Continuous-flow ozonation over modified ceramsite: implications for the degradation of cation red x-GRL

The dyes, based on aromatic and heterocyclic compounds with color groups (-N = N-, -N = O-) and polar groups (-SO₃Na, -OH, -NH₂), are derived from a series of chemical unit operations such as nitrification, sulfonation, amination, and diazotization (Moccia 2006; Zhang et al. 2008). Since the industrial revolution, extensive use of dyes in the printing and dyeing industry has become a serious issue to human health and the eco-environment. Along with sewage discharge standard advancement, traditional biological methods have been insufficient to meet requirements (Zee & Villaverde 2005). Meanwhile heterogeneous catalytic ozonation, as one of the advanced oxidation processes (AOPs), has recently gained significant attention in the field of wastewater treatment due to generation of highly reactive and non-selective ·OH (Park et al. 2004b).

Transition metal oxides, either loaded or unloaded, have been demonstrated to be efficient activators for ozone (Legube & Leitner 1999; Kasprzyk-Hordern et al. 2003; Nawrocki & Kasprzyk-Hordern 2010). Among these catalysts, synthetic iron oxides have drawn increasing attentions for catalytic ozonation due to low cost, environmental benignity and abundant reserves (Bing et al. 2015; Wu et al. 2016a; He et al. 2017; Zhu et al. 2017). Successive studies were conducted to reveal the catalytic performances and mechanisms. The reaction mechanism could be concluded into two pathways: ·OH generation mechanism and surface atomic oxygen or complex formation mechanism (Liu et al. 2016). It is reported that the chemisorbed ozone was converted into surface adsorbed ·OH and O²⁻ radicals at the Lewis acid sites of Fe₂O₃ (Bing et al. 2015). Similarly, high reactivity of ordered mesoporous Fe₃O₄, Cu_xO-Fe₃O₄ and Cu-Fe-O was also ascribed to their enhanced activities in decomposing ozone into ·OH (He et al. 2017; Zhu et al. 2017; Liu et al. 2019). Although the details of catalytic ozonation and related theories are extensively understood, its application is still limited to laboratory scale. The main cause is that the frequently-used semi-batch mode is quite different from real wastewater treatment in the sewage plant. Wastewater has a very complicated composition, and there has been a lack of studies on the continuous flow mode, which is close to the actual operating process. More importantly, many fine catalysts studied at present are difficult to collect and reuse. Therefore, an efficient synthetic strategy to prepare a large carrier catalyst consisting of several oxides during continuous-flow ozonation is highly desirable.

In our previous study, pyrite cinder has been confirmed as an efficient ozone catalyst with environmental benignity(Wu et al. 2016a). In this work, PyC, silty clay and kaolin were milled and shaped into spherical particles (namely modified ceramsite). The modified ceramsites mainly consisted of Fe₂O₃, Fe₃O₄, SiO₂ and Al₂O₃ and were easy to fill in and recycle from the reactor. The objectives of this study are to (i) investigate the performance of catalytic continuous-flow ozonation by modified ceramsite carrier; (ii) elaborate the specific radical chain reacting mechanisms of modified ceramsite catalytic ozonation.

Sodium thiosulfate pentahydrate (Na₂S₂O₇·5H₂O), cerium nitrate hexahydrate (Ce(NO₃)·H₂O), and potassium iodide (KI) were of analytical grade and supplied by Sinopharm Chemical Reagent Co., Ltd (Shanghai, China). 5-Dimethyl-1-pyrolin-N-oxide (DMPO) was purchased from Aladdin Chemistry Co., (Shanghai, China). Cation red x-GRL, pyrites cinder (PyC, <74 nm) and soluble starch were of industrial grade and purchased from Sinopharm Chemical Reagent Co., Ltd (Shanghai, China). The other reagents were at least of analytical grade and used without further purification. Ultrapure water (18.2 MΩ·cm⁻¹, Millipore) was used throughout the experiments. PTFE syringe filters (0.22 μm) were obtained from ANPEL Laboratory Technologies Inc. (Shanghai, China).

The crystalline structure of PyC was determined by X-ray diffractometer (XRD, D8 Advance X-ray Spectrometer, Bruker). The total organic carbon (TOC) of cation red x-GRL solution was directly determined on a TOC-L CPHCN 200. The pH was measured using a pH meter (pH-2s, Hach), and the pH medium was maintained via the addition of appropriate amounts of HCl (0.1 M) or NaOH (0.1 M). Dissolved Al and Fe ions were determined using inductively coupled plasma-atomic emission spectrometry (ICP-AES, Agilent 720ES).

The experimental setup is shown in Figure 1. Continuous-flow ozonation experiments were conducted in plexiglas column reactors with an internal diameter of 7 cm and working height of 75 cm under ambient conditions. Ozone was generated from pure oxygen through an ozone generator (CF-3-10 g, Guolin, Qingdao, China). Ozone gas at different flows, detected by an iodometry method (Bader & Hoigné 1981), was continuously bubbled into the reactor with a silica dispenser until the generator reached steady state and the flux rate was unchanged. Therefore, the concentration of ozone could be kept constant. The residual ozone was absorbed by iodide potassium solution. The total time of cation red x-GRL flowing through the whole reactor was noted as the hydraulic retention time (HRT). The effusive cation red x-GRL reflowed through the reactor; namely, cation red x-GRL experienced two cycles. Similarly, the third cycle means cation red x-GRL was treated three times. In other words, the cation red x-GRL experienced the whole column reactor from inlet to outlet three times. Furthermore, the modified ceramsite was reused to degrade initial cation red x-GRL after washing 15 min.

PyC, silty clay and kaolin at a mass ratio of 10:0.5:0.5 were milled at 500 rpm for 3.0 h. The milled mixtures were shaped into spherical particles before being dried at 105°C in air overnight. The modified ceramsite was prepared by calcining the spherical particles at 400 °C for 1.0 h and then at 1,030 °C in a furnace at a heating rate of 2 °C/min. After the annealing processes, the ceramsite was distributed with numerous pores owing to the gasification of the organic component in silty clay. As shown in Figure 2, all diffraction peaks of PyC are in good accordance with the standard diffraction patterns of Fe₂O₃, Fe₃O₄ and SiO₂ based on its characteristic diffraction peaks. Given that the inorganic components of silty clay and kaolin, the modified ceramsite was composed of Fe₂O₃, Fe₃O₄, SiO₂ and Al₂O₃.

To determine the impact of different porous microstructures in modified ceramsite, the four continuous adsorption experiments of 200 mg/L cation red x-GRL by modified ceramsite were conducted. For continuous experiment, the cation red x-GRL experienced the whole plexiglas column reactor from inlet to outlet for three times. HRT was 16 min, namely total time of continuous adsorption experiment was 48 min. As shown in Figure 3, TOC removal efficiency of different sampling sites during the first cycle were 5%, 4.8%, 5.1%. The results were nearly the same owing to the same HRT for every point in the continuous-flow system. Meanwhile, cation red x-GRL absorption increased with HRT increase and the highest TOC absorption reached up to 10% within 48 min. With cycle times of modified ceramsite increased, the adsorption performance decreased obviously. In the repeated fourth time of the ceramsite adsorption test, the removal rate of TOC was less than 1% within 48 min and no obvious promotion effect was noticed in further adsorption test. Therefore, the modified ceramsite alone had a slight absorbing effect toward cation red x-GRL and had no advantage in the long run.

The HRT, namely the ozone contact time, had a significant impact on the mineralization rate during heterogeneous ozonation system. This result could be feasibly concluded from Figure 4. TOC was selected as an analysis parameter because of the high decolorization ability of ozone toward dyes. After a basic comprehension of the modified ceramsite absorption properties, heterogeneous ozonation of 200 mg/L cation red x-GRL at a constant flow rate of 0.09, 0.18 and 0.27 L/min, corresponding to HRT = 16, 8 and 5 min, were conducted to investigate the HRT influence. Results showed that the longer the HRT, the higher the TOC removal efficiency. Compared with that of HRT = 5 min, mineralization efficiency after three cycles at HRT = 16 min increased by 30%. In addition, the TOC removal rate showed a reduced trend as cycle times increased. For example, TOC removal rates in three cycles during catalytic continuous-flow ozonation at HRT = 5 min were 20%, 15% and 5%, respectively. This is mainly because the original cation red x-GRL molecules containing certain functional groups (e.g. aromatic, unsaturated carbonate hydride) are apt to react with ozone. However, with the reaction going on, degradation intermediates (such as saturated hydrocarbons, alcohols, aldehydes, etc.) gradually increased. And these low-weight organic acids were refractory to be degraded by HO· (He et al. 2017). Mineralization efficiency at HRT = 5 min was obviously lower than that at HRT = 8, 16 min. Accordingly, HRT = 8, 16 min was employed in the following experiments.

Continuous-flow ozonation alone and catalytic ozonation experiments were conducted at HRT of 16 min to elucidate the catalytic activity of modified ceramsite towards cation red x-GRL degradation. The profile of TOC reduction of 200 mg/L cation red x-GRL at a constant flow rate of 0.09 L/min (HRT = 16 min) is depicted in Figure 5(a). It was observed that the TOC reduction in the heterogeneous continuous-flow ozonation process was much higher than that during the single ozonation process, indicating that modified ceramsite showed a remarkable catalytic activity. Specifically, without the modified ceramsite, the TOC removal rates were relatively lower, accounting for 7%, 22% and 32%. Cation red x-GRL experienced three cycles; with increase of cycle times, catalytic degradation of cation red x-GRL remained high, accounting for 45%, 60% and 65%, respectively. TOC reduction rise slowly in the third cycle, the main reason was high levels of intermediate products accumulated in the solution. In the other words, the TOC reduction in every cycling run was increased by 5.4, 1.7, 1.0 times through modified ceramsite catalytic ozonation. Therefore, modified ceramsite has presented great catalytic activity for cation red x-GRL mineralization by ozonation. Owing to the ·OH mechanism of Fe₂O₃, Fe₃O₄ and Al₂O₃ in heterogeneous ozonation (Kasprzyk-Hordern et al. 2006; Bing et al. 2015; Vittenet et al. 2015; Wu et al. 2016a; Zhu et al. 2017), thus the main species involved ·OH and the reaction mechanism probably ascribed to the improvement of ·OH produced from ozone decomposition.

As shown in Figure 5(a), when modified ceramsite was recycled four times, only a slight decrease was noticed in the mineralization efficiency, indicating that modified ceramsite shows potential as an effective and stable as well as low-cost catalyst for cation red x-GRL degradation catalyzed by ozonation. The profile of the TOC removal rate of 200 mg/L cation red x-GRL at a constant flow rate of 0.18 L/min (HRT = 8 min) after being reused four times was in line with that at HRT = 16 min, further manifesting the efficient and stable catalytic activity of modified ceramsite (Figure 5(b)).

The catalytic activity of the modified ceramsite was determined in the long run. As shown in Figure 6, TOC values of different sampling sites were nearly the same owing to the same HRT at every point in the continuous-flow system. It was observed that the TOC reduction in the heterogeneous continuous-flow ozonation process was much higher than that in the single ozonation process and adsorption, indicating that modified ceramsite showed a remarkable and stable catalytic activity in the long run.

Therefore, continuous-flow experiments were conducted at circumneutral, acid and alkali conditions to investigate the effect of pH during heterogeneous ozonation. As depicted in Figure 7, the profile of the TOC removal rate was studied under the experiment condition of a constant flow rate of 0.18 L/min (HRT = 8 min), 200 mg/L initial dye concentration. It can be seen that with pH increased, the TOC removal rate of cation red x-GRL in every cycling run gradually improved. Thus the higher initial pH is beneficial to the mineralization of cation red x-GRL. At initial pH 3, ozone was the main oxidant, leading to lower TOC reduction; while at initial pH 7 and 9, ozone and ·OH were the predominant oxidants, causing more TOC reduction. Although the initial pH has a certain effect on the removal efficiency of TOC, the removal rate of TOC still reached up to 50% at pH value of 3. These results may be related to the composition of modified ceramsite, because modified ceramic consisted of Fe₂O₃, Fe₃O₄, SiO₂ and Al₂O₃. These oxides would promote the decomposition of O₃ molecules to HO·. Furthermore, the trace dissolved Al and Fe ions were only 0.03 mg/L and 0.02 mg/L at pH value of 3, respectively. The leaching metal ions are much lower than those of previously reported Al-based catalysts and Fe-based catalysts (Lyu et al. 2015; Wu et al. 2016b; He et al. 2017). The low metal leaching as well as high mineralization efficiency of cation red x-GRL indicate that modified ceramsite is highly stable. Therefore, the continuous-flow modified ceramsite catalytic ozonation can maintain high TOC reduction in a wide range of pH under the synergistic catalysis of various oxides. That is to say, this process has great potential in application for industrial wastewater treatment.

It is well known that the intermediate oxidation products such as acetic acid are difficult to oxidize by ozone (pH > 5, k < 0.04 M⁻¹s⁻¹) (Hoigné 1985). However, the reaction rate constants of ·OH with acetic acid can reach up to 10⁶ M⁻¹s⁻¹ (Masten & Hoigné 1992), which is the reason why the key to high mineralization efficiency of contaminants during heterogeneous catalytic ozonation is the capacity of O₃ decomposition into ·OH. Therefore, the electron paramagnetic resonance spectrum (EPR) spin-trap technique was used to confirm the involvement of ·OH with 5,5-dimethyl-1-pyrroline N-oxide (DMPO) as a trapping agent. As shown in Figure 8, four characteristic peaks with the 1:2:2:1 quarter pattern of the DMPO-·OH spin adduct (a^H = a^N = 14.9 G) were obtained in the modified ceramsite catalytic ozonation system. The main species involving ·OH was confirmed and the reaction mechanism probably ascribed to the improvement of ·OH produced from ozone decomposition.

It has been reported that the degradation of organic matter by catalytic ozonation usually includes two stages (Beltrán et al. 2006). The first stage is decolorization of cation red x-GRL by direct oxidation of O₃. The second stage includes adsorption owing to the porous microstructure; transition metal oxides can accelerate the decolorization of cation red x-GRL and mineralization of organic pollutants. The adsorption capacity of modified ceramsite is very important for cation red x-GRL and intermediate products' removal. Meanwhile, the adsorption of organics on modified ceramsite can promote the contact of ·OH, which is formed by ozone decomposition at the interface of metal oxides, thereby improving the efficiency of ozone utilization and the mineralization rate of cation red x-GRL.

The efficiency of ozone utilization is directly related to the cost of ozone oxidation technology, which determines the prospect of its application. For this purpose, the ozone utilization efficiency of dye wastewater degradation at different HRT are studied and shown below in Table 1.

Compared with ozone oxidation alone, continuous modified ceramsite catalytic ozonation significantly improves ozone utilization efficiency (S.O.C). For HRT 16 min, ozone utilization efficiency in the first, second, and third cycling run during the catalytic ozonation process was 1.7, 2.0, and 1.4 times better than ozonation alone. Meanwhile, the more cycles, the faster the rate of ozone utilization efficiency. This is mainly because the content of small molecular organic acid gradually increases against reaction time. That is to say, the scavenger of ·OH is also increasing. It is worth noting that, with HRT 8 min as an example, the ozone utilization efficiency of dye wastewater by catalytic ozonation in the second cycling run (1.23 mgO₃/mgCOD) was still better than ozonation alone in the first cycle (1.67 mgO₃/mgCOD), indicating that modified ceramsite still has a high catalytic activity towards the degradation of intermediates by O₃. Therefore, the modified ceramsite catalytic ozonation process can increase ozone utilization. In other words, addition of modified ceramsite can save nearly half of the amount of ozone when degrading the same amount of COD. In addition, single ozonation, mainly functioned in direct oxidation, will be very limited for industrial wastewater, especially persistent organic pollutant wastewater. Perhaps no matter how much ozone dosing it cannot reach the discharge standard, but the catalytic ozonation can achieve deep treatment goals. Therefore, catalytic ozonation by modified ceramsite has great potential in application for industrial wastewater treatment.

The granular modified ceramsite was prepared with pyrite cinder and kaolin as the main raw materials. Compared with the process of continuous-flow ozonation alone, continuous-flow catalytic ozonation by modified ceramsite showed obviously higher performance towards dye degradation. TOC removal efficiency of wastewater is increased by 35% after three cycles of continuous flow. Meanwhile, the increase of pH is beneficial to the mineralization of dye wastewater and the continuous flow ceramsite ozonation can maintain high mineralization efficiency in a wide range of pH from 3 to 10. The continuous-flow ceramsites can remain stable in a long run. In addition, ·OH is the active species for the removal of cation red x-GRL and its intermediate products. More importantly, ozone utilization efficiency of catalytic ozonation under the same conditions is as high as 1.67 mgO₃/mgCOD, while that of ozonation alone is only 0.91 mgO₃/mgCOD. In a word, continuous ceramsite catalytic ozonation can greatly improve the utilization efficiency of ozone in terms of the saving in ozone dosage. The use of chemical waste PyC has obvious technical and economic advantages, therefore the large carrier catalytic ozonation can provide an innovative and important technical reference.

This study was financially supported by the Natural Science Foundation of China (Grant No. 21776223), and State Key Laboratory of Pollution Control and Resource Reuse Foundation (Grant No. PCRRT16001).