Treatment of wastewater derived from dinitrodiazophenol (DDNP) manufacturing by the Fe/Cu/O₃ process

Abstract

In this paper, the Fe/Cu bimetallic particles and ozone were combined to decompose or transform the toxic and refractory pollutants in dinitrodiazophenol (DDNP) wastewater. Firstly, operational parameters including theoretical Cu mass loading (TML_Cu), Fe/Cu dosage, initial pH, O₃ flow rate and treatment time were optimized respectively. The maximum COD removal efficiency (85.3%) and color removal efficiency (95.0%) were obtained under the optimal conditions (i.e., theoretical Cu mass loading (TML_Cu) = 0.02 g Cu per g Fe, Fe/Cu dosage = 20 g L⁻¹, initial pH = 5.0, O₃ flow rate = 0.3 L min⁻¹, treatment time = 15 min). Then, in order to confirm the superiority of the Fe/Cu/O₃ process, three control experiment systems including Fe/Cu, O₃ and Fe⁰/O₃ processes were set up under the same conditions. In addition, the UV-vis and FTIR results confirm the main pollutants in DDNP wastewater were oxidized and generated intermediates, which revealed the superiority of Fe/Cu/O₃ process. Thus, the DDNP wastewater could be treated by Fe/Cu/O₃ process in a promising way.

---

1. Introduction

Dinitrodiazophenol (DDNP) is widely used in the civil explosive industry for the production of primary explosive detonators, and it is mainly composed of a benzene ring, nitro group (NO₂⁻) and azo group (–N=N–).¹ During the DDNP production process, residual DDNP and its by-products would enter the wastewater stream, and they have carcinogenic, teratogenic and mutagenic characteristics. The color of DDNP wastewaters is extremely high (>10 000 times).² If DDNP wastewater is released into the receiving water directly, it would threaten ecological systems and public health. Physicochemical treatment processes including supercritical water oxidation,³ electro-catalytic oxidation,⁴ and active carbon adsorption⁵ could be used to treat the toxic and refractory DDNP wastewater. However, either their low removal efficiency or high cost limits their application. In our previous work, combined Fe⁰ (or Fe/Cu) and Fenton process was developed to treat DDNP wastewater and obtained a good treatment efficiency.^2,6 But it was a sophisticated treatment process with three processing units (e.g., 1^st Fe/Cu/air-2^nd Fenton-3^rd Fe/Cu/air), and the generated corrosion products was poorly settled because a plenty of Fe²⁺ ions were not oxidized to Fe³⁺. Meanwhile, a plenty of corrosion products (Fe²⁺/Fe³⁺) would cause secondary pollution. For example, total iron ions concentration in the effluent of 1^st Fe/Cu/air reached 890 mg L⁻¹.⁶ To overcome the above faults, therefore, it is necessary to develop a cost-effective process for the treatment of DDNP wastewater.

Fe⁰ used as an effective reductant (E₀ = −0.44 eV) has been successfully applied for the treatment of wastewater contaminated heavy metals, dyes, nitrate or phenol.^7–9 Furthermore, Fe/Cu bimetallic developed from Fe⁰ can overcomes the drawback of strait reacting pH and intrinsic passive layer.^10,11 Since the high standard potential difference (0.78 V) between Cu and Fe,^12–14 the metal copper could visibly improve the reactivity of Fe⁰. Two types of reaction mechanisms of the Fe/Cu bimetallic have been proposed in the previous work, (a) direct reduction by receiving electrons generated from the oxidation reaction,^13,15 (b) indirect reduction by the atomic hydrogen (H_abs) absorbed on the surface of bimetallic reductants.^16,17 In addition, a Fenton-like reaction would be occurred in the presence of dissolved oxygen (DO) due to the reduction of DO in Fe/Cu/air process.¹⁸ Fe/Cu bimetallic was mainly used to treat wastewater containing refractory organic substances^19,20 or lower concentration organohalides.²¹ Therefore, Fe/Cu bimetallic was superior to the conventional Fe⁰.

Ozone (O₃), a robust oxidizing agent (E₀ = 2.08 eV), is widely used in wastewater pretreatment. With the participation of catalyst, it generates superoxygen radical (O₂˙⁻) and hydroxyl radical (HO˙) with a strong oxidation capacity, which can decompose organic and inorganic molecules instantly.^22,23 In literature, ozonation can proceed via two routes: (a) direct reactions by molecular ozone, (b) indirect reactions by generating hydroxyl radical (HO˙) through ozone decomposition.²⁴ Since the reaction of sole ozonation could not achieve high mineralization for organic compounds, the catalytic ozonation was proposed as a promising way.^25,26 Furthermore, it has been reported that the ozonation could be integrated by internal micro-electrolysis filter (or ZVI), and have a high treatment efficiency for the pollutants.^27–29 Meanwhile, the synergistic effect between microscale Fe⁰ and ozonation was confirmed in our previous work, which play a leading role in the degradation of the pollutants. Since the reactivity of Fe/Cu bimetallic particles is far superior to Fe⁰, the ozonation in the present of Fe/Cu bimetallic particles would have a much better performance for degradation of the pollutants.

In literature, a microscale Fe⁰/O₃ process had been used to degrade p-nitrophenol,³⁰ the results suggested that high COD removal efficiency was obtained by mFe⁰/O₃ process. In the mFe⁰/O₃ process, Fe⁰ and O₃ had synergistic effect, p-nitrophenol was removed by homogeneous catalytic ozonation, heterogeneous catalytic ozonation, Fenton-like, adsorption and precipitate. Therefore, DDNP wastewater might be treated by the combined Fe/Cu and O₃ process.

In this paper, the combined Fe/Cu and ozone (Fe/Cu/O₃) process was set up to treat DDNP wastewater. Effects of theoretical Cu mass loading (TML_Cu), Fe/Cu dosage, initial pH, O₃ flow rate and reaction time on the treatment efficiency were evaluated systematically. Under the optimal operating conditions, three control experiments (i.e., Fe/Cu, O₃ and Fe⁰/O₃) were comparatively investigated to confirm the superiority of Fe/Cu/O₃ process. Finally, the treatment efficiency of DDNP wastewater were analyzed and evaluated through COD, BOD₅, B/C ratio, NO₃⁻ ions concentration, iron ions concentration, UV-vis and FTIR.

2. Materials and methods

2.1 Materials

The DDNP industry wastewater was came from an explosive industry in the southwest of China. Its pH, BOD₅, COD and color index were 9.6, 0, 1079 mg L⁻¹ and 12 000 times, respectively. Zero valent iron powders (mean particle size were approximately 120 μm, iron content above 98%), CuSO₄·5H₂O, NaOH and H₂SO₄ used in the experiment were came from Chengdu Kelong chemical reagent factory. All chemicals used in the experiment were of analytical grade. Throughout the whole experiment process, deionized water was used.

2.2 Experiment process

In the experiment of preparation, Fe/Cu bimetallic particles were prepared by displacement plating, via adding the iron particles to 200 mL CuSO₄ solution.^11,31 After adding the iron particles to 200 mL CuSO₄ solution, the slurry was stirred for 10 min by a mechanical stirrer with a stirring speed of 250 rpm. The displacement plating process was performed with certain temperature (30 ± 1 °C) through water bath heating. Next, the Fe/Cu bimetallic particles were departed from the supernatant after about 5 min precipitation. Ultimately, the Fe/Cu bimetallic particles were washed three times by deionized water. The mean particle size of Fe/Cu bimetallic particles were approximately 120–125 μm when the theoretical Cu mass loading increased from 0 to 0.24 g Cu per g Fe.¹¹

In each batch experiment, a 500 mL beaker was putted 300 mL DDNP wastewater. The prepared Fe/Cu particles were added meanwhile ozone gas was successively spread into the solution. Ozone gas produced by a laboratory ozone generator (the maximum generation flow rate of 10 g h⁻¹, Chengdu Yifeng, China) was generated from pure compressed dry oxygen (99.9%, v/v), and the ozone gas concentration was 5.42 mg L⁻¹. At the same time, to ensure the totally fluidization of Fe/Cu particles in beaker, the solution was stirred with a stirring speed of 300 rpm by a mechanical stirrer. The temperature of the experiment was controlled at 25 ± 1 °C by heating in a water bath. All the samples filtered through a PTFE syringe filter disc (0.45 μm). All experiments were run in triplicates.

Effects of TML_Cu (0–0.2424 g Cu per g Fe), Fe/Cu dosage (0–40 g L⁻¹), initial pH (3.0–8.0), O₃ flow rate (0–0.8 L min⁻¹) and reaction time (0–30 min) on the treatment efficiency of the Fe/Cu/O₃ process were investigated, respectively. Under the obtained optimal conditions, three control experiments including (a) Fe/Cu, (b) O₃, (c) Fe⁰/O₃ were set up to confirm the superiority of Fe/Cu/O₃ process. Furthermore, other operating conditions of control experiments were in accordance with those of the Fe/Cu/O₃ process. Meanwhile, UV-vis and FTIR were used to measure the influent and effluent.

2.3 Analytical methods

The UV-vis absorption spectra (Shimazu, Japan) of the influent and effluent were carried out in 10 mm quartz cuvettes, and it was recorded from 190 to 600 nm. The samples were diluted 15 times and filtered through a PTFE syringe filter disc (0.45 μm) before the analysis of UV-vis absorption spectra. Fourier transform infrared spectroscopy (FTIR) was used to confirm the degradation of the pollutants in DDNP wastewater by Fe/Cu/O₃ process. And each sample was scanned four times between the wavenumbers of 4000 to 400 cm⁻¹. Total iron ions concentration of the effluent was analyzed using an atomic absorption spectroscopy (AA-6300, Shimadzu, Japan). The NO₃⁻ ions concentration was analyzed by using an ion chromatography (IC, ICS1100, Dionex USA). The oxidation–reduction potential (ORP) of solution was detected by a redox electrode assembly (Sinomeasure, China). The pH, BOD₅ and COD of the samples were determined by using pHS-3C meter (Rex, China), BOD₅ analyzer (OxiTop IS12, WTW, Germany) and COD analyzer (Lianhua, China), respectively. The samples determined by COD analyzer were adjusted to pH 9 to ensure the total precipitation of iron ions, then filtered and adjusted to acidity. The color index of the samples was measured by using dilution multiple method.³² Ozone concentrations were measured by the iodometric and indigo method.³³

3. Results and discussion

3.1 Optimization of operational parameters

3.1.1 Effect of theoretical Cu mass loading (TML_Cu). Cu mass loading on the Fe⁰ surface was an important influencing factor for the reactivity of the Fe/Cu bimetallic particles.²¹ Subsequently, Fe/Cu bimetallic particles with different TML_Cu would affect the treatment efficiency of Fe/Cu/O₃ process for DDNP wastewater. Therefore, it is necessary to optimize the TML_Cu in this study.

COD and color removal of DDNP wastewater with different TML_Cu are shown in Fig. 1(a), which illustrates that COD removal enhanced rapidly from 49.2% to 77.2% with the TML_Cu increased from 0 to 0.0204 g Cu per g Fe, and then tardily improved to 80.9% with the TML_Cu of 0.2424 g Cu per g Fe. Color removal efficiencies have a similar variation trend with the COD removal efficiencies, increasing rapidly from 40.0% to 90.0% with the TML_Cu increasing from 0 to 0.0204 g Cu per g Fe and slowly increasing to 95.0% with the TML_Cu of 0.1096 g Cu per g Fe and then keeping stable.

Fig. 1 Effect of (a) TML_Cu, (b) Fe/Cu dosage, (c) initial pH, (d) O₃ flow rate and (e) reaction time on COD and color removal for DDNP wastewater by Fe/Cu/O₃ treatment.

The results can be explained that Cu plated on iron surface could accelerate the corrosion of Fe⁰ through the formation of a plenty of galvanic couples between Cu (cathode) and Fe⁰ (anode), which could release enough electrons and iron corrosion products (e.g., Fe²⁺/Fe³⁺, Fe₂O₃, FeOOH, etc.).^11,12 In Fe⁰/air process, the dissolved oxygen (DO) can accept the released electrons and generates H₂O₂, which further form the Fenton-like reaction with the catalysis of Fe²⁺/Fe³⁺.³⁴ Meanwhile, it could be deducted that ozone can also accept the released electrons and generate H₂O₂ or other active materials. In addition, iron corrosion products (e.g., Fe²⁺/Fe³⁺, Fe₂O₃, FeOOH, etc.) can catalyze the ozone decomposition and generate radicals (e.g., O₂˙⁻ and HO˙).²⁴ Furthermore, Cu, CuO and CuFe₂O₄ can also catalyze the ozonation reaction and improve the performance of Fe/Cu/O₃ process.^24,35 When the TML_Cu was above 0.0204 g Cu per g Fe, however, it might be not the key limiting factor for catalytic ozonation. This phenomenon also appeared in our previous work when the pollutants were treated by Fe/Cu/air process.¹¹ Thus, the optimal TML_Cu of Fe/Cu bimetallic particles used in Fe/Cu/O₃ system is 0.0204 g Cu per g Fe, and this optimal value would be performed in the subsequent batch experiments.

3.1.2 Effect of Fe/Cu dosage. Fig. 1(b) shows the effect of Fe/Cu dosage (i.e., 0–40 g L⁻¹) on COD and color removal efficiencies of DDNP wastewater by Fe/Cu/O₃ process. COD removal efficiencies increased from 25.7% to 77.2% with the increased dosage (from 0 g L⁻¹ to 20 g L⁻¹), while it changed slightly after the dosage was above 20 g L⁻¹. Color removal efficiencies have a similar variation trend, increasing from 24.0% to 90.0% with the Fe/Cu dosage increasing from 0 to 20 g L⁻¹ and then keeping steady when the Fe/Cu dosage increasing from 20 to 40 g L⁻¹. The similar results have been occurred in our previous work when the pollutants were treated by Fe⁰/air or Fe/Cu/air process.^2,11 Therefore, 20 g L⁻¹ was selected as the optimal Fe/Cu dosage to be used in the following experiments.

The rapid increase of COD and color removal efficiencies could be explained by the fact that the total surface area and reactive sites on the surface of Fe/Cu bimetallic particles increased rapidly with the increasing of Fe/Cu dosage, which facilitate the ozone decomposition to generate a plenty of hydroxyl radicals. In addition, the more Fe/Cu dosage was added in reaction solution, the more corrosion products would be generated rapidly, which could improve catalytic ozonation. Also, Fe/Cu bimetallic particles and their corrosion can remove pollutants by Fenton-like, adsorption, co-precipitation, and size-exclusion.^36,37 When the dosage was above 20 g L⁻¹, reaction rate might be limited by the mass transportation rate of pollutants, active materials, intermediates and products between Fe/Cu and solution phase.^38,39

3.1.3 Effect of initial pH. As shown in Fig. 1(c), the COD removal efficiencies increased from 69.7% to 78.4% with the increased initial pH (from 3.0 to 5.0), and then decreased to 47.8% with the increased initial pH from 5.0 to 8.0. Color removal efficiencies have a similar variation trend, it has a good efficiencies when the pH was 3.0 to 7.0. In addition, the highest color removal efficiency (i.e., 92.5%) was obtained when initial pH was 5.0. Therefore, the optimal initial pH should be 5.0, and this optimal value would be performed in the subsequent batch experiments.

The results suggest that the lower or higher pH (i.e., <4.0 or >7.0) would inhibit the performance of Fe/Cu/O₃ process. However, the acidic condition usually enhanced the treatment efficiency of Fe⁰, Fe/Cu, Fe⁰/air, Fe/Cu/air because a plenty of H⁺ ions could accelerate the iron corrosion process.^34,40 The alkaline condition can improve the performance of ozonation because hydroxide ions can initiate ozone decomposition.⁴¹ For example, Zhao and his colleagues 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 298 K.⁴² In a word, ozonation and Fe/Cu have different demand for the initial pH. Thus, the optimal initial pH of Fe/Cu/O₃ process would be an equilibrium point (i.e., pH of 5.0) of ozonation and Fe/Cu. The results could be attribute to the interaction, (i) the low pH can enhance the iron corrosion and generate enough corrosion products (e.g., Fe²⁺, Fe³⁺, Fe₃O₄, FeOOH, etc.), which could be used for catalytic ozonation.^43,44 In addition, consumption of H⁺ ions in solution by Fe/Cu could increase pH which facilitate the ozone decomposition.⁴². (ii) Degradation of the pollutants by ozonation would generate the organic acids that can take part in the conversion of initial pH, which could accelerate the iron corrosion.

3.1.4 Effect of O₃ flow rate. As shown in Fig. 1(d), the COD removal efficiencies rapidly improved from 59.1% to 78.4% with the increase of O₃ flow rate from 0.1 to 0.3 L min⁻¹, and then it gradually reached 87.1% when the O₃ flow rate further increased to 0.8 L min⁻¹. Color removal efficiencies rapidly increased from 60% to 92.5% with the O₃ flow rate increased from 0.1 to 0.3 L min⁻¹, and then it slowly reached 96.7% when the O₃ flow rate further increased to 0.8 L min⁻¹. The results suggest that the optimal O₃ flow rate should be 0.3 L min⁻¹, and the higher COD (78.4%) and color (92.5%) removal efficiencies were obtained.

With the increasing of the O₃ flow rate, more ozone was dissolved to the liquid phase, which could yield more hydroxyl radicals to decompose the pollutants.^45,46 On the one hand, the enough hydroxyl radicals were generated and could non-selectively oxidize the pollutants, when the O₃ flow rate reached 0.3 L min⁻¹. On the other hand, the excess hydroxyl radicals would be generated when the O₃ flow rate was above 0.3 L min⁻¹. The excess hydroxyl radicals would be scavenged through their recombination or reacting with ozone (eqn (1) and (2)).⁴⁵ In addition, the ozone concentration in solution would be closed to saturation when the O₃ flow rate was above 0.3 L min⁻¹. Therefore, 0.3 L min⁻¹ was selected as the optimal O₃ flow rate because of the high ozone process cost.

HO˙ + HO˙ → H₂O₂ (1)

HO˙ + O₃ → O₂ + HO₂˙ (2)

3.1.5 Effect of reaction time. Under the above optimal condition, effect of reaction time on the treatment of DDNP wastewater by Fe/Cu/O₃ process was investigated thoroughly. Fig. 1(e) shows that COD removal efficiencies increased rapidly from 22.4% to 85.3% with the increased time (from 1 to 15 min), then it was improved slightly to 88.6% when the time was increased to 30 min. The color removal efficiencies also rapidly increased from 16.7% to 95.0% with the time increased from 1 to 15 min, and then it slowly increased to 98.0% when the reaction time was further increased to 30 min.

The higher COD (85.3%) and color (95.0%) removal efficiencies could be obtained after only 15 min treatment by the Fe/Cu/O₃ process, while only a little improvement could be obtained when we doubled the reaction time (i.e., 30 min). However, the twice treatment time would cause twice treatment cost. Therefore, the optimal treatment time should be 15 min.

3.2 The superiority and synergistic effect of Fe/Cu/O₃ process

To prove the superiority of the Fe/Cu/O₃ process and synergistic effect between Fe/Cu and ozone, three control experiments including (a) Fe/Cu, (b) O₃, (c) Fe⁰/O₃, were setup under the same conditions (i.e., initial pH = 5.0, O₃ flow rate = 0.3 L min⁻¹, treatment time = 15 min, TML_Cu = 0.02 g Cu per g Fe, Fe/Cu or Fe⁰ dosage = 20 g L⁻¹). In particular, Fe/Cu and O₃ control experiments were used to confirm the synergistic effect between Fe/Cu and O₃ in Fe/Cu/O₃ process, while Fe⁰/O₃ control experiment was used to confirm the superiority of Fe/Cu/O₃ process.

3.2.1 COD and color removal. Fig. 2(a) indicates that the COD removal obtained by Fe/Cu/O₃ process (85.3%) was much higher than that of Fe/Cu (42.1%), O₃ (38.4%) or Fe⁰/O₃ (72.0%). Also, the sum of COD removal (i.e., 80.5%) obtained by Fe/Cu alone and O₃ alone was lower than the Fe/Cu/O₃ process (i.e., 85.3%), which shows the synergistic effect between Fe/Cu and O₃. Meanwhile, Fe/Cu/O₃ process (85.3%) was superior to Fe⁰/O₃ process (72.0%). Fig. 2(b) shows the similar results for the color removal. In a brief, the color removal obtained by Fe/Cu/O₃ process (95.0%) was much better than that of Fe/Cu (37.5%), O₃ (54.2%) or Fe⁰/O₃ (79.0%).

Fig. 2 COD removal efficiencies, color removal efficiencies, effluent pH variation and ORP variation during the degradation process of different methods (COD₀ = 1079 mg L⁻¹, TML_Cu = 0.02 g Cu per g Fe, Fe/Cu dosage = 20 g L⁻¹, initial pH = 5.0, O₃ flow rate = 0.3 L min⁻¹, treatment time = 15 min, stirring rate = 300 rpm).

3.2.2 pH and total iron ions concentration in the effluent. Fig. 2(c) shows the change of pH during the 15 min treatment process. It is clear that pH of reaction solution decreased rapidly from 5.0 to 2.7 after 15 min treatment by the O₃ alone process, while pH of reaction solution increased rapidly from 5.0 to 7.7 after only 7.5 min treatment by the Fe/Cu alone process. The organic pollutants was easy to be transferred to acidic organic compounds that were hard to be further decomposed by ozone alone, which would cause the accumulation of organic acids and decrease of pH.⁴⁷ The decrease of pH would seriously inhibit the ozonation decomposition.⁴² The pH increase of the Fe/Cu alone process was mainly attributed to consumption of H⁺ ions and generation of Fe(OH)₃/Fe(OH)₂. In addition, pH of reaction solution increased gradually from 5.0 to 7.4 after only 15 min treatment by the Fe/Cu/O₃ process. The results suggest that there was a strong synergistic effect between Fe/Cu and O₃. In a brief, Fe/Cu could consume the generated H⁺ ions and generate corrosion products, which facilitate catalytic ozonation. Meanwhile, the generated organic acids from the catalytic ozonation could accelerate the iron corrosion.

Fig. 2(c) also shows that the effluent pH of the Fe/Cu/O₃ process was higher than that of the Fe⁰/O₃ process. The results could be explained from two aspects, (i) Fe/Cu bimetallic particles have a stronger reactivity and a wider pH application range than that of Fe⁰. (ii) Corrosion rate of Fe/Cu was much higher than that of Fe⁰, and released more corrosion products (e.g., Fe(OH)₃/Fe(OH)₂) that could increase the pH of reaction solution. These explanations can be confirmed by the higher total iron ions concentration in the effluent of the Fe/Cu/O₃ process (Fig. 3).

Fig. 3 Variation of total iron ions concentration in reaction solution during 15 min treatment by Fe/Cu/O₃, Fe⁰/O₃ or Fe/Cu process.

3.2.3 ORP variation. Fe/Cu bimetallic particles had strong reduction reactivity which was reflected by the low and negative ORP values in Fig. 2(d). Residual oxygen in DDNP wastewater was consumed during reaction (eqn (3)), causing the decrease of ORP values in Fe/Cu process. Ozone was a robust oxidizing agent (E₀ = 2.08 eV), the ORP values were high and positive (Fig. 2(d)). With time increasing, the ORP values increased due to the continuously bubbled O₃/O₂ mixture (ozone were generated from pure compressed dry oxygen and there were some oxygen which had not be transformed to ozone). In addition, ozone and oxygen would be consumed (eqn (3) and (4))^12,30 in Fe⁰/O₃ and Fe/Cu/O₃ process, causing the relative low ORP values. Moreover, the ORP values in Fe/Cu/O₃ process were higher than that in Fe⁰/O₃ process (Fig. 2(d)). Fe/Cu had better reaction activity than Fe⁰,¹¹ releasing more electrons and generating more iron corrosion products (e.g., Fe²⁺/Fe³⁺, Fe₂O₃, FeOOH, etc.). Thus, the dissolved oxygen would more likely to accept the electrons and react with Fe²⁺/Fe³⁺ to form Fenton-like reaction. Also, more generated iron corrosion products would catalyze ozone to produce more hydroxyl radicals with strong oxidation capacity. This was the reason for the relative high ORP values of Fe/Cu/O₃ process, which showed its superiority.

Fe⁰ + O₂ + 2H⁺ → Fe²⁺ + H₂O₂ (3)

Fe⁰ + 2O₃ → Fe²⁺ + 2O₃˙⁻ (4)

3.2.4 UV-vis analysis. UV-vis spectra of the influent and effluent of Fe/Cu/O₃ process and three control experiments from 190 to 600 nm are shown in Fig. 4. The absorbance peak between 200 and 250 nm can be mainly assigned to the π–π* transition of benzene rings.^40,48 The absorbance peak between 300 and 500 nm is assigned to the common effect of azo group (i.e., –N=N–) and nitro group (i.e., –NO₂).^11,49 It can be observed from Fig. 4 that the peak between 250 and 600 nm was removed completely in the effluent of Fe/Cu/O₃ process, meanwhile the peak intensity between 190 and 250 nm was impaired remarkably. The results indicate that the azo group (i.e., –N=N–) and nitro group (i.e., –NO₂) of the pollutants were decomposed completely by the Fe/Cu/O₃ process, and their benzene ring structure also was broken remarkably.

Fig. 4 UV-vis spectra of the influent and effluent of the different methods (all the samples were diluted 15 times).

Fig. 4 also show that all the peak of the effluent of Fe⁰/O₃ process was higher than those of Fe/Cu/O₃ process, which further confirm the superiority of Fe/Cu/O₃ process. In addition, all the peaks only decreased a little after 15 min treatment by Fe/Cu alone or O₃ alone, which suggest that their synergistic effect plays a leading role in the pollutants decomposition.

3.2.5 Accumulation of NO₃⁻ in the treatment process. Fig. 5 shows that NO₃⁻ concentration rapidly increased to 73.4 mg L⁻¹ after only 10 min treatment by the Fe/Cu/O₃ process, and then it further increased to 74.8 mg L⁻¹ after 15 min treatment. The results suggest that azo group (i.e., –N=N–) and nitro group (i.e., –NO₂) of the pollutants were decomposed and transferred to NO₃⁻ ions. However, NO₃⁻ concentration (55.4 mg L⁻¹) in the effluent of Fe⁰/O₃ process was much lower than that (74.8 mg L⁻¹) of the Fe/Cu/O₃ process. The results suggest that the Fe/Cu/O₃ process has a stronger capacity to decompose the azo group (i.e., –N=N–) and nitro group (i.e., –NO₂) of the pollutants. In addition, the sum (45.7 mg L⁻¹) of the generated NO₃⁻ in the effluent of Fe/Cu alone and O₃ alone also was lower than the Fe/Cu/O₃ process (74.8 mg L⁻¹), which further suggest that their synergistic effect plays a leading role in the pollutants decomposition.

Fig. 5 Variation of total NO₃⁻ concentration in reaction solution during 15 min treatment by different methods.

3.3 FTIR analysis

Fig. 6 shows the FTIR absorption spectra of the influent and effluent of Fe/Cu/O₃ process after 15 min treatment. Fig. 6(a) is the FTIR absorption spectrum of the influent. The band at 3424 cm⁻¹ is attributed to the stretching vibration of O–H;⁵⁰ the bands at 1605, 1531, 1434 cm⁻¹ are attributed to the aromatic C=C skeletal vibration;⁵¹ the band at 1484 cm⁻¹ is attributed to the –N=N– bond vibrations;⁵² the bands at 1563, 1326 cm⁻¹ are attributed to the stretching vibration of aromatic –NO₂;⁵³ the band at 1130 cm⁻¹ is attributed to the stretching vibration of C–O–H;⁶ the bands from 820 to 620 cm⁻¹ are attributed to the out-of-plane N–H bond angular deformation.⁵⁴ Fig. 6(b) is the FTIR absorption spectrum of the effluent after 15 min treatment by Fe/Cu/O₃ process. The bands at 1563, 1531, 1484, 1434 cm⁻¹ almost removed, which means that nitro group, azo group and benzene ring were mostly decomposed. The intensity of the band at 1605 cm⁻¹ was decreased and shifted to 1650 cm⁻¹, which suggested that part of the benzene ring was removed. The intensity of the band at 1130 cm⁻¹ increased significantly and shifted to 1118 cm⁻¹ which means the pollutants were oxidized and generated intermediates.

Fig. 6 FTIR absorption spectra of 400–4000 cm⁻¹ region of (a) influent and (b) effluent of Fe/Cu/O₃ process.

3.4 Improvement of biodegradability

Fig. 7 shows that COD removal (85.3%) obtained by the Fe/Cu/O₃ process was much better than that of Fe/Cu alone (42.1%), O₃ (38.4%) or Fe⁰/O₃ (72.0%) process. Meanwhile, the B/C ratio of DDNP wastewater increased from 0 to 0.2 after 15 min treatment by the Fe/Cu/O₃ process under the optimal conditions, which was higher than that of Fe/Cu alone (0.09), O₃ (0.07) or Fe⁰/O₃ (0.16) process. The result indicated that the Fe/Cu/O₃ process could effectively degrade the pollutants in DDNP wastewater, and its biodegradability was improved. Since carbonate, sulfide, and nitrite hydrochloric acid were added during the DDNP manufacture process,⁵⁵ the salinity of DDNP wastewater reached about 2487 mg L⁻¹. The high salinity would inhibit the growth of microorganism,⁵⁶ which cause the lower B/C ratio (i.e., 0.2) even if the DDNP wastewater was treated by the Fe/Cu/O₃ process. Generally, salt-tolerant organism could significantly alleviate the adverse effects of high salt concentrations on microbial flora.⁵⁷ However, the microorganism we used to measure BOD index in this study were came from ordinary municipal sewage treatment plants. They were not like salt-tolerant organism which can grow normally in high salt concentrations, resulting in lower BOD index. In practical engineering design, the refractory and toxic nitrophenols in DDNP wastewater was decomposed into smaller organic molecules. The pretreated DDNP wastewater mixed with domestic sewage and then flowed into biological treatment pool for further processing. Since the pretreated industrial wastewater and domestic sewage had a similar discharge, the salinity was diluted double which could facilitate the subsequent biological treatment.

Fig. 7 COD removal efficiencies and B/C ratios of the effluent of different methods.

3.5 Analysis of operating cost

The operating cost of Fe/Cu/O₃ process was mainly derived from the consumption of Fe/Cu bimetallic particles and ozone gas. Since Cu²⁺ was not detected in the effluent, total iron ions concentration in effluent and TML_Cu (0.0204 g Cu per g Fe) were used to calculate the consumption of Fe/Cu bimetallic particles. Under the optimal conditions, the total COD removal was about 920 mg L⁻¹ and the total generated iron ions concentration was about 350 mg L⁻¹. So, it could be calculated that Fe/Cu consumption value was about 0.388 kg Fe/Cu per kg COD. In addition, the consumption of ozone consumption (OC) refers to a given COD removal amount of total ozone consumption.⁵⁸ According to eqn (5), it could be calculated that OC value was about 0.686 m³ O₃ per kg COD. The cost of Fe/Cu bimetallic particles and pure compressed dry oxygen were about 300 USD per t and 0.5 USD per m³. Then, the treatment cost was about 0.46 USD per treatment 1 kg COD. Therefore, the Fe/Cu/O₃ process is a promising treatment method to treat the DDNP 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.

In addition, the treatment time (15 min) of this new process was only one eighth of the treatment time (120 min) of combined Fe⁰/air and Fenton process in our previous work.² Thus this new process could remarkably save investment and floor area.

4. Conclusions

The DDNP wastewater was effectively treated by Fe/Cu/O₃ process, and its optimal operational parameters (i.e., TML_Cu = 0.02 g Cu per g Fe, Fe/Cu dosage = 20 g L⁻¹, initial pH = 5.0, O₃ flow rate = 0.3 L min⁻¹, treatment time = 15 min) were achieved by the batch experiments. Under the optimal conditions, COD removal efficiency (85.3%), color removal efficiency (95.0%) and B/C ratio (0.2) was obtained by Fe/Cu/O₃ process after only 15 min treatment. Additionally, its treatment ability was much higher than those of three control experiments. For instance, the COD removal efficiency by Fe/Cu/O₃ process (85.3%) was higher than the sum of COD removal (80.5%) obtained by Fe/Cu alone (42.1%) and O₃ alone (38.4%), meanwhile the sum (45.7 mg L⁻¹) of the generated NO₃⁻ in the effluent of Fe/Cu alone and O₃ alone also was lower than the Fe/Cu/O₃ process (74.8 mg L⁻¹). The results suggest the synergistic effect between Fe/Cu and O₃ plays a leading role in the degradation of the pollutants. Moreover, the analysis results of UV and FTIR further confirm the superiority of Fe/Cu/O₃ process. Finally, the operating cost of the Fe/Cu/O₃ process was about 0.46 USD per kg COD. Consequently, this technology could be considered as a cost-effective, feasible treatment method for the toxic and refractory DDNP wastewater.

Acknowledgements

The authors would like to acknowledge the financial support from National Natural Science Foundation of China (No. 21207094), and Fundamental Research Funds for the Central Universities (No. 2015SCU04A09).

References

1. Z.-W. Yang, Y.-C. Liu, D.-C. Liu, L.-W. Yan and J. Chen, J. Hazard. Mater., 2010, 177, 938–943.
2. Y. Yuan, B. Lai and Y.-Y. Tang, Chem. Eng. J., 2016, 283, 1514–1521.
3. X. Dong, Z. Gan, X. Lu, W. Jin, Y. Yu and M. Zhang, Chem. Eng. J., 2015, 277, 30–39.
4. G. Lv, D. Wu and R. Fu, J. Hazard. Mater., 2009, 165, 961–966.
5. Z. Ioannou and J. Simitzis, J. Hazard. Mater., 2009, 171, 954–964.
6. Y. Yuan, P. Cao, B. Lai, P. Yang and Y. Zhou, RSC Adv., 2016, 6, 35539–35549.
7. X. Qiu, Z. Fang, X. Yan, F. Gu and F. Jiang, Chem. Eng. J., 2012, 193–194, 358–365.
8. S. Luo, P. Qin, J. Shao, L. Peng, Q. Zeng and J.-D. Gu, Chem. Eng. J., 2012, 223, 1–7.
9. A. Shimizu, M. Tokumura, K. Nakajima and Y. Kawase, J. Hazard. Mater., 2012, 201–202, 60–67.
10. X. Guan, Y. Sun, H. Qin, J. Li, I. M. C. Lo, D. He and H. Dong, Water Res., 2015, 75, 224–248.
11. B. Lai, Y. Zhang, Z. Chen, P. Yang, Y. Zhou and J. Wang, Appl. Catal., B, 2014, 144, 816–830.
12. Y. Yuan, H. Li, B. Lai, P. Yang, M. Gou, Y. Zhou and G. Sun, Ind. Eng. Chem. Res., 2014, 53, 2605–2613.
13. L. M. Ma, Z. G. Ding, T. Y. Gao, R. F. Zhou, W. Y. Xu and J. Liu, Chemosphere, 2004, 55, 1207–1212.
14. A. Ghauch, H. A. Assi and S. Bdeir, J. Hazard. Mater., 2010, 182, 64–74.
15. W. Y. Xu and T. Y. Gao, J. Environ. Sci., 2007, 19, 792–799.
16. S. J. Bransfield, D. M. Cwiertny, K. Livi and D. H. Fairbrother, Appl. Catal., B, 2007, 76, 348–356.
17. C. J. Lin, S. L. Lo and Y. H. Liou, J. Hazard. Mater., 2004, 116, 219–228.
18. Z. Xiong, B. Lai, P. Yang, Y. Zhou, J. Wang and S. Fang, J. Hazard. Mater., 2015, 297, 261–268.
19. W. Y. Xu, T. Y. Gao and J. H. Fan, J. Hazard. Mater., 2005, 123, 232–241.
20. J.-H. Fan, H.-W. Wang, D.-L. Wu, Z.-G. Liu and L.-M. Ma, J. Chem. Technol. Biotechnol., 2011, 86, 1295–1302.
21. S. J. Bransfield, D. M. Cwiertny, A. L. Rorerts and D. H. Fairbrother, Environ. Sci. Technol., 2006, 40, 1485–1490.
22. S. Esplugas, J. Giménez, S. Contreras, E. Pascual and M. Rodríguez, Water Res., 2002, 36, 1034–1042.
23. B. Wang, C.-P. Zhu, R.-H. Gong, J. Zhu, B. Huang, F. Xu, Q.-G. Ren, Q.-B. Han and Z.-B. He, Water Sci. Eng., 2015, 8, 233–238.
24. J. Nawrocki and B. Kasprzyk-Hordern, Appl. Catal., B, 2010, 99, 27–42.
25. J. Wu, L. Ma, Y. Chen, Y. Cheng, Y. Liu and X. Zha, Water Res., 2016, 92, 140–148.
26. A. Ziylan and N. H. Ince, Catal. Today, 2015, 240, 2–8.
27. X.-C. Ruan, M.-Y. Liu, Q.-F. Zeng and Y.-H. Ding, Sep. Purif. Technol., 2010, 74, 195–201.
28. X.-B. Zhang, W.-Y. Dong, F.-Y. Sun, W. Yang and J. Dong, J. Hazard. Mater., 2014, 276, 77–87.
29. F. Pan, Y. Luo, J.-J. Fan, D.-C. Liu and J. Fu, Clean: Soil, Air, Water, 2012, 40, 422–427.
30. Z. Xiong, B. Lai, Y. Yuan, J. Cao, P. Yang and Y. Zhou, Chem. Eng. J., 2016, 302, 137–145.
31. B. Lai, Y.-H. Zhang, Y. Yuan, Z.-Y. Chen and P. Yang, Ind. Eng. Chem. Res., 2014, 53, 12295–12304.
32. SEPA, Standard methods for the examination of water and wastewater, China Environmental Science Press, Beijing, 2002.
33. H. Bader and J. Hoigné, Water Res., 1981, 15, 449–456.
34. 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, J. Hazard. Mater., 2010, 182, 886–895.
35. F. Qi, W. Chu and B. Xu, Chem. Eng. J., 2016, 284, 28–36.
36. C. Noubactep, Water Res., 2015, 85, 114–123.
37. L. Sun, H. Song, Q. Li and A. Li, Chem. Eng. J., 2016, 283, 366–374.
38. B. Lai, Z. Chen, Y. Zhou, P. Yang, J. Wang and Z. Chen, J. Hazard. Mater., 2013, 250–251, 220–228.
39. H. M. Hung, F. H. Ling and M. R. Hoffmann, Environ. Sci. Technol., 2000, 34, 1758–1763.
40. B. Lai, Y. Zhou, H. Qin, C. Wu, C. Pang, Y. Lian and J. Xu, Chem. Eng. J., 2012, 179, 1–7.
41. J. Staehelin and J. Hoigne, Environ. Sci. Technol., 1982, 16, 676–681.
42. L. Zhao, J. Ma, Z. Sun and X. Zhai, Environ. Sci. Technol., 2008, 42, 4002–4007.
43. M. Sui, L. Sheng, K. Lu and F. Tian, Appl. Catal., B, 2010, 96, 94–100.
44. F. J. Beltrán, F. J. Rivas and R. Montero-de-Espinosa, Water Res., 2005, 39, 3553–3564.
45. J. Wu, H. Gao, S. Yao, L. Chen, Y. Gao and H. Zhang, Sep. Purif. Technol., 2015, 147, 179–185.
46. X.-B. Zhang, W.-Y. Dong and W. Yang, Chem. Eng. J., 2013, 233, 14–23.
47. U. von Gunten, Water Res., 2003, 37, 1443–1467.
48. L. Xu, H. Zhao, S. Shi, G. Zhang and J. Ni, Dyes Pigm., 2008, 77, 158–164.
49. M. Stylidi, Appl. Catal., B, 2004, 47, 189–201.
50. Z. Droussi, V. D'orazio, M. R. Provenzano, M. Hafidi and A. Ouatmane, J. Hazard. Mater., 2009, 164, 1281–1285.
51. N. Sundaraganesan, B. Anand and B. Dominic Joshua, Spectrochim. Acta, Part A, 2006, 65, 1053–1062.
52. L. Hua, H. Ma and L. Zhang, Chemosphere, 2013, 90, 143–149.
53. M. Arivazhagan and S. Jeyavijayan, Spectrochim. Acta, Part A, 2011, 79, 376–383.
54. A. C. Lucilha, C. E. Bonancêa, W. J. Barreto and K. Takashima, Spectrochim. Acta, Part A, 2010, 75, 389–393.
55. J. Zhai and Y. Wang, Procedia Environ. Sci., 2013, 18, 632–637.
56. F. Kargi and A. R. Dincer, Enzyme Microb. Technol., 1996, 19, 529–537.
57. F. Kargi and A. R. Dincer, Bioprocess Eng., 1996, 15, 51–58.
58. S. S. Abu Amr, H. A. Aziz, M. N. Adlan and M. J. K. Bashir, Chem. Eng. J., 2013, 221, 492–499.

---