ABSTRACT
In this research, ascorbic acid (AA) was used to enhance Fe(II)/Fe(III)-activated permonosulfate (PMS) systems for the degradation of fluoranthene (FLT). AA enhanced the production of ROS in both PMS/Fe(II) and PMS/Fe(III) systems through chelation and reduction and thus improved the degradation performance of FLT. The optimal molar ratio in PMS/Fe(II)/AA/FLT and PMS/Fe(III)/AA/FLT processes were 2/2/4/1 and 5/10/5/1, respectively. In addition, the experimental results on the effect of FLT degradation under different groundwater matrixes indicated that PMS/Fe(III)/AA system was more adaptable to different water quality conditions than the PMS/Fe(II)/AA system. was the major reactive oxygen species (ROS) responsible for FLT removal through the probe and scavenging tests in both systems. Furthermore, the degradation intermediates of FLT were analyzed using gas chromatograph-mass spectrometry (GC-MS), and the probable degradation pathways of FLT degradation were proposed. In addition, the removal of FLT was also tested in actual groundwater and the results showed that by increasing the dose and pre-adjusting the solution pH, 88.8 and 100% of the FLT was removed for PMS/Fe(II)/AA and PMS/Fe(III)/AA systems. The above experimental results demonstrated that PMS/Fe(II)/AA and PMS/Fe(III)/AA processes have a great perspective in practice for the rehabilitation of FLT-polluted groundwater.
HIGHLIGHTS
The effect of AA on the circulation between Fe(II) and Fe(III) in the PMS system was investigated.
SO4•− was the primary ROS in PMS/Fe(II)/AA and PMS/Fe(III)/AA systems.
The effect of actual groundwater matrixes on FLT removal was investigated.
The possible degradation pathways of FLT were proposed.
Effective degradation of FLT in actual groundwater was achieved.
INTRODUCTION
Polycyclic aromatic hydrocarbons (PAHs) could be generated through industrial activities, energy production, and traffic exhausts, and enduringly exist in the soil of industrial sites and actual groundwater persistently. PAHs are getting more and more attention due to their complex composition, high concentration, and serious harm to the health of people. Fluoranthene (FLT) is frequently detected in contaminated sites in various countries as a typical PAH, which resulted in FLT being placed on the Environmental Protection Agency's (EPA) priority list for PAHs (Zhang & Chen 2017). FLT is mainly produced by incomplete combustion of organic matter and fossil fuels. As common PAHs, FLT has carcinogenic, teratogenic, and mutagenic properties, and most importantly, it is easily bio-accumulated and hard to biodegrade. FLT could enter the human body via breathing air containing FLT or by exposure to FLT-contaminated water and soil. In addition, FLT is photocytotoxic to human skin keratinocytes (Wang et al. 2007). Therefore, there is an urgent need to find an economic and efficient technology to restore FLT-contaminated groundwater and soil.
Currently, in situ chemical oxidation processes (ISCO) have been applied to degrade organic contaminants in contaminated soils and groundwater. It is reported that Fenton reagent, activated peroxydisulfate (PDS), activated peroxymonosulfate (PMS), O3, and permanganate are widely used in ISCO (Sunder & Hempel 1997; Sutton et al. 2011; Lyu et al. 2020; Bein et al. 2023). PMS and PDS are more adoptable than other oxidants due to the broader pH application range, the less effect of soil moisture, and the longer duration of action in groundwater (Cheng et al. 2016). Both PMS and PDS are solid and soluble in water, and the aqueous solutions are acidic. The peroxide bond in PMS is asymmetric in charge distribution, inducing a partial positive charge on the peroxide oxygen attached to the hydrogen, whereas in PDS it is symmetrical, leading to the difference in reactivity between PMS and PDS. Several studies have shown that the groundwater component fraction has less effect on the activation of PMS compared to PDS (Wang & Wang 2018).





Ascorbic acid (AA) is a ubiquitous vitamin found in various foods, particularly in plants and fruits. Due to its enediol structure conjugating with the carbonyl group in the lactone ring (Figure S1), it is also an important antioxidant, which plays a crucial role in inhibiting oxidative-induced cellular damage in both plant and animal tissues (Roig et al. 1993). AA is oxidated to non-toxic oxidation products such as hemihydro-L-ascorbate radical (SDA) and dehydro-L-ascorbate (DA) (Tripathi et al. 2009). AA is widely used in the reduction of metal activators in advanced oxidation processes. For example, some researchers have introduced AA into the H2O2/Cu(I) system to promote the Cu(I)/Cu(II) cycle in the system to improve the removal efficiency of pollutants (Zhou et al. 2016). Besides, the solution showed acidic properties with the addition of AA, which inhibited the iron precipitation and maintained the soluble iron stable (Roig et al. 1993). Importantly, iron chelates were formed after the addition of AA, in which Fe(II) was released more slowly with the generation of more stable chelated ions (Davies 1992).
Hence, in this study, AA was added to increase FLT removal by AA chelation with Fe(II) and Fe(III) in PMS/Fe(II) and PMS/Fe(III) processes. The outcomes of this study can provide efficient techniques for the remediation of FLT-contaminated groundwater. The purpose of this research was (1) to reveal the efficiency of FLT removal in PMS/Fe(II)/AA and PMS/Fe(III)/AA processes; (2) to demonstrate the major ROS for FLT degradation in both systems through the probe and scavenging experiments; (3) to investigate the effects of groundwater matrices and chemical agent dosage on FLT removal; (4) to propose the possible removal pathways of FLT based on the detected intermediates; and finally (5) to test the practicality of both systems for FLT removal in actual groundwater.
MATERIALS AND METHODS
Materials
Details of the chemical agents used in this study can be found in the supplementary material (Text S1).
Experimental procedures
All tests were performed in the glass reactor (250 mL) with stirring at 700 r min−1. The RPM used in this study was intended to make full contact with contaminants (FLT) between chemical agents added to the reactor so that the process of FLT removal could be performed quickly and efficiently. The treatment method used in this study has been adopted in several studies, and it can even reach 800 rpm (Cheng et al. 2016; Cao et al. 2019; Li et al. 2019a; Xu et al. 2022). Firstly, 0.004 M FLT stock solution was prepared with 0.02 g FLT dissolving in acetone (25 mL). Then, 1.0 μM FLT-contaminated solution was prepared by adding 0.248 mL of FLT stock solution in 1.0 L of ultrapure water, and the mixture was stirred for 9 h to ensure the sufficient dissolution of FLT. The reasons for the choice of acetone and the description of the stability of the aqueous FLT solution are given in Text S4. Two hundred and fifty mL FLT-contaminated solution was transferred into the reactor where only 0.062 mL acetone was presented, and the ratio of acetone to water was about 1/4,000, hence the influence of acetone on ROS was negligible. In addition, other predetermined dosage reagents (Fe(II), Fe(III), and AA) were added successively. Finally, PMS was added to start the test. At the desired time, 0.7 mL of the sample was removed into a brown vial (4 mL) containing 0.7 mL of methanol (stopping the reaction). After that, the samples were filtered through 0.22 μm microporous organic filters before analyzing the FLT concentration by high-performance liquid chromatography (HPLC). The reactions were performed three times at least and the results were averaged.
To investigate the influence of the pH of aqueous solution in both systems, the aqueous initial solution pH was adjusted to 2.0, 3.0, 5.0, 7.0, 9.0, and 11 by 0.1 M H2SO4 and 0.1 M NaOH. It is reported that the inorganic anions (Cl−, Br−, and ) broadly exist in natural water and groundwater, which could influence FLT removal in both systems (Carol & Alvarez 2016). Several representative inorganic anions (Cl−, Br−, and
) were selected to investigate the effect of inorganic anions on FLT removal, and the concentrations were set as 1.0, 20, and 50 mM based on the concentration present at the actual contaminated site (Carol & Alvarez 2016).
Natural organic matter (NOM) existing in actual groundwater could react with and
and further influence the removal of contaminants (Li et al. 2019b). Therefore, the effect of humic acid (HA) on FLT removal in PMS/Fe(II)/AA and PMS/Fe(III)/AA systems was investigated. Considering the concentrations of HA substances in groundwater, the chemical dosages of HA were set to 1.0, 3.0, 5.0, 10, 50, and 100 mg L−1 in the test to explore the influence of HA for FLT removal (Li et al. 2019c).
The specific scavengers were first added to the aqueous solution containing FLT in the scavenging experiments, and then other chemicals were added to initiate the reaction (the reaction rate constants of different scavengers with ROS are shown in Text S3). The dominant ROS in PMS/Fe(II)/AA and PMS/Fe(III)/AA processes were determined based on the degradation of FLT with the addition of different scavengers. In electron paramagnetic resonance (EPR) tests, 5,5-dimethyl-1-oxypyrrolidine (DMPO) was used as the ROS trap to detect the presence of different ROS. At the reaction time of 2.0 min, the 1.0 mL sample was added to a 4.0 mL vial pre-filled with 1.0 mL aqueous DMPO solution, and the mixture was analyzed by EPR.
In the probe experiments, degradation experiments were performed with probe compounds instead of FLT to investigate the types of reactive radicals in both systems. Since nitrobenzene (NB) has a high reactivity with (kHO• = 3.9 × 109 M−1 s−1) and is more reactive than with
, the reaction rate constant is three orders of magnitude higher, while
hardly reacts with NB, so NB was chosen as the chemical probe for
(Buxton et al. 1988). The rate constant for the reaction of carbon tetrachloride (CT) with
(1.6 × 1010 M−1 s−1) is higher than that of
(kHO• < 2 × 106 M−1 s−1), and therefore CT was chosen as the chemical probe for
in this experiment (Ahmad et al. 2015). Since all chemicals that can react with
can react with
, a separate probe suitable for
could not be selected (Teel & Watts 2002). The reaction rate constants of anisole (AN) with
and
are all greater than 109 M−1 s−1, AN was chosen as both
and
probe (Ahmad et al. 2012).
Analytical methods
The concentrations of FLT were detected through HPLC (LC-20AT, Shimadzu, Japan), which set 235 nm as the detection wavelength and 30 °C oven temperature. The constituents of the HPLC mobile phase were 1:9 ratio of methanol and water. The injection volume of the sampler was 100 μL. Other significant analytical methods of the tests are exhibited in the supplementary material (Text S2).
RESULTS AND DISCUSSION
The effectiveness of FLT removal in PMS/Fe(II)/AA and PMS/Fe(III)/AA systems
FLT degradation performance detected by HPLC in (a) PMS/Fe(II)/AA system ([FLT]0 = 0.2 mg L−1, [PMS]0 = 5.0 mg L−1, [Fe(II)]0 = 2.0 mg L−1, and [AA]0 = 2.8 mg L−1) and (b) PMS/Fe(III)/AA system ([PMS]0 = 12 mg L−1, [FLT]0 = 0.2 mg L−1, [Fe(III)]0 = 11 mg L−1, and [AA]0 = 3.5 mg L−1).
FLT degradation performance detected by HPLC in (a) PMS/Fe(II)/AA system ([FLT]0 = 0.2 mg L−1, [PMS]0 = 5.0 mg L−1, [Fe(II)]0 = 2.0 mg L−1, and [AA]0 = 2.8 mg L−1) and (b) PMS/Fe(III)/AA system ([PMS]0 = 12 mg L−1, [FLT]0 = 0.2 mg L−1, [Fe(III)]0 = 11 mg L−1, and [AA]0 = 3.5 mg L−1).
For the PMS/Fe(II)/AA process, FLT removal was merely 35.7% for the PMS/Fe(II) process, but the FLT degradation dramatically increased in the initial 5 min. It is worth noting that FLT removal increased significantly from 35.7 to 90.0% when AA was added to the PMS/Fe(II) process. For PMS/Fe(III)/AA process, when the molar ratios of PMS/Fe(III)/AA/FLT were set at 2/2/4/1, 5/5/10/1, and 5/10/10/1, the degradation of FLT was 57.6, 75.3, and 90.6%, respectively (Figure S3). Finally, in order to investigate the optimal AA dosage in this study, the molar ratio of PMS/Fe(III) was fixed to 5/10, FLT was removed by 91.8 and 87.7% for PMS/Fe(III)/AA process on the molar ratios of 5/10/5 and 5/10/20, respectively. The decrease of FLT degeneration with increasing molar ratio of AA was due to the scavenging effect of excess AA for ROS ( and
) (Xu et al. 2022). The degradation of FLT hardly changed after continuing to increase the amount of PMS based on the molar ratio of 5/10/5 for the PMS/Fe(II)/AA system. Therefore, 5/10/5 was chosen as the optimal molar ratio for the PMS/Fe(III)/AA system in this study considering the actual remediation cost and other factors.
Variation of Fe(II) concentration in PMS/Fe(II)/AA system and PMS/Fe(III)/AA system detected by o-phenanthroline spectrophotometry ([FLT]0 = 0.2 mg L−1, [PMS]0 = 0.08 mM, [Fe(II)]0 = 0.08 mM, and [AA]0 = 0.16 mM; [PMS]0 = 0.1 mM, [Fe(III)]0 = 0.2 mM, and [AA]0 = 0.1 mM).
Variation of Fe(II) concentration in PMS/Fe(II)/AA system and PMS/Fe(III)/AA system detected by o-phenanthroline spectrophotometry ([FLT]0 = 0.2 mg L−1, [PMS]0 = 0.08 mM, [Fe(II)]0 = 0.08 mM, and [AA]0 = 0.16 mM; [PMS]0 = 0.1 mM, [Fe(III)]0 = 0.2 mM, and [AA]0 = 0.1 mM).
The initial pH of PMS/Fe(II) and PMS/Fe(III) needed to be kept consistent with that of PMS/Fe(II)/AA and PMS/Fe(III)/AA processes with only Fe(II), Fe(III) and AA added to investigate whether the increase in FLT degradation was due to the influence of pH decrease or to the influence of reduction and chelation reactions after the addition of AA. The original solution pH of PMS/Fe(II) and PMS/Fe(III) processes was adjusted to 5.3 and 4.6 by 0.1 M H2SO4 which was consistent with the unadjusted pH of PMS/Fe(II)/AA and PMS/Fe(III)AA processes. As shown in Figure S4a and Figure S4b, the removal of FLT increased only 4.7 and 6.4% than the unadjusted one, confirming that the increase in FLT removal was mainly due to the chelation and reduction of iron by AA.
Previous studies have demonstrated the chelation reaction of AA with iron, showing that AA and Fe(III) form Fe(III)-AA chelates via an inner-sphere mechanism in milliseconds and also isolating a Fe(II)-AA chelate in the distorted octahedral environment realized in the polymeric structure (Kontoghiorghes et al. 2020). The full wavelength scan of UV-vis spectrum was used to investigate the chelating effect of AA (Ren et al. 2020). As shown in Figure S5, the characteristic absorption peak of AA was located at 253 nm. After the addition of Fe(II) and Fe(III), the characteristic absorption peak was red-shifted. In addition, after the addition of Fe(II), a new absorption band at 261–290 nm was formed, indicating that AA can chelate Fe(II). The absorption band at 261–648 nm shifted obviously and the intensity of the characteristic absorption peak increased significantly after the addition of Fe(III), which proved that AA has a strong chelating effect on Fe(III) compared with Fe(II).
Wu et al. (2015) investigated the degradation effects of hydroxylamine hydrochloride (HAH), sodium thiosulfate (STS), and sodium sulfite (SS) on trichloroethylene (TCE) removal in the peroxysulfate process and found that each reductant could promote the TCE removal. Li et al. (2019a) and Luo et al. (2021) also added HAH to the PMS/Fe(II) process to promote the Fe(II)/Fe(III) cycle by utilizing the reduction of HA, thus sufficiently activating the PMS to improve the pollutant removal rate increase. Rastogi et al. (2009a) investigated the effect of the different chelating agents (sodium citrate, sodium pyrophosphate, and ethylenediamine disuccinate (EDDS)) on the degradation of chlorophenols in the PMS system, and found that the pyrophosphate and EDDS chelated Fe(II) followed by slowly releasing Fe(II) to enhance the efficiency of activation of PMS (Rastogi et al. 2009a). Wu et al. (2014) also used CA-chelated Fe(II)-activated PDS to degrade TCE, and TCE could be degraded. In this research, the optimal molar ratio in PMS/Fe(II)/AA/FLT and PMS/Fe(III)/AA/FLT systems were 2/2/4/1 and 5/10/5/1, respectively, when the initial concentration of FLT was 1.0 μM. As shown in Table 1, The dosage of various agents for the different enhancement systems was significantly higher than in the present study since the addition of AA has both the reducing and chelating effect. In conclusion, AA was an efficient and economically enhanced agent in the persulfate system.
Comparative studies with other studies on iron-activated PMS and PDS systems
Process . | Pollutant . | Enhanced agent . | Experimental conditions . | Degradation efficiency . | Ref. . |
---|---|---|---|---|---|
PMS/Fe(II)/HAH | Benzoic acid (40 μM) | Hydroxylamine (HAH) | [HAH]0 = 0.40 mM, [Fe(II)]0 = 10.8 μM, [PMS]0 = 0.32 mM | 80% (pH0 = 3.0) | Zou et al. (2013) |
PDS/Fe(II)/HAH | Trichloroethylene (0.15 mM) | HAH | [HAH]0 = 1.5 mM, [Fe(II)]0 = 3.0 mM, [PDS]0 = 2.25 mM | 97.9% (pH0 = 4.0) | Wu et al. (2015) |
PMS/Fe3O4/HAH | Atrazine (23 μM) | HAH | [HAH]0 = 0.3 mM, [Fe3O4]0 = 0.69 g L−1, [PMS]0 = 0.4 mM | 94% (pH0 = 6.8) | Li et al. (2019a) |
PMS/FeOCl/HAH | Ciprofloxacin (5.0 mg L−1) | HAH | [HAH]0 = 1.0 mM, [FeOCl]0 = 0.2 g L−1, [PMS]0 = 1.5 mM | 74.5% (pH0 = 6.5) | Luo et al. (2021) |
PDS/Fe(II)/STS | Trichloroethylene (0.15 mM) | Sodium thiosulfate (STS) | [STS]0 = 1.5 mM, [Fe(II)]0 = 3.0 mM, [PDS]0 = 2.25 mM | 65.5% (pH0 = 5.5) | Wu et al. (2015) |
PDS/Fe(II)/SS | Trichloroethylene (0.15 mM) | Sodium sulfite (SS) | [SS]0 = 1.5 mM, [Fe(II)]0 = 3.0 mM, [PDS]0 = 2.25 mM | 81.1% (pH0 = 6.3) | Wu et al. (2015) |
PMS/Fe(II)/Citrate | 4-chlorophenols (0.396 mM) | Citrate | [citrate]0 = 0.4 mM, [Fe(II)]0 = 0.99 mM, [PMS]0 = 3.96 mM | 27.0% (pH0 = 7.0) | Rastogi et al. (2009a) |
PMS/Fe(II)/EDDS | 4-chlorophenols (0.396 mM) | Ethylenediaminedisuccinate (EDDS) | [EDDS]0 = 0.4 mM, [PMS]0 = [Fe(II)]0 = 0.99 mM | 91.5% (pH0 = 7.0) | Rastogi et al. (2009a) |
PMS/Fe(II)/Pyro | 4-chlorophenols (0.396 mM) | Pyrophosphate (Pyro) | [Pyro]0 = 0.40 mM, [Fe(II)]0 = 0.99 mM, [PMS]0 = 3.96 mM | 91.5% (pH0 = 7.0) | Rastogi et al. (2009a) |
PDS/Fe(II)/CA | Trichloroethylene (0.15 mM) | Citric acid (CA) | [CA]0 = 0.15 mM, [Fe(II)]0 = 3.0 mM, [PDS]0 = 2.25 mM | 100% (pH0 = 4.0) | Wu et al. (2014) |
Process . | Pollutant . | Enhanced agent . | Experimental conditions . | Degradation efficiency . | Ref. . |
---|---|---|---|---|---|
PMS/Fe(II)/HAH | Benzoic acid (40 μM) | Hydroxylamine (HAH) | [HAH]0 = 0.40 mM, [Fe(II)]0 = 10.8 μM, [PMS]0 = 0.32 mM | 80% (pH0 = 3.0) | Zou et al. (2013) |
PDS/Fe(II)/HAH | Trichloroethylene (0.15 mM) | HAH | [HAH]0 = 1.5 mM, [Fe(II)]0 = 3.0 mM, [PDS]0 = 2.25 mM | 97.9% (pH0 = 4.0) | Wu et al. (2015) |
PMS/Fe3O4/HAH | Atrazine (23 μM) | HAH | [HAH]0 = 0.3 mM, [Fe3O4]0 = 0.69 g L−1, [PMS]0 = 0.4 mM | 94% (pH0 = 6.8) | Li et al. (2019a) |
PMS/FeOCl/HAH | Ciprofloxacin (5.0 mg L−1) | HAH | [HAH]0 = 1.0 mM, [FeOCl]0 = 0.2 g L−1, [PMS]0 = 1.5 mM | 74.5% (pH0 = 6.5) | Luo et al. (2021) |
PDS/Fe(II)/STS | Trichloroethylene (0.15 mM) | Sodium thiosulfate (STS) | [STS]0 = 1.5 mM, [Fe(II)]0 = 3.0 mM, [PDS]0 = 2.25 mM | 65.5% (pH0 = 5.5) | Wu et al. (2015) |
PDS/Fe(II)/SS | Trichloroethylene (0.15 mM) | Sodium sulfite (SS) | [SS]0 = 1.5 mM, [Fe(II)]0 = 3.0 mM, [PDS]0 = 2.25 mM | 81.1% (pH0 = 6.3) | Wu et al. (2015) |
PMS/Fe(II)/Citrate | 4-chlorophenols (0.396 mM) | Citrate | [citrate]0 = 0.4 mM, [Fe(II)]0 = 0.99 mM, [PMS]0 = 3.96 mM | 27.0% (pH0 = 7.0) | Rastogi et al. (2009a) |
PMS/Fe(II)/EDDS | 4-chlorophenols (0.396 mM) | Ethylenediaminedisuccinate (EDDS) | [EDDS]0 = 0.4 mM, [PMS]0 = [Fe(II)]0 = 0.99 mM | 91.5% (pH0 = 7.0) | Rastogi et al. (2009a) |
PMS/Fe(II)/Pyro | 4-chlorophenols (0.396 mM) | Pyrophosphate (Pyro) | [Pyro]0 = 0.40 mM, [Fe(II)]0 = 0.99 mM, [PMS]0 = 3.96 mM | 91.5% (pH0 = 7.0) | Rastogi et al. (2009a) |
PDS/Fe(II)/CA | Trichloroethylene (0.15 mM) | Citric acid (CA) | [CA]0 = 0.15 mM, [Fe(II)]0 = 3.0 mM, [PDS]0 = 2.25 mM | 100% (pH0 = 4.0) | Wu et al. (2014) |
Influence of concentration (PMS, Fe(II), Fe(III), and AA) on FLT removal
Influence of PMS concentration
The initial concentrations of Fe(II), AA, and FLT were set at 2.0, 2.8, and 0.2 mg L−1, respectively, while changing the dosage of PMS from 2.0 to 40 mg L−1. Figure S6a shows that FLT removal was increased from 85.0 to 100% with the dosage of PMS increasing from 2.0 to 10 mg L−1, which suggested that an appropriate concentration of PMS can generate more ROS and improve FLT removal. However, FLT removal decreased to 81.1% with further increasing the dosage of PMS to 40 mg L−1, which indicated that the dosage of PMS was not proportional to the degradation of FLT.







Effect of Fe(II)/Fe(III) dosage
For PMS/Fe(II)/AA process, as illustrated in Figure S7a, with Fe(II) concentrations increasing to 4 mg L−1, FLT removal enhanced to 91.1% when PMS, AA, and FLT were set as 5.0, 2.8, and 0.2 mg L−1, respectively. It proved that PMS was activated efficiently to strengthen the production of ROS in the solution following the appropriate increase in Fe(II) concentration. However, the FLT removal was decreased to 38.7% when Fe(II) dosage was sequentially increased to 16 mg L−1.
For the PMS/Fe(III)/AA process, with the concentrations of Fe(III) increasing from 6.0 to 22 mg L−1, FLT removal increased from 80.5 to 96.3% when PMS, AA, and FLT were set to 12, 3.5, and 0.2 mg L−1 in Figure S7b, respectively. It was similar to Fe(II) in that the degradation of FLT barely changed when the concentrations of Fe(III) further increased to 88 mg L−1. For the PMS/Fe(II)/AA process, the excess Fe(II) could consume and
by the reaction of Fe(II) and ROS (Equations (4) and (5)), which can inhibit the degradation of FLT (Rastogi et al. 2009a). For the PMS/Fe(III)/AA process, the excess Fe(III) was first reduced to Fe(II) by AA, and the resulting Fe(II) was identical to the PMS/Fe(II)/AA process, which can act as a scavenger for free radicals. Xu et al. (2022) explored the influence of Fe(0) concentration on benzoic acid (BA) removal in the AA/Fe(0)/PMS process and found that appropriately increasing the dosage of Fe(0) could promote the degradation of BA, but the BA removal remained unchanged in the excess of Fe(0), which was attributed to the fact that the excess of Fe(II) scavenged the free radicals in the system. Rastogi et al. (2009b), in studying the effect of polychlorinated biphenyls (PCBs) removal Fe(II)/PMS process, found that the higher concentration of Fe(II) scavenged the free radicals and thus inhibited PCBs removal. The results of these experiments indicated that for both systems, the appropriate amount of iron was necessary for the effective activation of PMS.
Effect of AA concentration
For PMS/Fe(II)/AA process, the concentrations of PMS, Fe(II), and FLT were set to 5.0, 2.0, and 0.2 mg L−1, respectively. With the dosage of AA increasing from 0 to 2.8 mg L−1, FLT removal enhanced from 35.7 to 90.0%, as shown in Figure S8a. Fe(III) produced after Fe(II) oxidated was reduced by AA to restrict the precipitation of iron and continue activating PMS, which could explain why the degradation of FLT increased after AA was added. However, FLT removal decreased to 68.9% when the dosage of AA was increased to 5.6 mg L−1.






The mechanisms of FLT removal in PMS/Fe(II)/AA and PMS/Fe(III)/AA processes
Detecting ROS by the tests of probe and EPR
It is reported that a variety of ROS (,
, and
) could be produced in the PMS/Fe(II) process, which could be also produced in PMS/Fe(II)/AA and PMS/Fe(III)/AA processes as well (Dong et al. 2021). Anisole was chosen as both
and
probe compounds, CT and nitrobenzene were chosen as
, and
probe compounds to confirm the generated ROS in the process (Ahmad et al., 2012, 2015). In Figure S9a, CT removal indicated the presence of
. Figure S9b shows the presence of HO• because of NB removal. AN was also removed with the addition of excess tertiary butyl alcohol (TBA), which proved the existence of
in both processes (Figure S9c and S9d). In addition, the addition of AA could increase NB, AN, and CT removal, confirming AA can promote the production of ROS.
In this study, the presence of ROS was further demonstrated through EPR testing in both systems. In Figure S10a and Figure S10b, it indicated the presence of and
because of the formation of DMPO-OH and DMPO-SO4 adducts. However, the peaks of adducts composed (DMPO and
) were not detected because of the low stability and concentration of
(Fang et al. 2013).
The role of ROS in FLT degradation

















Effect of scavengers on FLT degradation detected by HPLC in (a) PMS/Fe(II)/AA system ([IPA]0 = [TBA]0 = [CF]0 = 10 mM, [FLT]0 = 0.2 mg L−1, [PMS]0 = 5.0 mg L−1, [Fe(II)]0 = 2.0 mg L−1, and [AA]0 = 2.8 mg L−1) and (b) PMS/Fe(III)/AA system ([IPA]0 = [TBA]0 = [CF]0 = 10 mM, [PMS]0 = 12 mg L−1, [FLT]0 = 0.2 mg L−1, [Fe(III)]0 = 11 mg L−1, and [AA]0 = 3.5 mg L−1).
Effect of scavengers on FLT degradation detected by HPLC in (a) PMS/Fe(II)/AA system ([IPA]0 = [TBA]0 = [CF]0 = 10 mM, [FLT]0 = 0.2 mg L−1, [PMS]0 = 5.0 mg L−1, [Fe(II)]0 = 2.0 mg L−1, and [AA]0 = 2.8 mg L−1) and (b) PMS/Fe(III)/AA system ([IPA]0 = [TBA]0 = [CF]0 = 10 mM, [PMS]0 = 12 mg L−1, [FLT]0 = 0.2 mg L−1, [Fe(III)]0 = 11 mg L−1, and [AA]0 = 3.5 mg L−1).
Influence of water quality conditions on FLT degradation
Effect of aqueous solution pH
For the PMS/Fe(II)/AA system, FLT was removed 58.0, 77.6, 93.4, 82.6, 51.6, and 45.6% with the pH of reaction solution pre-adjusting to 2.0, 3.0, 5.0, 7.0, 9.0, and 11 in Figure S11a, respectively. For PMS/Fe(III)/AA system, the degradations of FLT were 88.2, 100.0, 100.0, 98.0, 89.1, and 19.2% with the aqueous solution pH increasing from 2.0 to 3.0, 5.0, 7.0, 9.0 and 11 in Figure S11b, respectively. The above results show that FLT can be effectively removed for both systems when the initial pH of the solution was 3.0–9.0. FLT could be efficiently degraded under acidic conditions because it could increase the production of sulfate radicals that are more stable and can induce iron dissolution to benefit the Fenton-like oxidation (Pignatello et al. 2006). However, the degradations of FLT decreased at an initial solution pH of 2.0, which proved that the extremely acidic condition was unfavorable to FLT removal. There were explanations for this phenomenon. At extreme acidic pH, AA existed mainly in the protonated form, which was not conducive to chelated iron production (Torres et al. 1982). What is more, the lower efficiency of FLT removal was also attributed to the generation of the iron complex species ([Fe(H2O)6]2+), which was slower to react with PMS than other Fe(II) species (Xu et al. 2009).
In addition, FLT was removed unsatisfactorily in the alkaline environment. In the first place, more stable chelated ferrous ions formed by increasing the solution pH, which was unfavorable for Fe(II) release and thus led to the failure of PMS activation (Davies 1992). Then, Fe(II) and Fe(III) were precipitated at a high pH of the solution, reducing the concentration of dissolved Fe(II) and thus limiting the activation of the PMS. Furthermore, will be consumed by reacting with OH− under alkaline conditions (Yuan et al. 2011), which was the dominant radical for FLT degradation. Finally, AA was decomposed in the alkaline condition, which caused the significant decrease of chelated iron ions and the failure of reduction (Munialo & Kontogiorgos 2014). Munialo and Kontogiorgos (2014) conducted a kinetic study on the decomposition of AA at different solution pH values and indicated that in the aqueous solution, there was an equilibrium between AA and the oxidized product dehydroascorbic acid (DHAA), but increasing solution pH favored the transformation DHAA from AA, which caused the increase of AA decomposition. Tu et al. (2017) investigated the scavenging action of AA on free radicals and discovered that AA performed disproportionation reactions due to internal electron transfer at pH 7.
As shown in Figure S11, the solution pH changed significantly before and after the reaction at the initial pH of 7.0 and 9.0, so further experiments were conducted to investigate the change in solution pH with time after adding the agents. Figure S12 demonstrates that the solution pH decreased rapidly within 10 min after the addition of the agent and then stabilized, which was due to the acidic nature of the added PMS. In conclusion, the performance of FLT removal was better under acidic conditions, and by comparison, the PMS/Fe(III)/AA system was more adaptable to pH changes than the PMS/Fe(II)/AA system. On the one hand, this could be since the hydrolysis of Fe(III) can produce H+ to lower the pH, thus reducing the influence of the initial pH. On the other hand, this was because the results of chelation experiments in AA with iron showed the strong chelating effect of AA on Fe(III) compared to Fe(II) (Figure S5), thus preventing the rapid precipitation of Fe(III) at high initial pH of the solution.
Influence of anions on FLT removal
For the PMS/Fe(II)/AA process, Figure S13a shows that FLT was removed 90.9, 100.0, and 87.0% with the concentrations of Cl− increase from 1.0 to 50 mM, respectively. Analogously, FLT was removed 92.9, 74.0, and 73.1% in Figure S13b when the addition of Cl− increased from 1.0 to 50 mM in the PMS/Fe(III)/AA system. It suggested that FLT removal was hardly affected by Cl− and mildly increased with adding lesser dosages of Cl−. It could be explained by the reasons below. Firstly, Cl− could react with (Equation (1) in Table 2), but the reaction is reversible (Yuan et al. 2011). As shown in Equations (1) and (3) (Table 2), the reaction between Cl− and ROS could generate Cl• (2.4 V) and
(2.0 V) radicals, which could easily react with organic compounds (Yang et al. 2019). HOCl could be produced by the reaction of
,
, and Cl− (Equations (4) and (5) in Table 2), which can accelerate the degradation of FLT (Zhou et al. 2018). In addition,
could react with Cl− causing unnecessary consumption (Equation (2) in Table 2), which was the reason why FLT removal mildly decreased with the dosage of Cl− increasing to 50 mM (Baensch et al. 1991). The above results explained that FLT removal mildly increased with the small dosages of Cl−.
Reactions and rate constants on different anions
Equation . | Reaction . | Rate constant . | Ref. . |
---|---|---|---|
1 | ![]() | kf = (3.2 ± 0.4) × 108 M−1 s−1, kr = (2.1 ± 0.8) × 108 M−1 s−1 | Yuan et al. (2011) |
2 | Cl− + ![]() | kf = (4.3 ± 0.4) × 1010 M−1 s−, kr = (6.1 ± 0.8) × 109 M−1 s−1 | Yuan et al. (2011) |
3 | ![]() | kf = (7.8 ± 0.8) × 107 M−1 s−1, kr = (5.7 ± 0.4) × 104 M−1 s−1 | Yuan et al. (2011) |
4 | ![]() | k = 2.06 × 10−3 M−1 s−1 | Zhou et al. (2018) |
5 | ![]() | k = 3.8 × 10−4M−1 s−1 | Zhou et al. (2018) |
6 | ![]() | k = 3.5 × 109 M−1 s−1 | Wang et al. (2018) |
7 | Br− + ![]() | kf = 1.1 × 1010 M−1 s−1, kr = 3.3 × 107 M−1 s−1 | Wang et al. (2018) |
8 | Br• + Br• → Br2 | k = 1.0 × 109 M−1 s−1 | Wang et al. (2018) |
9 | ![]() | kf = 1.0 × 1010 M−1 s−1, kr = 1.0 × 105 M−1 s−1 | Wang et al. (2018) |
10 | ![]() | k = 1.9 × 109 M−1 s−1 | Wang et al. (2018) |
11 | ![]() | k = 1.9 × 109 M−1 s−1 | Lutze et al. (2015) |
12 | ![]() | k = 1.9 × 109 M−1 s−1 | Lutze et al. (2015) |
Equation . | Reaction . | Rate constant . | Ref. . |
---|---|---|---|
1 | ![]() | kf = (3.2 ± 0.4) × 108 M−1 s−1, kr = (2.1 ± 0.8) × 108 M−1 s−1 | Yuan et al. (2011) |
2 | Cl− + ![]() | kf = (4.3 ± 0.4) × 1010 M−1 s−, kr = (6.1 ± 0.8) × 109 M−1 s−1 | Yuan et al. (2011) |
3 | ![]() | kf = (7.8 ± 0.8) × 107 M−1 s−1, kr = (5.7 ± 0.4) × 104 M−1 s−1 | Yuan et al. (2011) |
4 | ![]() | k = 2.06 × 10−3 M−1 s−1 | Zhou et al. (2018) |
5 | ![]() | k = 3.8 × 10−4M−1 s−1 | Zhou et al. (2018) |
6 | ![]() | k = 3.5 × 109 M−1 s−1 | Wang et al. (2018) |
7 | Br− + ![]() | kf = 1.1 × 1010 M−1 s−1, kr = 3.3 × 107 M−1 s−1 | Wang et al. (2018) |
8 | Br• + Br• → Br2 | k = 1.0 × 109 M−1 s−1 | Wang et al. (2018) |
9 | ![]() | kf = 1.0 × 1010 M−1 s−1, kr = 1.0 × 105 M−1 s−1 | Wang et al. (2018) |
10 | ![]() | k = 1.9 × 109 M−1 s−1 | Wang et al. (2018) |
11 | ![]() | k = 1.9 × 109 M−1 s−1 | Lutze et al. (2015) |
12 | ![]() | k = 1.9 × 109 M−1 s−1 | Lutze et al. (2015) |
In contrast to Cl−, the degradation of FLT decreased to 14.1 and 11.5% with the concentrations of Br− increasing to 50 mM for PMS/Fe(II)/AA and PMS/Fe(III)/AA processes. First of all, the abundant presence of Br−, , and
can largely be consumed through the reaction of Equations (6) and (7) in Table 2, which caused the reduction of ROS in solution to terminate FLT removal. What is more, the radicals (Br• and
) generated by the reaction of Br− with ROS (
and
), further reacted with each other or themselves (Equations (8)–(10) in Table 2) to form Br2 and released into the air.
As shown in Figure S13a and Figure S13b, FLT removal was immensely restricted following the concentrations of increasing from 1.0 to 50 mM. For the PMS/Fe(II)/AA process, the degradation of FLT decreased from 90.0 to 47.6, 36.3, and 5.5% with the concentrations of
increasing from 0 to 50 mM. Similarly, FLT removal decreased from 91.8 to 38.6, 14.8, and 4.8% in the PMS/Fe(III)/AA system. The phenomenon can be explained in two ways. On the one hand,
could react with
and
to decrease the content of ROS through the reaction of Equations (11) and (12) in Table 2. On the other hand, the solution pH was increased in Table S1 when the dosages increased from 1.0 to 50 mM, which caused Fe(II) and Fe(III) to precipitate. In conclusion, for selected three inorganic anions in this study, the presence of
has the largest effect on FLT removal in both systems.
Effect of humic acid on FLT removal


Effect of HA on FLT degradation detected by HPLC in (a) PMS/Fe(II)/AA system ([FLT]0 = 0.2 mg L−1, [PMS]0 = 5.0 mg L−1, [Fe(II)]0 = 2.0 mg L−1, and [AA]0 = 2.8 mg L−1) and (b) PMS/Fe(III)/AA system ([PMS]0 = 12 mg L−1, [FLT]0 = 0.2 mg L−1, [Fe(III)]0 = 11 mg L−1, and [AA]0 = 3.5 mg L−1).
Effect of HA on FLT degradation detected by HPLC in (a) PMS/Fe(II)/AA system ([FLT]0 = 0.2 mg L−1, [PMS]0 = 5.0 mg L−1, [Fe(II)]0 = 2.0 mg L−1, and [AA]0 = 2.8 mg L−1) and (b) PMS/Fe(III)/AA system ([PMS]0 = 12 mg L−1, [FLT]0 = 0.2 mg L−1, [Fe(III)]0 = 11 mg L−1, and [AA]0 = 3.5 mg L−1).
The efficiency of FLT degradation in the groundwater


FLT degradation performance in actual groundwater detected by HPLC in (a) PMS/Fe(II)/AA system ([FLT]0 = 0.2 mg L−1, [PMS]0 = 5.0 mg L−1, [Fe(II)]0 = 2.0 mg L−1, and [AA]0 = 2.8 mg L−1) and (b) PMS/Fe(III)/AA system ([PMS]0 = 12 mg L−1, [FLT]0 = 0.2 mg L−1, [Fe(III)]0 = 11 mg L−1, and [AA]0 = 3.5 mg L−1).
FLT degradation performance in actual groundwater detected by HPLC in (a) PMS/Fe(II)/AA system ([FLT]0 = 0.2 mg L−1, [PMS]0 = 5.0 mg L−1, [Fe(II)]0 = 2.0 mg L−1, and [AA]0 = 2.8 mg L−1) and (b) PMS/Fe(III)/AA system ([PMS]0 = 12 mg L−1, [FLT]0 = 0.2 mg L−1, [Fe(III)]0 = 11 mg L−1, and [AA]0 = 3.5 mg L−1).
Sheng et al. (2022) investigated the degradation effect of sodium percarbonate (SPC)/Fe(II) system on FLT, and 90.3% of FLT in the actual groundwater was degraded after 15 min of reaction when the dosages of SPC, Fe(II) were 0.15 and 1 mM, respectively. In this study, for the PMS/Fe(II)/AA system, the degradation of FLT was 74.8% after 120 min when the dosages of PMS, Fe(II), and AA were 0.04, 0.04, and 0.16 mM, respectively. For the PMS/Fe(III)/AA system, the FLT degradation rate was 100% after 120 min when the dosages of PMS, Fe(III), and AA were 0.025, 0.05, and 0.025 mM, respectively. The above comparisons demonstrated that the PMS/Fe(II)/AA and PMS/Fe(III)/AA systems used significantly less reagent than the previously studied methods, implying that the AA-enhanced technology presents some advantages for the remediation of FLT contamination. Besides, most of the reagents used in the above research were procured from Macklin Biochemical Co. Ltd (Shanghai, China) and Aladdin Reagent Co. Ltd (Shanghai, China), PMS was about twice as expensive as SPC, Fe(III) and Fe(II) were about the same price. In conclusion, degradation of FLT using AA enhancement technology was more economical compared to previous studies.
Possible FLT removal pathways

Degradation pathways of FLT in PMS/Fe(II)/AA and PMS/Fe(III)/AA systems detected by GC–MS.
Degradation pathways of FLT in PMS/Fe(II)/AA and PMS/Fe(III)/AA systems detected by GC–MS.
CONCLUSIONS
In this study, the addition of AA significantly enhanced the degradation of FLT in PMS/Fe(II) and PMS/Fe(III) systems with molar ratios of 2/2/4 and 5/10/5 for PMS/Fe(II)/AA and PMS/Fe(III)/AA systems, respectively. It was elaborated that the degradation of FLT was enhanced by promoting the production of ROS due to the reduction and chelation of AA through the detection of Fe(II) concentration and the demonstration experiments of chelated iron. For PMS/Fe(II)/AA and PMS/Fe(III)/AA systems, the results from the probe and scavenging experiments suggested that was the dominant ROS for FLT removal. The validation and comparison of the effects of both systems on the degradation of FLT under different groundwater matrixes indicated that the PMS/Fe(III)/AA system had better adaptability than the PMS/Fe(II)/AA system. The intermediates of FLT degradation, such as 1-phenylnaphthalene, o-xylene, dibutyl phthalate, and naphthalene, were detected by GC–MS, and the possible pathways of FLT degradation were proposed. In the actual groundwater degradation experiments, the degradation rates of FLT in the PMS/Fe(II)/AA and PMS/Fe(III)/AA systems reached 88.8 and 100.0%, respectively, and the dosage of chemicals in PMS/Fe(III)/AA system was less than that required in PMS/Fe(II)/AA system due to the better adaptability of PMS/Fe(III)/AA system for different water conditions.
ACKNOWLEDGEMENTS
This study was financially supported by a grant from the National Natural Science Foundation of China (No. 41977164).
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
CONFLICT OF INTEREST
The authors declare there is no conflict.