Adsorptive removal of methylene blue dye from aqueous streams using photocatalytic CuBTC/ZnO chitosan composites

Dyeing, one of the most energy-intensive operations carried out in the textile sector, has expanded rapidly in the past few decades. With growing urbanization, the demand for textile goods and products has increased, leading to extensive growth of dyeing industries. As a result of industrial operations and activities, the textile industry wastewater has started concentrating the water bodies with various hazardous and toxic substances such as dyes, by-products, heavy metals etc. (Pattnaik et al. 2018; Shindhal et al. 2021). Dyes have been identified as one of the major contaminants and toxicants in water bodies and are linked with toxicity, and have highly negative environmental repercussions if not removed from the water bodies for a longer period (Bangari et al. 2021; Yadav et al. 2021a). Methylene blue dye, for example, is an azo dye that poses serious hazards to human health if exposed to a large quantity (Bangari et al. 2022b; Yadav & Dindorkar 2022). Methylene blue dye has highly toxic effects on the intravenous route and releases toxic gases like nitrogen dioxide on combustion and is also responsible for irritation of the eyes and nausea when it comes in direct contact with the body (Shakoor & Nasar 2017; Yadav et al. 2021b). The major challenge in removing these toxic dyes is the additional stability imparted to them during the synthetic process. They need to be retained on the fabric for a longer time and in a variety of conditions. This makes it difficult to design strategies for effective removal (Imran et al. 2015).

Various methods have been developed for removing toxic dyes from water sources, such as photodegradation, coagulation-flocculation, advanced oxidation processes, biodegradation, membrane techniques, filtration, and ion exchange (Yaseen & Scholz 2019; Yadav et al. 2022b). However, most treatment processes (filtration, ion exchange, photodegradation) involve tedious setup, complex chemical analysis, and huge expenses and are time-intensive, making it difficult to implement these strategies commercially (Dindorkar et al. 2022; Yadav et al. 2022a). Adsorption is a comparatively simple, cost-effective and easy to design process. The metal-organic frameworks (MOFs), because of their porous nature, semiconductor-like behaviour, and flexible structures, are among promising candidates for water treatment and have shown promising photocatalytic activity (Hussain et al. 2018). Among MOFs, the most popular and most investigated MOF so far is [copper(ii)-benzene-1,3,5-tricarboxylate] (CuBTC) due to its high thermal stability and well-defined structure (Peedikakkal & Aljundi 2020). Therefore, CuBTC may be favourable for the removal of dyes from wastewater. The adsorption using MOFs shows a great capacity in removing dyes from the water. Rajak et al. (2017) reported a novel cobalt-based MOF for the adsorption of Chicago sky blue and Congo red dyes. Liu et al. (2020) synthesized a poly threaded Mn^II-MOF based on a tridentate N-containing ligand that showed high adsorption capacity for methyl orange. Shahnawaz Khan et al. (2020) reported the factors that trigger the dye adsorption on MOFs. Mariyam et al. (2020) performed a topological analysis of tetrazole-based MOFs and examined their dye adsorption properties against methylene blue. However, it is evident from the reports that the MOF-based photocatalytic materials' regeneration studies after the first cycle of degradation are insufficient and lack evidence of regeneration (Zhao et al. 2017; Zhang et al. 2018). The major problem with adsorption is the reusability of adsorbents as the adsorbents get poisoned as the active sites get blocked. Zinc oxide (ZnO) is a well-known material that exhibits photocatalytic properties and can be used to regenerate adsorbents (Adeel et al. 2021). Cai et al. (2022) found that surface area and photocatalytic properties increased after incorporating ZnO in waste pine.

Despite having known the potential of CuBTC in the adsorption of dyes, their studies remained restricted. Moreover, very few reports are available wherein the attempts are made to design CuBTC-based composites that can adsorb the dyes and degrade them to less harmful products. This study used CuBTC and chitosan and photoactive ZnO nanoflowers to synthesize photocatalytic CuBTC/ZnO chitosan composites. The composite material was investigated for its efficiency to adsorb methylene blue from the aqueous streams. The composite beads were characterized using scanning electron microscopy (SEM) to analyze their morphological properties, X-ray diffraction (XRD) and Fourier transform infrared spectra (FT-IR) to gain insights into their structural properties, Brunauer-Emmett-Teller (BET) analysis to determine their surface area and thermogravimetric analysis (TGA) to study their thermal stability. The adsorption isotherms and the kinetic models were used to establish the regeneration mechanism. The regeneration studies of the CuBTC/ZnO chitosan composite for the photocatalytic degradation of the methylene blue dye under solar irradiation were also performed.

Copper nitrate (Cu(NO₃)₂), zinc acetate dihydrate (Zn(CH₃COO)₂.2H₂O), chitosan polymer, trimesic acid (H₃BTC), dimethylformamide (DMF), methylene blue dye, ethanol, and sodium hydroxide (NaOH) were procured from SRL chemicals. The solvent system chosen for the adsorption experiments was deionized water and obtained from Millipore Q BIOCEL unit, Millipore.

CuBTC was prepared by the method reported earlier (Xu et al. 2016). The solvothermal synthesis of CuBTC involved the addition of 1.45 g Cu(NO₃)₂, 0.84 g H₃BTC, and 80 ml DMF in a 100 ml beaker and subsequent stirring till it was homogenous. The suspension was transferred to a stainless-steel autoclave and held at 140 °C for 24 h. After cooling to room temperature, the light blue crystalline product was washed with DMF and methanol several times and dried overnight at 50 °C.

ZnO nanoflowers were prepared by the method reported earlier (Patel & Yadav 2022). 0.2 M zinc acetate dihydrate was dissolved in deionized water at room temperature (∼25 °C) followed by sonication for 2 h. A 0.02 M NaOH solution was prepared separately by dissolving the desired quantity of NaOH in deionized water. To the solution of 0.2 M zinc acetate dihydrate, the 0.02 M of NaOH solution was added dropwise with a constant stirring of 300 rpm. The resultant white precipitate was washed with ethanol multiple times. The filtered white precipitate was dried in a vacuum oven at 60 °C for 10 h to obtain ZnO nanoflowers.

CuBTC/ZnO chitosan composite beads were synthesized by the method reported earlier (Yadav et al. 2022c). The ionotropic gelation synthesis of CuBTC/ZnO chitosan composite beads involved the addition of CuBTC (500 mg) to 100 ml of deionized water and 100 mg of ZnO nanoflower followed by magnetic stirring for 10 min. 2 g chitosan and 2% w/v acetic acid were added to the solution and stirred until a homogeneous solution formed. Spherical beads of the solution were then dropped into 1 N NaOH solution using a syringe and incubated in the basic solution overnight at room temperature. After incubation, the beads were filtered and washed several times with deionized water until the solution's pH became neutral.

The details on adsorption experiments are included in section S1, ESI.

The details on characterization techniques are included in section S2, ESI.

The SEM images of as-synthesized CuBTC and ZnO nanoflowers are shown in Figure 1. The low magnification image of CuBTC showed that the units were individual with an octahedral geometry and did not agglomerate (Figure 1(a)). The high=magnification image of a single CuBTC unit (Figure 1(b)) shows that these octahedral units were elongated along an axis. This distortion from regular geometry was attributed to the Jhan-Tellar distortion (Schäfer et al. 2017). Further, the atomic percentage of Cu, C, O, and N in CuBTC were 6.42, 56.40, 34.31, and 2.88, respectively. A high proportion of C and O and low proportions of N was attributed to the rich organic profile of the BTC ligand. The high-magnification image of the ZnO reveals its flower-like shape (Figure 1(e)). The smaller nanoflower structures agglomerated and gave rise to comparatively larger nanoflowers. The corresponding EDS spectra shown in Figure 1(f) demonstrate the atomic percentage of Zn (53.52%) and O (46.48%), respectively, in the synthesized ZnO nanoflowers. It is evident from the EDS plots that the as-synthesized ZnO nanoflowers were of high purity.

The SEM image of CuBTC/ZnO composite beads reveals the rough and porous surface (Figure 2(a) and 2(b)) and confirms the high purity of as-synthesized composite beads. The atomic percentages of Cu, C, O were 2.27%, 48.59%, and 39.60%, respectively (Figure 2(c)). The atomic percentage of nitrogen was 7.39% was due to the copper nitrate used for the synthesis. A negligible percentage of impurities (∼2%) were found mainly due to the Si, Mg and Zn.

The XRD spectra of the ZnO nanoflowers, CuBTC, and the CuBTC/ZnO chitosan composite beads are depicted in Figure 3. The crystalline nature of the ZnO nanoflowers was confirmed from the XRD spectra. All the reflections were indexed to the wurtzite phase of ZnO (ICDD no. 36–1451). Two distinct characteristic peaks were observed for the CuBTC at 2θ values of 13.5° and 19.1°, respectively. Certain other peaks were also observed at higher 2θ values were due to CuO, Cu₂O and Cu. The XRD results for CuBTC and ZnO agreed with previous studies (Yedurkar et al. 2016; Cheng et al. 2018).

Figure 4(a) shows the FT-IR spectra of the synthesized ZnO nanoflowers. A sharp band at 488 cm⁻¹ was due to Zn-O stretching. In octahedral coordination, the Zn-O stretchings were observed at 654 cm⁻¹. Strong tetrahedral coordinations confirmed the wurtzite structure of ZnO (Mohd Yusof et al. 2020). The band at 1,370 cm⁻¹ was linked to the asymmetric stretching of C = O. At 2,855 cm⁻¹ and 2,916 cm^−1, bands arose due to C-H bending and C-H stretching vibrations, respectively. The FT-IR spectra of CuBTC are also shown in Figure 4(a). Due to Cu-O stretching, sharp bands appeared at 478 and 730 cm⁻¹ (Arul & Abraham John 2017). A band at 1,106 cm⁻¹ was ascribed to carboxylate stretching vibration. The bands that appeared between 1,450–1,350 cm⁻¹ and 1,625–1,550 cm⁻¹ were due to the asymmetric and symmetric stretching of carboxylate, respectively (Khoshhal et al. 2015). Figure 4(b) shows the FT-IR spectra of CuBTC/ZnO chitosan composite beads. The broad hump between 3,331 and 3,450 cm⁻¹ was ascribed to –OH and -NH stretching vibrations (Li et al. 2016). The characteristic out of plane vibrations for chitosan were observed at 562 and 1,174 cm⁻¹ (for the C–O–C stretching) due to the glucopyranose ring in chitosan polymer (Oh et al. 2019). At 1,589 cm⁻¹, a band corresponding to the N-H bending of the primary amine group was observed. The strong characteristic bands of the CuBTC were located at 1,646 cm⁻¹. A sharp band for Zn-O stretching in tetrahedral coordination was observed at 488 cm⁻¹.

To study the stability of the CuBTC and CuBTC/ZnO chitosan composite beads at high temperatures, TGA thermographs were obtained (Figure S1, ESI). A similar TGA trend was shown by both the composites. The composite bead without ZnO showed a steep decrease in mass composition, and around 800 °C, the majority of the composite beads (∼80%) got decomposed. However, after incorporating the ZnO nanoflowers, improved results were seen. The TGA curves revealed a residual mass of around 84% near 300 °C, followed by a loss of about 40% mass up to 350 °C. A continuous loss of mass was observed due to the rich organic profile of the composite beads that is mainly due to chitosan and the BTC ligand.

The effect of the CuBTC/ZnO chitosan composite dose on adsorption and removal efficiency is depicted in Figure 5. The removal efficiency decreased with the dosage of adsorbent up to 1.6 g L⁻¹. This was attributed to the availability of more active sites for adsorption on increased dosage. An equilibrium value of 1.6 g L⁻¹ was obtained, for which the adsorption capacity and the removal efficiency was maximum. The adsorption capacity decreased drastically in the initial stages, followed by a steady decrease. This increase in the adsorbent dose resulted in aggregation or overlapping of adsorption sites (Ndi Nsami & Ketcha Mbadcam 2013). As the adsorption experiments were carried out in polar aqueous media, the possibility of hydrogen bonding increased, which led to the agglomeration of the composites with an increase in dosage (Mahmoudzadeh et al. 2013).

Figure 6 revealed that when the initial concentration of the methylene blue dye was varied, the removal efficiency of around >98% was observed till the 60 ppm methylene blue dye concentration. A drastic decrease in removal efficiency was observed beyond 60 ppm methylene blue dye concentration. This was attributed to the increased concentration of methylene blue molecules and the same amount of adsorbent, meaning the number of active sites remained constant. Initially, when the concentration of the methylene blue dye was 20 ppm, the adsorption capacity was around 10 mg g⁻¹. As the methylene blue dye concentration increased, the adsorption capacity rose to about 50 mg g⁻¹ and remained constant.

pH plays a crucial role in the adsorption of the dyes. The variations in the pH of the methylene blue dye solution affected the adsorption capacity and the removal efficiency of CuBTC/ZnO chitosan composite beads, as shown in Figure 7. As the pH increased, the adsorption capacity and the removal efficiency increased. The removal efficiency (∼99%) and the adsorption capacity (12 mg g⁻¹) were maximum at the neutral pH. Hence, the neutral pH was considered optimum and maintained in the further batch adsorption experiments. At higher pH, methylene blue dye-containing chlorine anions react with NaOH to produce NaCl and methylene blue-S + OH (aqueous) through displacement reaction. NaCl salt decreases the adsorption of methylene blue-S + OH (aqueous) on the adsorbent surface (Amode et al. 2016).

Contact time refers to when the adsorbate molecules are in contact with the adsorbent. The more the contact time, the greater the chances of the adsorbate molecule to contact the adsorbent's active sites (Hao et al. 2021). From Figure 8, it was observed that the removal efficiency and the adsorption capacity increased with contact time. This was attributed to the fact that initially, more sites were available, and the number decreased during the adsorption process. An equilibrium value of 90 min was obtained where their removal efficiency reached ∼100%.

The maximum adsorption capacity (q_m) of various adsorbents reported in the literature against methylene blue dye are compared in Table S1, ESI. Parameters such as adsorption capacity (q_m), adsorbent dose, pH and contact time formed the basis of the comparison. The results from the present study were comparable with previous works, suggesting that methylene blue could be easily adsorbed on the as-synthesized CuBTC/ZnO composite beads prepared in this work. The additional advantage over the other reported studies was the photocatalytic degradation under visible sunlight.

To propose a plausible kinetic mechanism of the adsorption process, adsorption kinetics was investigated based on pseudo-first-order (PFO) and pseudo-second-order (PSO) models. Table 1 depicts the adsorption capacities, kinetic coefficients, reduced chi-square, and residual sum of squares for PFO and PSO kinetics for adsorption of methylene blue dye on CuBTC/ZnO composite beads. The correlation coefficient (R²) for PFO (0.9653) model deviated more from the unity value compared to the PSO (0.9968) model. The closeness of the R² value for the PSO model indicated that the non-linear second-order model fits well for the adsorption of methylene blue on CuBTC/ZnO chitosan beads. The better fitting of the PSO model confirmed that the adsorption was chemical in nature (Kuwer et al. 2021). PSO showed a lower value for the residual sum of the square and reduced chi-square than the PFO model. This suggested that the difference between the theoretical and experimental values for PSO and PFO was minimum. The initial adsorption of methylene blue was faster for PFO than PSO (Figure 9).

The Freundlich and Langmuir isotherm models were used to investigate the adsorption of methylene blue dye on CuBTC/ZnO composite beads. The regression coefficient (R²), reduced chi-square, and residual sum of squares were evaluated to differentiate between two isotherm models (Table 2). The R² value was close to unity in the case for the Langmuir model (Freundlich: 0.8036 and Langmuir: 0.9914), confirming that the Langmuir isotherm model fit. The Langmuir model gives a fraction of the surface covered by adsorbate molecules at a particular temperature and pressure. A typical Langmuir isotherm shows a characteristic horizontal asymptote indicating saturation after the monolayer adsorption. This indicated that the adsorption mechanism of methylene blue dye on CuBTC/ZnO chitosan composite beads was chemisorption (Bangari et al. 2022a). The closeness of experimental and predicted data points was also observed (Figure 10).

To comment upon the removal of methylene blue dye by the CuBTC/ZnO chitosan composite beads, the role of the individual component in the composite must be studied. The BTC ligand or the benzene tricarboxylate ion of the CuBTC is rich in charge density due to the availability of π-electrons of the benzene ring and the carboxylate groups (Hu et al. 2015). The electron-rich profile of the ligand in the CuBTC makes the adsorption easier by attracting the cationic methylene blue dye molecule. Further, CuBTC has large octahedral cavities that facilitate adsorption. The methylene blue dye molecules are present at a higher rate constant than the single molecules. Chitosan, a sugar polymer, adsorbs the methylene blue dye molecules in aggregate form on its surface (Kellner-Rogers et al. 2019). Since the adsorption process follows the PSO model, methylene blue dye was adsorbed in aggregated form. Hence, CuBTC present in the CuBTC/ZnO chitosan composite beads is majorly responsible for providing the vacant sites for the adsorbed methylene blue dye molecules. In contrast, the chitosan provides a surface where the methylene blue molecules aggregated and adsorbed. The CuBTC/ZnO chitosan composite beads’ degradation efficiency was tested to degrade methylene blue dye under solar irradiation. ZnO is well known for its photocatalytic activity and thus imparts photocatalytic properties to the CuBTC/ZnO composite beads (Biswal et al. 2021). ZnO has a semiconducting nature at room temperature with a bandgap of 3.37 eV (Razavi-Khosroshahi et al. 2017). Irradiating ZnO with sunlight is sufficient to promote the electrons from the filled valence band to the empty conduction band. This excitation results in the formation of electron-hole pairs that bears the ability to delocalize throughout the crystal lattice of ZnO. Water molecules are split into H⁺ ion and OH^• radical by holes while the electron reacts with oxygen molecules to generate superoxide ions. Further, these superoxide ions react with H⁺ formed to give hydroperoxyl radicals (⁠⁠). Two hydroperoxyl radicals yield hydrogen peroxide, which reacts with superoxide ions yielding hydroxyl free radicals. The oxidizing ability of these free radicals is very high. The attack of resulting hydroxyl radicals on adsorbed methylene blue dye molecules results in the rapid formation of degradation products via intermediate formation. The conversion of methylene blue dye into less harmful degradation products was traced back to the formation of active oxygen species (Figure S3, ESI). The photodegradation mechanism confirmed that in the due course of degradation process H₂O₂ is generated and combines with the superoxide ions. The resulting hydroxyl radical (OH^•) breaks methylene blue into intermediates that further decompose to CO₂ + H₂O (Ayodhya & Veerabhadram 2018). The stability and reusability of the CuBTC/ZnO chitosan composites were tested over methylene blue dye after recycling to replicate the experiments under identical conditions (Figure S4, ESI). More specifically, under sunlight irradiation, CuBTC/ZnO chitosan composites displayed sustained and consistent behaviour up to the 5th cycle (around 80%), which proves the composite's stability and recyclability.

This study reports the facile synthesis of CuBTC/ZnO chitosan composite beads and adsorption of methylene blue dye on it, followed by its photocatalytic regeneration. The SEM micrograph showed the nanoflower morphology of the ZnO and the octahedral units of CuBTC. The surface of the beads was rough and porous. XRD and the FT-IR spectra confirmed the structural properties of the CuBTC and ZnO nanoflowers. Further, the BET surface area of the CuBTC/ZnO chitosan composite beads was 5.91 m² g⁻¹. The TGA analysis showed that incorporating ZnO imparts thermal stability to the composites. At pH 7, 1.6 g L⁻¹ of the adsorbent gave a maximum adsorption capacity of 98.75% at the equilibration time of 90 min. The factors that governed these values were mainly the formation of NaCl during the adsorption process, decrease in the active sites on the adsorbent, and the aggregation of composites. The PSO models were found to best fit for the adsorption process. Of the Freundlich and the Langmuir models used to analyze the adsorption process, the Langmuir value was more useful, as was evident from the R² values. The Langmuir and PSO models’ fitting indicated that the adsorption interaction between methylene blue and composites was chemisorption. It was clear from the mechanism that the role of chitosan is crucial in both the adsorption of methylene blue and in determining the kinetics of the process. The photocatalytic degradation was mainly due to ZnO, where the active oxygenated species were responsible for degrading the dye. The regeneration studies showed that the composites could be used up to 5 cycles without significant loss in the efficiency towards removing methylene blue.

The CSIR-CSMCRI PRIS number for this manuscript is 031/2022. The authors are grateful for partial funding support from the Council of Scientific and Industrial Research, India (MLP-0043). The authors acknowledge AED&CIF, CSMCRI for providing instrumental facilities. The authors also thank Ms Pratibha Yadav, IEHE Bhopal for help in synthesizing materials. The comments from anonymous reviewers and the editor have greatly improved the content.

We certify that the authors are not affiliated with or involved with any organization or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this paper.

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