Enhanced catalytic activity and stability of composite of cellulose film and nano zero-valent iron on Juncus effusus for activating peroxydisulfate to degrade Rhodamine B dye

With the rapid development of industrialization, emissions of organic wastewater are on the rise, causing serious harm to the ecological environment and human health (Lee et al. 2020; Ledakowicz & Pazdzior 2021). As one of the main organic waste waters, dye wastewater has the characteristics of large chromaticity, high content, stable structure, and large change in pH (Matzek & Carter 2016; Collivignarelli et al. 2019). At present, various methods such as microbial degradation (Hameed & Ismail 2021), photodegradation (Zheng et al. 2006), adsorption (Zhang et al. 2021), chemical precipitation (Mateus et al. 2020), membrane process (Nawaz et al. 2021) and advanced oxidation method (Elmobarak et al. 2021), have been applied in the degradation of dye wastewater. Among them, advanced oxidation processes (AOPs) have currently become a research hotspot because they can effectively degrade macromolecular organic pollutants into small molecules and finally achieve their mineralization (Ma et al. 2021; Khajouei et al. 2022).

In the reported AOPs (e.g., ozonation, UV irradiation, photocatalysis, Fenton reaction, persulfate, etc), persulfate can generate sulfate radical (SO₄^·−), which has high redox potential, a longer half-life and wider pH range compared with •OH based on Fenton system (Yin et al. 2018; Gao et al. 2020). Peroxydisulfate (PDS) and peroxymonosulfate (PMS) are widely used as SO₄^·− precursors. At present, the common methods for activating persulfate mainly include UV light (Lau et al. 2007), heating (Waldemer et al. 2007), ultrasound (Kyzas et al. 2022), transition metals and their oxides (Anipsitakis & Dionysiou 2004), etc.

Because of the wide source, low price and environmental friendliness, iron-based materials have been widely used to activate persulfate for removing organic pollutants in wastewater (Dong et al. 2017; Guo et al. 2020). Especially, nZVI exhibits higher reactive due to the high specific surface area and small size, which could effectively activate persulfate to degrade organic pollutants (Fu et al. 2014; Chokejaroenrat et al. 2015). However, nZVI has the problems of easy agglomeration, easy oxidation, poor stability and difficult separation from the solution after use, which limits its application (Xiao et al. 2020; Wang et al. 2021). To solve the above problems, some researchers loaded nZVI onto the solid materials like carbon (Li et al. 2022), bentonite (Baldermann et al. 2021), kaolinite (Lakkaboyana et al. 2021) and zeolite (Eljamal et al. 2021) for removing pollutants more effectively. However, most of the reported materials are powder, which are still difficult to recover after use. By contrast, the macro-materials possessing a three-dimensional (3D) structure not only are easy to separate from aqueous solution but also can expose more active sites (Tan et al. 2021).

Juncus effusus, a perennial herbaceous fiber, has been cultivated extensively in many places in China (Wang et al. 2009). It has a natural 3D structure and high specific surface area, being a good carrier for nZVI (Zhou et al. 2020). Nevertheless, JE fiber has strong hydrophobicity, which brings some obstacles to the introduction of active groups. According to the recent research from our group, the ethanol/aqueous solution can help to overcome the hydrophobicity of JE without destroying its 3D structure (Zhang et al. 2022). But to the best of our knowledge, there is no report on loading nZVI on the JE.

Sodium borohydride (NaBH₄), ferrous sulfate (FeSO₄·7H₂O), Rhodamine B (Rh B), potassium peroxydisulfate (K₂S₂O₈), sodium chloride (NaCl), sodium sulfate (Na₂SO₄), sodium sulfate (NaNO₃), t-butyl alcohol (TBA, C₄H₁₀O), methanol, ethanol and dimethylsulfoxide (DMSO) were purchased from Chengdu Kelong Chemical Co., Ltd, which were analytical grade and used without any purification. Nitrogen was purchased from Chengdu Shimao Gas Co., Ltd JE was purchased from Guangxi Jintu Ecological Agriculture Co., Ltd Cotton pulp cellulose was provided by Bailu Chemical Fiber Group Ltd, China. Aqueous solution of tetrabutylammonium hydroxide (TBAH, 40 wt%) was obtained from Shanghai Meryer Chemical Technology Co. Ltd, China.

First, JE was treated in 50% ethanol/aqueous solution for 10 min, and then immersed in a certain concentration of FeSO₄ solution for 30 min under N₂ gas protection. Afterwards, the immersed JE was transferred onto polyethylene webs (Figure S1), and 8 g/L of NaBH₄ solution was dropwise added until no obvious bubbles appeared. The obtained sample was denoted as nZVI-JE. Next, the nZVI-JE was put in a 2% cellulose solution followed with regeneration and washing by deionized water with dissolved O₂ excluded. The resultant sample was denoted as CF-nZVI-JE.

C₀ (mg/L) and C_e (mg/L) are the initial concentration and equilibrium concentration of Rh B, m (g) is the dosage of the activator, and V (L) is the volume of the solution. All experiments were conducted in triplicate.

The influence of different anions including Cl⁻, NO₃⁻ and SO₄²⁻ on Rh B degradation was investigated by fixing Rh B concentration at 50 mg/L. Their concentrations were 200 mg/L, respectively.

Cycle experiment was performed as follows: take out the CF-nZVI-JE from the first reaction system at 20 min, and then add it to the second reaction system. After that, further add to the next reaction system (Figure S2).

The physical morphologies of the JE and CF-nZVI-JE were observed on a Apreo 2C scanning electron microscope (Thermo Scientific., Holland) equipped with X-ray energy dispersive spectroscopy (OXFORD ULTIM Max65). To improve the conductivity of the samples and the quality of the SEM images, a layer of gold film was coated on the surface of the samples. The crystal structure of the JE and CF-nZVI-JE was analyzed by a D8 Advanced X-ray diffractometer (Bruker, Germany) using Cu-Kα radiation. The surface functional groups of the JE and CF-nZVI-JE were measured by a FTIR spectroscope (Nicolet iS50, Thermo Fisher Scientific, America) with an ATR accessory. The data in the range of 4,000–400 cm⁻¹ were recorded at a resolution of 4 cm⁻¹.

Figure S3 showed the XRD patterns of the JE and CF-nZVI-JE. Compared with the XRD pattern of JE, the characteristic diffraction peak of Fe⁰ can be observed at 45.84 ° on the XRD pattern of the CF-nZVI-JE. Furthermore, it was weak and broad, verifying that the Fe⁰ particles loaded on the JE had a small grain size.

Figure S4 displayed the FTIR spectra of the JE and CF-nZVI-JE. Compared with the spectrum of JE, the adsorption peaks at 3,340, 1,245 and 1,030 cm⁻¹ related to the O-H stretching and bending vibrations gave a decrease after Fe⁰ was loaded. The reason was due to the formed hydrogen bond between -OH and Fe nanoparticles (He et al. 2007). The band at 1,725 cm⁻¹ assigned to the C = O group displayed a red shift because Fe formed strongly-bound inner-sphere complex (-COOFe) with the carboxyl group (Dong et al. 2016). At the same time, the strong adsorption was observed between 400 and 500 nm due to the stretching vibrations of Fe-O (Zhang et al. 2019). As far as the peaks at 1,600 and 1,510 cm⁻¹ originating from aromatic ring C = C stretching of lignin disappeared, it was because of the dissolution of lignin in the alkali solution.

Fe²⁺ concentration in the preparation of the CF-nZVI-JE directly influenced the nZVI amount loaded on the JE, which further affected the degradation rate of Rh B. Hence, the effect of Fe²⁺ concentration at the immersing stage on Rh B degradation was investigated. From the results shown in Figure 3(b), it can be seen that the degradation rate of Rh B firstly increased and then decreased with increasing Fe²⁺ concentration from 10 to 30 g/L. At 20 min, the degradation rates were 72, 92 and 85%, respectively. The reason was that the lower Fe²⁺ concentration was, the less the generated nZVI, which led to insufficient SO₄^·− to participate in the reaction (Wang et al. 2018). However, too high Fe²⁺ concentration would generate too much nZVI. As a result, it could consume SO₄^·−, reducing the degradation rate of Rh B (Fu et al. 2020). Hence, 30 g/L of Fe²⁺ concentration was used in subsequent experiments.

The effect of CF-nZVI-JE dosage on Rh B degradation was shown in Figure 4(b). With the increasing of the CF-nZVI-JE dosage, the degradation rate of Rh B increased. It was due to the fact that more CF-nZVI-JE provided more nZVI which can activate PDS to produce more free radicals in the reaction system.

When other activation conditions were fixed, the effect of PDS concentration on Rh B degradation is shown in Figure 4(c). It was evident that the degradation rate of Rh B showed a gradual increase with increasing PDS concentration. For instance, when PDS concentrations were 1, 2 and 8 g/L, the degradation rates of Rh B were 74, 89 and 92% at 10 min, respectively. The reason was that increasing PDS concentration can provide more SO₄^·− in the presence of CF-nZVI-JE, facilitating the Rh B degradation. Of course, excess PDS could become the scavenger of SO₄^·− produced by itself (Yang et al. 2011; Wang et al. 2017).

Figure 4(d) shows the effect of initial Rh B concentration on the degradation. As the initial concentration increased from 25 to 100 mg/L, the degradation rates of Rh B were decreased from 90% to 47% at 10 min. Further delaying reaction time to 60 min, 99% Rh B was degraded for the initial concentration of 25 and 50 mg/L, while 90% for 100 mg/L. It was because fixing PDS concentration and CF-nZVI-JE dosage, the amount of the produced free radicals was certain (Barzegar et al. 2018). Thus, the higher concentration of Rh B was not entirely degraded by the limited free radicals (Zhang et al. 2009).

The recycling performance of persulfate activator is very important for its application, so that of the CF-nZVI-JE was investigated as presented in Figure 5(b). It was clear that the degradation rate of Rh B reached 98, 97, 96, 91 and 86% from the first to fifth cycles at 60 min, respectively, suggesting that the CF-nZVI-JE had good reusability. Nevertheless, a small difference can be observed at the initial stage. At 10 min, the degradation rate of Rh B was 75, 65, 60, 56 and 43%, respectively. It could be related to the reduce of the active sites on the JE-ZVI-CF due to the dissolution of Fe⁰.

The fitting curves and related parameters were shown in Figure S5 and Table 1, respectively. It was obvious that the correlation coefficient R² obtained by the pseudo-second-order kinetic model was higher than that of the pseudo-first-order kinetic model. Moreover, the q_e,cal values were also closer to the experimental values (q_e,exp). Hence, Rh B degradation in the CF-nZVI-JE/PDS system fitted the pseudo-second-order kinetic model, suggesting that the degradation process was involved in chemical adsorption. A similar result has been reported in previous literature (Chen et al. 2021). Meanwhile, the values of k₂ gave a decrease with increasing the initial concentration of Rh B, suggesting that the degradation of higher concentration of Rh B needed longer time.

Ethanol can react with SO₄^·− and ·OH rapidly, while the reaction rate constant of t-butyl alcohol with ·OH is much greater than SO₄^·−. Therefore, the combination of ethanol and t-butyl alcohol is commonly used to detect SO₄^·− and ·OH free radicals. In the present study, the free radicals that were mainly responsible for Rh B degradation in the CF-nZVI-JE/PDS system were determined via adding excessive free radical quenchers ethanol and t-butyl alcohol.

In additon, compared with the nZVI-JE/PDS system, the release amount of iron ions gave an obvious decrease in the CF-nZVI-JE/PDS system as shown in Figure 6(c). The phenomenon could be related to the following two aspects: One was that the existence of the CF reduced the dissolution of Fe⁰, the other was that the O-containing functional groups like –COOH and –OH on the CF fixed part of the released iron ions.

To visually exhibit the degradation process of Rh B in the CF-nZVI-JE/PDS and nZVI-JE/PDS systems, the solution at different reaction time was irradiated under 365 UV light and the results are shown in the inset of Figure 7. It was evident that the decolorization of Rh B was quickly completed within 10 min in the JE-ZVI/PDS system, but since then the solution emitted a green fluorescence at 492 nm. By contrast, in the CF-nZVI-JE/PDS system, fluorescence color of the solution changed from red to orange and to green during the initial 30 min. Afterward, the green fluorescence became weak with delaying the reaction time, suggesting the gradual degradation of the formed intermediate products (Wu et al. 2020). This was consistent with the UV-Vis result that Rh B was degraded into some intermediate products firstly and then the intermediate products were further mineralized in the CF-nZVI-JE/PDS system.

In this study, a new PDS activator CF-nZVI-JE was facilitated, which possessed the same 3D structure as JE. Due to the stabilization from the JE, the nZVI in-situ loaded on the JE existed in two forms: single nanoparticles with a size of 60–100 nm and chain-type agglomeration of nanoparticles. The cellulose film wrapped outside the CF-nZVI-JE prevented the leakage of nZVI. The CF-nZVI-JE can efficiently activate PDS to degrade Rh B. Under the optimal activating conditions (T = 25 °C, [Rh B] = 50 mg/L, [K₂S₂O₈] = 1 g/L, JE = 0.5 g/L, pH = 4), the degradation rate of Rh B reached 99% within 30 min. The degradation process fit the pseudo-second-order kinetic model, indicating that the degradation process was involved in chemical adsorption. Compared with NO₃⁻ and SO₄²⁻, the coexistence of Cl⁻ had a significant negative effect on Rh B degradation. Cycling experiment displayed that the degradation rate of Rh B still reached 86% after five cycles, suggesting that the CF-nZVI-JE had good reusability. Mechanism study demonstrated that the activation of PDS by the CF-nZVI-JE included three ways: (1) redox action of Fe²⁺/Fe³⁺; (2) nZVI mediated activation; (3) O-containing groups as an activator of the electron-transfer mediator. The degradation of Rh B in the CF-nZVI-JE/PDS system was involved in SO₄^·− and ·OH, and SO₄^·− was the main activation radical. The introduction of CF was conducive to the mineralization of Rh B. In addition, compared with the nZVI-JE, the CF-nZVI-JE obviously reduced the release amount of iron ions.

This work was financially supported by Science & Technology Program of Sichuan Province (2022YFS0503, 2021YFS0284).

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.