The photocatalytic rate of ZnO supported onto natural zeolite nanoparticles in the photodegradation of an aromatic amine

Aniline and its derivate are critical environmental pollutants, and thus, the introduction of an eco-friendly catalyst for removing them is an important research future. The ZnO supported on the ball-mill prepared clinoptilolite nanoparticles (CNPs) was prepared via an ion-exchange process followed by the calcination process. The amount of loaded ZnO in the ZnO-CNP (CZ) samples varied as 0.54, 0.63, 0.72, and 0.86 meq/g as the Zn(II) concentration in the ion-exchange solution varied from 0.1 to 0.5 M. The ZnO-CNP catalyst was briefly characterized by XRD, FTIR, and DRS techniques. The pHpzc value for the various ZnO-CNPs was about 7.1 that had no change with the ZnO loading. By applying the Scherrer equation on the XRD results, a nano-dimension of about 50 nm was obtained for the catalyst. Bandgap energy of the ZnO-CNP samples was estimated by applying the Kubelka-Munk equation on the DRS reflectance spectra. The value for the CZ2 catalyst was about 3.64 eV. The supported ZnO-CNP sample was then used in the photodegradation of 2,4-dichloroaniline (DCA). Raw zeolite showed a relatively low photocatalytic activity. The degradation efficiency was followed by recording the absorbance of the DCA solution by UV-Vis spectrophotometer. The effects of the essential critical operating factors on the degradation efficiency were kinetically studied by applying the Hinshelwood equation to the results. The ZnO-CNP catalyst with 2 w% ZnO showed the best photocatalytic rate in the optimal conditions of 0.75 g/L, CDCA: 15 ppm, and the initial pH: 5.8. Finally, HPLC analysis of the blank and the photodegraded DCA solutions at 180 and 300 min confirmed 74 and 87% of DCA molecules were degraded during these times. The results confirm that supported ZnO onto clinoptilolite caused enhanced photocatalytic activity because the zeolite internal electrical field prevents the e−/h+ recombination.


Introduction
Nowadays, an increased need for purified water is essential for human life due to the increased world population and industrial activities  a,b Karimi-Maleh et al. 2021a, b). The increased industrial activities drastically polluted the aquatic resources due to the discharge of the wastewater into the environment ( c,d Karimi-Maleh et al. 2020a, 2021c. In contrast, such industrial activities are essential for human life (Naderi Asrami and Karimi-Maleh 2021). The population increase also needs increased health and care activities and drugs ( e,f Karimi-Maleh et al. 2020b, 2021d. Unfortunately, there are no balances between these activities and the increased world population, and thus, the pollution of water supplies by various organic/ inorganic pollutants is a major global problem at this time) Segundo et al. 2019;Ahmad Bhat et al. 2020;Huang et al. 2019).
An important class of chemicals with broad applications in pharmacological, biological, and industrial activities is aniline and aniline derivate. These chemicals have adopted broad uses to manufacture pesticides, dyes, cosmetics, medicines, synthetic polymers, rubber, etc. These are also used as intermediates to prepare many chemical compounds (Rappoport 2007;Akyuz and Ata 2006;Seiied Bonakdar et al. 2017: Massah et al. 2008Abbasi et al. 2017). Thus, high amounts of these aromatic amines have been discharged into the aquatic Responsible Editor: Sami Rtimi environments, resulting in high water resource pollution. Aromatic amines have been known as highly toxic and carcinogenic agents for body life ) Akyuz and Ata 2006). The absorbed chloroanilines and chloramine via inhalation, ingestion, or cutaneous cause vertigo, headache, cyanosis, and mental confusion decease. This also causes anorexia, anemia, weight loss, and cutaneous lesions as some chronic toxicities (Pande and Dwivedi 2011). 2,4-dichloroaniline (DCA) is the degradation product of some herbicides known as a pollutant agent for the environment (Pande and Dwivedi 2011;Zh and Lin 2009).
Thus, the need for an eco-friendly removal method for removing such pollutants is of great attention to environmental engineering and human health. This eco-friendly, cheap, and effective method is semiconducting-based photocatalysis, in which four main reactive species are responsible for destroying the organic pollutants from water. The initial photogenerated electron/hole pairs (e/h) can be produced by the illumination of arrived UV or visible photons in the conduction (CB) and valence (VB) of the semiconductor. These photoinduced e − /h + pairs can attack the organic pollutants directly. They can also react with the dissolved oxygen and water molecules and produce superoxide and hydroxyl radicals, respectively, as other main reactive species for destroying the pollutants (Habibi and Askari 2011;Shabani and Aliyan 2016;Hassani et al. 2018;Honarmand et al. 2020;Khanmohammadi et al. 2020;Shah and Hao 2016;Mohamed JafferSadiq and Samson Nesaraj 2014;Rashmi et al. 2020;Bordbar et al. 2016;Pankaj et al. 2020;Vishal et al. 2020;. Unfortunately, recombination of these e − /h + pairs drastically decreases the overall efficiency of typical heterogeneous photodegradation catalysis. This can be overcome by various technologies such as doping of metals and non-metals into the semiconducting material, use the catalysts with nano-dimension, coupling of two or more semiconductors, supporting the catalyst into supports, and the use of plasmonic systems. Detailed mechanism pathways for these technologies have been illustrated in the literature (Vafayi and Gharibe 2015;Palma et al. 2020;Rostami-Vartooni et al. 2019;Giahi and Hoseinpour Dargahi 2016;Naderpour et al. 2013;Pouretedal et al. 2017;Derikvandi and Nezamzadeh-Ejhieh 2017;Bordbar et al. 2015;Derikvandi et al. 2020;Khodadadi and Bordbar 2016;Pouretedal and Sohrabi 2016;Thirumalai et al. 2016;Sheetal et al. 2021;Hassani et al. 2020;). Generally, catalysis is an important field in various chemical reactions (Massah et al. 2013;Javad Kalbasi et al. 2012: Zarei et al. 2018Nobahari et al. 2017;Ehsani et al. 2016). Photocatalysis can also be used for removing/determining some hazardous heavy metals via the photoreduction process (Majidnia and Fulazzaky 2016;Majidnia and Fulazzaky 2017).
In this work, for enhancing the photocatalytic activity of ZnO, it was supported on the clinoptilolite nanoparticles (CNP). Generally, zeolites have an internal electric field that helps prevent the photogenerated e − /h + pairs in the photoexcited supported semiconductor (Behin et al. 2019;Divband et al. 2019;Rahimi et al. 2019;Karimi-Shamsabadi et al. 2017). The n-type ZnO semiconductor is a highly transparent material with a bandgap energy of about 3.3 eV for the direct electronic transition. It also has an exciton binding energy of about 60 meV (Matei et al. 2012). Accordingly, it is a transparent material in the wavelength range between 400 and 800 nm. This transparency can change by techniques, such as doping the various atoms (Yuonesi and Pakdel 2010). Clinoptilolite used here is an eco-friendly material with high depositions in Iran. Thus, the use of this material drastically decreases the overall cost of the process.
The supported ZnO-CNP was used in the photodegradation of 2,4-dichloroaniline (DCA). The catalysts were briefly characterized and used for the study of the kinetics of the photodegradation process. For this goal, the Hinshelwood equation was applied to the results, and the kinetics of some critical experimental variables such as supporting, the size of clinoptilolite particles used, the amount of the loaded ZnO, solution pH, the concentration of DCA, and the dose of the ZnO-CNP catalyst was studied on the DCA photodegradation. In our previous works, the photodegradation of DCA by NiO-CNP and CuO-CNP has also been reported (Iazdani and Nezamzadeh-Ejhieh 2020;Iazdani and Nezamzadeh-Ejhieh 2021).
The natural clinoptilolite tuff was changed into the micronized (CP) and nano(CNP) particles by the mechanical ballmill method. Detailed procedures for pretreating the obtained powders in removing the magnetic and water-soluble impurities have been illustrated in the literature (Faghihian et al. 2008;Arabpour and Nezamzadeh-Ejhieh 2015). Briefly, about 10 g of the CNP powder was added to a round flask, and about 200 mL of water and a magnet were added. The suspension temperature reached 70°C, and it was magnetically stirred under reflux conditions for 8 h. Then, the solid material was separated and dried at 80°C for 1 h.
For preparing the ZnO-CNP sample, the experiments were carried out as below. About 0.5 g of the pretreated CNP powder was added to a closed bottle containing 10 mL 0.1 M Zn(II) solution and shaken for 24 h to complete the ionexchange reaction. After the separation of the solid powder by centrifugation at 13000 rpm, it was washed many times with water and then dried in air. The dried powder was calcined at 450°C for 12 h to obtain the ZnO-CNP sample. A similar procedure was used to prepare of ZnO-CP or ZnO-CNP sample with other Zn(II) concentrations.

Characterization methods
Fourier transformation infrared (FTIR) spectra of the samples were obtained with a Nicolet FTIR (Impact 400D) spectrometer by using KBr pellets. The X-ray diffraction patterns for the samples were obtained using a diffractometer Bruker, D8ADVANCE, X-ray tube anode: Cu-Kα wavelength: 1.5406 Å, Filter: Ni. A transmission electron microscope (TEM) Zeiss EM 900 at 80 kV was used for recording TEM images. Diffuse reflectance spectra (DRS) of the samples were also obtained using JASCO V-670 (JASCO Co. Japan) equipped with an integrating sphere and a Hg lamp (75W) as the UV light source. The absorption spectra of DCA solutions were recorded on a double-beam spectrophotometer (Varian Carry 100 Scan).
HPLC chromatograms of DCA solutions were recorded by an Agilent Technologies 1200 Series instrument with Quaternary pump, column XDB-C18 (length = 15 cm, internal diameter = 4.6 mm, and particle size = 5 mm) and a UV detector. An atomic absorption spectrometer (AAS) 121 PerkinElmer Analyst (Air-C 2 H 2 , λ = 213.9 nm) was used to determine Zn in the solutions. For this goal, 0.1000 g of the solid catalyst was added into a Pt crucible and digested in 1 mL HF by heating on a sand bath until dried. Then, 1 mL (HClO 4 + HNO 3, 1:1 V) mixture was added and heated until dried. Finally, 1 mL HCl and 10 mL water were added, and after the dissolution of the solid, it was filtered in a 25-mL volumetric flask and reached the mark with water. The resulted solution was introduced to the AAS instrument (Nosuhi and Nezamzadeh-Ejhieh 2018;Sharifian and Nezamzadeh Ejhieh 2016).

Photodegradation experiments
To achieve an equilibrated adsorption/desorption process, the suspensions were shaken at dark for 10 min before the irradiation process. In a typical photodegradation process, 10 mL DCA aqueous suspension (15 mg/L DCA at pH 5.8 containing 0.75 g/L of the catalyst) in a 25-mL quartz beaker was irradiated by a 35 W Hg lamp on a magnetic stirrer. At definite times, the suspension was sampled out and centrifuged (>13000 rpm). The absorbance of the supernatant (Ao for before and A for after irradiation process) was recorded at λ max = 290 nm. The recorded absorbance values were then used for calculating of degradation percent of DCA by the following formula:

Results and discussion
Characterization results

Determination of loaded ZnO in the prepared ZnO-CNP catalysts
The used ZnO-CNP catalysts prepared in this work are listed in Table 1. The chemical composition of the samples is summarized in this table, which shows that with an increase in the Zn(II) concentration in the ion-exchange solution, the amount of Zn(II) loaded in CNP ion-exchange sites was relatively increased. The photodegradation experiments (see the following sections) showed that the best photodegradation rate could be achieved by CZ2 catalyst. Thus, this catalyst was applied for characterization methods.
Crystallite phase purity studies X-ray diffraction (XRD) was well known to identify the crystalline structure of a solid crystalline powder (Mortazavi and Aghaei 2020). The diffraction patterns of some CZ samples are shown in Fig. 1. In all patterns, the diffraction peaks of clinoptilolite have appeared because it consists of a majority of the samples. According to JCPDS No. 39-1383, these peaks agree with the standard pattern assigned in the literature (Olad et al. 2011

FTIR and TEM study
FTIR is a helpful tool for chemical structure characterization of chemical compounds by the appearance of absorption peaks of functional groups (Khozeymeh Nezhad and Aghaei 2021). FTIR spectra of the raw CNP and the ZnO-loaded CNP samples are shown in Fig. 2. The main peak positioned at 1092 cm −1 belongs to the Si-O-Si asymmetric stretching vibration. This peak usually appears in FTIR spectra of all zeolitic structures, and its position changes depending on the Si/ Al ratio of the framework. Furthermore, this peak overlaps with the Al-O-Si and Al-O stretching vibration modes. Here, some slight shifts in the range of 1082-1074 cm −1 appeared for the loaded CZ samples. Besides, the tetrahedra and double rings 4 and 6 atoms tetrahedral of the zeolite showed typical absorption peaks for stretching vibrations at 804-616 cm −1 and 476 cm −1 , respectively (Shahwan et al. 2005;Karimi Shamsabadi and Behpour 2021). In the ZnO-loaded samples, the absorption peak at 804 cm −1 was shifted in the range of 797 to 795 cm −1 , while the peak at 616 cm −1 in the range of 610 to 602 cm −1 .
It has been reported that the Zn-O bond can show an FTIR absorption peak in the range of 500 to 400 cm −1 (Nagaraju et al. 2017). For example, the non-calcined ZnO sample showed a Zn-O absorption peak at about 457 cm −1 which shifted and splitted to 518 cm −1 and 682 cm −1 when calcined at 300 and 500°C, respectively (SowriBabu et al. 2013). Here, the absorption peak of ZnO overlapped with the peak of zeolite at 476 cm −1 and caused the broadening of the peak and slight shifts in the range of 482 to 474 cm −1 . Figure 3C shows typical TEM of the as-prepared ZnO-CNP sample, that confirms the nano-dimension for the sample.

Estimation of pHpzc of ZnO-CNP
The accumulated charge on the surface of a catalyst/adsorbent plays a vital role in the efficiency of the proposed catalyst/ adsorbent. On the other hand, the net charge present on the catalyst's surface affects the adsorbed pollutant molecules on the surface, where hydroxyl, superoxide, and other reactive species formed. The most crucial point for evaluating the surface charge of the catalyst is the point of zero charges pH (pHpzc), in which the surrounding solution neutralizes the accumulated charges on the surface. The method used to determine this pH has been illustrated in the literature (Eslami et al. 2018;Nezamzadeh-Ejhieh and Zabihi-Mobarakeh 2014). Briefly, 5 mL suspensions of about 0.05 g ZnO-CNP sample in 0.01 M NaCl solution (as the ionic strength buffer) were prepared, and the pH was adjusted between 2 to 12 (pH I : initial pH). After 24-h shaking, the final pH (pH F ) was recorded. Based on the results, two plots were constructed for the estimation of pHpzc of the sample (Fig. 3). Figure 3A shows the bisector method, in which the plot of pH I versus pH I gives the bisector of the plot. The crossing point of the plot of pH F versus pH I gives the pHpzc for the sample. Figure 3B shows the plot of ΔpH versus pH I in which when ΔpH is zero, the crossing point of the curve with the x-axis gives the pHpzc value for the sample.
As shown in Fig. 3A, the initial pHs of 2, 4, and 6 increased to about 4.75. 6.63, and 6.93 after shaking with the catalysts. Accordingly, in the ΔpH curve, the positive values were obtained. These pHs are located below pHpzc values. In these pHs, the catalyst's surface has negatively charged due to the adsorption of protons from the contacted aqueous solution. This process occurs due to the native negatively charged the surface or its native alkaline property at such pHs. After the pHpzc, the native positively charged surface or its native acidic property caused to adsorption of hydroxyl anions from the adjacent aqueous solution. Thus, the surface got a net negative charge. Due to the decrease in the concentration of hydroxyl anions in the solution, the pH of the adjacent solution was decreased. Thus, the initial pHs of 8, 10, and 12 were changed to 7.08, 7.25, and 11, respectively. Accordingly, in the Here, all samples have a pHpzc value of about 7.1 (or 7.2), which confirms that this value is not affected by the amount of ZnO loaded on the surface.

DRS analysis
Some prepared CZ samples were subjected to diffuse reflectance spectroscopy (DRS) to study the energy required for carrying out the electronic transitions in the samples (or estimate the bandgap energy). The obtained reflectance spectra are shown in Fig. 4A, which are used for further investigations by the Kubelka-Munk equation and Tauc plots. In the Kubelka-Munk model (Eq. 2), the absolute reflectance (R) in the reflectance spectra should be changed to the transformed values (K). Then, the obtained K-values should be changed to Eq. 3 by the procedures defined in the literature (Sobhani-Nasab et al. 2020;Azimi and Nezamzadeh-Ejhieh 2015).
In Eq. 3, the bandgap energy (Eg) for various electronic transitions (depending on the n-values) can be estimated by plots of (α υ) n versus photon energy (hυ). When the n-values of 1/2 and 2 use, the bandgap energy of the allowed direct and indirect electronic transitions can be estimated, respectively. But, when the n-values of 3/2 and 3 are substituted in Eq. 3, the Eg-values for the forbidden direct and indirect transitions can be estimated by the corresponding plots, respectively. In this equation, "α" shows the absorption coefficient of the sample, which varies with the change in the sample transmission (T) and thickness (t) (1/t=ln (1/T)) (Dianat 2018; Balakrishnan and John 2020; Ng 2011).
Typical Tauc plots for the various electronic transitions of the samples are shown in Fig. 4B-E. The rising slopes of the curves were extrapolated toward the x-axis. The crossing point shows the Eg-value for the investigated sample. The results are summarized in Table 2. As the results show, a slight

Theory of the photodegradation kinetics
Here, the effects of the influencing variables were evaluated kinetically. Thus, the theory of the kinetics of typical heterogeneous photodegradation catalysis will study here. The best model for this goal is the Langmuir-Hinshelwood (L-H) model constructed based on monolayer adsorption of both the pollutant and the oxidant molecules on the catalyst surface. The formation of this monolayer at the solid-liquid interface controls the rate of the overall process. On the other hand, the formation of this monolayer is a crucial factor and is the ratelimiting step of the process. The general L-H model is shown by Eq. 4 in which the essential factors in the reaction rate (r: in mg/L min) are the specific reaction rate constant (k: in mg/L min), the equilibrium constant of the reactant (K: in L/mg), and the concentration of the pollutant (C). When concentrations above 5 mM use, a zero-order model is supposed, but at concentrations below 1 mM, an apparent first-order reaction is supposed.
To simplify this equation, its integrated or the logarithmic form has been used that is shown by Eq. 5. In this apparent first-order equation, the initial concentration of the pollutant (C o ) changes to concentration at time t (C t ) by proceeding with the photodegradation process. Thus, the constant k calls the apparent first-order rate constant (Dharmraj Khairnar et al. 2018;Assi et al. 2017;Mohammadi et al. 2020;Ghattavi and Nezamzadeh-Ejhieh 2019;Pourshirband et al. 2021).
In the following sections, the kinetics of the effects of some important experimental variables will be discussed.
It is important to note that the mass transfer kinetics can be evaluated by the mass transfer factor (MTF) models which has been developed by Fulazzaky . This suggested the model capable to distinguish the different kinetics of the external and internal mass transfer in the removal of a solute via the adsorption process.
The effects of clinoptilolite, ZnO supporting, and loaded ZnO amount Figure 5A shows typical Hinshelwood plots obtained in the DCA removal by the supported ZnO onto clinoptilolite comparing the unsupported one. The slopes of these curves as a measure of the apparent first-order rate constants of the DCA removal are summarized in Table 3. As mentioned in the "Photodegradation experiments" section, before the photodegradation experiments, the suspension was shaken at dark for 10 min to remove the surface adsorption effects in DCA removal. Thus, the following photodegradation results will be obtained and reported in the next sections. The raw CP and CNP obtained a slight photodegradation efficiency. As we know, all zeolitic frameworks contain high amounts of AlO 4 and SiO 4 tetrahedra. Thus, their SiO 2 and Al 2 O 3 may act as semiconducting materials that can be excited against the arrived photons to produce e − /h + pairs and the abovementioned reactive radicals. When both CP and CNP ionexchanged in Zn(II) solution, the resulted Zn(II)-CP and Zn(II)-CNP showed relatively faster rates in DCA photodegradation concerning the raw zeolites. In this case, some weak Zn-O bonds formed in the ion-exchange sites of clinoptilolite act as weak semiconductors.
When the calcined ZnO-zeolite samples were used, the photodegradation rate was drastically enhanced concerning all raw and the Zn(II)-exchanged clinoptilolite and even bulk ZnO. In the bulk ZnO sample, aggregation of ZnO active centers caused the lesser absorption of the arrived photons. Thus, lesser  Fig. 3 The plots used for the estimation of pHpzc of ZnO-CNP samples Nezamzadeh-Ejhieh 2020, 2021) reactive species, including the e − /h + pairs and the radicals mentioned above, can be produced. In general, in bulk samples such as bulk ZnO, the path length for traveling the photoinduced e − / h + pairs from bulk to the surface of the semiconductor has increased, resulting in the e − /h + recombination. Thus, when ZnO has supported on the surface of both CP and CNP, its photocatalytic activity was increased because of the decrease in the aggregation of ZnO centers. Besides, zeolites have a permanent internal electric field that can cause higher separation of e − /h + pairs in the photoexcited ZnO species (Koc et al. 2016;Millot et al. 2004). In CNP, higher photodegradation rate can also be related to the higher effective surface area (Chahkandi et al. 2019;Norouzi Esfahani et al. 2020).  Figure 5B shows that in the supported ZnO-CNP, the amount of loaded ZnO was changed. No considerable change in the rate of DCA photodegradation was observed. At higher loading, some partial aggregation between ZnO species maybe happened, and thus, only the outer layers of the ZnO aggregates participate in the photon absorption. Based on the observed results, the CZ2 catalyst was used for the following steps.

Effects of the catalyst dosage and the pollutant concentration
The results in Fig. 5C and Table 3 show that when the concentration of DCA was increased from 5 to 15 mg/L its photodegradation rate was increased in the applied conditions and after that was decreased. However, the collision probability between the photoinduced reactive species (the e − /h + pairs, superoxide, and hydroxyl radicals) and the pollutant molecules plays a key role in the photodegradation rate. This is due to the very short lifetime of these reactive species (about a few nanoseconds) and their fast participation in the side effect reactions if there are no molecules of pollutants present near the catalyst surface, where these reactive species produce. The collision probability depends on the concentration of the pollutant, and by an increase in its value to a definite level, the collision probability can increase. Here, this optimum value is 15 mg/ L. At higher concentrations, despite the high collision probability between the species as mentioned above, the screening ef-

A B
Time ( In the next step, the effects of the change in the dose of the CZ2 catalyst in the photodegradation rate of DCA were studied. The results in Fig. 6A and Table 3 confirm that by an increase in the CZ2 dose from 0.5 to 0.75 g/L, the photodegradation rate of DCA was increased and after this optimum value, the photodegradation rate was decreased. The increased dose from 0.5 to 0.75 g/L provided more active ZnO centers to be countered with photons. Thus, more reactive species are produced in this range, causing an increase in the photodegradation rate. At higher doses, beyond this optimum value, the light scattering effects caused lesser photons in the inner parts of the suspension. Thus, the photoexcitation of ZnO was decreased. Furthermore, at higher doses, aggregation of the solid particles was increased. The resulted decrease in the effective surface area caused the decrease in the photoexcitation of the ZnO species (Zebardast et al. 2018;Movahedi et al. 2009). Thus, at higher dosages above 0.75 mg/L, the photodegradation rate was decreased.

Effect of the solution pH
One of the most critical operating variables on the photodegradation yield of a general heterogeneous photocatalysis is the pH solution. This factor simultaneously affects the charges aggregated on the surface of the catalyst and the various ionic species for the pollutant molecules present in the media. Such charged species in the media result in the electrical attraction or repulsive forces that drastically determine the overall photodegradation efficiency (Senobari and Nezamzadeh Ejhieh 2018).
Thus, the effects of the change in solution pH were evaluated by testing some photodegradation experiments in various pHs ranging from acidic to alkaline pHs. The resulted Hinshelwood plots are shown in Fig. 6B and the obtained rate constants in Table 3. The results show that the best DCA photodegradation rates were obtained in the relatively acidic pHs of 5-6.
In the "Estimation of pHpzc of ZnO-CNP" section, the pHpzc about 7.1 was obtained for the CZ2 catalyst. At this pH, the surface of the CZ2 catalyst has a net zero charge. At pHs below this value, the surface of CZ2 catalyst has a net positive charge. In contrast, at acidic pHs about 3, more DCA molecules have protonated because the conjugated acid of DCA has a pKa-value The bolded values show the optimized variable of 2.97. Thus, in acidic pHs about 3, a repulsive force has resulted from the repulsive force between these positively charged species. By increasing in pH toward 5-6, more DCA molecules are present as the neutral form while the surface of the CZ2 catalyst has positively charged yet. Thus, an attraction force can form between the positive charges accumulated on the catalyst surface and free electrons on the nitrogen atoms of neural DCA molecules in this pH range caused to a higher rate for DCA photodegradation. When the pH of the solution was increased to pHs above the pHpzc, the charge on the surface of the catalyst changes to negative. In these pHs, a repulsive force between the free electron pairs on DCA molecules and the negatively charged surface. Thus, the photodegradation rate of DCA was decreased by increasing pH toward the stronger alkaline pHs. Detailed discussion on the various effects of solution pH on photodegradation efficiency has been illustrated in the literature (Bunluesak et al. 2020;Nezamzadeh-Ejhieh and Salimi 2011).

HPLC study
We used the HPLC technique for confirming the DCA photodegradation. Thus, some photodegradation experiments were performed based on the optimal parameters at two photodegradation times of 180 and 30 min. The resulted phodegradaded solutions were separated by centrifugation and injected into the HPLC column. Such HPLC analysis was also applied to the blank DCA solution. The obtained HPLC chromatograms are shown in Fig. 7 that show a sharp peak for the blank DCA solution at a retention time of about 2.87 min. The intensity of this peak was drastically decreased after the photodegradation for 180 min and 300 min. The decrease in the peak intensities showed that 74% and 87% of DCA molecules were degraded during these times. Furthermore, new weak peaks formed at a retention time of 2.45 min that belongs to the formation of the degradation intermediates. These intermediates were further degraded by elongation of the process because their intensity was decreased during the time. No further investigation was done to detect degradation intermediates in this step, and only HPLC was used for confirming the DCA photodegradation.

Conclusion
The mechanical ball-mill process successfully prepared clinoptilolite nanoparticles (CNPs). ZnO was supported on the CNP via the ion exchange and calcination processes. pHpzc of the ZnO-CNP samples with various loadings of ZnO did not change from 7.1, confirming that the surface of CNP covered by a ZnO layer and this layer controls the accumulated charges on the surface. The change in the ZnO loading also observed no considerable change in the bandgap. The relatively low photocatalytic activity of the raw zeolite can be related to the presence of structural AlO 4 and SiO 4 tetrahedra  Nezamzadeh-Ejhieh 2020, 2021) units that may act as semiconductor centers for the e − /h + production under illumination. The ion-exchanged Zn(II)-CNP has also enhanced the photodegradation efficiency rather than the raw CNP. This may relate to the resulted Zn(II)-O bond in the surface ion-exchange sites that can act as weak semiconductor centers. The highly increased photocatalytic activity of ZnO-CNP concerning the bulk ZnO can be related to the zeolite's permanent internal electrical field, which interacts with the photoinduced e − /h + pairs to separate them. Also, the zeolitic support prevents the ZnO aggregation. The kinetics of the photodegradation process obeyed the apparent first-order kinetic model by applying the Hinshelwood model. The best photodegradation rate was achieved in moderate acidic pHs about 5-6 in which the DCA is present in the neutral form, and the catalyst surface has a net positive charge.