Mineralization of ammunition wastewater by a micron-size Fe⁰/O₃ process (mFe⁰/O₃)

Abstract

A micron-size Fe⁰/O₃ process (mFe⁰/O₃) was set up to mineralize the pollutants in ammunition wastewater, and its key operational parameters (e.g., initial pH, ozone flow rate, and mFe⁰ dosage) were optimized by the batch experiments, respectively. Under the optimal conditions, COD removal efficiency obtained by the mFe⁰/O₃ process (i.e., 92.6% after 30 min treatment) was much higher than those of ozone alone (46.5%), mFe⁰ alone (38.3%) or mFe⁰/air (58.5%), which confirm the synergetic effect between mFe⁰ and ozone. In addition, the BOD₅/COD (B/C) ratio was elevated from 0 to 0.54 after 30 min treatment by the mFe⁰/O₃ process, which indicates the significant improvement of biodegradability. Furthermore, the analysis results of the UV-vis and excitation–emission matrix (EEM) fluorescence spectra further confirm that the toxic and refractory pollutants in ammunition wastewater had been completely decomposed or transformed into smaller molecule organic compounds. Meanwhile, the superiority of the mFe⁰/O₃ process has been confirmed according the analysis results of COD removal, B/C ratio, UV-vis and EEM. Therefore, the mFe⁰/O₃ process could be proposed as a promising treatment technology for toxic and refractory ammunition wastewater.

---

1. Introduction

Ammunition wastewater is mainly generated from the manufacturing and demilitarization of ammunition, which usually contains nitroaromatic compounds such as TNT (2,4,6-trinitrotoluene), RDX (1,3,5-trinitro-1,3,5-triazacyclohexane), or HMX (1,3,5,7-tetranitro-1,3,5,7-tetrazocane).^1–4 Due to the toxicity, carcinogenicity and mutagenicity of the nitroaromatic compounds in ammunition wastewater, this wastewater is generally recalcitrant to biological treatment.^5,6 With the development of the ammunition manufacturing industry, a large amount of ammunition wastewater could be generated and discharged into the environment. If the untreated ammunition wastewater is directly discharged into the receiving water, it would severely threaten the ecosystem and human health. Therefore, adequate treatment of the ammunition wastewater is necessary before discharge to the environment.

In recent years many treatment technologies have been used to decompose these nitroaromatic compounds in wastewater, such as solvent extraction,⁷ adsorption,⁸ combined ultrasound and Fenton (US-Fenton) process,¹ electro-Fenton oxidation,⁹ alkaline hydrolysis,¹⁰ vacuum distillation,¹¹ supercritical water oxidation.¹² Although these methods could obtain a high performance for the pollutants removal, their engineering application for the ammunition wastewater was limited by rigorous reaction conditions or high operating cost. In addition, the conventional biological treatment processes are not effective to mineralize the nitroaromatic compounds in ammunition wastewater because the electron withdrawing nitro constituents in these compounds inhibit the electrophilic attack through enzymes.¹³ Thus it is necessary to develop a cost-effective ammunition wastewater treatment system.

Zero valent iron (ZVI or Fe⁰) has been extensively applied for the remediation and treatment of groundwater or industry wastewater contaminated with nitroaromatics, chlorinated organic compounds, dyes, or nitrate.^14–17 Furthermore, Fe⁰ has been combined with other methods (e.g., ultrasound, Fenton, or ozone) to decompose the refractory pollutants.^18–20 In particular, combined Fe⁰ and Fenton has been successfully used to treat explosive chemicals producing wastewater (e.g., 2,4-dinitroanisole (DNAN), RDX or TNT).^21–23 In recent years, several Fe-based materials as the catalysts were investigated in catalytic ozonation process, which shown approving performance for organic contaminants degradation.^24–26 Degradation of azo dyes in aqueous solution by ozone integrated with ZVI or internal micro-electrolysis have been reported in literatures, which suggest that the combined Fe⁰ and ozone was a high-effective process.^20,27–29 However, there is no report about the mineralization of nitroaromatic compounds in ammunition wastewater by using a combined Fe⁰ and ozone process (i.e., Fe⁰/O₃).

In this study, therefore, a micron-size Fe⁰/O₃ process (mFe⁰/O₃) was developed to treat the ammunition wastewater. The primary objective of this study is to investigate the contribution of synergetic effect between mFe⁰ and ozone for the mineralization of toxic and refractory pollutants in ammunition wastewater. In addition, the key operating parameters (i.e., initial pH, ozone flow rate, mFe⁰ dosage and treatment time) of the mFe⁰/O₃ process were investigated, respectively. Finally, mineralization of the pollutants in ammunition wastewater was investigated thoroughly by analyzing the variation of COD, BOD₅, UV-vis and excitation–emission matrix (EEM) fluorescence spectra before and after mFe⁰/O₃ system treatment.

2. Materials and methods

2.1. Materials

Ammunition wastewater used in this study was obtained from an ammunition factory in southwest China. Its COD and initial pH was 503.8 mg L⁻¹ and 7.8, respectively. Micron-size zero valent iron (mFe⁰) powders from Chengdu Kelong chemical reagent factory was used in this experiment. The mFe⁰ powders have mean particle size of approximately 0.12 mm, and their iron content is above 99%. The mFe⁰ powder presents a Brunauer–Emmet–Teller (BET) specific surface area of 3.11 m² g⁻¹ and an average pore width of 3.27 nm. Additionally, X-ray diffraction (XRD) pattern of mFe⁰ powder as shown in Fig. S1 in ESI,† which indicates that almost all the composition of the mFe⁰ powder is zero valence iron due to the iron content is above 99%. Other chemicals used in the experiment were of analytical grade. Deionized water was used throughout the whole experiment process.

2.2. Batch experiments

Treatment of ammunition wastewater by the mFe⁰/O₃ process was carried out by batch experiments. In each batch experiment, mFe⁰ powders and 300 mL ammunition wastewater were added into a 500 mL beaker, and the slurry was stirred by a mechanical stirrer with a stirring speed of 300 rpm. Meanwhile, ozone gas was dispersed into the reacting solution through a gas diffuser at the bottom of the beaker. Ozone gas was generated onsite from pure compressed dry oxygen (99.9%, v/v) by a laboratory model ozone generator (10 g h⁻¹, Chengdu Yifeng Co., Ltd., China), and the concentration of the generated ozone was about 5.42 mg L⁻¹. The whole experiment process was performed at a running temperature of 25 ± 1 °C by heating in water bath. Effects of initial pH (1.0–10.0), ozone flow rate (0–600 mL min⁻¹), mFe⁰ dosage (0–40 g L⁻¹) and treatment time (0–60 min) on the treatment efficiency of the mFe⁰/O₃ process were investigated, respectively.

To confirm the superiority of mFe⁰/O₃ process and the synergistic effect between mFe⁰ and ozone, three control experiments were set up, (a) O₃, (b) mFe⁰, (c) mFe⁰/air. Meanwhile, the operating conditions of control experiments were in accordance with the obtained optimal conditions of the mFe⁰/O₃ process. In particular, mFe⁰ dosage of (b) mFe⁰ and (c) mFe⁰/air was same to that of the mFe⁰/O₃ process. Ozone flow rate of (a) O₃ was same to that of the mFe⁰/O₃ process. Air flow rate of (c) mFe⁰/air was same to ozone flow rate of the mFe⁰/O₃ process. The samples were withdrawn at pre-determined time intervals, and diluted by deionized water and filtered through a PTFE syringe filter disc (0.45 μm). Finally, COD, BOD₅, UV-vis and EEM fluorescence spectra of the samples were measured, respectively.

2.3. Analytical methods

COD, BOD₅ and pH of the samples were measured by COD rapid analyzer (Lianhua, China), BOD₅ analyzer (OxiTop IS12, WTW, Germany) and pHS-3C meter (Rex, China), respectively. The structure of mFe⁰ was detected based on the BET method using a surface area analyzer (ASAP2020, USA). The compound composition of mFe⁰ was analyzed by X-ray diffraction (XRD) EMPYREAN diffractometer (PANalytical B.V., Holland). The UV-vis absorption spectra (Shimazu, Japan) of the influent and effluent were carried out in 10 mm quartz cuvettes, and the UV-vis spectra were recorded from 190 to 500 nm using deionized water as blank. The excitation–emission matrix (EEM) fluorescence spectra were measured by a luminescence spectrometry (F-7000 spectrophotometer, Hitachi, Japan), and the EEM spectra were collected with the corresponding scanning emission spectra from 200 to 550 nm at 5 nm increments by varying the excitation wavelength from 200 to 400 nm at 5 nm sampling intervals. The excitation and emission slits were kept at 10 nm and the scanning speed was set at 1200 nm min⁻¹.³⁰ The samples were diluted 5 times and filtered through a PTFE syringe filter disc (0.45 μm) before the analysis of EEM florescence spectroscopy. Chromaticity of the samples was measured by using dilution multiple method. Ozone concentrations in solution were measured by the iodometric method.³¹

3. Results and discussion

3.1. Optimization of operational parameters

3.1.1. Effect of initial pH. On the one hand, it is clear that the initial pH can significantly affected the concentration of residual ozone, offgas, and the utilization efficiency of ozone.³² In particular, hydroxide ions can initiate ozone decomposition to generate hydroxyl radicals (HO˙) and this process can be artificially enhanced by increasing the pH.³³ For example, Zhao and his co-workers have found that degradation efficiency of nitrobenzene in aqueous solution is greatly enhanced with increasing pH from 2.89 to 12.46 in the ozone alone system at reaction temperature 25 °C.³² On the other hand, the initial pH is an important parameter for the mFe⁰ process because the increase of H⁺ ions concentration can significantly accelerate the corrosion rate of mFe⁰. Furthermore, it can facilitate the generation of corrosion products and Fenton-like reaction under oxic condition.^34,35 The previous studies find that the low initial pH can help to remove the pollutants.^36,37 Therefore, it is necessary to investigate the influence of initial pH on the mineralization efficiency of ammunition wastewater by the mFe⁰/O₃ process.

Fig. 1(a) shows the COD removal efficiency of ammunition wastewater is gradually enhanced with decreasing pH from 10.0 to 2.0 in the mFe⁰/O₃ process. However, it is well known that oxidation capacity of ozone alone can be artificially accelerated by increasing the pH because hydroxide ions can initiate ozone decomposition to generate HO˙.³³ The converse variation trend can be explained from two aspects, (a) the lower initial pH can accelerate the iron corrosion rate and rapidly generate enough corrosion products (e.g., Fe²⁺/Fe³⁺, FeOOH, Fe₂O₃ or Fe₃O₄), which can catalyze ozone decomposition to generate HO˙;^38–41 (b) the lower initial pH can also facilitate the formation of Fenton-like reaction under oxic condition (eqn (1) and (2)),⁴² which can improve the mineralization of the pollutants in ammunition wastewater.

Fe⁰ + O₂ + 2H⁺ → Fe²⁺ + H₂O₂ (1)

Fe²⁺ + H₂O₂ → HO˙ + OH⁻ + Fe³⁺ (2)

Fig. 1 Effect of (a) initial pH, (b) ozone flow rate, (c) mFe⁰ dosage and (d) treatment time on the COD removal efficiencies and effluent pH after mFe⁰/O₃ treatment (influent COD = 503.8 mg L⁻¹, O₃ concentration = 5.42 mg L⁻¹).

However, COD removal efficiency began to decrease when the initial pH was below 2.0 (Fig. 1(a)), which can be explained from three aspects, (a) the excess H⁺ ions present in the solution can act as hydroxyl radical scavenger;⁴³ (b) the too low initial pH (<2.0) results in the excess Fe²⁺ that also can act as hydroxyl radical scavenger;⁴⁴ (c) the excess H⁺ ions also can inhibit the decomposition of ozone.³²

Fig. 1(a) also shows the effect of initial pH on the effluent pH. In particular, the effluent pH was higher than the initial pH when the initial pH was below 6.0. On the contrary, the effluent pH was lower than the initial pH when the initial pH was above 6.0. The results can be explained from three aspects, (a) the iron corrosion or Fenton-like reaction could be limited seriously under the neutral or alkaline condition,⁴⁵ thus the ozonation played a leading role in the mineralization of the pollutants in ammunition wastewater. Meanwhile the acidic intermediates (e.g., carboxylic acids) would be generated during the ozonation process, which could decrease the pH of reaction solution;³² (b) under acidic condition, H⁺ ions could be rapidly consumed by iron corrosion reaction, which facilitate the formation of Fenton-like reaction and corrosion products;⁴² (c) with the increase of pH, the corrosion products and hydroxide ions could catalyze the decomposition of ozone to generate much more HO˙.^32,46 In other words, the maximum COD removal efficiency (92.6%) obtained at the initial pH of 2.0 was mainly attributed to the synergistic effect of mFe⁰ (e.g., Fenton-like) and ozone (e.g., HO˙ from catalytic ozonation). As a result, the optimal initial pH of 2.0 was selected to optimize the following parameters.

3.1.2. Effect of ozone flow rate. The operating cost of the mFe⁰/O₃ process is mainly attributed to the ozone dosage, which can be control through optimizing the ozone flow rate. Fig. 1(b) shows that COD removal efficiency was rapidly improved from 38.3% to 92.6% with the increasing of ozone flow rate from 0 to 300 mL min⁻¹, and then it was gradually increased to 95.6% when ozone flow rate was further increased to 600 mL min⁻¹. The results suggest that the ozone dosage was not a limited factor when the ozone flow rate was above 300 mL min⁻¹. The increase of ozone flow rate can increase the dissolved ozone in the liquid phase, which facilitates to increase the yield of HO˙. Thus mineralization of the pollutants in ammunition wastewater could be significantly improved with the increase of ozone flow rate. The COD of ammunition wastewater was decreased from 503.8 to 37.3 mg L⁻¹ after 30 min treatment by the mFe⁰/O₃ process with an ozone flow rate of 300 mL min⁻¹, and only a few of small molecule intermediates were remained in the effluent. The small molecule intermediates were hard to be mineralized completely by catalytic ozonation.⁴⁶ In addition, it would be an uneconomic strategy if the mineralization was performed through further increasing ozone flow rate (i.e., >300 mL min⁻¹). Fig. 1(b) also shows that all the effluent pH (>4.0) of the mFe⁰/O₃ process with different ozone flow rate (0–600 mL min⁻¹) was higher than the initial pH (2.0), but the effluent pH (5.0) of the mFe⁰ process without ozone aeration was higher than that obtained under ozone aeration condition. The conversion of pH was performed from two aspects, (a) some organic acids were generated during the ozonation process; (b) the corrosion of iron could consume the H⁺ ions in reactive solution. In addition, the effluent pH was close to 5.0 when the ozone flow rate was 500 or 600 mL min⁻¹, which might be attributed to the rapid mineralization of the organic acids by the higher ozone flow rate. Thus the optimal ozone flow rate of 300 mL min⁻¹ was selected to further optimize the following parameters.

3.1.3. Effect of mFe⁰ dosage. Fig. 1(c) shows the influence of mFe⁰ dosage on the COD removal efficiencies of the ammunition wastewater by the mFe⁰/O₃ process. In particular, the low COD removal efficiency (46.5%) was obtained after 30 min treatment by the ozonation process without mFe⁰, while it could rapidly increase to 83.7% even if a few of mFe⁰ powders (2.5 g L⁻¹) were added in treatment system. Furthermore, the COD removal would gradually increase to the maximum (92.6%) with the further increasing of mFe⁰ dosage from 2.5 g L⁻¹ to 20 g L⁻¹. And then it would not be further improved when the mFe⁰ dosage was above 20 g L⁻¹. The results can be explained from four aspects, (a) ozonation could be enhanced significantly through adding mFe⁰ in reactive system; (b) the increased reactive sites, surface area and microscopic galvanic cells formation with the increasing of mFe⁰ dosage, which could enhance the formation of Fenton-like reaction under oxic condition;⁴² (c) adsorption of mFe⁰ and its corrosion products would be enhanced with the increase of mFe⁰ dosage; (d) the effluent pH increased from 1.9 to 5.7 with the increasing of mFe⁰ dosage from 0 to 40 g L⁻¹ (Fig. 1(c)), which also indicate that the yield of corrosion products is proportional to the mFe⁰ dosage. Subsequently, the more mFe⁰ dosage was added in treatment system, the more corrosion products (e.g., Fe²⁺/Fe³⁺, FeOOH, Fe₂O₃ and Fe₃O₄) were generated, which could improve the catalytic ozonation.⁴⁶ The maximum COD removal efficiency (92.6%) was obtained by the mFe⁰/O₃ process with mFe⁰ dosage of 20.0 g L⁻¹, thus the optimal mFe⁰ dosage of 20.0 g L⁻¹ was chosen to optimize the following parameter.

3.1.4. Effect of treatment time. Fig. 1(d) shows that COD removal efficiency rapidly increased to 56.9% in the initial 2 min, and then it gradually increased to 92.6% in the following 28 min, finally it only further increased a little at 30–60 min. The rapid removal of the pollutants in the initial 2 min was mainly attributed to adsorption of the mFe⁰ and its corrosion products, and then the pollutants were gradually mineralized by the mFe⁰/O₃ process. Fig. 1(d) also shows that in the initial 3 min, pH of reaction solution rapidly increased from 2.0 to 4.0, and then it began to decrease from 4.0 to 3.5 (3–15 min), finally it gradually increased to 5.6 (15–60 min). The rapid increase of pH in the initial 3 min was mainly attributed to the rapid consume of H⁺ ions by the iron corrosion reaction. Furthermore, the pH decrease at 3–15 min was mainly resulted from the generated and accumulated acidic intermediates (e.g., carboxylic acid) in reaction solution. Finally, the pH increase at 15–60 min could be explained from two aspects, (a) the generated and accumulated acidic intermediates were further gradually mineralized; (b) the H⁺ ions would be consumed by the corrosion of mFe⁰. Since the increase of treatment time would increase the ozone dosage and operating cost, the optimal treatment time was 30 min.

3.2. Synergetic effect between mFe⁰ and ozone

To evaluate the superiority of the mFe⁰/O₃ process and synergetic effect between mFe⁰ and ozone, three control experiments including (a) ozone alone, (b) mFe⁰ alone, (c) mFe⁰/air were set up and performed under the same conditions (i.e., initial pH of 2.0, gas flow rate (air or ozone) of 300 mL min⁻¹, mFe⁰ dosage of 20 g L⁻¹, and treatment time of 30 min).

As shown in Fig. 2(a), during the 30 min treatment process, the COD removal efficiencies obtained by the mFe⁰/O₃ process were much higher than those of three control experiments. In particular, after 30 min treatment, COD removal of the mFe⁰/O₃ process (92.6%) was much higher than that of mFe⁰ alone (38.3%), ozone alone (46.5%), or mFe⁰/air (58.5%). In addition, Fig. 2(b) shows that the COD removal efficiencies obtained by the mFe⁰/O₃ process were higher than the sum of mFe⁰ alone and ozone alone. The logarithmic plots of residual concentration of COD in reaction solution versus the reaction time are shown in Fig. 2(c), which illustrates that a good linear fitting was observed in each batch experiment. The results indicate that the COD removal efficiencies obtained by the different treatment processes are all in accordance with the pseudo-first-order kinetics model, but they have different K_obs. According to Fig. 2(c) and (d), it is clear that K_obs (0.322 min⁻¹) obtained by the mFe⁰/O₃ process was much higher than those of three control experiments at the initial phase (0–3 min). Meanwhile, the variation of K_obs at the second phase (3–30 min) is similar to that at the initial phase (0–3 min).

Fig. 2 COD removal efficiencies and degradation kinetics of different treatment processes (influent COD = 503.8 mg L⁻¹, mFe⁰ dosage = 20 g L⁻¹, initial pH = 2.0, O₃ concentration = 5.42 mg L⁻¹, ozone flow rate = 300 mL min⁻¹, reaction time = 30 min).

In the mFe⁰/air process, mineralization of the pollutants is mainly attributed to the Fenton-like reaction that is formed from the reduction of oxygen through accepting the electrons released from iron corrosion (eqn (1) and (2)).⁴² Besides the Fenton-like reaction, however, catalytic ozonation also plays a leading role in the mineralization process. In addition, the similar results were reported when the dyes (e.g., reactive red X-3B, azo dye RR2, and Reactive Black 5) were degraded by ozone integrated with internal micro-electrolysis.^27–29 In the mFe⁰/O₃ process, the iron corrosion products (e.g., Fe²⁺, Fe³⁺, Fe₂O₃, Fe₃O₄, FeOOH, etc.) could be used as catalyst for the decomposition of ozone, and generate free radicals to mineralize the pollutants.^39,47–49 In addition, the organic acids would be generated when the pollutants was decomposed by the catalytic ozonation,⁴⁶ and these acids could enhance the corrosion rate of Fe⁰ which facilitate the Fenton-like reaction.^34,50 Therefore, it could be concluded that there is strong synergetic effect between mFe⁰ and ozone, and the mFe⁰/O₃ process is much superior to the existing mFe⁰/air process.

In addition, the sequential experiments including ozone (30 min) + mFe⁰ (30 min) and mFe⁰ (30 min) + ozone (30 min) were setup and performed under the identical conditions (i.e., initial pH of 2.0, ozone flow rate of 300 mL min⁻¹, mFe⁰ dosage of 20 g L⁻¹) to further confirm the synergetic effect of mFe⁰ and ozone. As shown in Fig. S3,† the results demonstrate that the COD removal obtained by the mFe⁰/O₃ process (i.e., 92.6%, 30 min of treatment) was much higher than that of sequential ozone + mFe⁰ (i.e., 67.1%, 60 min of treatment) or mFe⁰ + ozone (i.e., 84.1%, 60 min of treatment), even the reaction time of the sequential experiments was twice of mFe⁰/O₃ process. The results further confirm that high treatment efficiency of mFe⁰/O₃ process was mainly attributed to the synergetic effect between mFe⁰ and ozone. Additionally, the released of iron ions (e.g., Fe²⁺ or Fe³⁺) could catalyze the decomposition of ozone and accelerate the degradation efficiency, which induce the rapidly increasing of COD removal by mFe⁰ + ozone treatment during 30–60 min.

3.3. UV-vis spectra analysis

To further confirm the superiority of the mFe⁰/O₃ process, the effluent from the mFe⁰/O₃ process and other three control experiments (i.e., ozone alone, mFe⁰ alone, and mFe⁰/air) was determined by using an UV-vis absorption spectrometer. As shown in Fig. 3, the UV absorbance peak of ammunition wastewater was mainly located between 190 and 300 nm, which was attributed to the nitroaromatic compounds (e.g., TNT).⁸ It can be observed that the absorbance peak intensity of the effluent from these four experiments all dropped with respect to the influent, and they have the following trend towards the intensity: ozone alone > mFe⁰ alone > mFe⁰/air > mFe⁰/O₃. The results suggest that the mFe⁰/O₃ process is more effective than other three treatment processes, and it could completely decompose the nitroaromatic pollutants in ammunition wastewater.

Fig. 3 UV-vis spectra of ammunition wastewater by different treatment process (experimental conditions were same to Fig. 2).

3.4. EEM spectra analysis

Three-dimensional EEM fluorescence spectroscopy is a rapid and sensitive technique for the analysis of dissolved organic matter (DOM), and it had been successfully used to detect the changes and transformation of aromatic pollutants in our previous work.^30,51 Thus the changes and transformation of nitroaromatic pollutants in ammunition wastewater during the degradation process can also be rapidly analyzed by the EEM fluorescence spectroscopy.

The obtained EEM fluorescence spectra are shown in Fig. 4 and S1 (ESI†). Meanwhile, the fluorescence parameters of the spectra including peak locations, fluorescence intensity, peak intensity ratios and total intensity are summarized in Tables 1 and 2. Fig. 4(a) and Table 1 show that only a peak with low intensity of 283 a.u. was observed at the Ex/Em of 300/410 nm in the EEM fluorescence spectrum of raw ammunition wastewater. However the aromatics usually have two strong fluorescence peaks located at the similar emission wavelength according to our previous works.^51,52 The low fluorescence intensity of ammunition wastewater were mainly attributed to the fluorescence quenching of the nitro group in the molecular structure of nitroaromatic pollutants.⁵³

Fig. 4 Variation of EEM fluorescence spectra, (a) influent, (b) effluent of mFe⁰/air process (30 min), (c) effluent of mFe⁰/O₃ process (30 min), (d) variation of total intensity during 30 min treatment process.

Table 1 Fluorescence spectral parameters of DOM of the influent and effluent of the mFe⁰/air process at different time points

Time (min)	Peak A		Peak B		Peak C		Peak D		Peak intensity ratio			Total intensity (a.u.)
	Ex/Em (nm)	Intensity (a.u.)	Ex/Em (nm)	Intensity (a.u.)	Ex/Em (nm)	Intensity (a.u.)	Ex/Em (nm)	Intensity (a.u.)	A/B	A/C	A/D
0	300/410	283	—	—	—	—	—	—	—	—	—	283
1	300/420	1689	245/395	1067	275/390	1020	—	—	1.58	1.66	—	3776
2	300/410	1084	240/400	912	—	—	—	—	1.19	—	—	1996
3	300/410	1146	240/395	949	—	—	—	—	1.21	—	—	2095
4	295/410	1143	400/395	924	—	—	—	—	1.24	—	—	2067
5	295/410	1134	240/395	885	—	—	—	—	1.28	—	—	2019
10	300/410	1208	240/400	993	—	—	—	—	1.22	—	—	2201
15	300/410	1332	235/405	1155	—	—	—	—	1.15	—	—	2487
20	295/410	1405	240/395	1207	275/300	1217	—	—	1.16	1.15	—	3829
25	300/410	1503	240/400	1342	275/390	1421	270/300	1710	1.12	1.06	0.88	5976
30	295/410	1565	240/395	1395	275/385	1592	270/300	1710	1.12	0.98	0.92	6262

---

Table 2 Fluorescence spectral parameters of DOM of the influent and effluent of the mFe⁰/O₃ process at different time points

Time (min)	Peak A		Peak B		Peak C		Peak D		Peak intensity ratio			Total intensity (a.u.)
	Ex/Em (nm)	Intensity (a.u.)	Ex/Em (nm)	Intensity (a.u.)	Ex/Em (nm)	Intensity (a.u.)	Ex/Em (nm)	Intensity (a.u.)	A/B	A/C	A/D
0	300/410	283	—	—	—	—	—	—	—	—	—	283
1	315/430	1027	250/400	707	275/395	776	—	—	1.45	1.32	—	2510
2	300/415	876	245/400	650	—	—	—	—	1.35	—	—	1526
3	305/430	941	244/400	632	—	—	—	—	1.49	—	—	1573
4	300/415	858	240/400	562	—	—	—	—	1.53	—	—	1420
5	295/410	801	240/400	528	—	—	—	—	1.52	—	—	1329
10	295/410	754	235/405	475	—	—	355/445	542	1.59	—	1.39	1771
15	295/405	563	235/395	367	—	—	355/445	454	1.53	—	1.24	1384
20	295/410	443	—	—	—	—	360/445	314	—	—	1.41	757
25	295/410	319	—	—	—	—	355/440	207	—	—	1.54	526
30	295/410	229	—	—	—	—	—	—	—	—	—	229

---

Fig. 4(b) and Table 1 show that 4 new peaks were occurred in the fluorescence spectrum of the effluent after 30 min treatment by the mFe⁰/air process, and its total fluorescence intensity reached up to 6262 a.u. As shown in Fig. 4(d), during the 30 min treatment process by the mFe⁰/air, its total fluorescence intensity was rapidly elevated to 3776 a.u. in the initial 1 min, and then it quickly decreased to about 2000 a.u. and maintained 4 min. After 5 min treatment, it began to increased from 2019 to 6262 a.u. during the following 25 min (i.e., 5–30 min). The results could be explained that the fluorescence quenching problem was resolved due to the removal of nitro groups in the molecular structure of nitroaromatic pollutants. The high fluorescence intensity of the effluent suggests that the intermediates with conjugative effect were accumulated during the treatment process by the mFe⁰/air. In the treatment process by the mFe⁰/O₃, its total fluorescence intensity has a similar trend to the mFe⁰/air in the initial 10 min, but it began to decreased from 1771 to 229 a.u. in the following 20 min (i.e., 10–30 min). The results indicate that the intermediates with conjugative effect could be further mineralized effectively by the mFe⁰/O₃ process. In other words, the mFe⁰/O₃ process is much more effective than the mFe⁰/air process for ammunition wastewater treatment.

According to the above results, it can be concluded that the degradation process of the nitroaromatic pollutants in ammunition wastewater can be rapidly monitored by using EEM fluorescence spectral technology. In addition, the EEM fluorescence spectra analysis also confirm that a high COD removal efficiency (92.6%) obtained by mFe⁰/O₃ system was valid.

3.5. Discharge standard and improvement of biodegradability

After 30 min treatment by the mFe⁰/O₃ system under the optimal conditions, COD, BOD₅ and colority of the effluent were 37.3 mg L⁻¹, 20.0 mg L⁻¹ and 40 times, respectively. Meanwhile, the nitroaromatic pollutants (e.g., TNT) were not detected in the effluent of the mFe⁰/O₃ system. Thus the main water quality indexes of the effluent come up to the effluent standards of water pollutants for ammunition loading industry (GB 14470.3-2011, COD ≤ 60.0 mg L⁻¹, BOD₅ ≤ 20.0 mg L⁻¹, colority ≤ 40, TNT ≤ 0.5).

As shown in Fig. 5, the B/C ratio of ammunition wastewater increased from 0 to 0.54 after 30 min treatment by the mFe⁰/O₃ process under the optimal conditions, which was approximately triple of that obtained by the mFe⁰/air process (0.20). The results indicate that the toxic and refractory pollutants in ammunition wastewater could be effectively mineralized or transformed by the mFe⁰/O₃ process, and its biodegradability was improved significantly.

Fig. 5 The COD removal efficiency and B/C ratio of effluent after mFe⁰/O₃ or mFe⁰/air treatment ([COD]₀ = 503.8 mg L⁻¹, influent B/C ratio = 0).

3.6. Reuse of mFe⁰ in the reaction system

To evaluate the reusability of the mFe⁰ in Fe⁰/O₃ process, the reuse of mFe⁰ was investigated by running five successive catalytic ozonation experiments under the same optimal conditions (i.e., initial pH of 2.0, ozone flow rate of 300 mL min⁻¹, and mFe⁰ dosage of 20 g L⁻¹). At the end of each cycle, mFe⁰ powder was separated from the reaction solution and collected, and then was gently rinsed three times with deionized water and dried at 25 °C in the vacuum drying oven. As shown in Fig. 6, the COD removal efficiencies were all above 89% in each cycle, and no evident decline was observed after five successive mFe⁰/O₃ experiments, which manifest the approving recyclability of the mFe⁰.

Fig. 6 The COD removal for five successive mFe⁰/O₃ experiments.

3.7. Analysis of operating cost

The operating cost of the mFe⁰/O₃ process was mainly from the consumption of mFe⁰ and ozone. Under the optimal operating conditions, the total iron concentration in the effluent of the mFe⁰/O₃ system was about 74 mg L⁻¹, and the COD removal was about 485 mg L⁻¹. Thus it can be calculated that mFe⁰ consumption values was about 0.15 kg mFe⁰ per kg COD. In addition, ozone with the gas flow rate of 300 mL min⁻¹ was continuously dispersed into the 300 mL reacting solution for 30 min, and ozone gas was generated onsite from pure compressed dry oxygen (99.9%, v/v) by a laboratory model ozone generator. Ozone consumption (OC) is defined as the amount of ozone consumed for a given mass of COD removed.^54,55 According to eqn (3), it can be calculated that OC value was about 2.61 m³ O₃ per kg COD after 30 min treatment by the mFe⁰/O₃ process. The market prices of iron powders and pure compressed dry oxygen (99.9%, v/v) were about 280 USD per t and 0.5 USD per m³, respectively. Thus the operating cost was about 1.34 USD per kg COD. As a results, the mFe⁰/O₃ process is an cost-effective treatment technology for the ammunition wastewater.where Q_G is the ozone gas flow rate, V is the sample volume, C_AG is the off gas ozone concentration, C_AG0 is the input ozone concentration, t is the reaction time, and COD₀ and COD are the initial and effluent COD, respectively.

4. Conclusions

In this study, the toxic and refractory ammunition wastewater was treated by the mFe⁰/O₃ process, and its crucial operational parameters including initial pH, ozone flow rate, and mFe⁰ dosage were optimized by the batch experiments. Under the optimal conditions (i.e., initial pH of 2.0, ozone flow rate of 300 mL min⁻¹, and mFe⁰ dosage of 20 g L⁻¹), COD removal efficiency obtained by the mFe⁰/O₃ process (i.e., 92.6%) was much higher than those of ozone alone (46.5%), mFe⁰ alone (38.3%) or mFe⁰/air (58.5%). Meanwhile, the COD removal efficiencies obtained by the mFe⁰/O₃ process were higher than the sum of mFe⁰ alone and ozone alone during the whole 30 min treatment process. The results indicate that high treatment efficiency was mainly attributed to the synergetic effect between mFe⁰ and ozone. In addition, the B/C ratio was elevated from 0 to 0.54 after 30 min treatment by the mFe⁰/O₃ process, which indicates the significant improvement of biodegradability. Furthermore, the analysis results of UV-vis and EEM fluorescence spectra further confirm the superiority of the mFe⁰/O₃ process. Finally, the operating cost of the mFe⁰/O₃ process was about 1.34 USD per kg COD. Therefore, the mFe⁰/O₃ process is a cost-effective treatment technology for the ammunition wastewater.

Acknowledgements

The authors would like to acknowledge the financial support from National Natural Science Foundation of China (No. 21207094), Fundamental Research Funds for the Central Universities (No. 2015SCU04A09) and Special S&T Project on Treatment and Control of Water Pollution (No. 2012ZX07201-005).

References

1. Y. Li, W. P. Hsieh, R. Mahmudov, X. Wei and C. P. Huang, Combined ultrasound and Fenton (US-Fenton) process for the treatment of ammunition wastewater, J. Hazard. Mater., 2013, 244–245, 403–411.
2. S. C. Ahn, D. K. Cha, B. J. Kim and S.-Y. Oh, Detoxification of PAX-21 ammunitions wastewater by zero-valent iron for microbial reduction of perchlorate, J. Hazard. Mater., 2011, 192, 909–914.
3. M. Astratov, A. Preiβ, K. Levsen and G. Wünsch, Identification of pollutants in ammunition hazardous waste sites by thermospray HPLC/MS, Int. J. Mass Spectrom. Ion Processes, 1997, 167–168, 481–502.
4. E. Atikovic, M. T. Suidan and S. W. Maloney, Anaerobic treatment of army ammunition production wastewater containing perchlorate and RDX, Chemosphere, 2008, 72, 1643–1648.
5. B. R. Flokstra, B. V. Aken and J. L. Schnoor, Microtox® toxicity test: detoxification of TNT and RDX contaminated solutions by poplar tissue cultures, Chemosphere, 2008, 71, 1970–1976.
6. B. Lachance, A. Y. Renoux, M. Sarrazin, J. Hawari and G. I. Sunahara, Toxicity and bioaccumulation of reduced TNT metabolites in the earthworm Eisenia andrei exposed to amended forest soil, Chemosphere, 2004, 55, 1339–1348.
7. W.-S. Chen, W.-C. Chiang and C.-C. Lai, Recovery of nitrotoluenes in wastewater by solvent extraction, J. Hazard. Mater., 2007, 145, 23–29.
8. Q. Zhao, Y. Gao and Z. Ye, Reduction of COD in TNT red water through adsorption on macroporous polystyrene resin RS 50B, Vacuum, 2013, 95, 71–75.
9. W.-S. Chen and S.-Z. Lin, Destruction of nitrotoluenes in wastewater by electro-Fenton oxidation, J. Hazard. Mater., 2009, 168, 1562–1568.
10. A. Saupe, H. J. Garvens and L. Heinze, Alkaline hydrolysis of TNT and TNT in soil followed by thermal treatment of the hydrolysates, Chemosphere, 1998, 36, 1725–1744.
11. Q. Zhao, Z. Ye and M. Zhang, Treatment of 2,4,6-trinitrotoluene (TNT) red water by vacuum distillation, Chemosphere, 2010, 80, 947–950.
12. S.-j. Chang and Y.-c. Liu, Degradation mechanism of 2,4,6-trinitrotoluene in supercritical water oxidation, J. Environ. Sci., 2007, 19, 1430–1435.
13. B. Clark and R. Boopathy, Evaluation of bioremediation methods for the treatment of soil contaminated with explosives in Louisiana Army Ammunition Plant, Minden, Louisiana, J. Hazard. Mater., 2007, 143, 643–648.
14. Y. L. Jiao, C. C. Qiu, L. H. Huang, K. X. Wu, H. Y. Ma, S. H. Chen, L. M. Ma and D. L. Wu, Reductive dechlorination of carbon tetrachloride by zero-valent iron and related iron corrosion, Appl. Catal., B, 2009, 91, 434–440.
15. Y. Mu, H. Q. Yu, J. C. Zheng, S. J. Zhang and G. P. Sheng, Reductive degradation of nitrobenzene in aqueous solution by zero-valent iron, Chemosphere, 2004, 54, 789–794.
16. M. F. Hou, F. B. Li, X. M. Liu, X. G. Wang and H. F. Wan, The effect of substituent groups on the reductive degradation of azo dyes by zerovalent iron, J. Hazard. Mater., 2007, 145, 305–314.
17. T. Suzuki, M. Moribe, Y. Oyama and M. Niinae, Mechanism of nitrate reduction by zero-valent iron: equilibrium and kinetics studies, Chem. Eng. J., 2012, 183, 271–277.
18. M. Barreto Rodrigues, F. T. Silva and T. C. B. Paiva, Optimization of Brazilian TNT industry wastewater treatment using combined zero-valent iron and Fenton processes, J. Hazard. Mater., 2009, 168, 1065–1069.
19. B. Lai, Z. Y. Chen, Y. X. Zhou, P. Yang, J. L. Wang and Z. Q. Chen, Removal of high concentration p-nitrophenol in aqueous solution by zero valent iron with ultrasonic irradiation (US-ZVI), J. Hazard. Mater., 2013, 250–251, 220–228.
20. F. Pan, Y. Luo, J.-J. Fan, D.-C. Liu and J. Fu, Degradation of disperse Blue E-4R in aqueous solution by zero-valent iron/ozone, Clean: Soil, Air, Water, 2012, 40, 422–427.
21. J. Shen, C. Ou, Z. Zhou, J. Chen, K. Fang, X. Sun, J. Li, L. Zhou and L. Wang, Pretreatment of 2,4-dinitroanisole (DNAN) producing wastewater using a combined zero-valent iron (ZVI) reduction and Fenton oxidation process, J. Hazard. Mater., 2013, 260, 993–1000.
22. M. Barreto Rodrigues, F. T. Silva and T. C. B. Paiva, Combined zero-valent iron and Fenton processes for the treatment of Brazilian TNT industry wastewater, J. Hazard. Mater., 2009, 165, 1224–1228.
23. S.-Y. Oh, P. C. Chiu, B. J. Kim and D. K. Cha, Enhancing Fenton oxidation of TNT and RDX through pretreatment with zero-valent iron, Water Res., 2003, 37, 4275–4283.
24. A. Ziylan and N. H. Ince, Catalytic ozonation of ibuprofen with ultrasound and Fe-based catalysts, Catal. Today, 2015, 240, 2–8.
25. A. Cano Quiroz, C. Barrera-Díaz, G. Roa-Morales, P. Balderas Hernández, R. Romero and R. Natividad, Wastewater Ozonation Catalyzed by Iron, Ind. Eng. Chem. Res., 2011, 50, 2488–2494.
26. R. Huang, H. Yan, L. Li, D. Deng, Y. Shu and Q. Zhang, Catalytic activity of Fe/SBA-15 for ozonation of dimethyl phthalate in aqueous solution, Appl. Catal., B, 2011, 106, 264–271.
27. X. B. Zhang, W. Y. Dong, F. Y. Sun, W. Yang and J. Dong, Degradation efficiency and mechanism of azo dye RR2 by a novel ozone aerated internal micro-electrolysis filter, J. Hazard. Mater., 2014, 276, 77–87.
28. X.-C. Ruan, M.-Y. Liu, Q.-F. Zeng and Y.-H. Ding, Degradation and decolorization of reactive red X-3B aqueous solution by ozone integrated with internal micro-electrolysis, Sep. Purif. Technol., 2010, 74, 195–201.
29. X. Guo, Y. Cai, Z. Wei, H. Hou, X. Yang and Z. Wang, Treatment of diazo dye C.I. Reactive Black 5 in aqueous solution by combined process of interior microelectrolysis and ozonation, Water Sci. Technol., 2013, 67, 1880–1885.
30. B. Lai, Y. X. Zhou, J. L. Wang, Z. S. Yang and Z. Q. Chen, Application of excitation and emission matrix fluorescence (EEM) and UV-vis absorption to monitor the characteristics of Alizarin Red S (ARS) during electro-Fenton degradation process, Chemosphere, 2013, 93, 2805–2813.
31. J. A. Roth and D. E. Sullivan, Solubility of Ozone in Water, Ind. Eng. Chem. Fundam., 1981, 20, 137–140.
32. L. Zhao, J. Ma, Z. Sun and X. Zhai, Mechanism of Influence of Initial pH on the Degradation of Nitrobenzene in Aqueous Solution by Ceramic Honeycomb Catalytic Ozonation, Environ. Sci. Technol., 2008, 42, 4002–4007.
33. J. Staehelin and J. Hoigne, Decomposition of ozone in water: rate of initiation by hydroxide ions and hydrogen peroxide, Environ. Sci. Technol., 1982, 16, 676–681.
34. W. Yin, J. Wu, P. Li, X. Wang, N. Zhu, P. Wu and B. Yang, Experimental study of zero-valent iron induced nitrobenzene reduction in groundwater: the effects of pH, iron dosage, oxygen and common dissolved anions, Chem. Eng. J., 2012, 184, 198–204.
35. A. Shimizu, M. Tokumura, K. Nakajima and Y. Kawase, Phenol removal using zero-valent iron powder in the presence of dissolved oxygen: roles of decomposition by the Fenton reaction and adsorption/precipitation, J. Hazard. Mater., 2012, 201–202, 60–67.
36. J. Dong, Y. S. Zhao, R. Zhao and R. Zhou, Effects of pH and particle size on kinetics of nitrobenzene reduction by zero-valent iron, J. Environ. Sci., 2010, 22, 1741–1747.
37. J.-L. Chen, S. R. Al-Abed, J. A. Ryan and Z. Li, Effects of pH on dechlorination of trichloroethylene by zero-valent iron, J. Hazard. Mater., 2001, 83, 243–254.
38. A. Lv, C. Hu, Y. Nie and J. Qu, Catalytic ozonation of toxic pollutants over magnetic cobalt and manganese co-doped γ-Fe₂O₃, Appl. Catal., B, 2010, 100, 62–67.
39. A. Lv, C. Hu, Y. Nie and J. Qu, Catalytic ozonation of toxic pollutants over magnetic cobalt-doped Fe₃O₄ suspensions, Appl. Catal., B, 2012, 117–118, 246–252.
40. Y. Nie, C. Hu, N. Li, L. Yang and J. Qu, Inhibition of bromate formation by surface reduction in catalytic ozonation of organic pollutants over β-FeOOH/Al₂O₃, Appl. Catal., B, 2014, 147, 287–292.
41. R. Sauleda and E. Brillas, Mineralization of aniline and 4-chlorophenol in acidic solution by ozonation catalyzed with Fe²⁺ and UVA light, Appl. Catal., B, 2001, 29, 135–145.
42. K. S. Wang, C. L. Lin, M. C. Wei, H. H. Liang, H. C. Li, C. H. Chang, Y. T. Fang and S. H. Chang, Effects of dissolved oxygen on dye removal by zero-valent iron, J. Hazard. Mater., 2010, 182, 886–895.
43. L. Gomathi Devi, S. Girish Kumar, K. Mohan Reddy and C. Munikrishnappa, Photo degradation of methyl orange an azo dye by advanced Fenton process using zero valent metallic iron: influence of various reaction parameters and its degradation mechanism, J. Hazard. Mater., 2009, 164, 459–467.
44. M. S. Lucas and J. A. Peres, Decolorization of the azo dye Reactive Black 5 by Fenton and photo-Fenton oxidation, Dyes Pigm., 2006, 71, 236–244.
45. B. Lai, Y. X. Zhou, H. K. Qin, C. Y. Wu, C. C. Pang, Y. Lian and J. X. Xu, Pretreatment of wastewater from acrylonitrile–butadiene–styrene (ABS) resin manufacturing by microelectrolysis, Chem. Eng. J., 2012, 179, 1–7.
46. J. Nawrocki and B. Kasprzyk-Hordern, The efficiency and mechanisms of catalytic ozonation, Appl. Catal., B, 2010, 99, 27–42.
47. F. J. Beltran, F. J. Rivas and R. Montero-de-Espinosa, Iron type catalysts for the ozonation of oxalic acid in water, Water Res., 2005, 39, 3553–3564.
48. L. Yang, C. Hu, Y. Nie and J. Qu, Surface acidity and reactivity of β-FeOOH/Al₂O₃ for pharmaceuticals degradation with ozone: in situ ATR-FTIR studies, Appl. Catal., B, 2010, 97, 340–346.
49. T. Zhang, C. Li, J. Ma, H. Tian and Z. Qiang, Surface hydroxyl groups of synthetic α-FeOOH in promoting OH generation from aqueous ozone: property and activity relationship, Appl. Catal., B, 2008, 82, 131–137.
50. I. A. Katsoyiannis, T. Ruettimann and S. J. Hug, pH Dependence of Fenton Reagent Generation and As(III) Oxidation and Removal by Corrosion of Zero Valent Iron in Aerated Water, Environ. Sci. Technol., 2008, 42, 7424–7430.
51. B. Lai, Y. X. Zhou and P. Yang, Treatment of wastewater from acrylonitrile–butadiene–styrene (ABS) resin manufacturing by biological activated carbon (BAC), J. Chem. Technol. Biotechnol., 2013, 88, 474–482.
52. B. Lai, Y. X. Zhou, P. Yang, J. H. Yang and J. L. Wang, Degradation of 3,3′-iminobis-propanenitrile in aqueous solution by Fe⁰/GAC micro-electrolysis system, Chemosphere, 2013, 90, 1470–1477.
53. J. Xu and Z. Wang, Fluorimetry, Science Press, Beijing, 2006.
54. C. Tizaoui, L. Bouselmi, L. Mansouri and A. Ghrabi, Landfill leachate treatment with ozone and ozone/hydrogen peroxide systems, J. Hazard. Mater., 2007, 140, 316–324.
55. S. S. Abu Amr, H. A. Aziz, M. N. Adlan and M. J. K. Bashir, Pretreatment of stabilized leachate using ozone/persulfate oxidation process, Chem. Eng. J., 2013, 221, 492–499.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra06135d

---