1 Introduction

Heavy metals and industrial organic dyes are regarded as the most vital sources of water pollution [1]. The discharge of dye residues by the dye manufacturing and textile industries is highly carcinogenic and toxic to human health [2]. Since dyes are discharged in high concentration, their removal by the conventional techniques, such as adsorption on activated carbon, biological oxidation, coagulation and filtration is ineffective and expensive [3]. Therefore, it is more significant to design an economical, simple and eco–friendly method that can remove toxic dyes from contaminated water resources [4]. Environmental remediation based on visible light induced semiconductor photocatalysis has been shown as an efficient and cost–effective technique for removing dyes from wastewater [5,6,7,8].

Up to now, most of the materials used in photocatalysis are semiconductors, such as ZrO2, ZnO and TiO2. Environmental friendly zirconia (ZrO2) is among one of the important metal oxides with cost–effective, high chemical inertness, biocompatible, less toxicity, high thermal stability and electrochemical nature [2, 9]. The surface of ZrO2 has both basic and acidic properties, and therefore, display reduction and oxidation properties, as well as photoactivity [10]. As a typical functional and structural material, ZrO2 has been extensively used in solid oxide fuel cells [11], adsorbents [12], catalyst supports [13] and photocatalysis [14]. Despite the good chemical stability and activity of ZrO2, its photoactivity was restricted due to its large bandgap energy (5–5.8 eV) and faster rate of electron–hole pairs recombination [15]. In an attempt to extend its visible light photoactivity and minimise the electron–hole pairs recombination rate, several approaches such as elemental doping, dye sensitisation, hybridising with visible light bandgap semiconductors or carbon materials and others have been used [16].

The design of heterojunction materials has been used as an effective strategy in water treatment to decrease the electron–hole pairs recombination rate and extend the visible light absorption edge [17]. Up to now, graphene oxide (GO) with a fascinating sp2 carbon arrangement in the two–dimensional carbon framework has attracted much attention because of its high Young’s modulus, large surface area, high electron conductivity and charge carrier mobility [18]. Again, studies have shown that GO is an exceptional absorber of organic pollutants because of its π–π loading interaction and aromatic nature [19]. Based on the aforementioned properties, graphene oxide–metal oxide nanocomposites with excellent photochemical properties and structural morphology have been used in water treatment via the elimination of heavy metals and degradation of toxic organic contaminants [20]. The decoration of ZrO2 on GO sheets as a photocatalyst and fuel cell transistor has been reported [21], where ZrO2 nanoparticles showed endowed properties when coupled with GO sheets [22].

Besides the design of heterojunction nanomaterials, doping of lanthanide ions has been demonstrated to further promote the separation of charge carriers and visible light performance [23]. The presence of lanthanide ions also promotes the semiconductor spectra to shift from a lower energy range leading to the enhancement in their photoactivity [24]. Due to the unique spectral characteristics and 4f electronic configuration, lanthanide ions are ideal dopants for modifying the surface adsorption, optical properties, electronic and crystal structures of several semiconductors [25]. Moreover, the introduction of lanthanide ions could form complexes with several organic molecules by concentrating the pollutants on the photocatalyst surface and hence, promote the photocatalytic reactivity [26]. As one of the lanthanide ions, several studies have shown that the use of cerium ion as dopant leads to excellent photoactivity [27]. Because of the higher separation of charge carriers, and the adsorption equilibrium constant, Ce3+–TiO2 showed an enhanced photoactivity towards 2-mercaptobenzothiazole degradation. Also, Xie and Yuan [28] reported that Ce4+–TiO2 exhibited enhanced photoactivity towards X-3B degradation.

Though studies on the visible light performance of Ce–ZrO2 and ZrO2–GO nanocomposite have been reported, studies on Ce–ZrO2–GO as a visible light material is unknown. Therefore, this study was aimed to prepare several hybrid visible-light-driven Ce–ZrO2–GO (0.3, 0.6 and 1.00% Ce) nanocomposites using a co-precipitation technique and examine its photocatalytic performance via the photodegradation of eosin yellow (EY) dye under solar light irradiation. In the Ce–ZrO2–GO nanocomposite, the introduction of Ce3+ as dopants can create energy levels just below the conduction band (CB) of ZrO2, thereby promoting its visible light performance, while the introduction of GO sheets with ZrO2 can promote the lifetime of the charge carriers by forming a heterojunction. In this study, the transfer pathway of charge carriers, electronic and structural properties of hybrid Ce–ZrO2–GO nanocomposites were theoretically investigated to assist in explaining the origin of the improved photocatalytic activity and stability. This study could offer novel insights for a better understanding of the role of Ce3+ ions and GO sheets for the removal of organic pollutants from wastewater.

2 Materials and Methods

2.1 Materials and Reagents

All the chemicals obtained from Sigma–Aldrich was used without further purification.

2.2 Fabrication of GO

The GO was prepared following our earlier report [29] and employed the modified Hummers method [30].

2.3 Preparation of Ce–ZrO2–GO Photocatalyst

A literature procedure [31], which was adopted for the synthesis of the composites, involved the dispersion and sonication of 9.81 g of Zr(NO3)2.6H2O and a specific quantity of Ce(NO3)3.6H2O (0.04527, 0.09054 and 0.1505 g) in 50 mL of deionised water to obtain 0.3, 0.6 and 1.0% of Ce–doped ZrO2 nanocomposites. The mixtures were stirred magnetically for an hour. After the addition of 0.5% of GO, stirring was continued at 50 °C for a further 2 h. A solution of 3M KOH was then added dropwise under stirring conditions into the Ce–ZrO2–GO mixture until a precipitate was formed at a pH of 10. The obtained nanocomposites [Ce–ZrO2–GO (0.3, 0.6 and 1.0% Ce)] were rinsed with both water and ethanol using a centrifuge to get rid of any contaminants. Finally, the isolated nanocomposites were dried overnight at 90 °C and calcined at 300 °C for 3 h. Pure ZrO2 and ZrO2–GO nanocomposite were also fabricated by the same technique, but without the addition of either Ce(NO3)3.6H2O or both GO and Ce(NO3)3.6H2O, respectively.

2.4 Characterisation

Perkin Elmer Raman microscope (Raman Micro 200) and Perkin Elmer FTIR spectroscopy (Spectrum 100) were used to measure the Raman and FTIR spectrum, respectively. The microstructure, crystal phase and morphology of the fabricated samples were characterised by Philip’s X-ray diffractometer at 30 mA and 40 kV. The surface morphology of the photocatalyst was evaluated by the scanning electron microscopy on TESCAN (Vega 3XMU), while transmission electron spectroscopy was used to measure the size and internal morphology of the prepared samples (JEOL, JEM-2100F). The Shimadzu UV-2540 Spectrophotometer with BaSO4 as a reflectance standard was used to evaluate the optical properties.

2.5 Photocatalysis Study

The photocatalytic degradation reaction of the as-synthesised nanocomposite was examined using a Full-Spectrum Solar Simulator (Newport, Model: 9600). The solar simulator was equipped with 150 W ozone free xenon lamp, where a collimated beam of 33 mm diameter, which is an equivalent of 1.3 suns was produced. The EY dye concentration in the sample of 5 mL suspension was evaluated by a Shimadzu UV-2450 spectrophotometer at a λ of 516 nm. In this process, 100 mg of the photocatalyst was suspended in an aqueous solution of 20 mg/L of IC dye and then magnetically stirred in the dark, prior to irradiation, for 1 h and at 25 C in a quartz beaker (150 mL) and this was carried out to create an adsorption equilibrium of IC dye with the semiconductor photocatalyst. After the adsorption equilibrium, the light sources were switched on for photodegradation of IC dye at 30 min intervals by the various photocatalysts and aliquots (5 mL) was withdrawn from the solution using a disposal syringe containing filtered membrane of 4 µm for a period of 210 min.

2.6 Density Functional Theory Calculations

The electronic and structural properties were performed using a plane-wave DFT calculation implemented in the CASTEP code [32] of the Materials Studio 2016 [33]. The geometry optimisation was carried out using the conjugate gradient method [34] and the generalized gradient approximation (GGA) with the Perdew–Burke–Ernzerhof (PBE) functional [35] for the exchange-correlation potential. The norm-conserving pseudopotential was used to describe the core-valence electrons interaction [36]. A valence states of Zr (4d2 5s2), O (2s2 2p4), C (2s2 2p2) and Ce (4f2 6s2) were considered to construct the potential. In this study, the electronic property was considered using the hybrid HSE06 functional [37], since the GGA–PBE functional normally underestimate the electronic properties of most photocatalysts [38]. The convergence criteria for maximum force and energy, and the cut-off was set at 0.01 eV/Å, 10−6 eV/atom and 500 eV, respectively. A large vacuum space of 20 Å was used to prevent the interactions between the periodic images. The Brillouin zone was described by the Monkhorst–Pack and a k-point grid of 2 × 2 × 1, 12 × 12 × 1 and 3 × 3 × 1 were used for the ZrO2, GO sheet, ZrO2–GO and Ce–ZrO2–GO nanocomposites, respectively [39].

3 Results and Discussion

3.1 XRD Pattern Analysis

The crystal structures and phase purity of pure ZrO2, GO sheets, ZrO2–GO and Ce–doped ZrO2–GO (with varied Ce concentrations) nanocomposites were evaluated using the X–ray diffractometer, and the corresponding X-ray diffraction pattern is given in Fig. 1a.

Fig. 1
figure 1

a XRD pattern and b FTIR spectra of pure ZrO2, GO sheets, ZrO2–GO and Ce–ZrO2–GO (with variable concentrations of Ce) nanocomposites

Full size image

The characteristic XRD peaks of the tetragonal phase observed at 30.2°, 34.5°, 50.2° and 60.2° corresponding to the (101), (200), (220), and (311) reflection planes, respectively [40]. However, the diffraction peaks corresponding to 16.4°, 24.5°, 28.1° and 32.7° indexed to Miller indices of (011), (110), (111) and (110), respectively, agreed with the monoclinic phase of ZrO2 (JCPDS No. 89-7710). The broad XRD pattern obtained for ZrO2 nanoparticles was due to the small crystallite size of the chemically precipitated ZrO2 [41], as well as the presence of tetragonal and monoclinic crystal structure [42]. The GO sheets showed a sharp diffraction peak around 2θ = 10.72° with a 0.77 nm interlayer distance, which corresponds to the (002) plane [43]. Due to the low concentration of GO in the nanocomposites, a low-intensity diffraction peak was observed for GO sheets in ZrO2–GO nanocomposite and this was in agreement with earlier studies on XRD characterisation of GO-based nanocomposites [44]. No peaks relating to Ce2O3 were observed in the Ce–ZrO2–GO nanocomposite. This revealed that Ce remained as a dispersed phase on the ZrO2 surface or substituted Zr in the lattice sites [45]. The higher ionic radius of Ce3+ (1.03 Å) than that of Zr4+ (0.72 Å) indicated that surface doping occurred on the ZrO2 surface [45]. In the Ce–ZrO2–GO (with variable concentrations of Ce) nanocomposites, peaks corresponding to monoclinic ZrO2 and GO sheets were identified, confirming the successful fabrication of the nanocomposites.

3.2 FTIR Analysis

The functional groups of ZrO2, GO sheets, ZrO2–GO and Ce–doped ZrO2–GO nanocomposite (with varied Ce concentrations) were determined by the FTIR analysis, as given in Fig. 1b. The peak observed at 3460 and 1635 cm−1 relates to hydroxyl stretching and bending vibration bands of the absorbed water and hydrated oxide surface [46], while the broad absorption band situated at 576 and 750 cm−1 were ascribed to the Zr–O vibration [46]. The broad and strong peak at 3416 cm−1 in GO sheets could be related to the O–H stretching vibration. The broad absorption peak around 1614 and 1732 cm−1 were due to C=O and carbonyl CO stretching, respectively, while the peaks at 1420 and 1055 cm−1 corresponded to the carboxyl OH stretching and C–O epoxy group at the edges of graphene oxide, respectively [43]. Due to the small concentration of GO sheets in the nanocomposite, the FTIR spectra of the ZrO2–GO nanocomposite showed bands corresponding to only ZrO2. However, the FTIR spectra of Ce–ZrO2–GO (with variable concentrations of Ce), which exhibit bands associated with ZrO2 and GO, confirmed the successful fabrication of the nanocomposite.

3.3 Raman Studies

To check the different phases in the as-prepared nanocomposite samples, Raman spectra were plotted, and the results are given in Fig. 2.

Fig. 2
figure 2

Raman spectra of a pure ZrO2, b GO sheets and ZrO2–GO and (c) the Ce–ZrO2–GO nanocomposites

Full size image

Pure ZrO2 and ZrO2–GO (Fig. 2a) showed peaks at 178, 334, 380, 476, 540, 558, 618 and 638 cm−1 and these pertained to the Ag, Bg, Ag, Ag, Ag, Ag and Ag modes of the monoclinic phase, respectively [45]. The GO and ZrO2–GO nanocomposite showed two distinct D (1366 cm−1) and G (1599 cm−1) Raman bands [47]. Raman spectra of Ce–ZrO2–GO nanocomposites showed peaks related to monoclinic ZrO2 and GO sheets and this indicated that ZrO2 and GO sheets were present in the nanocomposites, which were consistent with the XRD results. The D and G band of GO was due to the breathing mode of A1g and E2g mode of ordered sp2-bonded C atoms, respectively. The degree of disorder in GO was obtained by the ratio of the intensity of D-band to that of the G-band [48]. Based on the different reduction levels, the intensity ID/IG ratio is a measure of defect concentration in graphene sheets, where a smaller intensity of ID/IG ratio suggests fewer sp3 defects/disorder and better average size of the in-plane graphitic crystallite sp2 domains [49]. The ID/IG ratio of GO sheets (0.96) in the present study agreed with an earlier study (1.04) [50]. Also, ZrO2–GO and Ce–ZrO2–GO nanocomposites showed a slight decrease in the ID/IG ratio of 0.75 and 0.72, respectively, compared with GO sheets. This suggests a strong interaction in the nanocomposites.

3.4 SEM and TEM Analysis

The nanostructure, morphology and uniformity of the as-fabricated nanocomposites were determined using TEM and SEM analysis (Fig. 3).

Fig. 3
figure 3

SEM images of a ZrO2, b ZrO2–GO and c Ce–ZrO2–GO; TEM images of d ZrO2, e GO and f Ce–ZrO2–GO; g EDX spectrum of Ce–ZrO2–GO and h Elemental mapping of Ce–ZrO2–GO

Full size image

Figure 3a–c shows the SEM image of ZrO2, ZrO2–GO and Ce–ZrO2–GO nanocomposites. The SEM image revealed a distinct pebble–like particles of ZrO2 nanoparticle, which was irregular in shape (Fig. 3a). The incorporation of GO aids ZrO2 nanoparticle to disperse smoothly in ZrO2–GO nanocomposite, where GO sheets were in the form of flakes. The separation between the agglomerated nanoparticles increased in the Ce–ZrO2–GO nanocomposite. Figure 3e showed that the TEM image of GO appeared as a 2D transparent structure, which confirmed the 2D nanosheets of GO with micrometre-long wrinkles. Due to the interfacial interactions and heterogeneous nucleation, ZrO2 nanocrystals were densely deposited on the wrinkled paper–like structure of GO sheets (Fig. 3f). The nanomaterial composition is obtained by EDX analysis. The presence of Zr, Ce, O and C elements in the Ce–ZrO2–GO nanocomposite was confirmed using the EDX spectrum (Fig. 3g). The atomic percentage was about 70.35 and 24.21% of O and C in the EDS spectrum, while the Ce concentrations were ca. 0.36, 0.64 and 1.06 wt%, respectively. The elemental mapping obtained showed that Zr, Ce, O and C were uniformly dispersed in the Ce–ZrO2–GO nanocomposite (Fig. 3h).

3.5 Bandgap Determination

The bandgap energy (Eg), which is the difference between the valence band (VB) and CB was used to characterised the band structure of Ce–ZrO2–GO nanocomposites [51]. The bandgap energy of the as–fabricated materials was obtained from the Kubelka–Munk formula [52]:

$${\left( {\alpha hv} \right)^{1/n}}=A (hv - {E_{\text{g}}})$$
(1)

where h is Planck’s constant, v is the photon’s frequency, α is the absorption coefficient, Eg is the band gap and A is the slope of the Tauc plot in the linear region. The value of the exponent denotes the nature of the electronic transition, whether allowed or forbidden and whether direct or indirect [53]. n = 1/2 and 2 for direct and indirect transitions, respectively [53]. The optical transition for pure ZrO2 was indirectly allowed since the n value was 2. Therefore, the bandgap energies were evaluated from a plot of \({\left(\alpha hv\right)}^{1/2}\)versus photon energy, and the intercept of the tangent to the x-axis gave a better estimation of the bandgap energy for pure ZrO2, GO sheets, ZrO2–GO and Ce–ZrO2–GO nanocomposite [17]. The bandgap energies of pure ZrO2 and ZrO2–GO nanocomposite were calculated as 5.68 and 3.29 eV, respectively (Fig. 4).

Fig. 4
figure 4

Plot of (αhv)1/2 against hv for pure ZrO2, ZrO2–GO and Ce–ZrO2–GO (with variable concentrations of Ce)

Full size image

The reduced Eg is because of the chemical bonding between the active sites of GO and ZrO2 [54]. The bandgap energy of ZrO2–GO nanocomposite was very close to the 3.25 eV previously reported [55]. As obtained from the Tauc plot, the Eg values of Ce–ZrO2–GO (0.3% Ce), Ce–ZrO2–GO (0.6% Ce) and Ce–ZrO2–GO (1.0% Ce) nanocomposites were 2.82, 2.87 and 2.93 eV, respectively (Fig. 4). The narrow Eg of Ce–ZrO2–GO nanocomposite might be ascribed to the defective energy levels and electronic interaction [56]. Based on the above results, the favourable Eg values of Ce–ZrO2–GO nanocomposite compared with the pure ZrO2 and ZrO2–GO nanocomposite clearly revealed that the Ce–ZrO2–GO nanocomposites could have a better electron migration property, and therefore, could act as an efficient photocatalyst material with improved photoactivity under simulated light irradiation [51].

3.6 Optical Analysis

The performance of visible light absorption is a significant contributor to accessing the optical property of semiconductor-based photocatalyst materials [57]. The optical performance of the as-synthesised nanocomposites was measured with the UV–Vis absorbance spectroscopy. The absorbance spectra of ZrO2 and Ce–ZrO2–GO nanocomposite with variable concentrations of Ce are given in Fig. 5a.

Fig. 5
figure 5

a UV–Vis absorbance spectra and b photoluminescence (PL) spectra of ZrO2, ZrO2–GO and Ce–ZrO2–GO (with variable concentrations of Ce)

Full size image

Both pure ZrO2 and ZrO2–GO nanocomposite showed weak absorption, which was attributed to their wide bandgap energy (Fig. 5a). The Ce–ZrO2–GO nanocomposites showing a broader optical-response range compared with pure ZrO2 and ZrO2–GO was due to the π \(\to\) π* transitions of the C=C bond [21]. The improved photoactivity was because of the highly visible light absorption edge and reduce electron–hole pairs recombination rate of ZrO2 by the introduced Ce and GO sheets. The incorporation of different concentrations of Ce significantly influenced the optical performance of visible light absorption of Ce–ZrO2–GO nanocomposites (Fig. 5a). The Ce–ZrO2–GO (0.3% Ce) nanocomposite showed the highest visible light absorption and this suggested that the optimum Ce concentration for optimum visible light absorption was 0.3%. Thus, the obtained Ce–ZrO2–GO (0.3% Ce) nanocomposite was expected to show an improved photoactivity than bulk ZrO2 under simulated visible light irradiation.

3.7 The Photoluminescence (PL) Analysis

In the photocatalysis process, the photogenerated charge carrier recombination is among the key issue responsible for the low photocatalysis efficiency. Generally, PL analysis can offer in-depth insights into the separation and the recombination rate of photoexcited charge carriers in the photocatalysts [58]. Stronger PL intensity usually indicates a faster charge carrier recombination rate. Figure 5b shows the PL spectra of the fabricated photocatalyst. Due to π–π* electronic transitions, a strong peak at 416 nm was detected in all the as-prepared photocatalysts. The PL spectra of pure ZrO2 showed a weak emission peak in the visible spectrum, which was because of the fast recombination of charge carriers. The PL intensity of ZrO2 was reduced after the introduction of GO sheets, whereas with the addition of Ce, a further strong reduction in PL was observed. This significant PL quenching was attributed to the efficient electron migration from ZrO2 onto GO sheets via Ce ion. Thus, the PL spectrum of ZrO2–GO nanocomposite was significantly quenched compared to pure ZrO2, which was because of the interactions between the GO sheets and the excited state of ZrO2, and the interfacial charge migration in ZrO2–GO nanocomposite. Moreover, the hybrid Ce–ZrO2–GO nanocomposite showed more quenching of PL emission compared to ZrO2–GO, which revealed that Ce–ZrO2–GO had a better electrical conductivity than ZrO2, thus leading to better separation of charge carriers. Therefore, the interaction among GO sheets, ZrO2 and Ce ion in the hybrid Ce–ZrO2–GO nanocomposite could contribute to the efficient separation, thereby prolonging the lifetime of charge carriers, and further improved the photodegradation efficiency. The highest PL quenching occurred in Ce–ZrO2–GO (0.3% Ce) nanocomposite, suggested that the 0.3% Ce was the optimal concentration to obtain the best efficiency of charge carrier migration. Herein, when the concentration of Ce ion was excessive, the number of active sites capturing the photoexcited electron was reduced by increasing the size of Ce nanoparticles. Thus, excessive Ce ions could cover the ZrO2 surface, leading to an increased recombination rate of charge carriers and reduced photoactivity of the photocatalysts [59].

3.8 Assessment of Photocatalytic Performance

The photoactivity of the as-fabricated pure ZrO2, ZrO2–GO and Ce–ZrO2–GO (with variable concentrations of Ce) were tested by the photodegradation of EY dye under UV light irradiation. In addition, dark absorption of EY dye was accessed for 60 min to evaluate self-degradation. The absorbance remained virtually the same for bulk ZrO2, ZrO2–GO and Ce–ZrO2–GO (with variable concentrations of Ce) photocatalysts, suggests that there is no absorption in the dark. The photocatalytic degradation efficiency of the as-fabricated photocatalyst was evaluated following Eq. (2) [60]:

$${\text{Photodegradation}} \left( {\text{\% }} \right)=(1 - C/{C_o}) \times 100{\text{\% }}$$
(2)

where C and C0 represent the residual concentration of EY dye at different light intervals and the concentration of EY dye at adsorption equilibrium, respectively [60]. Pure ZrO2 nanoparticles showed the lowest photodegradation performance (Fig. 6a).

Fig. 6
figure 6

a Photocatalytic degradation of eosin yellow dye under UV light irradiation of pure ZrO2, ZrO2–GO and Ce–ZrO2–GO (with variable concentrations of Ce), b plot of ln(C0/C) vs. light irradiation time for photodegradation of eosin yellow solution containing pure ZrO2, ZrO2–GO and Ce–ZrO2–GO (with variable concentrations of Ce), c nitrogen adsorption–desorption of the fabricated photocatalysts, d visible light irradiation of eosin yellow solution in the presence of Ce–ZrO2–GO (0.3% Ce) photocatalyst and e role of active species on the photodegradation of eosin yellow solution

Full size image

For pure ZrO2 nanoparticles, the degradation rate was only 16% (see Table 1) after 210 min of simulated light irradiation.

Table 1 Some parameters for pure ZrO2, ZrO2–GO and Ce–ZrO2–GO (with variable concentrations of Ce)
Full size table

The introduction of GO sheets significantly accelerated the photodegradation process of ZrO2–GO nanocomposite. The percentage photodegradation was increased to 31% of EY dye when GO was added, signifying the substantial effect of GO in increasing the photodegradation of ZrO2–GO nanocomposite. The promoted photoactivity of ZrO2–GO nanocomposite was ascribed to the close contact between the ZrO2 and GO sheets. In contrast, Ce–ZrO2–GO nanocomposites with several concentrations of Ce showed improved photoactivity for EY dye degradation than pure ZrO2 and ZrO2–GO. When the concentration of Ce was less than 0.3%, the photodegradation was enhanced. Nonetheless, due to the accessibility of active sites and efficient penetration of light on the ZrO2 surface, any further increase in the amount of Ce during the co-precipitation fabrication reduced the photodegradation performance. This is comparable with other studies [61]. Interestingly, the Ce–ZrO2–GO nanocomposites with 0.3% of Ce showed the highest photoactivity with almost 100% photodegradation of EY dye in 210 min. This high efficiency for EY degradation by Ce–ZrO2–GO nanocomposites was found to be slightly higher compared with that of GO–Fe3O4@ZrO2 (98%) [21], but comparable to the GO–Ag3PO4 nanocomposite with 0.26 wt% GO [50]. This clearly showed that the Ce–ZrO2–GO (0.3% Ce) nanocomposite rendered the best catalytic efficiency. The photodegradation efficiency of the catalyst was due to the ratio of Ce and Zr in the nanocomposite. As expected, the Ce seems to act as a charge carriers scavenger to minimise their recombination rate [31].

3.9 Kinetics Studies

The kinetics of EY dye degradation under light irradiation was also investigated. As revealed in Fig. 6b, the kinetics of EY dye degradation was pseudo first-order reaction, since pure ZrO2, ZrO2–GO and Ce–ZrO2–GO (with variable concentrations of Ce) showed linear lines [60]. The first-order kinetics indicated that the photocatalytic reaction rate was mainly reliant on the absorption efficiency and light intensity of the EY dye.

The photodegradation rate constant, k is evaluated from the kinetic method:

$${\text{ln}}\left( {{\text{C}}/{{\text{C}}_0}} \right){\text{ }}={\text{ }}-kt$$
(3)

where C and C0 are the concentration of a reaction time t and initial concentration, respectively [60]. The rate constants are estimated as 8.95 × 10−4 and 3.11 × 10−3 min−1 for pure ZrO2 and ZrO2–GO nanocomposite, respectively. The photoactivity of ZrO2–GO nanocomposite is about 3.47-fold compared with pure ZrO2 nanoparticles. The rate constants for Ce–ZrO2–GO (0.3% Ce), Ce–ZrO2–GO (0.6% Ce) and Ce–ZrO2–GO (1.0% Ce) nanocomposites were calculated as 1.43 × 10−2, 9.40 × 10−3 and 7.30 × 10−3 min−1, respectively. The apparent rate constants of Ce–ZrO2–GO (0.3% Ce) is 15.98, 1.52 and 1.96 times higher compared with ZrO2 nanoparticles, Ce–ZrO2–GO (0.6% Ce) and Ce–ZrO2–GO (1.0% Ce) nanocomposites, respectively. Generally, a higher BET surface area (SBET) could promote the photoactivity of the photocatalyst [62], since the photocatalytic process is linked with the adsorption and desorption of pollutants on the photocatalyst surface [63]. To evaluate the effect of GO sheets and Ce dopant on the BET surface area of ZrO2, nitrogen adsorption–desorption isotherm is employed (Fig. 6c).

The Ce–ZrO2–GO (0.3% Ce) nanocomposite has SBET of 86.45 m2 g−1 (Table 1) and this is about 3.23-fold higher compared with pure ZrO2 (26.79 m2 g−1). The higher surface area was ascribed to the coupled GO sheets, which had a large exposed area, leading to improved adsorption of pollutants and most active sites on the nanocomposite surface [64]. The high k-value of Ce–ZrO2–GO (0.3% Ce) nanocomposite is ascribed to the combined synergy of both the optimum Ce concentration and GO in the nanocomposite. In conclusion, the incorporation of optimum amounts of Ce and GO on the ZrO2 nanoparticle resulted in an enhancement of photocatalytic pollutant degradation performance.

3.10 EY Photodegradation Profile

Figure 6d shows the absorption spectra of EY dye solution of the as-prepared Ce–ZrO2–GO and exposure to the irradiation of light under different durations. The absorption peak of 515 nm corresponding to the EY dye was used to check the photodegradation process. This absorption peak was comparable with other studies [63]. The EY photodegradation profile (Fig. 6d) showed a constant decrease in the highest absorption wavelength (λ = 515 nm) of EY over a period of exposure time (0–180 min), indicating the direct relationship between the dye degrading and exposure time. This occurrence was ascribed to the presence of Ce, which scavenge the photoexcited electrons and reduce their recombination with the holes, thereby promoting the photoactivity of the photocatalyst.

3.11 Experiment for Radical Scavenging

The radical trapping experiment was carried out to comprehend the role of key active species involved in the photodegradation of EY dye over ZrO2 photocatalyst. This was done using benzoquinone (BQ) [65] and isopropanol (IPA) as O2−• and HO• scavengers, respectively [66]. As given in Fig. 6e, the photoactivity of Ce–ZrO2–GO (0.3% Ce) nanocomposite is clearly suppressed by adding BQ or IPA. The degradation rate constant of EY dye decrease of 8.73 × 10−3 min−1 (42% degradation) in the existence of BQ (Table 2) compared with the 1.43 × 10−2 min−1 (58% degradation) when BQ was absent.

Table 2 The rate of photocatalytic degradation in the presence or absence of radical scavenger
Full size table

The maximum degradation in the presence of BQ is lower than the 58.2% reduction in photodegradation efficiency with the addition of BQ to Bi–TiO2–ZrO2 nanocomposite [45]. Likewise, adding an IPA decreased the photocatalytic degradation rate to 7.26 × 10−3 min−1, which was ~ 1.96 times slower compared to the photodegradation without IPA. The obtained result signifies that both HO• and O2−• radicals are responsible for the degradation of EY dye. This is similar to other studies [67].

3.12 Total Organic Carbon analysis (TOC) Analysis

General, photodegradation does not indicate complete mineralization since the pollutants could be photocatalytically degraded into small molecules. Fusion UV/persulfate TOC analysis was performed to study the extent of mineralisation of eosin yellow dye by the photocatalyst [68]. The experiment was performed over a period of 5 h. According to Fig. 7a, 76% TOC removal was reached after 5 h showing that a considerable degree of mineralisation of the dye by the nanocomposite was achieved within 5 h.

Fig. 7
figure 7

a Percentage Total Organic Carbon removal and b recycle test of Ce–ZrO2–GO (0.3% Ce) nanocomposite for eosin yellow degradation under simulated light irradiation

Full size image

This indicated that the reactive species produced by Ce–ZrO2–GO (0.3% Ce) nanocomposite could not only photodegrade EY dye, but also can efficiently decompose EY dye into CO2 and water [69]. Although the Ce–ZrO2–GO (0.3% Ce) nanocomposite exhibited an enhanced photoactivity, higher stability is vital in view of its industrial applications. Therefore, to promote the reusability efficiency from the industrial perspective, the photocatalyst was anticipated to be photostable throughout the reaction. Under the same condition, the recyclability experiment was performed in a five-run cycling test to study the performance of the Ce–ZrO2–GO (0.3% Ce) nanocomposite. The as-prepared nanocomposite was recycled, cleaned and dried for each batch of the experiment. The photodegradation of EY dye in the first cycle was 87%, which was then reduced in the next cycles and finally reached 70% after the fifth cycle (Fig. 7b). The reduction in the photocatalytic efficiency after its recycled was induced by the reduction of the number of active sites accessible on the Ce–ZrO2–GO (0.3% Ce) nanocomposite surface after each cycle of photodegradation. However, there was no much loss of performance after five cycles of photodegradation reaction, which signifies the stable structure of Ce–ZrO2–GO (0.3% Ce) nanocomposite during the degradation process. This indicated that the Ce–ZrO2–GO (0.3% Ce) nanocomposite maintained a good performance towards the degradation of EY dye even after the fifth cycle.

3.13 Theoretical Study

3.13.1 Catalytic Stability

The Ce–ZrO2–GO nanocomposite was constructed with a 4 × 2 supercells of ZrO2 and 4 × 4 supercells of GO sheets. This gave rise to a 3.13% lattice mismatch, which established the fabrication of the ZrO2–GO and Ce–ZrO2–GO photocatalyst [70]. The interactions between the ZrO2 (or Ce–ZrO2) and GO sheets determines the photocatalytic efficiency and the origin of charge transfer of the as-prepared nanocomposite. Therefore, the interface adhesion energy and interlayer spacing were accessed to examine the physical properties of the as-fabricated nanocomposite using a hybrid DFT study. In this study, the interlayer spacing was determined as the distance between the bottom-most and top-most atoms of pure ZrO2 (or Ce–ZrO2) and GO sheets, respectively. For the ZrO2–GO and Ce–ZrO2–GO nanocomposites, the interlayer spacing was calculated as 3.19 and 3.06 Å, respectively. This agreed with other GO-based photocatalysts [71, 72]. Since the separation characterises the strength of interfacial interactions, the Ce–ZrO2–GO nanocomposite with a smaller interlayer spacing suggest a stronger interfacial interaction than the ZrO2–GO nanocomposite [73].

Since the ZrO2–GO and Ce–ZrO2–GO nanocomposites were successfully fabricated, the catalytic stability and possibility of the as-prepared nanocomposites were evaluated by comparing the interface adhesion energy:

$${E_{ad}}=[{E_{Zr{O_2} - GO}} - {E_{Zr{O_2}}} - {E_{GO}}]/{\text{S}}$$
(4)
$${E_{ad}}=[{E_{Ce - Zr{O_2} - GO}} - {E_{Zr{O_2}}} - {E_{GO}} - {\mu _{{\text{Ce}}}}+{\mu _{{\text{Zr}}}}]/{\text{S}}$$
(5)

where \({E}_{Zr{O}_{2}-GO}\), \({E}_{Zr{O}_{2}}\), \({E}_{GO}\) and \({E}_{Ce- Zr{O}_{2}-GO}\) represent the formation energies of ZrO2–GO, ZrO2, GO sheets and Ce–ZrO2–GO nanocomposite, respectively. \({\mu }_{\text{Ce}}\) and \({\mu }_{\text{Zr}}\)are the chemical potential of the Ce dopant and host Zr ion, respectively. S is the surface area of the plane.

3.13.2 Electronic Band Structures and Projected Density of States (PDOS)

To examine the effect of GO hybridisation on the electronic structures of ZrO2, the electronic band structure and PDOS of ZrO2 and GO before and after the formation of the as–fabricated photocatalysts (Fig. 8).

Fig. 8
figure 8

The PDOS and band structure of a bulk ZrO2, b graphene oxide, c ZrO2–GO and d Ce–ZrO2–GO photocatalysts

Full size image

For pure ZrO2, the bandgap energy of 5.01 eV was indirect with the valence band maximum (VBM) positioned at Γ point and primarily contributed by the O–2p state, while the conduction band minimum (CBM) was located on the B point and mainly dominated by Zr–4d state (Fig. 8a). The band structure of GO showed that the CBM and VBM was located at Y and Γ points, respectively, which confirmed an indirect bandgap semiconductor and the bandgap energy was calculated as 2.75 eV (Fig. 8b). In addition, as revealed in Fig. 8b, the CBM consisted of C–2p and C–2 s states, whereas the VBM mostly dominated by O–2p and O–2 s states. The theoretical calculations confirmed the indirect bandgap character of ZrO2–GO nanocomposite where the VBM resides on the Γ point, while the CBM positioned at some point between Γ and Y points. The bandgap energy was calculated as 3.28 eV. For ZrO2–GO, the CBM consisted of O-2p state mixed with C-2p state of GO sheets, while the VBM was contributed by O-2p state of ZrO2, see Fig. 8c. In the band structure of Ce–ZrO2–GO nanocomposite, the CBM was situated along the Z–D line, while the VBM was located along the Y–A line in the Brillouin Zone. This reveals an indirect bandgap semiconductor with an Eg of 2.81 eV. Figure 8d showed that Ce-4f state significantly hybridised with the C-2p and O-2p state of GO sheets at the CBM. Moreover, the VBM was contributed by the O-2p state of ZrO2 with hybridisation of Ce-4f and Zr-4d states. The calculated Eg values of pure ZrO2, ZrO2–GO and Ce–ZrO2–GO nanocomposites were consistent with the experimental values and this confirmed the accuracy and reliability of the theoretical method and parameters employed in this study.

3.13.3 Charge Density Distribution

The orbital contributions of the CBM and VBM can be further understood from the charge density distribution of the highest occupied level (HOL) and the lowest unoccupied level (LUL), see Fig. 9.

Fig. 9
figure 9

The electrons and holes density distribution of highest occupied level (a, b) and lowest unoccupied level (c, d) with an isovalue of 0.07 e Å−3 for ZrO2–GO and Ce–ZrO2–GO nanocomposites

Full size image

The charge density distribution suggested that the orbital density of the HOL and LUL for ZrO2 and GO sheets highlighted the occurrence of electron or hole donor–acceptor states in the nanocomposites. In the ZrO2–GO, the lowest unoccupied level was localised around the GO sheets with the involvement of O-2p and C-2p states, whereas the Zr-4d and O-2p states of ZrO2 constituted the major hole density distribution of the HOL (Fig. 9a, c). The charge density distribution of Ce–ZrO2–GO nanocomposite predicted the contribution of Ce ions (Ce-4f state) as part of the HOL and LUL orbitals, which was different from that of ZrO2–GO. Both nanocomposites predicted that Zr-4d and O-2p states constitute the HOL and LUL, respectively, which represent hole and electron acceptor states (Fig. 9b, d). This suggested that when Ce–ZrO2–GO nanocomposite was irradiated with solar light, the Zr-4d and O-2p electrons in the HOL were photoexcited into the O-2p and C-2p states in the LUL via the Ce-4f state. Thus, ZrO2 could serve as a visible-light sensitizer in the nanocomposite to enhance not only the separation of charge carriers, but also the visible light photoactivity. The electron density obviously predicted the key constituents of energy levels and was in agreement with the DOS results.

3.13.4 Charge Transfer and Mechanism Analysis

The calculated Fermi energies of pure ZrO2, GO sheets, and the ZrO2–GO and Ce–ZrO2–GO nanocomposites were − 4.17, − 2.24, − 1.95 and − 1.49 eV, respectively. Such a shift in the Fermi energy level revealed a charge density redistribution of charges after ZrO2 or Ce–ZrO2 was adsorbed onto the GO sheets and this was confirmed by plotting the three–dimensional (3D) charge, see Fig. 10.

Fig. 10
figure 10

The 3D charge density differences of a ZrO2–GO and b ZrO2–GO–GO. The isosurface value was set at 0.006 e Å−3

Full size image

This was evaluated by subtracting the charge density of the nanocomposites from that of the individual monolayers making up the nanocomposites. The green and purple regions represent charge density depletion and accumulation, respectively. As given in Fig. 10, the charge redistribution density led to the polarisation of electrons, and the formation of an interface dipole. Moreover, the charge redistribution at the surface suggested an interfacial charge migration owing to the coupling of GO and ZrO2 or Ce–ZrO2. Figure 10a revealed that the charge density largely accumulated in the interface region of ZrO2–GO and reduced a bit in ZrO2 phase, while there was virtually no or little charge migration of the O atoms in the graphene oxide far away from the interface. However, in the Ce–ZrO2–GO nanocomposites, charge redistribution was observed on the O atoms in the GO sheets far away from the interface, since the Ce dopant improved the intermolecular interaction (Fig. 10b). In both nanocomposites, a strong charge accumulation was observed above the oxygen atoms close to the interface region of GO sheets and Zr atoms at the bottom of ZrO2, while the charge depletion appeared on the surface facing the ZrO2 [74]. The charge density redistribution obviously revealed that the charge density redistribution increase on the ZrO2, while the redistribution at GO was less intense when Ce ion was present. Thus, the interface attracted both electrons and holes, which is vital for the separation of charge carriers at the interface. The Mulliken charge analysis revealed a 0.35 electrons migration at the Ce–ZrO2–GO interface, while only 0.09 electrons were transferred from GO sheets to ZrO2. This was due to the strong interfacial adhesion of Ce–ZrO2–GO nanocomposite.

3.14 Electrochemical Impedance Spectroscopy Analysis

The separation efficiency of charge carriers at the interface is a vital factor in photocatalysis. To determine the migration and separation of charge carriers between the photocatalysts, electrochemical impedance spectroscopy was used [75]. In this study, an EIS Nyquist plot was used to examine the charge separation and migration processes at the interface region of electrode–electrolyte [76]. Generally, a smaller arc radius signifies higher migration of charge carriers [77]. Figure 11a shows the Nyquist plot of bulk ZrO2, ZrO2–GO and Ce–ZrO2–GO nanocomposites with variable concentrations of Ce.

Fig. 11
figure 11

a Electrochemical impedance spectroscopy, b Mott–Schottky plot and c the band alignment of the prepared photocatalysts

Full size image

As shown in Fig. 11a, incorporation of GO into ZrO2 dramatically enhanced the separation and migration efficiency of charge carriers via an interfacial interaction between GO and ZrO2. Moreover, the arc radius of Ce–ZrO2–GO photocatalyst with variable concentrations of Ce was smaller than pure ZrO2 and ZrO2–GO nanocomposite, signifying the lower resistance value for charge migration, as well as effective separation and fast interfacial transfer of charge carriers. The Ce–ZrO2–GO (0.3% Ce) nanocomposite showed the lowest semicircle diameter among all the Ce–ZrO2–GO nanocomposites, which resulted enhanced the efficiency of charge carrier migration and photoactivity [78]. This agreed with the above EY photodegradation and PL results. From the above EIS study, the amount of the Ce ions influence the photoactivity and charge migration kinetics of Ce–ZrO2–GO nanocomposites.

3.15 Mott–Schottky Plot Analysis

The effects of Ce doping and GO sheets with the flat–band potential (\({E}_{\text{fb}}\)) of ZrO2 was evaluated using Mott–Schottky equation [79]:

$$\frac{1}{{{C^2}}}= \frac{2}{{e\varepsilon {\varepsilon _0}{N_D}}} \, \left( {E - {E_{{\text{fb}}}} - \frac{{{k_B}T}}{e}} \right)$$
(14)

where C, ε, ε0, ND, E, Efb, kB, T and e are the space charge capacitance, dielectric constant, permittivity of free space, electron carrier density, external applied potential, flat band potential, Boltzmann constant, temperature and elemental charge of the photocatalyst, respectively. The \({E}_{\text{fb}}\) values were obtained by taking the x-intercept of a linear fit to Mott–Schottky plot against the applied potential, as shown in Fig. 11b. The \({E}_{\text{fb}}\) of pure ZrO2, ZrO2–GO, Ce–ZrO2–GO (0.3% Ce), Ce–ZrO2–GO (0.6% Ce) and Ce–ZrO2–GO (1.0% Ce) were ca. − 1.27, − 1.15, − 1.11, − 1.01 and − 0.92 V vs. SCE, and are ca. − 1.03, − 0.91, − 0.87, − 0.77 and − 0.68 V versus normal hydrogen electrode (NHE), respectively, where NHE = SCE + 0.24 V. With the assumption that the \({E}_{\text{fb}}\) lied at the same potential with that of CB edge [80], the VB level was obtained by subtracting the Eg energy from the CB level. The VB positions of pure ZrO2, ZrO2–GO, Ce–ZrO2–GO (0.3% Ce), Ce–ZrO2–GO (0.6% Ce) and Ce–ZrO2–GO (1.0% Ce) were calculated as 4.65, 2.38, 2.06, 2.10 and 2.14 eV vs. NHE, respectively. The band alignment of the as-fabricated photocatalysts were schematically shown in Fig. 11c. Moreover, the photoinduced electrons in the CB of as-fabricated nanocomposites could reduce the dissolved O2 to produce active O2−• anion radical owing to the more negative potential of ECB of the as-fabricated nanocomposites compared to the O2/O2−• potential (− 0.33 eV). Moreover, the holes in the VB of the as-fabricated nanocomposites could effectively oxidise the HO− ion to generate HO• radical, since ECB was more positive compared to the potential of HO−/HO• (+ 1.99 eV). Clearly, the results suggest that the h+, HO• and O2−• radicals could be the reactive species requires for the photodegradation of EY dye.

3.16 Photocatalytic Mechanism

The VB and CB edge potential of ZrO2 were evaluated by the Mulliken electronegativity theory, as given in Eqs. (11) and (12) [81]:

$${E_{VB}}=\chi - {E^o}+0.5{E_g}$$
(11)
$${E_{CB}}={E_{VB}} - {E_g}$$
(12)

where \({E}_{VB}\), \({E}_{CB}\), \(\chi\), \({E}^{o}\) and \({E}_{g}\)represent the VB edge, CB edge, electronegativity of the studied semiconductors, energy of free electrons on the hydrogen scale (ca. 4.5 eV) and the calculated bandgap from Kubelka–Munk, respectively. The electronegativity value of ZrO2 is 5.91 [82]. The CB and VB potentials of ZrO2 were − 1.12 and 3.94 eV, respectively.

In this study, the mechanistic pathways were presented in the form of direct semiconductor photoexcitation of EY dye. The direct semiconductor photoexcitation photocatalytic mechanism of Ce–ZrO2–GO nanocomposite was proposed based on the interfacial charge migration and change in the Fermi energy level, as given in Fig. 12.

Fig. 12
figure 12

Schematic illustration of the photocatalytic mechanism of Ce–ZrO2–GO nanocomposite under simulated visible light irradiation

Full size image

As shown in Fig. 12, when the Ce–ZrO2–GO nanocomposite was exposed to visible light, the photogenerated electrons on the CB of ZrO2 tend to migrate to GO sheets via metallic Ce ion, since the CB potential of ZrO2 was higher than that of GO sheets [83]. Here, GO sheets exhibited an excellent electron storage capacity and act as an electron acceptor to promote the charge carrier separation, while Ce served as an electron/hole mediator. Thus, the combined effects of GO sheets and Ce ion contributed to the improved photoactivity of Ce–ZrO2–GO nanocomposite with regards to visible light absorption and electron transfer. Since O2−• and HO• played a key influence on the degradation of Ce–ZrO2–GO nanocomposite, the major routes in the photodegradation of EY dye mechanism under solar light irradiation were discussed, as given in Eqs. (13)–(24).

$${\text{GO}}+hv \to {\text{GO}} \, \left( {\text{e}}^- + {\text{h}}^+ \right)$$
(13)
$${\text{GO}}\, \left( {{{\text{e}}^ - }} \right)+{\text{Ce}} \to {\text{GO}}+{\text{Ce}} \, \left( {{{\text{e}}^ - }} \right)$$
(14)
$${\text{Ce}} \, \left( {{{\text{e}}^ - }} \right)+{\text{Zr}}{{\text{O}}_2} \to {\text{Ce}}+{\text{Zr}}{{\text{O}}_2} \, \left( {{{\text{e}}^ - }} \right)$$
(15)
$${\text{Ce}} \, \left( {{{\text{e}}^ - }} \right)+{{\text{O}}_2} \to {\text{Ce}}+{\text{O}}_{2}^{{ - \bullet }}$$
(16)
$${\text{ZrO}}_2 \, \left( {{{\text{e}}^ - }} \right)+{{\text{O}}_2} \to {\text{ZrO}}_2+{\text{O}}_{2}^{{ - \bullet }}$$
(17)
$${\text{O}}_{2}^{{ - \bullet }}+{{\text{H}}_2}{\text{O}} \to {\text{HO}}^{\bullet}$$
(18)
$${\text{GO}}\,\left( {\text{h}}^+ \right)+{{\text{H}}_2}{\text{O}} \to {\text{HO}}^\bullet +{{\text{H}}^+}$$
(19)
$${\text{O}}_{2}^{{ - \bullet }}+{{\text{H}}^+} \to {{\text{HO}}_2}^{ \bullet }$$
(20)
$$2{{\text{HO}}_2}^{ \bullet } \to {{\text{H}}_2}{{\text{O}}_2}+{{\text{O}}_2}$$
(21)
$${{\text{H}}_2}{{\text{O}}_2}+{\text{O}}_{2}^{{ - \bullet }} \to {{\text{HO}}^\bullet }+{\text{HO}}^-+{{\text{O}}_2}$$
(22)
$${\text{GO}} \, \left( {{{\text{h}}^+}} \right)+{\text{HO}}^- \to {\text{GO}}+{\text{HO}}^\bullet $$
(23)
$${\text{HO}}^{\bullet}+{\text{EY \, dye}} \to {\text{degradation \, products}}$$
(24)

The electrons and holes accumulated on the CB and VB of GO sheets and ZrO2 could reduce O2 to O2−• and HO− to HO•, respectively. Afterwards, the O2−• radical anion could combine with the H+ to generate additional HO• radical. Therefore, aqueous EY dye solution could be readily adsorbed onto the surface of Ce–ZrO2–GO because of the large surface area and then attacked by the reactive species to successfully photodegraded into CO2, H2O and other by products. Therefore, the improved photoactivity was ascribed to the improvement of direct semiconductor photoexcitation. The HO• radical is very non-selective in nature and rapidly reacts with numerous organic species with relatively higher rate constants between 108 and 1010 M−1 s−1, which result in the formation of degradation products such as water, carbon dioxide and inorganic ions and/or biodegradable compounds [84, 85]. According to previous studies, the HO• radical can oxidise EY to environmentally benign products, such as water, carbon dioxide and inorganic salts, such as bromide ions [86]. Therefore, in this study, HO• radical participates as an active oxidizing species in the degradation of eosin yellow into water, carbon dioxide and inorganic salts, such as bromide ions, as the rate of degradation was appreciably reduced in the presence of hydroxyl radical scavenger.

4 Conclusion

In this study, the co–precipitation technique was employed to synthesise a nanostructure made up of Ce–doped–ZrO2 anchored on graphene oxide. The photocatalytic activity of Ce–ZrO2–GO nanocomposite was assessed by the photodegradation of EY dye under solar light irradiation. The interfacial charge migration, electronic structure, orbital composition, and the mechanism of the excellent visible light photocatalytic stability and activity was examined in the novel Ce–ZrO2–GO nanocomposite using a hybrid DFT calculation. The EY dye photodegradation results showed that the Ce–ZrO2–GO nanocomposite showed an enhanced activity and stability compared to pure ZrO2. The optimum concentration of Ce ion in the nanocomposite was found to be 0.3%, beyond which the performance reduces. This was because of the improved adsorption of EY dye on its surface, narrow bandgap energy, an increased lifetime of charge carriers, enhanced visible light absorption and high surface area. The linear nature of the kinetics curves indicated that the photodegradation was a pseudo–first–order kinetics for the EY dye. Moreover, a 76% decrease in TOC was observed in the photocatalytic process and a high photodegradation efficiency was achieved after five cycles with remarkable stability. Superoxide radical anion and hydroxyl radicals were the key active species involved in the photodegradation of EY dye using Ce–ZrO2–GO (0.3% Ce) nanocomposite. We showed that the HOL of Ce–ZrO2–GO nanocomposite was dominated by ZrO2 through the first–principles study and therefore, after contact, the ZrO2 possessed a negative charge to induce a higher electrostatic interaction with GO sheets. Our calculations showed that Ce ions serve as an acceptor and a donor to the photogenerated electrons, which led to electron–rich surface. This study not only revealed that Ce–ZrO2–GO nanocomposite could be a potential photocatalyst for photodegradation EY dye with a high stability and activity, but also offers a theoretical reference, and the role of GO and Ce ions in the nanocomposite can offer new insights in designing low-cost and highly efficient ZrO2-based photocatalyst materials for applications in environmental remediation.