, as calculated by means of Scherrer equation, is 14 nm. Figure 1B and 1C respectively show a representative SEM and TEM image of LNO. Figure 1C reveals a uniform texture of spherical particles. The particle sizes are in nanoscale which is in consistent with the crystallite sizes calculated by Scherrer equation. 10    Zeta potential measurements ( Figure 2) revealed that the LNO surface is more negatively charged at pH > 10. It should be noted that measurements obtained under acidic conditions cannot be considered trustworthy due to catalyst dissolution [15,19,20]. The specific surface area of LNO was found equal to~5 m 2 /g, characteristic of as-prepared perovskite materials [15,35]. Zeta potential measurements ( Figure 2) revealed that the LNO surface is more negatively charged at pH > 10. It should be noted that measurements obtained under acidic conditions cannot be considered trustworthy due to catalyst dissolution [15,19,20]. The specific surface area of LNO was found equal to ~5 m 2 /g, characteristic of as-prepared perovskite materials [15,35].

SPS Activation
The effect of LNO, SPS, and SMX concentration on SMX degradation was examined and the results are presented in Figure 3. Figure 3A shows that 65% SMX removal was achieved after 25 min using only 25 mg/L LNO, proving its ability to effectively activate SPS for SMX degradation. An increase in catalyst dosage enhanced SMX degradation ( Figure 3A); complete SMX removal occurs after 25, 15, and 10 min of reaction using 50, 100, and 250 mg/L, respectively, in the presence of 100 mg/L SPS. The plot of -Ln(C/C0) versus time in all tests resulted in a straight line ( Figure S1) with the linear regression coefficient, R 2 , greater than 0.90, indicating that these experimental data fit well to the pseudo-first-order kinetic model (Equation (1)). The computed apparent rate constants decreased as follows: 0.2267 min −1 (CLNO 250 mg/L) > 0.1319 min −1 (CLNO= 100 mg/L) > 0.1247 min −1 (CLNO= 50 mg/L) > 0.0400 min −1 (CLNO= 25 mg/L). This observation is in accordance with other studies, since an increase in catalyst dosage produces more active surface sites that are available for SPS activation and SMX adsorption, thus favoring pollutant degradation [15,19,20,36,37]. In the absence of SPS, the adsorption capacity of SMX onto the catalyst surface was negligible (ca. 10%). Additionally, the presence of 100 mg/L SPS alone (without catalyst) showed that SPS oxidative ability was insufficient for SMX degradation ( Figure 3A).

SPS Activation
The effect of LNO, SPS, and SMX concentration on SMX degradation was examined and the results are presented in Figure 3. Figure 3A shows that 65% SMX removal was achieved after 25 min using only 25 mg/L LNO, proving its ability to effectively activate SPS for SMX degradation. An increase in catalyst dosage enhanced SMX degradation ( Figure 3A); complete SMX removal occurs after 25, 15, and 10 min of reaction using 50, 100, and 250 mg/L, respectively, in the presence of 100 mg/L SPS. The plot of -Ln(C/C 0 ) versus time in all tests resulted in a straight line ( Figure S1) with the linear regression coefficient, R 2 , greater than 0.90, indicating that these experimental data fit well to the pseudo-first-order kinetic model (Equation (1)). The computed apparent rate constants decreased as follows: 0.2267 min −1 (C LNO 250 mg/L) > 0.1319 min −1 (C LNO = 100 mg/L) > 0.1247 min −1 (C LNO = 50 mg/L) > 0.0400 min −1 (C LNO = 25 mg/L). This observation is in accordance with other studies, since an increase in catalyst dosage produces more active surface sites that are available for SPS activation and SMX adsorption, thus favoring pollutant degradation [15,19,20,36,37]. In the absence of SPS, the adsorption capacity of SMX onto the catalyst surface was negligible (ca. 10%). Additionally, the presence of 100 mg/L SPS alone (without catalyst) showed that SPS oxidative ability was insufficient for SMX degradation ( Figure 3A).   Figure 3B depicts the effect of the initial SPS concentration on SMX degradation. The results demonstrate that an increase in SPS concentration from 50 to 250 mg/L leads to a progressive enhancement in SMX removal, probably because more reactive species can participate in SMX degradation. For example, at 6 min of treatment time, 46%, 72%, and 100% SMX degradation was observed for 50, 100, and 250 mg/L initial SPS concentration, respectively. The computed rate constants were 0.1343 min −1 at 50 mg/L, 0.2267 min −1 at 100 mg/L and 0.4032 min −1 at 250 mg/L.   Figure 3B depicts the effect of the initial SPS concentration on SMX degradation. The results demonstrate that an increase in SPS concentration from 50 to 250 mg/L leads to a progressive enhancement in SMX removal, probably because more reactive species can participate in SMX degradation. For example, at 6 min of treatment time, 46%, 72%, and 100% SMX degradation was observed for 50, 100, and 250 mg/L initial SPS concentration, respectively. The computed rate constants were 0.1343 min −1 at 50 mg/L, 0.2267 min −1 at 100 mg/L and 0.4032 min −1 at 250 mg/L. Considering both the economic aspects and complete SMX removal, a LNO dosage of 250 mg/L and 100 mg/L SPS were applied in the following experiments.
The performance of the LNO/SPS system as a function of SMX concentration is given in Figure 3C. The removal of SMX distinctly decelerates as the pollutant concentration increases (within a range of 250 to 1000 mg/L). Furthermore, the corresponding rate constants were 0.3408 min −1 , 0.2267 min −1 and 0.1566 min −1 at an initial SMX concentration of 250, 500, and 1000 mg/L, respectively. The increased competition between the intermediates and SMX for reactive species and catalyst surface, at a higher initial concentration of SMX, decreased the apparent k values. This observation is a well-known phenomenon occurring in different heterogeneous AOPs. Thus, comparable results have been reported in several previous studies [15,[38][39][40], indicating that the oxidation rate could be significantly affected by a change in the initial concentration of pollutants.
In order to obtain an insight into the efficacy of the present system towards SMX degradation, Table 1 summarizes the main characteristics of other heterogeneously activated persulfate systems employed for SMX degradation. Magioglou et al. [6], who studied the degradation of 0.5 mg/L SMX by biochar-activated persulfate, reported significantly longer time periods for complete SMX degradation than those reported in Figure 3. Alexopoulou et al. [14] used Cu 3 P as a heterogeneous persulfate activator. They found that fast degradation of SMX occurred in UPW with the rate depending on parameters such as SMX, SPS and catalyst concentrations. Their results are of the same order of magnitude as the present study. Moreover, Gkika et al. [15] studied the heterogeneous activation of SPS by La 0.8 Sr 0.2 CoO 3-δ , reporting k values between 0.022 and 0.061 min −1 (depending on the catalyst dosage) for the degradation of 0.5 mg/L SMX with 500 mg/L SPS. In another study, Fe(II)-based metal-organic frameworks (MOFs) were synthesized for the degradation of SMX [41]. The results showed that 500 mg/L Fe(II)MOFs could effectively activate 500 mg/L persulfate leading to~97% removal of 10 mg/L SMX in 180 min, with a k value equal to 0.056 min −1 , lower than the present study. Wu et al. studied the heterogeneous activation of persulfate using Fe 3 O 4 nanoparticles [42], showing promising results in the application of Fe 3 O 4 in water decontamination systems. The effect of several experimental parameters, as well as the water matrix effect were thoroughly analyzed in their study [42]. They observed that the rate constant increased from 8.07 ± 0.58 × 10 −5 s −1 to 1.11 ± 0.10 × 10 −4 s −1 , increasing Fe 3 O 4 concentration from 0.3 to 1.0 g/L, while further increase to 2.0 g/L inhibited SMX degradation [42]. Moreover, SMX degradation efficiency decreased with increasing pH value, while chloride ions hindered SMX degradation at pH 4.0, but had no impact on SMX degradation at pH 10.5 [42]. Magnetic graphitized biochar (GMBC) synthesized from pine wood-derived biochar has been also proposed as heterogeneous catalyst for persulfate activation showing good performance for complete SMX removal from water [43]. In this case, almost 100% degradation of 10 mg/L SMX was recorded within 60 min, with the corresponding k value equal to 0.179 min −1 [43].  [44]. Specifically, Oh et al. found that the mixed metal system had greater stability and activity than single metal systems (Co 3 O 4 and Fe 2 O 3 ) with 5 mg/L SMX being degraded in 30 min using 60 mg/L sodium persulfate [44]. High efficiency towards persulfate activation was also achieved using CoFe 2 O 4 mixed oxide immobilized on Al 2 O 3 ceramic membrane [45]. In this case, 10 mg/L SMX was degraded after 60 min [44]. In an attempt to enhance the activity of carbon materials towards persulfate activation, Yanan et al. synthesized carbon nanotubes (NCNTs) encapsulating Zn and Fe hybrid metal particles (FeZn@NC), and tested them for SMX degradation [46]. They observed that an almost complete removal of 5 mg/L SMX took place after 90 min, with the degradation rate being highly dependent on the mass ratio of zinc to iron in the FeZn@NC nanoparticles [46]. The maximum k value was 0.082 min −1 , i.e., an order of magnitude lower than that reported in Figure 3. The authors paid special attention to the reactive oxygen species produced through persulfate activation, suggesting that 1 O 2 was the main species in the FeZn@NC/PS system for SMX degradation [46]. Another significant contribution to the enlightenment of persulfate activation mechanism was made by Wang et al. [47] studying the biochar-induced Fe(III) reduction for SMX degradation. They pointed out that persistent free radicals contained in biochars and other pyrogenic carbonaceous materials play a key role in these degradation processes. A value of 0.019 min −1 was recorded for the degradation of 10 mg/L SMX with 500 mg/L biochar and 500 mg/L SPS.
Comparing the results of the present work to previous studies of our group [6,14,15], it is evident that LNO is more effective than other materials for SMX degradation. Moreover, comparing the efficiency of the most studied Co, Fe, Cu-based oxides, as well as carbon-based materials with the marginally studied Ni-based perovskite oxide proposed by the present study, it is observed that Ni-based materials show great potential towards persulfate activation, thus deserving further investigation.
In other studies on the heterogeneous activation of persulfate for micropollutant degradation, the time period for complete degradation varied depending on SPS or catalyst dosage. Characteristic examples are summarized in Table 2. MnO x loaded on ordered mesoporous carbon composites (MnO x @OMC) was found effective for phenol degradation by persulfate activation [10]. Lei et al. [13] stated that almost 100% mineralization of 0.1 mM phenol could be achieved in 120 min using 0.3 g/L CuO-Fe 3 O 4 and 5.0 mM SPS at pH 11.0, while Miao et al. [20] demonstrated that SrCo 1−x Ti x O 3−δ exhibited a fast catalytic activity for phenol oxidation in the presence of peroxymonosulfate in a wide pH range. Jiang et al. [22] showed that the addition of reduced graphene oxide (RGO) effectively prevented agglomeration of Ni 2 SnO 4 nanoparticles and improved its catalytic performance towards SPS activation to degrade bisphenol A.

pH Effect
The effect of medium pH is a well-known factor for the degradation of organic pollutants. Therefore, the impact of two initial pH values (6.0, 9.0) on SMX degradation was studied in the LNO/SPS system, and the results are depicted in Figure 4. It should be noted that the buffer solution was not used to constrain any potential interferences of buffering agents; however, pH values did not change throughout the course of the reaction. As seen, the LNO/SPS system is effective at near-neutral pH, leading to complete removal of SMX within 10 min. An increase of the pH value to 9 led to a total inhibition of SMX removal, implying that basic conditions change the surface of perovskite oxide and the target compound's physicochemical properties. Indeed, SMX has two pK a values (pK a,1 = 1.85 ± 0.30 and pK a,2 = 5.60 ± 0.04) [48], suggesting that it is positively charged at pH < pK a,1 , not charged at pK a,1 < pH < pK a,2 and negatively charged at pH > pK a,2 . Moreover and according to zeta potential measurements (Figure 2), LNO is negatively charged at basic pH, resulting in repulsive forces between the catalyst surface and SMX. An additional experiment was performed at acidic conditions (pH = 3) resulting in severe LNO dissolution (data not shown). Similar results regarding the instability of perovskite materials at acidic environments have been reported elsewhere [15,19,20].

Αpplicability of LaNiO3/SPS System for the Degradation of Various Micro-Pollutants
The LNO/SPS system performance in eliminating other organic pollutants commonly detected in the aquatic environment was additionally evaluated. Therefore, experiments were carried out to degrade 500 μg/L of piroxicam (PIR) or methylparaben (MeP). These compounds have been recently classified as emerging pollutants by the US Environmental Protection Agency [49,50]. However, they are still widely used in several applications; MeP is employed as a preservative in foodstuff, cosmetics, and personal care products [49], while PIR is used mainly as a painkiller in order to decrease fever and to prevent blood clots [50]. As seen in Figure 5, PIR was rapidly degraded in less than 2 min, while MeP was eliminated within 10 min. The satisfactory degradation of several other micropollutants suggests a robust catalytic performance of the applied system.

Applicability of LaNiO 3 /SPS System for the Degradation of Various Micro-Pollutants
The LNO/SPS system performance in eliminating other organic pollutants commonly detected in the aquatic environment was additionally evaluated. Therefore, experiments were carried out to degrade 500 µg/L of piroxicam (PIR) or methylparaben (MeP). These compounds have been recently classified as emerging pollutants by the US Environmental Protection Agency [49,50]. However, they are still widely used in several applications; MeP is employed as a preservative in foodstuff, cosmetics, and personal care products [49], while PIR is used mainly as a painkiller in order to decrease fever and to prevent blood clots [50]. As seen in Figure 5, PIR was rapidly degraded in less than 2 min, while MeP was eliminated within 10 min. The satisfactory degradation of several other micropollutants suggests a robust catalytic performance of the applied system.

Αpplicability of LaNiO3/SPS System for the Degradation of Various Micro-Pollutants
The LNO/SPS system performance in eliminating other organic pollutants commonly detected in the aquatic environment was additionally evaluated. Therefore, experiments were carried out to degrade 500 μg/L of piroxicam (PIR) or methylparaben (MeP). These compounds have been recently classified as emerging pollutants by the US Environmental Protection Agency [49,50]. However, they are still widely used in several applications; MeP is employed as a preservative in foodstuff, cosmetics, and personal care products [49], while PIR is used mainly as a painkiller in order to decrease fever and to prevent blood clots [50]. As seen in Figure 5, PIR was rapidly degraded in less than 2 min, while MeP was eliminated within 10 min. The satisfactory degradation of several other micropollutants suggests a robust catalytic performance of the applied system.

Water Matrix Effect
To further assess the applicability of the LNO/SPS system, its activity in different water matrices was investigated. A set of experiments were carried out in wastewater (WW), wastewater diluted with an equal volume of UPW (50% WW), and bottled water (BW), and the results are depicted in Figure 6. It is observed that the presence of various organic and inorganic substances in actual water matrices (see Table S1) dramatically affects degradation leading to less than 25% SMX removal after 15 min. Apart from the complexity of these matrices, their slightly alkaline nature could significantly affect SMX removal, as has already been discussed in Section 2.3 (see Figure 4). For this reason, the experiment in WW was repeated after having adjusted its initial pH value of 8 (Table S1) to 6. As seen in Figure 6, no evident change was recorded. A similar adverse effect of real water matrices, e.g., wastewater, tap water, and river water, on the degradation rate of various pollutants using perovskite materials for SPS activation has also been reported elsewhere [4,15]. 15 min. Apart from the complexity of these matrices, their slightly alkaline nature could significantly affect SMX removal, as has already been discussed in Section 2.3 (see Figure 4). For this reason, the experiment in WW was repeated after having adjusted its initial pH value of 8 (Table S1) to 6. As seen in Figure 6, no evident change was recorded. A similar adverse effect of real water matrices, e.g., wastewater, tap water, and river water, on the degradation rate of various pollutants using perovskite materials for SPS activation has also been reported elsewhere [4,15]. The effect of natural organic matter and inorganic anions such as HCO and Cl on SMX degradation in the LNO/SPS system was investigated (Figure 7) [51]. As seen, the addition of HCO in the system had a detrimental influence on SMX degradation at both concentration levels tested (i.e., 125 and 250 mg/L which are typical of those found in BW and WW-see Table S1); on the other hand, the addition of 250 mg/L Cl did not practically affect SMX degradation. These results could be attributed to the possible formation of radicals with lower redox potentials. For instance, HCO can rapidly be transformed to CO , both of which can eventually react with sulfate and hydroxyl radicals, forming carbonate radicals (HCO • and CO • ) of lower oxidation potential [15,36,52]. Sulfate and hydroxyl radicals can also react with Cl forming Cl • which can further react with chloride forming Cl • . Consequently, these radicals Cl • , Cl • are likely to contribute to the degradation of the target compound [15,52,53]. According to previous studies concerning the use of perovskite materials for SPS activation, diverse results have been reported about the effect of common inorganic anions, e.g., HCO and Cl , on pollutant degradation. Lin et al. [53] reported that the presence of NaCl from 250 to 1000 mg/L noticeably decreased the degradation kinetics of dyes in the presence of LaMO3. Comparable results were obtained using CaMnO , La . Sr . CoO δ and LaCoO nanocomposites as persulfate activators for the removal of phenol and pharmaceutical compounds [15,36,51]. A 4-fold decrease was observed in the rate constant of ciprofloxacin degradation when 20 mmol/L Cl were added in water in the presence of Sr-doped BiFeO3 [54]. The detrimental effect of bicarbonate on the degradation of SMX and phenol with persulfate was reported when SPS was The effect of natural organic matter and inorganic anions such as HCO − 3 and Cl − on SMX degradation in the LNO/SPS system was investigated (Figure 7) [51]. As seen, the addition of HCO − 3 in the system had a detrimental influence on SMX degradation at both concentration levels tested (i.e., 125 and 250 mg/L which are typical of those found in BW and WW-see Table S1); on the other hand, the addition of 250 mg/L Cl − did not practically affect SMX degradation. These results could be attributed to the possible formation of radicals with lower redox potentials. For instance, HCO − 3 can rapidly be transformed to CO 3 2− , both of which can eventually react with sulfate and hydroxyl radicals, forming carbonate radicals (HCO 3 • and CO 3 •− ) of lower oxidation potential [15,36,52]. Sulfate and hydroxyl radicals can also react with Cl − forming Cl • which can further react with chloride forming Cl 2 •− . Consequently, these radicals Cl • , Cl 2 •− are likely to contribute to the degradation of the target compound [15,52,53]. According to previous studies concerning the use of perovskite materials for SPS activation, diverse results have been reported about the effect of common inorganic anions, e.g., HCO − 3 and Cl − , on pollutant degradation. Lin et al. [53] reported that the presence of NaCl from 250 to 1000 mg/L noticeably decreased the degradation kinetics of dyes in the presence of LaMO 3 . Comparable results were obtained using CaMnO 3 , La 0.8 Sr 0.2 CoO 3−δ and LaCoO 3 nanocomposites as persulfate activators for the removal of phenol and pharmaceutical compounds [15,36,51]. A 4-fold decrease was observed in the rate constant of ciprofloxacin degradation when 20 mmol/L Cl − were added in water in the presence of Sr-doped BiFeO 3 [54]. The detrimental effect of bicarbonate on the degradation of SMX and phenol with persulfate was reported when SPS was activated with La 0.8 Sr 0.2 CoO 3−δ and CaMnO 3 , respectively [15,36]. Guo et al. [51] observed that the presence of bicarbonate could significantly retard carbamazepine oxidation in the LaCoO 3 /SPS system. Wang et al. [54] also reported a negative effect on ciprofloxacin degradation in the BiFeO 3 /SPS system with the addition of HCO − 3 .
Catalysts 2020, 10, x FOR PEER REVIEW 11 of 17 activated with La . Sr . CoO δ and CaMnO , respectively [15,36]. Guo et al. [51] observed that the presence of bicarbonate could significantly retard carbamazepine oxidation in the LaCoO3/SPS system. Wang et al. [54] also reported a negative effect on ciprofloxacin degradation in the BiFeO3/SPS system with the addition of HCO . As can also be seen in Figure 7, the presence of humic acid (HA) at two concentrations (i.e., 10 and 20 mg/L representative of the level of total organic carbon found in WW-see Table S1) completely suppressed the SMX degradation rate. This negative result may be due to either the scavenging of reactive oxygen species by HA [15,55,56] or competition for the catalyst surface. This is consistent with the reported adverse effect of HA on SMX degradation using lanthanum strontium cobaltite as catalyst for SPS activation [15]. In contrast, Guo and et al. [51] found that the presence of HA did not affect carbamazepine degradation using lanthanum cobaltite perovskite as the activator.

Reactive Species and Catalyst Reusability
Trapping experiments were conducted to investigate the dominant reactive species and their contributions to the present system. For this reason, tert-butyl alcohol (TBA) and methanol (MeOH) were selected as OH • and SO • scavengers. TBA is an effective OH • quenching agent (k(t-ButOH, HO•)= 6 × 10 8 M −1 s −1 ), while MeOH can react with both OH • and SO • [k(MeOH,HO•)= 9.7 × 10 8 M −1 s −1 k(MeOH,SO4•ˉ)= 1.1 × 10 7 M −1 s −1 ] [57,58]. As seen in Figure 8, a high dosage of each scavenger relative to the SPS concentration (TBA:SPS = MeOH:SPS = 100:1) resulted in different inhibitory effects on 500 μg/L SMX degradation. More specifically, SMX degradation was substantially decreased to nearly 50% after 10 min, in the presence of either TBA or MeOH. However, complete SMX degradation was achieved in the case of TBA at the end of the 25-min long experiment, while only 70% removal was recorded in the case of MeOH. This indicates that both OH • and SO • were generated in the LNO/SPS system, with SO • being the primary radical species. Numerous studies on persulfate activation with perovskite oxides also reported that oxidation reactions were mainly driven by SO • and, to a lesser extent, by OH • [15,17,19,21,36,37,51,53,59,60]. As can also be seen in Figure 7, the presence of humic acid (HA) at two concentrations (i.e., 10 and 20 mg/L representative of the level of total organic carbon found in WW-see Table S1) completely suppressed the SMX degradation rate. This negative result may be due to either the scavenging of reactive oxygen species by HA [15,55,56] or competition for the catalyst surface. This is consistent with the reported adverse effect of HA on SMX degradation using lanthanum strontium cobaltite as catalyst for SPS activation [15]. In contrast, Guo and et al. [51] found that the presence of HA did not affect carbamazepine degradation using lanthanum cobaltite perovskite as the activator.
Catalysts 2020, 10, x FOR PEER REVIEW 12 of 17 In the final set of experiments, the recyclability and reusability of LNO was tested in four consecutive cycles. As seen in Figure 9, LNO retained its activity leading to complete SMX degradation within the same period of time. These results indicate that LaNiO3 can be a promising heterogeneous SPS activator for the treatment of SMX-containing wastewater in real life applications.  In the final set of experiments, the recyclability and reusability of LNO was tested in four consecutive cycles. As seen in Figure 9, LNO retained its activity leading to complete SMX degradation within the same period of time. These results indicate that LaNiO 3 can be a promising heterogeneous SPS activator for the treatment of SMX-containing wastewater in real life applications. In the final set of experiments, the recyclability and reusability of LNO was tested in four consecutive cycles. As seen in Figure 9, LNO retained its activity leading to complete SMX degradation within the same period of time. These results indicate that LaNiO3 can be a promising heterogeneous SPS activator for the treatment of SMX-containing wastewater in real life applications.

Chemicals and Water Matrices
Wastewater samples were collected from the wastewater treatment plant located in the University of Patras campus, Greece. Commercially bottled water was used for experiments in the drinking water matrix. Properties of the water matrices are given in Table S1.

Catalyst Synthesis and Characterization
For LNO preparation, appropriate amounts of La(NO 3 ) 3 ·6H 2 O, nickel (II) nitrate and Ni(NO 3 ) 2 ·6H 2 O were dissolved in UPW. NH 4 NO 3 (NH 4 NO 3 to metal ions molar ratio 1:1) and citric acid C 6 H 8 O 7 (C 6 H 8 O 7 to metal ions molar ratio 2:1) was then added. The pH of the solution was adjusted to 9 with the addition of NH 3 . The solution was heated at 200 • C, and after water evaporation, heating temperature was raised to 400 • C with the help of a heating gun. Finally, the as-obtained powder was calcined at 700 • C for 5 h. Details can be found elsewhere [35].
A Brucker D8 Advance (Cu Kα) device (Billerica, MA, USA) was used for XRD analysis. Crystallographic phases naming was based on JCPDS cards. Nanocrystals primary crystallite size was calculated according to Debye-Scherrer's equation [53]. Specific surface area of LNO was determined with the Brunaueur-Emmett-Teller (BET) method (N 2 physisorption at the temperature of liquid nitrogen (77 K)) using a Micromeritics (Gemini III 2375, Norcross, GA, USA). Zeta potential measurements were carried out via laser doppler micro-electrophoresis with a use of a Malvern Zetasizer (Nano-ZS, Malvern, UK). SEM images were obtained with the use of a JEOL 6300 (JEOL USA, Inc.) microscope equipped with an energy dispersive spectrometer (EDS) allowing element distribution of LNO. TEM images were recorded by means of an Erlangshen CCD Camera (Gatan Model 782 ES500 W, CA, USA). The specimens were prepared by dispersion in water and spread onto a carbon-coated copper grid (200 mesh). Details about physicochemical characterization can be found elsewhere [15].

Catalytic Degradation Experiments
Typically, 15 mg of catalyst were added into a 60 mL solution containing 500 µg/L SMX in a 100 mL beaker open to the atmosphere and the mixture was stirred for 15 min to achieve adsorption-desorption equilibrium. Then, 100 mg/L SPS was added into the suspension to initiate the reaction under continuous magnetic stirring. Experiments were, in most cases, conducted in UPW and pH was adjusted with an appropriate aliquot of NaOH or H 2 SO 4 when needed. At pre-set periods, 1.2 mL of aliquot was withdrawn, and immediately added 300 µL MeOH to quench the reaction. Afterwards, samples were filtered through a Whatman Uniflo Syringe Filter with a polypropylene overmold housing (0.22 µm, Whatman, UK), and the filtrate was analyzed using liquid chromatography. In most cases, experiments were repeated twice, and mean values are quoted as results. It is worth noting that the variance between measurements never exceeded 5%.
For the recycle/reusability experiments, the catalyst was collected at the end of each cycle, filtered, and washed with deionized water several times. Then, it was dried at 60 • C overnight and employed into the next cycle.
The kinetics of SMX degradation were fitted to a pseudo-first-order model and the apparent rate constant (k) was calculated as follows (Equation (1)): −Ln(C/C 0 ) = kt (1) where C 0 and C t are the SMX concentrations at the beginning and at a specific reaction time (t), respectively.

Chemical Analysis
The residual SMX concentration was determined by high performance liquid chromatography (HPLC, Waters Alliance 2695, MA, USA) equipped with a photo diode array detector (PDA, Waters 2996). SMX was detected at 270 nm using a mobile phase consisting of ACN:0.1% H 3 PO 4 UPW (15:85, v/v) mixture at a flow rate of 200 µL/min on a Kinetex C18 column (50 mm × 2.1 mm, i.d. 2.6 µm).

Conclusions
The present study proposed a nickel-containing perovskite material as SPS activator. The major findings are as follows: • LNO possesses high efficiency as a persulfate activator and can be adopted in water decontamination processes.

•
The SMX degradation rate can be effectively tuned by adjusting SMX, SPS, and LNO loadings.

•
The efficiency is favored in almost neutral conditions (pH 6), thus making it an environmentally friendly process.

•
Bicarbonate ions and organic matter present in aqueous media suppress the activity of LNO/SPS system. • LNO can be effectively used in repeated experimental runs, showing excellent stability thus proving the true catalytic nature of the proposed system.