Photodegradation of tartrazine dye favored by natural sunlight on pure and (Ce, Ag) co-doped ZnO catalysts

Growing consumption coupled with the increasingly declining supplies of water has driven society and governments to consider the decontamination of contaminated water as the source of the usable water for selected household, industrial, and agriculture usage. The decontamination of wastewater using advanced oxidation processes (AOP) (Oller et al. 2011), phytoremediation (Salt et al. 1998), photocatalysts (Herrmann et al. 1993), electrochemical techniques (Brillas & Martínez-Huitle 2015), biosorption (Vieira & Volesky 2000), and nanotechnology (Baruah et al. 2015) has been reported as possible techniques for wastewater purification and solar wastewater treatment (SOWAT) (Igoud et al. 2020). Many scientists, industrialists, and policymakers have collaborated to develop materials and technology for a safe world with the intention of successfully eliminating dye effluents (Salthammer et al. 2010; Vaiano et al. 2016). The majority of water contaminants are residual colors found in the effluents from various industries, e.g. paper and pulp, textile, dye, pharmaceutical, cosmetics, rubber, paints, and printing industries, etc. (Ali et al. 2018a). Discharge of pollutants from these industries into water sources poses a major environmental threat across the world. Not only do pollutants harm the esthetic quality of water, but they also destroy marine flora and fauna (Kaur & Singhal 2014). Tartrazine (C.I. Acid Yellow 23, AY23) is chosen for the present research as a model pollutant for such dyes an dius called by other names including E102 (EFSA), FD&C Yellow 5 (FDA-US Food and Drug Administration, FDA) or C.I. 19140 (Color Index International). It is approved as a dye by the FDA (Bhatt et al. 2018) and is a widely used azoic dye in the textile, cosmetics, medicinal, and food industries (Aoudjit et al. 2018). Many studies have been published on tartrazine hazards in recent times, describing its potentially deleterious impact, such as food allergy, mutagenic, carcinogenic, and photototoxicity (Abdullah Hashim et al. 2001; dos Santos et al. 2014; Chekir et al. 2017; Aoudjit et al. 2018; Ali et al. 2018b).

Tartrazine's chemical name is trisodium-5 hydroxy-1-(4-sulfonatophenyl)-4-(4-sulfonatophenylazo)-H-pyrazole-3-carboxylate and incorporates azo bonding (–N = N–) in its form, believed to be metabolized by intestinal bacteria into sulfanilic acid and aminopyrazolone (Rehman et al. 2019). The chemical structure of the tartrazine is presented in Figure 1.

Advanced oxidation processes (AOPs have been defined in recent years as effective procedures for obtaining high oxidation yields from several types of organic compounds. These approaches are focused on the production of very reactive agents such as hydroxyl radicals (OH) that are highly reactive and fast oxidizing agents (E^o = 2.8 V), capable of mineralizing organic contaminants (Daneshvar et al. 2005; Saien & Soleymani 2007). Semiconductors based on metal oxides have been widely developed in recent decades for photocatalytic applications (Guo et al. 2011; Kamegawa et al. 2011, 2012; Chen et al. 2017). The photocatalysts may capture photons with energy equal to or greater than their band gap energy during the photocatalytic phase to form electrons and holes on the catalyst surface. It is known that superoxide and hydroxyl radicals are generated to oxidize organic contaminants by interacting between holes or electrons and water molecules (Fu et al. 2015). The photocatalysts in the spectral region can be classified into two standardized classes according to their different photocatalytic properties: ultraviolet (UV) and visible light-responsive photocatalysts (Su et al. 2019). Among the various semiconductors, zinc oxide (ZnO) is a photocatalyst with a large direct band gap (3.3 eV) and has a high binding free-excision capacity of 60 meV (Nair et al. 2011). Because of these characteristics, ZnO is regarded as one of the main active photocatalysts, including high initial operation levels, multiple active sites with high surface reactivity, low price, and environmental protection (Lam et al. 2012). Different strategies were therefore adopted to enhance photocatalytic activity of ZnO. Doping is well known to be an effective and easy way of improving the photocatalytic properties. Surface area variability and dopant ion incorporation produce lattice defects and band gap energy transition (Ullah & Dutta 2008; Kao et al. 2011), which play an important role in metal oxides photocatalytic operations. Doping transition metals, noble metals, and non-metals is therefore a very expedient way to improve the photocatalytic activity (Poongodi et al. 2015). Modification of noble metal semiconductors has gained tremendous interest because they improve the reduction cycle and hence the mechanism of photocatalytic degradation. Silver (Ag) was selected among the noble metals because of its extraordinary catalytic ability, non-toxicity, and comparatively cost-effectiveness. Ag can trap the photogenerated electrons from the semiconductor and allow holes to form hydroxyl radicals, which results in the degradation reaction of the present organic species (Saravanan et al. 2013).

Rare earth (RE) metals are also strong doping elements for modifying ZnO's electronic structure and extending its visible light absorption. Doping with RE metals is a localized level of impurity in ZnO's band structure, and changes the band structure (Khatamian et al. 2012). Cerium (Ce) has received considerable attention from various lanthanides as a promising dopant for semiconducting metal oxide materials due to several unique properties and applications (Chang et al. 2014; Lamba et al. 2015; Li et al. 2015). Ce doping has been found to increase the efficiency of ZnO in photocatalytic applications. Ce doping facilitates the formation of crystallite to minimize particle size and increase the surface area, thereby helping to improve the efficiency of photodegradation (Rezaei & Habibi-Yangjeh 2013). Recently, simultaneous doping into semiconductor materials with two groups of atoms (co-doping) has drawn significant interest, as it may contribute to higher photocatalytic behavior and different characteristics relative to single element doping into semiconductor oxides (Zhao et al. 2011).

In this work, in conjunction with ZnO and (Ce, Ag) co-doped ZnO catalysts, solar-light irradiation has been successfully used to efficiently remove tartrazine from polluted water. The remediation of dyes in effluents employs methods such as electrochemical treatment, membrane filtration, adsorption, and hybrid technologies, among others. However, these technologies lack efficiency and cost-effectiveness. As a possible solution, this work presents a photodegradation of tartrazine dye favored by natural sunlight on pure and (Ce, Ag) co-doped ZnO catalysts.

Tartrazine (M = 534.36 g/mol, assay: ≥85%) also known as Yellow 23 (Figure 1), with the chemical formula C₁₆H₉N₄Na₃O₉S₂ and a maximum absorption at the wavelength at 427 nm was purchased from ACROS organics (USA). Hydrochloric acid (HCl) and sodium hydroxide (NaOH) used in this study were laboratory grade products. Deionized water was used for all experiments.

A 100 mL borosilicate glass Erlenmeyer flask was used as the reactor. Global solar radiation intensity was measured with a Kipp and Zonen CMP11 pyranometer, pH value was measured using CONSORT C3010 multi-parameter analyzer, tartrazine dye concentration was determined using Shimadzu UV1800 Spectrophotometer and chemical oxygen demand (COD) was measured by type (DR 1900 LANGE HACH, Düsseldorf, Germany) spectrophotometer.

The experiments were performed using solar irradiation at Bou-Ismail, Algeria, during April 2020 between 10:00 and 16:00 at the Solar Equipment Development Unit, situated 30 km west of Algiers (longitude: 2 W 420 N, latitude: 36 W 390 E, altitude: 5 m). The entire system is linked to photovoltaic panels providing energy for the alimentation (agitation) of our system for degradation of tartrazine.

A sufficient quantity of Zn(NO₃)₂•4H₂O was dissolved in bi-distilled water. The pH of the solution was brought to 7.5 by drop-wise addition of ammonium hydroxide solution (25%). The reaction mixture was then stirred for 2 h at room temperature with 100 rpm and transferred to a steel autoclave coated with Teflon and kept at 120 °C for 22 h. The product was washed several times using a vacuum filtration unit with bi-distilled water, filtered and dried at 80 °C overnight. The co-doped ZnO nanoparticles (heterojunction) were also prepared according to the same procedure, except that the products of cerium chlorides and silver nitrates (CeCl₃ and AgNO₃) were added.

The X-ray diffraction (XRD) patterns and scanning electron microscope (SEM) micrographs of the samples prepared as mentioned above were recorded. The XRD patterns were recorded on a Siemens D-5000 diffractometer with Cu-Kα radiation (λ = 1.5418 Å) and surface morphology was studied using SEM. The UV–Vis absorption spectra were measured by UV–Vis spectrophotometer (Shimadzu UV-2401).

XRD analysis of samples was carried out to determine the crystal structures and phase purities of ZnO and (Ce, Ag)/ZnO. The XRD patterns are shown in Figure 3. The main diffraction peaks for pure ZnO were observed at 2θ values of 31.73, 34.37, 36.21, 47.48, 56.53, 62.77, 66.30, 67.86, 69.00, 72.46, and 76.86 corresponding to the lattice plane (100), (002), (101), (102), (110), (103), (200), (212), (201), (004), and (202)planes of wurtzite ZnO hexagonal phase (space group, P63mc), which are in good agreement with the literature values (PDF 89-1397) (Kumar et al. 2015).

For (Ce, Ag)/ZnO, diffraction peaks well suited to (100), (002), (101), (102), (110), (103), (200), (212), (201), (004), and (202) planes at 2θ values were same as that of wurtzite ZnO hexagonal phase. There were additional diffraction peaks observed corresponding to the Ag and CeO₂. The diffraction peaks at 2θ values 38.11, 44.30, 64.44, and 77.39, labeled with ‘◊’ matched with the (111), (200), (220), and (311) crystalline planes of cubic phase (space group, Fm-3 m) of metallic Ag (PDF 65-2871) (Güy & Özacar 2019), and the diffraction peaks at 2θ values 27.79, 32.20, 46.18, and 56.09, marked with ‘*’ matched with the (222), (400), (440), and (631) planes of cubic phase (Ia-3) of CeO₂ (PDF 89-8429) (Wen et al. 2018). XRD analysis therefore suggests that the wurtzite ZnO hexagonal phase is present in the pure ZnO samples and in the (Ag, Ce) co-doped samples, wurtzite ZnO hexagonal phase along with the metallic Ag in cubic phase, and CeO₂ in cubic phase is also present.

The surface morphology of the samples was obtained by SEM and is presented in Figure 4. From Figure 4(a) and 4(b), hexagonal rods of ZnO and (Ce, Ag) co-doped ZnO catalysts can be observed. However, the average size of the crystallites in the pure ZnO sample is considerably larger compared to the average size in the (Ce, Ag) co-doped ZnO sample. The decrease in crystallite size with the co-doping of Ce and Ag leads to an increase in the specific surface area compared to the undoped samples. Additionally, the surface of (Ce, Ag) co-doped ZnO samples have numerous fibrous columns anchored from the ZnO surface, which is conducive to forming the heterojunction between ZnO and (Ce, Ag) dopants. The presence of such fibrous columns on the ZnO surface increases the total surface area of co-doped samples, and therefore increases the catalytic performance of the samples immensely.

Figure 5(a) shows the UV-Vis absorption spectra of the ZnO and (Ce, Ag)/ZnO. Interestingly, with the Ag and Ce co-doping, the absorption peak changes from 283 to 352 nm. This change in the absorption peak to the higher wavelength results in the decrease of band gap of (Ce, Ag)/ZnO. A single absorption peak for (Ce, Ag)/ZnO shows that Ce and Ag are doped effectively into ZnO lattices and that the sample has strong optical properties (Kumar et al. 2015).

The Tauc plot to calculate the optical band gap of the samples is presented in Figure 5(b). The band gap is determined by extrapolating the linear portion of the plot between (αhν)² and hv to intersect the x-axis. The value of the optical band gap energy of the pure and (Ce, Ag) co-doped ZnO samples decreases from 3.08 to 2.82 eV, respectively. The decrease in the band gap of the semiconductor is due to the additional impurity levels generated in the doped samples. The band gap energy value was decreased for the (Ce, Ag)/ZnO samples. This result reveals that the co-doped ZnO absorbs UV and visible light. Therefore, the optical absorption property attributes that the (Ce, Ag)/ZnO photocatalyst could be promising in visible light photocatalysis (Poongodi et al. 2015).

Preliminary experiments were carried out to determine the photocatalysis performance. Three main removal procedures were tested during our analysis for the efficacy of tartrazine degradation in polluted water. The first procedure was carried out in the presence of catalyst without radiation based on the adsorption; the second procedure was performed in the presence of solar radiation without catalyst based on the photolysis; the third procedure was carried out under solar radiation (with sunlight) and in the presence of catalyst based on photocatalysis. Degradation of tartrazine under these processes for pure ZnO and (Ce-Ag) co-doped ZnO is presented in Figure 6. It can be observed that photocatalysis is impactful in photodegradation of tartrazine, whereas the effect of adsorption and photolysis is negligible. These results are in good agreement with similar works (Tanaka et al. 2000; Chekir et al. 2017; Aoudjit et al. 2018). In our azo dye degradation studies, the degradation of tartrazine under photocatalytic conditions in the presence of pure ZnO catalysts for a reaction time of 90 min is 44.82% and 98.91% for (Ce-Ag) co-doped ZnO. The degradation in presence of ZnO increases rapidly for 30 min until it reaches 24%. Thereafter, it slows down for 90 min until reaching 44.82%. For (Ce–Ag) co-doped ZnO, the degradation of the azo dye saturates in about 45 min to ∼95% and takes almost 90 min for 100% degradation. However, during photolysis or adsorption, no signs of removal were observed, as indicated in Figure 6. Further investigations and refinement of different process parameters and optimal conditions used for the effective extraction of tartrazine from polluted water were taken.

The effect of irradiation type on the photocatalytic degradation of tartrazine by ZnO and (Ce, Ag) co-doped ZnO was studied by artificial irradiation (PHILPS PL-L 24 W/10/4P UV lamps, λ_max = 365 nm and I = 18.6 W/m²), and solar UV radiation using a global UV radiometer (Kipp & Zonzn, CMP11, I = 853 W/m²). As seen in Figure 7(b), after 90 min of exposure, the degradation percentage of tartrazine due to (Ce, Ag) co-doped ZnO photocatalysts was 98.91%, and 98.88%, under sunlight and UV lamp, respectively. The corresponding values for the degradation of the tartrazine due to ZnO are found to be 42.02% and 13.79%, respectively, as shown in Figure 7(a). Comparison of the photocatalytic efficiency ZnO and (Ce, Ag) co-doped ZnO irradiated with the sunlight and UV radiation under identical experimental condition is presented in Figure 8. These results indicate a remarkable photocatalytic activity of (Ce, Ag) co-doped ZnO photocatalyst. The possible mechanism for the degradation efficiency of the photocatalyst is presented as follows: upon absorption of light by the photocatalyst, the electrons from the valence band are transferred to the conduction band, along with the production of holes in the valence band. The photo-created electrons reduce the oxygen to form superoxide radicals (O²⁻) and at the same time holes present in the valence band oxidize the water molecules/OH⁻ ions to produce hydroxyl radicals (OH^.). These superoxide and hydroxyl radicals participate in the degradation process of tartrazine dye. It was previously observed that photogenerated electrons that move to the conduction band are mostly unstable and quickly go back to the valence band. Therefore, they can recombine with the holes by going to the valence band and decrease the quantum efficiency of photocatalysis. In this study, enhancement in photocatalytic performance of (Ag, Ce) co-doped ZnO can be attributed to the combined effect of Ag and Ce co-doping. The localized electronic states of dopant ions (Ag⁺,Ce³⁺) may be acting as charge trap sites for photo-created electron-hole pairs (Khalid et al. 2019).

The use of natural sunlight, instead of the artificial UV lamp, may help greatly in reducing the cost of photocatalytic oxidation compared to some alternative technologies. The availability of solar UV-irradiation at a specific geographical area, like Algeria, represents a major factor in determining the cost of a solar-based system. Algeria is characterized by an important solar potential and an annual duration of sunshine exceeding (3,000 h/y). Solar photocatalysis can be recommended for its low-cost energy and high efficiency. These results are in good agreement with similar works reported in the literature (Aoudjit et al. 2018; Zioui et al. 2019; Ghribi et al. 2020).

Photodegradation is well known to be influenced by the adsorption of the dye on the catalyst surface. The dye's adsorption is strongly influenced by the pH of the solution (Velmurugan & Swaminathan 2011; Krishnakumar et al. 2012; Subash et al. 2013). Thus, in this study the effect of pH on tartrazine photodegradation was studied in an aqueous medium. The experiments were conducted at a concentration of 10 mg/L tartrazine to cover a 3.0 ≤ pH ≤ 9.0 for 90 min of sunlight irradiation. The effect of the pH on the photodegradation of tartrazine in the presence of ZnO and (Ce, Ag)/ZnO exposed to the sunlight is presented in Figure 9. Looking at the Figure 9(a), it can be observed that the degradation of tartrazine molecule is very high in the presence of (Ce, Ag)/ZnO catalyst throughout the pH range. The degradation percentage of tartrazine increases almost linearly as a function of time at different pH values, with a maximum at pH = 9.0 as shown in Figure 8(b).

The interpretation of pH effects on dye photodegradation process output is a very challenging task due to its dependence on various physical and chemical parameters. First, the influence of pH of the solution on the rate of degradation can be explained primarily by the processes of sorption–desorption in the semiconductor particle sheet (Chekir et al. 2017). The azo dye used in the present experiment was anionic dye, and under laboratory conditions was negatively charged. For this reason, it is best to use the solution's free pH. The photocatalytic degradation of tartrazine at free pH is 98.91% for (Ce, Ag) co-doped ZnO catalyst.

Figure 10 shows the effect of initial concentration of tartrazine on the rate of photodegradation at free pH = 6.0 for a reaction time of 90 min. The increase in the amount of tartrazine from 10 to 30 mg/L shows decreased degradation percentage, indicating slower degradation kinetics for ZnO catalyst in the presence of sunlight.

For the (Ce-Ag) co-doped ZnO catalyst, the degradation percentage decreases with an increase in the tartrazine concentration. With an increase in the concentration of tartrazine, the degradation behavior changes from initial rapid degradation at 10 mg/L to linear at 30 mg/L. After 90 min, 12% reduction in the degradation of the 30 mg/L tartrazine solution was observed. This could be because of insufficient quantity of generated OH^. radicals present in the solution, which influence the photodegradation process favorably (Li et al. 2010; Farzadkia et al. 2014; Tran et al. 2018). Degradation rate is proportional to the available surface of the catalyst for the development of electron-hole pairs to generate hydroxyl radicals. In the present study, for the constant volume of catalyst, the number of hydroxyl radicals produced remains the same, although the concentration of dye in the solution increases. Consequently, the ratio of hydroxyl radicals to tartrazine molecules decreases with the increase in dye concentration (Jamil & Sharaf El-Deen 2016). Additionally, the decrease in dye degradation has been attributed to a decrease in light penetration into the solution (Krishnakumar et al. 2012). The amount of the dye adsorbed on the catalytic surface increases with the increase in the dye concentration in solution. This higher particle density of the dye molecules in the solution reduces the penetration depth of incident photon in the dye solution. At high dye concentrations, the adsorbed dyemolecules absorb the majority of the incident visible light, effectively preventing the light reaching the catalyst. This shielding of the catalyst from the sunlight can therefore also decrease photocatalytic degradation (Davis et al. 1994).

The influence of the catalyst dose on the tartrazine photodegradation is illustrated in Figure 11. The catalyst dose varies from 0.5 g/L to 2 g/L. The degradation increases with the increase in the dose and the maximum degradation occurs for an optimal dose of 1 g/L. This is due to increased availability of active sites on the photocatalyst surface and hydroxyl radicals (OH^.) in the reaction mixture. However, a further increase of catalyst dose beyond 1 g/L decreases the catalytic degradation. In fact, an excess amount of catalyst could notably inhibit the penetration of solar light into the tartarazine solution, and an increase in the photocatalyst suspension leads to a decrease in the generation of oxidizing agents. Consequently, 1 g/L catalyst dose was chosen as the optimum dose for further experiments.

The recyclability of ZnO and (Ce, Ag) co-doped ZnO (1 g/L) catalysts used in the photodegradation of tartrazine (C_TART = 10 mg/L) at free pH = 6.0 has also been studied under sunlight irradiation over the course of 90 min and the results are presented in Figure 12. To reuse the photocatalyst in multiple cycles, the photocatalyst was collected, washed with distilled water, dried, and then used in a new experimental cycle. After six cycles of treatment, the catalyst still maintained good photodegradation capacity. The degradation percentage of tartrazine during the six cycles were 98.91%, 98.89%, 98.36%, 97.86%, 97.36%, and 97.16% for (Ce, Ag) co-doped ZnO, as shown in Figure 12(b), and 42.02%, 42.02%, 40.86%, 40.30%, 39.87%, and 39.12% for ZnO as shown in Figure 12(a). This confirms the excellent reusability of (Ce, Ag) co-doped ZnO photocatalysts. The high degradation activity could be attributed to the highly stable nature of (Ce, Ag) co-doped ZnO, while its surface remained resistant to agglomeration or sintering under the experimental conditions specified.

Su et al. (2019) reported similar findings for ZnO/Ag–Ag₂O based catalyst on Congo Red degradation after five cycles. Subash et al. (2012) reported strong recyclability of catalysts when using Ce co-doped Ag–ZnO under natural sunlight for Naphtol Blue Black dye photodegradation. In metronidazole photodegradation, Dong et al. (2014) reported strong recyclability of catalysts when using ZnO/reduced graphene oxide under visible light.

The degradation of the tartrazine was also analyzed using COD values to confirm the mineralization of tartrazine. All parameters of the reaction such as (tartrazine concentration (C_TART = 10 mg/L), solar UV radiation (853 W/m²), dose of catalysts (1 g/L), and free pH = 6 conditions over the course of 90 min) were kept constant during the experiments.

Table 1 shows the percentage of COD reductions obtained in this study. After 90 min, the value of COD with pure ZnO and (Ce, Ag) co-doped ZnO irradiation was 30.39% and 78.43%, respectively, indicating the dye mineralization and the higher potential of (Ce, Ag) co-doped ZnO system for the removal of dyes from wastewater.

In general, photocatalytic degradation by semiconductors involves three steps: (1) irradiation of the semiconducting catalyst causes the transfer of electrons from the valence band (VB) to the conduction band (CB), creating equal number of empty sites (holes) in the valance band; (2) the excited electrons and holes then migrate to the catalyst surface; and (3) the excited electrons and holes react with free radicals such as OH⁻, and O₂⁻ resulting in oxidation and reduction processes, respectively. Such free radicals further react with organic dye and degrade them in the process to components, harmless compounds such as CO₂, H₂O, and inorganic by-products (Beura & Thangadurai 2017; Kumar et al. 2019).

Figure 13 demonstrates the mechanistic interpretation of the photocatalytic degradation reaction of tartrazine dye in the presence of photocatalyst ZnO under solar radiation onto the semiconductor.

The chemical reactions producing reactive chemical species mentioned above during the photocatalytic degradation of tartrazine by ZnO are also generated during the presence of (Ce, Ag) co-doped ZnO. In the case of (Ce, Ag)/ZnO catalyst, the improvement of photocatalysis in the presence of Ag and Ce has been presented as follows.

Figure 14 demonstrates the mechanistic interpretation of the photocatalytic reaction for the photocatalyst (Ce, Ag) co-doped ZnO. When solar radiation illuminates the semiconductor. The presence of Ag and Ce simultaneously traps the electron from the CB of ZnO which suppresses the recombination of the electron-hole. The photocatalytic process occurs in which oxygen adsorbed on the photocatalyst surface traps the photogenerated electron (Mills & McGrady 2008). The transition of the electron to oxygen can be the rate-determining step in the photocatalytic reaction of semiconductors (Mura et al. 2002). The role of Ag in the photocatalytic process of trapping the electrons from ZnO's CB during solar irradiation is reported in the literature (Chen & Nickel 1996). The doped Ag in the ZnO photocatalyst can function as an electron collector to increase the lifespan of charge carriers because the Fermi energy level of the noble metal (Ag) is still greater than that of the photocatalyst semiconductor (Zhang et al. 2013). Improvement of (Ce, Ag) co-doped ZnO photocatalytic reaction is due to a synergistic interaction between Ag and Ce in ZnO. Photocatalytic performance of (Ce, Ag) co-doped ZnO is found to be better than that of ZnO.

It can also be shown that the Ce 4f level in (Ce, Ag) co-doped ZnO plays a significant role in the conversion and inhibition of electron-hole recombination of interfacial charges. The UV–Vis-spectrum reveals that Ag and Ce co-doping increases the visible light absorption of ZnO and produces more electron-hole pairs under solar light irradiation, which aims to enhance the photocatalytic function of the photocatalyst (Ce, Ag) co-doped ZnO.

Ce⁴⁺ quickly captures the electron, and it serves as an electron scavenger. The coexistence of Ce³⁺ and Ce⁴⁺ in (Ce, Ag) co-doped ZnO affect the photo-reactivity by altering the recombination of electron-hole pairs. In addition, the trapping aspect of Ag and Ce⁴⁺/Ce³⁺ sites is consequently passed to the adjacent adsorbed O₂ to create a significant number of radical superoxide anions. At the same time, ZnO's VB holes react with water to generate highly reactive hydroxyl (OH⁻) radical compounds. Dye oxidation is caused by the strongly reactive superoxide radical anion and hydroxyl radicals. The increase in the production of O₂ and OH by Ce and Ag increases the photocatalytic activity of (Ce, Ag)/ZnO (Subash et al. 2012).

With the synthesis of pure and (Ce, Ag) co-doped ZnO by solvothermal method, the presence of Ce and Ag co-doped ZnO has been revealed by XRD. The pure ZnO exhibited good photocatalytic activity with a degradation of 77.77% for tartrazine at pH = 9. The (Ce, Ag)/ZnO catalyst exhibited outstanding photocatalytic activity for the photodegradation of tartrazine under sunlight at different pH = 3, 6, and 9, and the different concentrations of the pollutant (10.20 mg/L and 30 mg/L). The improved photocatalytic output of (Ce, Ag)/ZnO could be due to the narrowing of ZnO band gap energy resulting from the efficient high charge separation. The (Ce, Ag)/ZnO catalyst was found to be highly efficient and recyclable and therefore reusable. The excellent photocatalytic properties of (Ce, Ag)/ZnO for degradation of tartrazine to the extent of 98.91% has been found within 90 min under solar irradiation. (Ce, Ag)/ZnO therefore has potential and can act as an efficient catalyst for the degradation of tartrazine.

The authors of this manuscript would like to thank the General Directorate for Scientific Research and Technological Development (DGRSDT) and the Ministry of Higher Education and Scientific Research (MESRS), Algeria, for funding this project. The authors would also like to thank Dr Fouzia Touahra, CRAPC, for his valuable input in the project.

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