Abstract

Degradation of naproxen (NAP) by persulfate (PS) activated with zero-valent iron (ZVI) was investigated in our study. The NAP in aqueous solution was degraded effectively by the ZVI/PS system and the degradation exhibited a pseudo-first-order kinetics pattern. Both sulfate radical (SO4•−) and hydroxyl radical (HO) participate in the NAP degradation. The second-order rate constants for NAP reacting with SO4•− and HO were (5.64 ± 0.73) × 109 M1 s1 and (9.05 ± 0.51) × 109 M1 s1, respectively. Influence of key parameters (initial pH, PS dosage, ZVI dosage, and NAP dosage) on NAP degradation were evaluated systematically. Based on the detected intermediates, the pathways of NAP degradation in ZVI/PS system was proposed. It was found that the presence of ammonia accelerated the corrosion of ZVI and thus promoted the release of Fe2+, which induced the increased generation of sulfate radicals from PS and promoted the degradation of NAP. Compared to its counterpart without ammonia, the degradation rates of NAP by ZVI/PS were increased to 3.6–17.5 folds and 1.2–2.2 folds under pH 7 and pH 9, respectively.

HIGHLIGHTS

  • Naproxen (NAP) could be efficiently degraded by persulfate activated with zero-valent iron.

  • The rate constants of NAP reacting with SO4•− and HO were determined.

  • The possible oxidation pathways was elaborated for NAP degradation.

  • Ammonia enhanced the degradation of NAP by accelerating the release of dissolved iron.

Graphical Abstract

Graphical Abstract
Graphical Abstract

INTRODUCTION

Naproxen (NAP), a typical nonsteroidal anti-inflammatory drug (NSAIDs), was frequently detected in wastewater treatment plant effluents with concentration ranging from 25 ng L1 to 33.9 μg L1 (Kanakaraju et al. 2015). The toxicological study indicated that adverse impact on Vibrio fischeri was caused by NAP with an EC50 of 21.2 μg L1 (Luo et al. 2018a). Moreover, long-term ingestion of NAP may increase the risk of heart attack in humans (Chi et al. 2019a). Unfortunately, it is still a challenge for conventional water treatment to eliminate NAP efficiently at a short treatment period (Xu et al. 2019), and thus, it is urgent to exploit appropriate approaches to degrade these contaminants in the aqueous environment.

In the past decade, advanced oxidation processes (AOPs) based on the sulfate radical (SO4•−, E0 = 2.5–3.1 V) were found to be excellent for degradation of refractory organic contaminants such as ranitidine, sulfamethoxazole, ibuprofen, ketoprofen, chloramphenicol, theophylline, etc. (Ghauch et al. 2012, 2013, 2017; Ayoub & Ghauch 2014; Naim & Ghauch 2016; Amasha et al. 2018; Hakim et al. 2019, 2020). SO4•− could be generated from persulfate (PS) or peroxymonosulfate (PMS) by numerous activation methods, such as heat, UV, ultrasound, transition metal ion, and metal oxides (Xiao et al. 2014; Ghauch et al. 2017; Lai et al. 2018, 2020; Hong et al. 2019; Ji et al. 2019; Li et al. 2019a, 2019b; Yan et al. 2019). Among these, transition metal ion activation has the superiorities of high efficiency and cost effectiveness (Rao et al. 2014; Chen et al. 2018). Iron is of special interest due to its abundance and environmentally friendly properties (Zhao et al. 2016). SO4•− can be efficiently generated from the reaction of Fe2+ with PS.

Unlike homogeneous Fe2+ activation, zero-valent iron (ZVI) could react as an alternative source of Fe2+ to gradually release Fe2+ (Equations (1)–(3)) (Girit et al. 2015; Zhao et al. 2016), causing minimal quenching of SO4•− induced by excessive Fe2+ (Equation (4)) (Ghauch et al. 2013; Wei et al. 2016). ZVI has been used as stand-alone material to eliminate pharmaceuticals from water such as diclofenac (Ghauch et al. 2010). However, the excess of iron sludge formation inspired researchers to use it as a sacrificial catalyst in AOPs as, for example, the degradation of carabamazepine in ZVI/H2O2 systems, sulfamethoxazole in bimetalllics and trimetallics iron–PS based systems, or ranitidine using iron scrap multi-metallic systems (Ghauch et al. 2011; Ayoub & Ghauch 2014; Naim & Ghauch 2016). Moreover, Fe3+ could be recycled to produce Fe2+ on the surface of ZVI through Equation (5) (Hussain et al. 2012). Therefore, utilization efficiency of PS is enormously enhanced especially in oxic solutions where total organic carbon (TOC) removal was improved compared to anoxic conditions as reported earlier (Ghauch et al. 2013; Ayoub & Ghauch 2014). Recently, there has been interest in using the combination of ZVI/PS for refractory organics decontamination. Wei et al. (2016) examined the roles of key parameters in the ZVI/PS process. A study on fenitrothion (FNT) degradation in the ZVI/PS process revealed that the addition of ZVI increased the FNT degradation by PS (Liu et al. 2019). However, there were several defects in the reaction between Fe2+ and PS, particularly in the limits of a strict pH range as in the traditional Fenton-like system (Wu et al. 2015), which results in reduced removal efficiency of target pollutants. 
formula
(1)
 
formula
(2)
 
formula
(3)
 
formula
(4)
 
formula
(5)

As far as we know, NAP has been investigated in thermally activated PS systems (Ghauch et al. 2015); however, there is no information about NAP degradation in the ZVI/PS system. The major purpose of this work were (1) to elucidate the superiority and kinetics of ZVI/PS system on NAP degradation; (2) to make clear the primary reactive species contribution to NAP degradation and to determine the reaction rates between NAP and reactive species; (3) to explore the impacts of pivotal parameters, including initial pH, PS dosage, ZVI dosage, and NAP concentration on NAP degradation; (4) to propose the NAP degradation pathways by the detected intermediates analysis; and (5) to show an unexpected finding that better removal of NAP were achieved with adding ammonia in ZVI/PS system under neutral and basic conditions.

MATERIALS AND Methods

Materials

All chemicals were of reagent grade without any further purification. The solutions were prepared using deionized (DI) water produced from a Milli-Q Academic water purification system and stored at 4 °C before being used. Except for NAP, all the stock solutions were prepared daily. NAP (purity >99.0%), 5,5-dimethyl-1-pyrroline N-oxide (DMPO) and ZVI (purity >99%) were purchased from Aladdin Chemistry Co. Ltd. (Shanghai, China). Approximately 50% of these ZVI particles had a diameter size less than 36.54 μm (shown in Fig. S1). Methanol (high performance liquid chromatography (HPLC) grade) and all other chemicals used were purchased from Sinopharm Chemical Reagent Co. Ltd. (Beijing, China).

Experimental procedure

Experiments were carried out in a 1-L glass beaker to investigate the degradation efficiency of NAP in the ZVI/PS process. For each run, 500 mL of prepared NAP (25 μM) solution was poured into the 1-L glass beakers and the initial solution pH (3.0–11.0) was adjusted using 0.1 M H2SO4 and/or NaOH. Experiments were initiated by adding the appropriate amount of ZVI (0.25–1.50 mM) and PS (0.1–0.5 mM) to the aforementioned solution. The temperature of experiments was maintained at 25 ± 1 °C. The solution was mixed uniformly by a mechanical stirrer, the rotate speed of agitation bar was kept at 300 rpm to ensure a complete mixing state. Samples were withdrawn at predetermined time intervals with 5.0 mL syringes and quickly filtered with 0.22 μm membrane. At the same time, 800 μL filtered sample was collected and 200 μL ethanol (EtOH) was injected into each sample to terminate the reaction.

The scavenging experiments with tertbutyl alcohol (TBA) and EtOH were conducted by adding desired alcohols into the reaction solution before the addition of PS. To disclose the potential effect of ammonia, NAP solution was mixed with NH4Cl (0–10 mM) first, and then the pH was adjusted to 7.0 and 9.0 using the procedures described above. All tests were repeated at least twice.

Analytical methods

The concentrations of benzoic acid (BA), nitrobenzene (NB), and NAP were measured with HPLC (Shimadzu LC-2010AHT, Kyoto, Japan), and the detailed parameters are shown in Text S1. PS was quantified spectrophotometrically (Liang et al. 2008). The concentration of Fe2+ was quantified at 510 nm according to the 1,10-phenanthroline method and total iron was determined by atomic absorption spectroscopy (Shimadzu AA-7000, Kyoto, Japan). TOC was measured with a TOC analyzer (liqui TOC II, Elementar, Germany). The pH value was measured with a pH meter (Mettler Toledo Five Easy Plus pH Meter, Shanghai, China). The electron spin resonance (ESR) was carried out using a JES-FA 200 ESR spectrometer. The ESR experiments were performed under the following conditions: a center field of 335 mT, a sweep width of 10 mT, a microwave frequency of 9,425.572 MHz, a microwave attenuator of 30 dB, a microwave power of 0.998 mW, and a sweep time of 60 s. The intermediates of NAP were detected using ultra-performance liquid chromatography (UPLC)–quadrupole time-of-flight mass spectrometry (QTOF–MS)/mass spectrometry (MS) (AB SCIEX ExionLC AD UPLC coupled with AB SCIEX 5600 + Q-TOF, USA), with details shown in Text S2.

To determine the utilization efficiency of PS, the reaction stoichiometric efficiency (RSE) was used in previous articles (Ghauch et al. 2015, 2017; Amasha et al. 2018), which is defined as the ratio between the number of NAP degraded versus the number of PS consumed.

RESULTS AND DISCUSSION

Degradation of NAP in different systems

The results of NAP degradation in different systems can be seen in Figure 1. Neither ZVI nor PS alone had a notable effect on the NAP degradation with the extremely low NAP removal efficiency (<2%). In contrast, the ZVI/PS process showed high levels of removal efficiency (93.5%) for NAP. Similar results have also been observed in the ZVI/PS system for degradation of sulfamethoxazole (Ghauch et al. 2013; Ayoub & Ghauch 2014) and bentazon (Wei et al. 2016). As NAP was difficult to be degraded by only PS oxidation and ZVI, it could be concluded that SO4•− and/or HO were the main reactive species in NAP degradation during the ZVI/PS process (Equations (6)–(9)) (Oh et al. 2010; Wu et al. 2014). However, NAP degradation in PS thermally activated system showed that SO4•− were the dominant species responsible for NAP degradation. The structure of NAP has a naphtalenic ring favoring more electron transfer than hydrogen abstraction (Ghauch et al. 2015). The average % RSE calculated was close to 44.5%, which is smaller than that obtained in thermally activated PS systems (Ghauch et al. 2015); however, it exceeded the RSE values in chemically activated systems (Amasha et al. 2018). The Fe2+/PS system was also set up in this study. For comparison, the additional amount of Fe2+ was the total amount of Fe dissolved in the ZVI/PS system: 1.07 mg/L. As shown in Figure 1, NAP was promptly degraded in the first 5 min, while the NAP removal efficiency remained unchanged afterwards. The final degradation efficiency of NAP after 30 min treatment was 59.9%. The reason for this phenomenon may be that the initially added Fe2+ reacts rapidly with PS to generate a large amount of sulfate radicals, while radicals produced in the system would be quenched by excessive Fe2+ (Equation (4)) (Ghauch et al. 2013; Naim & Ghauch 2016), causing a lower final removal efficiency of NAP. Thus, the ZVI/PS system exhibited the best performance on NAP degradation. 
formula
(6)
 
formula
(7)
 
formula
(8)
 
formula
(9)
Figure 1

NAP degradation during the different systems. Experimental conditions: [NAP]0 = 25 μM, [PS]0 = 0.25 mM, [ZVI]0 = 0.5 mM, [Fe2+]0 = 1.07 mg/L, T = 298.15 K, initial pH = 5.0.

Figure 1

NAP degradation during the different systems. Experimental conditions: [NAP]0 = 25 μM, [PS]0 = 0.25 mM, [ZVI]0 = 0.5 mM, [Fe2+]0 = 1.07 mg/L, T = 298.15 K, initial pH = 5.0.

As the plot of ln(Ct/C0) versus time shows a linear relationship (inset of Figure 1), the NAP degradation well follows a pseudo-first-order kinetics model (Equation (10)): 
formula
(10)
where C0 is the initial concentrations of NAP; Ct is the concentrations of NAP at reaction time t, and kobs is the degradation rate constant. In the inset of Figure 1, the degradation rate constant of NAP was determined to be 0.08418 min−1 under the certain experimental condition.

Reactive species identification

It has been demonstrated that reactive oxidants (i.e. SO4•−, HO) could be produced in the decomposition of PS induced by catalysts (Zou et al. 2013; Li et al. 2016). EtOH with α-H is a suitable quencher for both SO4•− (k = 2.5 × 107 M1 s1) and HO (k = 9.7 × 108 M1 s1), while TBA without α-H is an effective scavenger for HO (k = 6.0 × 108 M1 s1) but not for SO4•− (k = 8.0 × 105 M1 s1) (Buxton et al. 1988; Neta et al. 1988). Based on these properties, EtOH and TBA, were used to differentiate the role of active species in NAP degradation.

In order to inhibit the oxidation adequately, the radical scavengers were added to obtain a concentration of 0.25 M:scavengers/PS/NAP (10000/10/1). The results taken from the scavenging experiments are shown in Figure 2(a). It is obvious that either TBA or EtOH presented strong inhibiting impact on NAP degradation at pH 5.0 and the addition of 0.25 M EtOH almost completely inhibited NAP degradation. When TBA and EtOH were added, the value of kobs decreased from 0.10475 min1 to 0.03952 min1 and 0.00386 min1, respectively, indicating that both SO4•− and HO are the primary reactive species facilitating NAP degradation (as shown in Table 1). In addition, Fig. S2 and Table 1 shows that the addition of TBA and EtOH exerted negative effect on the NAP degradation under both pH 7.0 and pH 9.0 conditions. These results demonstrated that both SO4•− and HO are responsible for the removal of NAP in the ZVI/PS system. However, there was no difference in PS decomposition between the results with and without scavengers (Figure 2(b)). These results suggest that both SO4•− and HO almost react with scavengers rather than NAP. Although SO4•− was usually recognized as one of the dominant active species, it was controversial whether HO was one of the reactive species (Liang & Su 2009; Rastogi et al. 2009; Wang & Chu 2011; Ghauch et al. 2015). To test the role of HO in the ZVI/PS process, nitrobenzene (NB) was selected as the probe for efficient reactivity toward •OH (k = 3.9 × 109 M1 s1) (Neta et al. 1988) but insignificant toward SO4•− (k ≤ 1.0 × 106 M1 s1) (Buxton et al. 1988). As depicted in Fig. S3, approximately 70% of NB was oxidized within 30 min, while NB oxidation was inhibited with the presence of TBA (10 mM). The findings further verify that HO was also one of the dominant active species in the ZVI/PS system.

Table 1

The results from radical scavenging experiments on NAP degradation

pHDegradation rate (kobs × 10−3 (min−1)) with addition of
Contribution to NAP degradation of
No scavengerEtOHTBASO4•−HO
5.0 104.75 3.86 39.52 60.8% 39.2% 
7.0 8.36 1.29 5.03 28.9% 71.1% 
9.0 7.08 0.82 3.03 51.6% 48.4% 
pHDegradation rate (kobs × 10−3 (min−1)) with addition of
Contribution to NAP degradation of
No scavengerEtOHTBASO4•−HO
5.0 104.75 3.86 39.52 60.8% 39.2% 
7.0 8.36 1.29 5.03 28.9% 71.1% 
9.0 7.08 0.82 3.03 51.6% 48.4% 
Figure 2

Effect of different radical scavengers on (a) NAP degradation and (b) PS decomposition. Experimental conditions: [NAP]0 = 25 μM, [PS]0 = 0.25 mM, [ZVI]0 = 0.5 mM, [TBA]0 = 0.25 M, [EtOH]0 = 0.25 M, T = 298.15 K, initial pH = 5.0.

Figure 2

Effect of different radical scavengers on (a) NAP degradation and (b) PS decomposition. Experimental conditions: [NAP]0 = 25 μM, [PS]0 = 0.25 mM, [ZVI]0 = 0.5 mM, [TBA]0 = 0.25 M, [EtOH]0 = 0.25 M, T = 298.15 K, initial pH = 5.0.

ESR technology was conducted to detect SO4•− and HO directly. DMPO was employed as the spin-trapping agent, and recent studies reported that SO4•− and HO can be identified through the detected signal of DMPO–SO4 adducts and DMPO–OH adducts, respectively (Zou et al. 2013; Li et al. 2016). As shown in Figure 3, a specific four-peak spectrum with intensity ratios of 1:2:2:1 was detected, and the special hyperfine coupling constants of a(N) = 1.50 mT and a(H) = 1.49 mT were consistent with DMPO-OH adducts (Zou et al. 2013; Li et al. 2019b), indicating that HO was generated. The ESR spectra at pH 5.0 showed the signal of DMPO–OH adducts with higher intensity than at pH 3.0. The typical spectrum of DMPO–SO4 adducts was not found. The results might suggest that DMPO–SO4 adducts had been converted to DMPO–HO adducts by nucleophilic substitution (Zou et al. 2013; He et al. 2020).

Figure 3

ESR spectra with the existence of DMPO in the ZVI/PS system. Experimental conditions: [PS]0 = 5.0 mM, [ZVI]0 = 3.0 g/L, [DMPO]0 = 100 mM, T = 298.15 K.

Figure 3

ESR spectra with the existence of DMPO in the ZVI/PS system. Experimental conditions: [PS]0 = 5.0 mM, [ZVI]0 = 3.0 g/L, [DMPO]0 = 100 mM, T = 298.15 K.

Reaction rate constants of NAP with SO4•− and HO

A competition kinetic method was utilized to measure the reaction rate constants of NAP with SO4•− and HO. The reaction rate constants between BA with HO (5.9 × 109 M1 s1) and SO4•− (1.2 × 109 M1 s1) were determinated (Wang et al. 2019), thus, BA was selected as the reference. Because the contribution of direct oxidation by oxidants can be ignored (Fig. S4), the reaction rate constant for NAP reacting with SO4•− can be calculated using Equation (11) (see Text S3 for details): 
formula
(11)

The plot of ln([NAP]0/[NAP]t) versus ln([BA]0/[BA]t) drew a straight line with zero intercept. As seen from Figure 4(a), the value of the slope was 4.70 ± 0.61, thus the reaction rate constant of SO4•− with NAP was calculated as (5.64 ± 0.73) × 109 M1 s1. Applying the same means, the value of the slope was 1.54 ± 0.09 (Figure 4(b)), and the rate constant of NAP with HO was calculated to be (9.09 ± 0.53) × 109 M1 s1.

Figure 4

Determination of the reaction rate constants of NAP reacting with (a) SO4•− and (b) HO. Experimental conditions: [NAP]0 = 5 μM, [BA]0 = 5 μM, [ZVI]0 = 1 mM, [PS]0 = 1 mM, [H2O2]0 = 1 mM, [TBA]0 = 0.1 M (a), 10 mM acetic acid buffer, pH = 5.0, T = 298.15 K.

Figure 4

Determination of the reaction rate constants of NAP reacting with (a) SO4•− and (b) HO. Experimental conditions: [NAP]0 = 5 μM, [BA]0 = 5 μM, [ZVI]0 = 1 mM, [PS]0 = 1 mM, [H2O2]0 = 1 mM, [TBA]0 = 0.1 M (a), 10 mM acetic acid buffer, pH = 5.0, T = 298.15 K.

Effects of parameters on NAP degradation

Effect of initial pH

The impacts of different initial pH (3.0, 5.0, 7.0, 9.0, 11.0) on the ZVI/PS degradation of NAP were evaluated and the experimental results are shown in Figure 5(a) and 5(b), At initial solution pH ≤ 5.0, the ZVI/PS proccess can effectively degrade NAP within 30 min. In particular, kobs at initial pH 3.0 was the largest (0.27155 min−1). The kobs of NAP removal markedly decreased from 0.08418 to 0.00815 min−1 with pH0 increasing from 5.0 to 7.0, then indistinctively decreased to 0.00691 min−1 at pH0 9.0; however, the NAP degradation was marginal at pH0 11.0 (kobs = 0.00201 min−1), indicating that the ZVI/PS process cannot degrade NAP in strong alkaline condition. The results suggest that (1) acidic conditions favor the formation and maintenance of Fe2+ (Fig. S5) and Fe2+ plays a major role in the production of SO4•− (Equation (7)). The pitting corrosion of ZVI decreased under neutral conditions (Ghauch et al. 2013), and the passivation on the ZVI surface was accelerated in alkaline solution, and thus hindered the further release of Fe2+, which results in less reactive oxygen species generated. (2) SO4•− tended to be converted to HO (Equations (8)–(9)), and the oxidation–reduction potential of HO is significantly decreased to 1.8 V under alkaline conditions (Li et al. 2018a). HO-induced oxidation was unselective and previous reports showed that SO4•− can oxidize organic pollutants that cannot be oxidized by HO (Monteagudo et al. 2016). Moreover, SO42- could accumulate during the treatment and the gradually increased SO42- could impede the HO-driven oxidation reported by Wei et al. 2016. (3) The speciation of NAP will vary with solution pH. With a pKa value of 4.15, its proportion was calculated and results are shown in Fig. S6, When the solution pH is over the pKa, the fraction of the deprotonated form increases. The reaction rate constant of SO4•− with deprotonated organic contaminants may be lower due to electric repulsion (Rickman & Mezyk 2010; Gu et al. 2019; Wang et al. 2019).

Figure 5

Effect of (a and b) initial pH, (c and d) PS dosage, (e and f) ZVI dosage, and (g and h) NAP concentration on NAP degradation. Experimental conditions: (a) [NAP]0 = 25 μM, [PS]0 = 0.25 mM, [ZVI]0 = 0.5 mM; (c) [NAP]0 = 25 μM, [ZVI]0 = 0.5 mM, initial pH = 5.0; (e) [NAP]0 = 25 μM, [PS]0 = 0.25 mM, initial pH = 5.0; (g) [PS]0 = 0.2 mM, [ZVI]0 = 0.5 mM, initial pH = 5.0.

Figure 5

Effect of (a and b) initial pH, (c and d) PS dosage, (e and f) ZVI dosage, and (g and h) NAP concentration on NAP degradation. Experimental conditions: (a) [NAP]0 = 25 μM, [PS]0 = 0.25 mM, [ZVI]0 = 0.5 mM; (c) [NAP]0 = 25 μM, [ZVI]0 = 0.5 mM, initial pH = 5.0; (e) [NAP]0 = 25 μM, [PS]0 = 0.25 mM, initial pH = 5.0; (g) [PS]0 = 0.2 mM, [ZVI]0 = 0.5 mM, initial pH = 5.0.

Effect of PS dosage

Different PS dosages (0.1–0.5 mM) on the NAP removal rate was examined and the results are shown in Figure 5(c) and 5(d). The removal rate of NAP increased with an increase of PS dosage. Figure 5(d) exhibited a linear relationship between kobs and PS dosage (R2 = 0.99). The value of kobs increased from 0.02844 min−1 to 0.17915 min−1 as the PS dosage increased from 0.1 mM to 0.5 mM. The degradation of NAP was distinctly improved by increasing the PS dosage because more reactive species were generated in the case of higher oxidants, which is in accordance with other reports (Rao et al. 2014; Chen et al. 2018). However, the PS depletion occurred in the ZVI/PS process, and only approximately 20% of PS was consumed (Fig. S7). The maximum % RSE (82.5%) obtained was at [PS]0 = 0.1 mM, which is higher than that obtained in thermally activated PS systems (Ghauch et al. 2015; Amasha et al. 2018). As [PS]0 increased from 0.1 mM to 0.5 mM, the RSE decreased from 82.5% to 24.0%, probably due to both Fe2+ and PS acting as scavengers of the sulfate radical (Equations (4) and (12)) and the recombination between two sulfate radical (Equation (13)) (Amasha et al. 2018; Gu et al. 2019). 
formula
(12)
 
formula
(13)

Effect of ZVI dosage

The effect of ZVI dosage (0.25–1.50 mM) on the degradation of NAP by the ZVI/PS process was evaluated and the results are shown in Figure 5(e) and 5(f). As the [ZVI]0 increased from 0.25 to 1.50 mM, the rate and extent of NAP removal were both greater and the corresponding kobs increased from 0.03675 min−1 to 0.18835 min−1. Increased ZVI dosage resulted in increased NAP removal efficiency. As discussed earlier, ZVI could act as an alternative source of Fe2+ reacting with other substrates (e.g. PS, O2, and H+) (Equations (1)–(3) and (12)). PS and/or the generated H2O2 (Equation (14)) can be activated to produce SO4•− and HO through Equations (7)–(9) and (15) (Wang et al. 2010; Tan et al. 2018; Cao et al. 2019). As [ZVI]0 increased, the dosage of Fe2+ was increased and more PS was consumed (Fig. S8); therefore, more radicals (SO4•− and HO) were generated to degrade NAP (Wang et al. 2019; Zhu et al. 2019), resulting in higher efficiency of NAP removal. However, the RSE dropped from 43.9% to 17.3% with increasing [ZVI]0 from 0.25 to 1.50 mM, likely due to the side reaction (Equation (4)) that Fe2+ could quench SO4•−. 
formula
(14)
 
formula
(15)

Effect of NAP dosage

Figure 5(g) and 5(h) shows the impact of different NAP dosage (5–25 μM) on NAP degradation in the ZVI/PS system. As NAP dosage was increased from 5 μM to 25 μM, the kobs within 30 min gradually decreased from 0.19671 min−1 to 0.07324 min−1 (Figure 5(h)), consistent with studies concerning other organic contaminants (Wang & Zhou 2016; Wei et al. 2016; Zhu et al. 2019). Theoretically, the overall yields of generated reactive species could be relatively fixed once the PS and ZVI dosages were fixed. With increasing concentration of NAP, the molar ratio of NAP to reactive species increased, causing a reduction in NAP degradation at a specific reaction time.

Degradation intermediates and degradation pathway of NAP

The NAP mineralization during the ZVI/PS system was also examined and the results are plotted in Fig. S9. As the PS concentration (0.1–0.5 mM) increased, the TOC removal efficiency also increased, which was in accordance with other reports (Ayoub & Ghauch 2014; Ghauch et al. 2015, 2017; Amasha et al. 2018). The highest removal efficiency of TOC could reach 29.12% in the experiment. In contrast to the higher level of NAP removal (Figure 5(c)), most of the NAP was only degraded to intermediates. In order to further understanding the degradation pathways of NAP in the ZVI/PS system, UPLC–QTOF–MS/MS was used to identify the intermediates. All the detected intermediates are listed in Table 2. The MS/MS spectra of nine degradation products are presented in Fig. S10 to S18.

Table 2

Degradation intermediates of NAP detected by UPLC-QTOF-MS/MS

Oxidation products[M + H]+ (m/z)Molecular formulaStructural formula
NAP 231 C14H14O3  
OP-246 247 C13H14O4  
OP-214 215 C14H14O2  
OP-186 187 C13H14 
OP-202 203 C13H14O2  
OP-184 185 C13H12O3  
OP-158 159 C11H10 
OP-216 217 C13H12O3  
OP-200 201 C13H12O2  
OP-218 219 C13H14O3  
Oxidation products[M + H]+ (m/z)Molecular formulaStructural formula
NAP 231 C14H14O3  
OP-246 247 C13H14O4  
OP-214 215 C14H14O2  
OP-186 187 C13H14 
OP-202 203 C13H14O2  
OP-184 185 C13H12O3  
OP-158 159 C11H10 
OP-216 217 C13H12O3  
OP-200 201 C13H12O2  
OP-218 219 C13H14O3  

It has been investigated that both SO4•− and HO were responsible for the destruction of NAP in this study. SO4•− shares similar reaction mechanisms, including hydrogen abstraction and hydroxyl addition with HO, while single electron transfer is more prevailing in SO4•−-driven reactions (Dulova et al. 2017; Luo et al. 2018b). According to the structure of NAP, it may be attacked by SO4•− and/or HO with an addition on the aromatic ring or by hydrogen attraction on the saturated α-carbon (Waldemer et al. 2007; Chi et al. 2019b). Based on the identified oxidation products, the possible degradation pathways of NAP in the ZVI/PS system are proposed in Figure 6. For pathway I, OP-246 was identified as a hydroxylated product, which was formed by HO attacking the methyl position (Luo et al. 2018a; Ray et al. 2018). OP-184 may be formed by the dehydration and decarboxylation of OP-246 (Jallouli et al. 2016; Chi et al. 2019a; Xu et al. 2019). Additionally, the NAP molecule is ionized on the naphthalene ring upon electron abstraction by SO4•− (Ghauch et al. 2015; Chi et al. 2019a). The ionized NAP molecule can be transformed to OP-184 followed by decarboxylation, which was consistent with several studies (Marotta et al. 2013; Kanakaraju et al. 2015; Coria et al. 2016; Chi et al. 2019c; Tu et al. 2019). The unsaturated carbon in the OP-184 molecular structure is easily attacked by SO4•− and HO, which makes it further oxidized to OP-200 (Chi et al. 2019a, 2019b; Xu et al. 2019). In pathway II, SO4•−/HO attack the methyl position of the naphthalene ring to break the C—Ccarboxyl bond, which leads to the decarboxylation of NAP to form OP-186 (Jallouli et al. 2016; Tu et al. 2019; Jung et al. 2020). OP-186 can transform to OP-202 with hydrogen extraction reaction and hydroxylation (Jallouli et al. 2016; Tu et al. 2019). In the presence of dissolved oxygen, OP-186 can further form the alkylperoxide derivative OP-218 (Kanakaraju et al. 2015; Jallouli et al. 2016; Tu et al. 2019). During the production of OP-200 from OP-218, the hydrogen atom on the C atom in the C—O bond in OP-218 moves toward distal oxygen and then dehydrates to form a ketone derivative OP-200 (Tu et al. 2019). In pathway III, OP-214 was produced by eliminating the hydroxyl group (—OH) from the carboxyl group (—COOH) of NAP (Ray et al. 2018). During formation of fragment OP-202 from OP-214, the carbonyl group is also eliminated followed by hydroxylation (Ray et al. 2018). OP-202 undergoes a hydrogen extraction reaction and a free radical coupling reaction to form a secondary glycol derivative and then produces OP-200 through its intramolecular dehydration condensation reaction (Arany et al. 2013; Jallouli et al. 2016; Tu et al. 2019). Moreover, in OP-202, the cleavage of the C—C bond to the naphthalene ring attacked by SO4•−/HO causes the formation of OP-158 (Méndez-Arriaga et al. 2008; Kanakaraju et al. 2015). For pathway IV, OP-216 was possibly formed by the attack of SO4•− and/or HO on the methoxyl group on the naphthalene ring, and this is followed by demethoxylation, which was consistent with previous researches (Méndez-Arriaga et al. 2008; Chin et al. 2014; Chi et al. 2019a, 2019b).

Figure 6

Proposed degradation pathways of NAP in ZVI/PS system.

Figure 6

Proposed degradation pathways of NAP in ZVI/PS system.

Figure 7

Effect of ammonia on the ZVI/PS oxidation of NAP. Experimental conditions: [NAP]0 = 25 μM, [PS]0 = 0.25 mM, [ZVI]0 = 0.5 mM, initial pH = 7.0, T = 298.15 K.

Figure 7

Effect of ammonia on the ZVI/PS oxidation of NAP. Experimental conditions: [NAP]0 = 25 μM, [PS]0 = 0.25 mM, [ZVI]0 = 0.5 mM, initial pH = 7.0, T = 298.15 K.

Ammonial-enhanced oxidation of NAP under neutral and basic conditions

In this work, the effect of chloride on the degradation of NAP was marginal (data not shown), and therefore ammonia is defined as the total of NH3 + NH4+ in solutions. The results of NAP degradation during the ZVI/PS process with different dosages of ammonia at pH0 7.0 are plotted in Figure 7. With the increase of ammonia dosages, the removal efficiency of NAP increased. The value of kobs in the absence of ammonia was 0.00815 min−1, while the value of kobs varied linearly from 0.02976 min−1 to 0.14276 min−1 with increasing concentration of ammonia (the inset of Figure 7). Next, experiments were also conducted at pH0 9.0 and a similar enhanced impact of ammonia on NAP degradation is presented in Fig. S19. When the pH0 was increased from 7.0 to 9.0, the promotion effect of ammonia on NAP degradation was weakened, which might be interpreted as the equilibrium of ammonia (Equation (16)). The favorable effect of ammonia for NAP oxidation in ZVI/PS could be explained by high production of Fe2+ in the case of ammonia. Passivation films that formed and covered the surface of ZVI can be destructed by depassivators (Ansaf et al. 2016; Ling et al. 2018). In addition, NH4Cl has shown a promising ability to activate the hydrogen ion, thus it is a good anti-passivation corrosive chemical for many kinds of metal (Forse et al. 1993; Baranwal & Venkatesh 2017; Li et al. 2018b; Baranwal & Rajaraman 2019). 
formula
(16)
In order to test the impact of ammonia on PS activation by ZVI, the change of dissolved total iron (Fetot) with different ammonia dosages was plotted in Fig. S20. Apparently, ammonia could enhance the corrosion of ZVI. Once the ammonia was applied, more dissolved iron was observed after 30 min treatment. At pH0 = 7.0, for example, with ammonia added, the generation of Fetot increased from 0.51 mg L−1 at 0 mM to 1.363 mg L−1 at 10 mM. The enhanced production of Fe2+ generated more SO4•− and HO and thus increased the degradation of NAP. Previous publications also observed that the dissolution rate of carbon steel increased with increasing NH4Cl concentration and the dissolved species was detected as Fe2+ (Baranwal & Venkatesh 2017; Baranwal & Rajaraman 2019). The results also indicated the promising role of ammonia in NAP removal in the system due to faster and more Fe2+ being generated.

CONCLUSIONS

In this paper, the degradation of NAP by ZVI-activated PS was surveyed in detail. The ZVI/PS system performs betters at degrading NAP than the Fe2+/PS system, and the NAP degradation by the ZVI/PS system showed a pseudo-first-order kinetics pattern. The degradation of NAP was mainly attributable to SO4•− and HO using radical scavenging experiments and ESR study. The second-order rate constants for NAP with SO4•− and HO were estimated to be (5.64 ± 0.73) × 109 M1 s1 and (9.05 ± 0.51) × 109 M1 s1, respectively. NAP oxidation rate increased with an increase in PS dosage and ZVI concentration, but it decreased with increasing solution pH and NAP concentration. The oxidation products of NAP were detected and the pathways of NAP degradation were analyzed. The addition of ammonia into the ZVI/PS system could significantly improve the removal efficiency of NAP under neutral and basic conditions. This work proves that ZVI is an efficient and convenient PS activator and thus the ZVI/PS technique can be introduced to eliminate refractory pollutants in the aqueous environment. In order to use ZVI/PS technology for wastewater application, for example hospital and pharmaceutical industries effluents, further research is needed into the cost of implementing ZVI/PS on a larger scale at the source of pollution. Work in this direction is currently in progress in our laboratory.

ACKNOWLEDGEMENTS

This work was supported by the Department of Science and Technology of Jilin Province (No. 20170204016GX). The authors also appreciate anonymous reviewers for their thoughtful comments and suggestions on this work.

SUPPLEMENTARY MATERIAL

The Supplementary Material for this paper is available online at https://dx.doi.org/10.2166/wst.2020.263.

REFERENCES

REFERENCES
Arany
E.
,
Szabó
R. K.
,
Apáti
L.
,
Alapi
T.
,
Ilisz
I.
,
Mazellier
P.
,
Dombi
A.
&
Gajda-Schrantz
K.
2013
Degradation of naproxen by UV, VUV photolysis and their combination
.
Journal of Hazardous Materials
262
,
151
157
.
Baranwal
P. K.
&
Rajaraman
P. V.
2019
Electrochemical investigation on effect of sodium thiosulfate (Na2S2O3) and ammonium chloride (NH4Cl) on carbon steel corrosion
.
Journal of Materials Research and Technology
8
(
1
),
1366
1378
.
Buxton
G. V.
,
Greenstock
C. L.
,
Helman
W. P.
&
Ross
A. B.
1988
Critical review of rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals (•OH/•O−) in aqueous solution
.
Journal of Physical and Chemical Reference Data
17
(
2
),
513
886
.
Chin
C. J. M.
,
Chen
T. Y.
,
Lee
M.
,
Chang
C. F.
,
Liu
Y. T.
&
Kuo
Y. T.
2014
Effective anodic oxidation of naproxen by platinum nanoparticles coated FTO glass
.
Journal of Hazardous Materials
277
,
110
119
.
Coria
G.
,
Sirés
I.
,
Brillas
E.
&
Nava
J. L.
2016
Influence of the anode material on the degradation of naproxen by Fenton-based electrochemical processes
.
Chemical Engineering Journal
304
,
817
825
.
Ghauch
A.
,
Assi
H. A.
&
Bdeir
S.
2010
Aqueous removal of diclofenac by plated elemental iron: bimetallic systems
.
Journal of Hazardous Materials
182
(
1
),
64
74
.
Ghauch
A.
,
Baydoun
H.
&
Dermesropian
P.
2011
Degradation of aqueous carbamazepine in ultrasonic/Fe0/H2O2 systems
.
Chemical Engineering Journal
172
(
1
),
18
27
.
Ghauch
A.
,
Tuqan
A. M.
&
Kibbi
N.
2012
Ibuprofen removal by heated persulfate in aqueous solution: a kinetics study
.
Chemical Engineering Journal
197
,
483
492
.
Ghauch
A.
,
Ayoub
G.
&
Naim
S.
2013
Degradation of sulfamethoxazole by persulfate assisted micrometric Fe0 in aqueous solution
.
Chemical Engineering Journal
228
,
1168
1181
.
Ghauch
A.
,
Tuqan
A. M.
&
Kibbi
N.
2015
Naproxen abatement by thermally activated persulfate in aqueous systems
.
Chemical Engineering Journal
279
,
861
873
.
Ghauch
A.
,
Baalbaki
A.
,
Amasha
M.
,
Asmar
R. E.
&
Tantawi
O.
2017
Contribution of persulfate in UV-254 nm activated systems for complete degradation of chloramphenicol antibiotic in water
.
Chemical Engineering Journal
317
,
1012
1025
.
Girit
B.
,
Dursun
D.
,
Olmez-Hanci
T.
&
Arslan-Alaton
I.
2015
Treatment of aqueous bisphenol A using nano-sized zero-valent iron in the presence of hydrogen peroxide and persulfate oxidants
.
Water Science and Technology
71
(
12
),
1859
1868
.
Gu
D.
,
Guo
C.
,
Hou
S.
,
Lv
J.
,
Zhang
Y.
,
Feng
Q.
,
Zhang
Y.
&
Xu
J.
2019
Kinetic and mechanistic investigation on the decomposition of ketamine by UV-254 nm activated persulfate
.
Chemical Engineering Journal
370
,
19
26
.
Hakim
S. A.
,
Jaber
S.
,
Eddine
N. Z.
,
Baalbaki
A.
&
Ghauch
A.
2019
Data for persulfate activation by UV light to degrade theophylline in a water effluent
.
Data in Brief
27
,
104614
.
Hakim
S. A.
,
Jaber
S.
,
Eddine
N. Z.
,
Baalbaki
A.
&
Ghauch
A.
2020
Degradation of theophylline in a UV254/PS system: matrix effect and application to a factory effluent
.
Chemical Engineering Journal
380
,
122478
.
Hussain
I.
,
Zhang
Y.
,
Huang
S.
&
Du
X.
2012
Degradation of p-chloroaniline by persulfate activated with zero-valent iron
.
Chemical Engineering Journal
203
,
269
276
.
Jallouli
N.
,
Elghniji
K.
,
Hentati
O.
,
Ribeiro
A. R.
,
Silva
A. M.
&
Ksibi
M.
2016
UV and solar photo-degradation of naproxen: TiO2 catalyst effect, reaction kinetics, products identification and toxicity assessment
.
Journal of Hazardous Materials
304
,
329
336
.
Jung
S. C.
,
Bang
H. J.
,
Lee
H.
,
Kim
H.
,
Ha
H. H.
,
Yu
Y. H.
&
Park
Y. K.
2020
Degradation behaviors of naproxen by a hybrid TiO2 photocatalyst system with process components
.
Science of The Total Environment
708
,
135216
.
Kanakaraju
D.
,
Motti
C. A.
,
Glass
B. D.
&
Oelgemöller
M.
2015
Tio2 photocatalysis of naproxen: effect of the water matrix, anions and diclofenac on degradation rates
.
Chemosphere
139
,
579
588
.
Lai
L.
,
Zhou
H.
,
Zhang
H.
,
Ao
Z.
,
Pan
Z.
,
Chen
Q.
,
Xiong
A.
,
Yao
G.
&
Lai
B.
2020
Activation of peroxydisulfate by natural titanomagnetite for atrazine removal via free radicals and high-valent iron-oxo species
.
Chemical Engineering Journal
387
,
124165
.
Li
R.
,
Kong
J.
,
Liu
H.
,
Chen
P.
,
Su
Y.
,
Liu
G.
&
Lv
W.
2018a
Removal of indomethacin using UV–vis/peroxydisulfate: kinetics, toxicity, and transformation pathways
.
Chemical Engineering Journal
331
,
809
817
.
Li
J.
,
Li
D.
,
Xu
Z.
,
Liao
C.
,
Liu
Y.
&
Zhong
B.
2018b
Selective leaching of valuable metals from laterite nickel ore with ammonium chloride-hydrochloric acid solution
.
Journal of Cleaner Production
179
,
24
30
.
Liang
C.
&
Su
H. W.
2009
Identification of sulfate and hydroxyl radicals in thermally activated persulfate
.
Industrial & Engineering Chemistry Research
48
(
11
),
5558
5562
.
Liang
C.
,
Huang
C. F.
,
Mohanty
N.
&
Kurakalva
R. M.
2008
A rapid spectrophotometric determination of persulfate anion in ISCO
.
Chemosphere
73
(
9
),
1540
1543
.
Luo
S.
,
Gao
L.
,
Wei
Z.
,
Spinney
R.
,
Dionysiou
D. D.
,
Hu
W. P.
,
Chai
L.
&
Xiao
R.
2018a
Kinetic and mechanistic aspects of hydroxyl radical–mediated degradation of naproxen and reaction intermediates
.
Water Research
137
,
233
241
.
Luo
S.
,
Wei
Z.
,
Spinney
R.
,
Villamena
F. A.
,
Dionysiou
D. D.
,
Chen
D.
,
Tang
C. J.
,
Chai
L.
&
Xiao
R.
2018b
Quantitative structure–activity relationships for reactivities of sulfate and hydroxyl radicals with aromatic contaminants through single–electron transfer pathway
.
Journal of Hazardous Materials
344
,
1165
1173
.
Méndez-Arriaga
F.
,
Gimenez
J.
&
Esplugas
S.
2008
Photolysis and TiO2 photocatalytic treatment of naproxen: degradation, mineralization, intermediates and toxicity
.
Journal of Advanced Oxidation Technologies
11
(
3
),
435
444
.
Neta
P.
,
Huie
R. E.
&
Ross
A. B.
1988
Rate constants for reactions of inorganic radicals in aqueous solution
.
Journal of Physical and Chemical Reference Data
17
(
3
),
1027
1284
.
Oh
S. Y.
,
Kang
S. G.
&
Chiu
P. C.
2010
Degradation of 2, 4-dinitrotoluene by persulfate activated with zero-valent iron
.
Science of the Total Environment
408
(
16
),
3464
3468
.
Rao
Y. F.
,
Qu
L.
,
Yang
H.
&
Chu
W.
2014
Degradation of carbamazepine by Fe(II)-activated persulfate process
.
Journal of Hazardous Materials
268
,
23
32
.
Rastogi
A.
,
Al-Abed
S. R.
&
Dionysiou
D. D.
2009
Sulfate radical-based ferrous–peroxymonosulfate oxidative system for PCBs degradation in aqueous and sediment systems
.
Applied Catalysis B: Environmental
85
(
3–4
),
171
179
.
Waldemer
R. H.
,
Tratnyek
P. G.
,
Johnson
R. L.
&
Nurmi
J. T.
2007
Oxidation of chlorinated ethenes by heat-activated persulfate: kinetics and products
.
Environmental Science & Technology
41
(
3
),
1010
1015
.
Wang
K. S.
,
Lin
C. L.
,
Wei
M. C.
,
Liang
H. H.
,
Li
H. C.
,
Chang
C. H.
,
Fang
Y. T.
&
Chang
S. H.
2010
Effects of dissolved oxygen on dye removal by zero-valent iron
.
Journal of Hazardous Materials
182
(
1–3
),
886
895
.
Wang
S.
,
Wu
J.
,
Lu
X.
,
Xu
W.
,
Gong
Q.
,
Ding
J.
,
Dan
B.
&
Xie
P.
2019
Removal of Acetaminophen in the Fe2+/persulfate system: kinetic model and degradation pathways
.
Chemical Engineering Journal
358
,
1091
1100
.
Wei
X.
,
Gao
N.
,
Li
C.
,
Deng
Y.
,
Zhou
S.
&
Li
L.
2016
Zero-valent iron (ZVI) activation of persulfate (PS) for oxidation of bentazon in water
.
Chemical Engineering Journal
285
,
660
670
.
Wu
X.
,
Gu
X.
,
Lu
S.
,
Xu
M.
,
Zang
X.
,
Miao
Z.
,
Qiu
Z.
&
Sui
Q.
2014
Degradation of trichloroethylene in aqueous solution by persulfate activated with citric acid chelated ferrous ion
.
Chemical Engineering Journal
255
,
585
592
.
Wu
Y.
,
Bianco
A.
,
Brigante
M.
,
Dong
W.
,
de Sainte-Claire
P.
,
Hanna
K.
&
Mailhot
G.
2015
Sulfate radical photogeneration using Fe-EDDS: influence of critical parameters and naturally occurring scavengers
.
Environmental Science & Technology
49
(
24
),
14343
14349
.
Xiao
R.
,
He
Z.
,
Diaz-Rivera
D.
,
Pee
G. Y.
&
Weavers
L. K.
2014
Sonochemical degradation of ciprofloxacin and ibuprofen in the presence of matrix organic compounds
.
Ultrasonics Sonochemistry
21
(
1
),
428
435
.
Zhao
L.
,
Ji
Y.
,
Kong
D.
,
Lu
J.
,
Zhou
Q.
&
Yin
X.
2016
Simultaneous removal of bisphenol A and phosphate in zero-valent iron activated persulfate oxidation process
.
Chemical Engineering Journal
303
,
458
466
.
Zhu
J. P.
,
Lin
Y. L.
,
Zhang
T. Y.
,
Cao
T. C.
,
Xu
B.
,
Pan
Y.
,
Zhang
X. T.
&
Gao
N. Y.
2019
Modelling of iohexol degradation in a Fe(II)-activated persulfate system
.
Chemical Engineering Journal
367
,
86
93
.
Zou
J.
,
Ma
J.
,
Chen
L.
,
Li
X.
,
Guan
Y.
,
Xie
P.
&
Pan
C.
2013
Rapid acceleration of ferrous iron/peroxymonosulfate oxidation of organic pollutants by promoting Fe(III)/Fe(II) cycle with hydroxylamine
.
Environmental Science & Technology
47
(
20
),
11685
11691
.