Controlled crystal phase and particle size of loaded-TiO₂ using clinoptilolite as support via hydrothermal method for degradation of crystal violet dye in aqueous solution

2.1 Materials

Natural clinoptilolite with a Si/Al ratio of 10.63 was supplied by Guo TouShengShi Science and Technology Co. Ltd. TiCl₄ (99% purity), CV dye (high purity biological stain) were obtained from J & K Co. Ltd. Deionized water with resistivity of 18.25 MΩ∙cm was used for making all kind of solutions.

2.2 TiO₂/clinoptilolite preparation

The water soluble and para-magnetic impurities present in finely divided natural clinoptilolite were removed by heating its suspension made in water at 70 °C under continues stirring for 8 h. The resulting suspension was then filtered and dried at 70 °C for overnight. This process was repeated for three times. The purified clinoptilolite was then treated with aqueous solution of TiCl₄.

25 mL of 0.25 M TiCl₄ solution made in ice cold water were taken in three different 50 mL Teflon lined vessels along with 1 g of obtained above clinoptilolite and enclosed in stainless steel autoclaves. The mixtures were then heated on oil baths under continuous magnetic stirring for 3 h at 100, 150 and 200 °C, respectively. After that, the suspension was filtered, washed with enormous amount of deionized water until the pH of rinsing water became almost neutral and finally dried in an oven at 70 °C for overnight. The composites were marked as RTZ1, RTZ2 and RTZ3 obtained at 100, 150 and 200 °C, respectively.

Similarly, the prepared composites were named as RTZ4, RTZ5 and RTZ6 corresponding to the 1, 2 and 3 M HCl, respectively, used for 0.25 M TiCl₄ hydrolysis. Another set of composites were named as RTZ7, RTZ8 and RTZ9 corresponding to 0.08, 0.15 and 0.5 M TiCl₄, respectively, each one made in 3 M HCl solution. All the composites RTZ4-RTZ9 were synthesized at 200 °C using 1 g clinoptilolite, 25 mL TiCl₄ solution and reaction time interval of 3 h.

Another set of composites were prepared by taking 12.5 mL of 0.5 M TiCl₄ solution made in ice cold water in six different Teflon lined vessels. The pH of these solutions was adjusted to 0.05, 0.50, 1.00, 2.00, 3.00 and 7.00, using concentrated NH₄OH solution and then diluted to 25 mL. 1 g of clinoptilolite was added to each 25 mL (0.25 M) of the above mentioned solutions and then heated hydrothermally at 200 °C under constant stirring for 3 h. The TiO₂/clinoptilolite composites obtained at pH values of 0.05, 0.5, 1, 2, 3, and 7 were named as ATZ1, ATZ2, ATZ3, ATZ4, ATZ5, and ATZ6, respectively. Pure rutile and anatase-rutile mixed phase TiO₂ were prepared by the same method mentioned above without adding clinoptilolite and were named as R-TiO₂ and AR-TiO₂, respectively.

2.3 Characterizations

The crystal structure of the resultant TiO₂/clinoptilolite catalysts was determined via XRD (Beijing Purkinje General Instrument Co. Ltd) with a CuKα radiation source in a 2θ range of 5–75°. Scherrer’s equation (D = kλ/βcosθ) was used to calculate the crystallite size of the supported TiO₂ (McGehee and Renault, 1972), where D is the size (nm) of the particle, k is a constant (0.9) called crystallite shape factor, λ is the wavelength of X-rays (0.15432 nm) from CuKα radiation source, β is the peak width at half of the maximum and 2θ is the diffraction angle. The elemental composition of clinoptilolite and its composites having TiO₂ were measured via ICP (Perkin Elmer Optima 2000 DV). The FT-IR spectroscopy in the wave number range of 400 ∼ 4000 cm⁻¹ was used to examine their chemical structures. N₂ adsorption-desorption isotherm measured at 77 K using JWGB JW-BK300 from Beijing Sci. & Tech. Co. Ltd, was used to find out their textural properties including pore size, pore volume, and specific surface area. Each sample was degassed at 200 °C for 6 h prior to measurement. The SEM (JEOLJEM-220) and TEM (JEOL-2010) operating at 15.0 kV and 200 kV respectively were used to observe their morphology and microstructures. The UV–vis spectrophotometer (SHIMADZU UV-2600) was used to find out the band gaps. The pH of point of zero charge (pH_pzc) was determined using Malvern Zetasizer (Malvern Instruments Ltd, Malvern, UK). The photoluminescence (PL) spectra for photocatalysts were investigated on F-700 FL (Hitachi High-Tech Science Corporation Tokyo Japan) fluorescence spectrophotometer.

2.4 Photo-catalytic degradation experiments

The photo-catalytic degradation of CV dye in aqueous solution was carried out at room temperature and 100 mL solution was taken in 250 mL glass beaker. The light source used was a high pressure UV-Hg source (80 W) enclosed in a rectangular steel box. The calibration of UV-irradiation source was carried out with iodide-iodate actinometry and the UV-fluence rate was found to be 140.2 mW/cm². Aliquots of used sample was taken at regular interval, centrifuged to remove the suspended solid particles and then analyzed using UV–Vis spectrophotometer. Each experiment was carried out in duplicate in order to validate the results. At the end of each cycle, the used catalysts were regenerated by centrifugation, washed with deionized water and dried at 60 °C overnight to study their photo-stability.

The degradation percentages, X % of the CV dye were calculated using the following Eq. (1):

3.1 Structure and texture of TiO₂/clinoptilolites

The XRD patterns of original clinoptilolite, typical ATZ1 and RTZ6 are shown in Fig. 1A, whereas the others (RTZ1-RTZ9 and ATZ1-ATZ6) are shown in Figs. S1 and S2 of the Supporting Information (SI) section. As can be seen, the HEU structure of clinoptilolite and the crystal phase of TiO₂ could be confirmed by comparison with JCDPSs values, in which, the characteristic peaks at 2θ degree of 9.98, 11.27, 22.29, 26.82, 30.07, and 32.23 in Fig. 1A and Figs. S1 and S2 are ascribed to the main features of clinoptilolite (Liu et al., 2014, Davari et al., 2017), others (RTZ6, as shown in Fig. 1A-c) at 2θ value of 27.79, 36.11, 41.39, and 54.19 in Fig. 1A-c and Fig. S1 are assigned to rutile phase, while that (ATZ1, as shown in Fig. 1A-b) at 2θ value of 25.3, 38.07, 48.07 and 53.91 (Fig. 1A-b and Fig. S2) are attributed to the anatase phase (Davari et al., 2017, Sene et al., 2017, Tang et al., 2016). Furthermore, as shown in Table 1, the crystal size calculated on the basis of Scherer’s equation presented the increasing tendency of from 13 to 26 nm for rutile phase in the composite RTZ1-RTZ9 along with enhancements of temperature, acidity, and concentration of TiCl₄, but almost the same around 8–10 nm for anatase phase in the composites ATZ1-ATZ3. However, the decline phenomenon in intensity of all characteristic peaks for clinoptilolite in ATZ samples and RTZ samples was due to the large amounts and big particles of loaded TiO₂. The XRD patterns of both ATZ1 and RTZ6 after photocatalytic degradation (Fig. S4) presented declined intensity of characteristic peaks belonging to anatase, rutile, and clinoptilolite, as compared that pristine catalysts (Fig. 1A). The main reason behind it might be due to the adsorption of dye molecules on the surface of catalyst.

Close

Fig. 1

(A) XRD patterns, (B) FT-IR spectra, (C) Zeta potentials of clinoptilolite and related composites measured by zetasizer, (D) Zeta potentials of clinoptilolite and related composites determined via classical method. (a): clinoptilolite, (b): ATZ1, and (c): RTZ6.

Export to PPT

FT-IR spectra of ATZ1 and RTZ6, as shown in Fig. 1B, exhibit almost the same profiles as that of clinoptilolite, suggesting that the skeleton framework of clinoptilolite are remained intact after TiO₂ loading. In details, the most intense peak at around 1050 cm⁻¹ is associated to asymmetric stretching vibration of O—Si(Al)—O (Trujillo et al., 2013), which shifts to a relatively higher wavenumber in RTZ6 (1090 cm⁻¹, as shown in Fig. 1B-c) and ATZ1 (1080 cm⁻¹, as shown in Fig. 1B-a) as compared to clinoptilolite (Fig. 1B-a). The bands appeared at around 1650 and 3500 cm⁻¹ are assigned to the stretching vibration of OH group of water present in the zeolite structure (Rahmani et al., 2015). Other peaks at around 570, 720, and 800 cm⁻¹ are attributed to the stretching vibrations of tetrahedral atoms and dual rings 4 and 6 tetrahedral atoms present in zeolite structure (Davari et al., 2017, Najafabadi and Taghipour, 2012). The broadening of bands below 1000 cm⁻¹ observed in ATZ1 (Fig. 1B-b) and RTZ6 (Fig. 1B-c) after TiO₂ loading may be due to the overlapping of Ti—O stretching band at 690 cm⁻¹ of TiO₂ and Si—O—Si bands of zeolite (Yener et al., 2017).

Fig. 1C and D show the zeta potential of clinoptilolite and related components as a function of pH of the solutions in the range of 2–10 as determined by zetasizer and classical method, respectively. As can be seen, the pH values of point of zero charge (pH_pzc) for clinoptilolite (Fig. 1C-a), ATZ1 (Fig. 1C-b), and RTZ6 (Fig. 1C-c) measured by zetasizer were found to be 2.23, 6.66, and 6.15, respectively. While the pH_pzc for clinoptilolite (Fig. 1D-a), ATZ3 (Fig. 1D-b), and RTZ6 (Fig. 1D-c) were found to 2.5, 6.5, and 5.6, respectively, as determined by classical method. The relatively high value of pH_pzc of ATZ1 (Fig. 1C-b) and RTZ6 (Fig. 1C-c) as compared to the pure clinoptilolite (Fig. 1C-a) may be due to their synthesis procedure. The surface at pH below pH_pzc reveals positively charged, whereas negatively charged at pH above pH_pzc. Therefore, it is concluded that the surface of clinoptilolite throughout the pH range of 0.05–7.00 remains highly negatively charged.

UV–vis spectra of TiO₂ supported clinoptilolite were taken to study the band gap variation during the process. Band gap of TiO₂/clinoptilolite was determined using Tauc relation (Ahυ)ⁿ = C(hυ-Eg), where A is the absorbance, h is the Planck’ s constant, υ is the frequency and Eg is the band gap energy. The n has value of 1/2, 2, 3/2, and 3 (Karimi-Shamsabadi and Nezamzadeh-Ejhieh, 2016) for the allowed direct, indirect, and forbidden direct, and indirect, respectively. Accordingly, (Ahυ)ⁿ (y-axis) was plotted against hυ (x-axis). Extrapolating the linear portion of (Ahυ)ⁿ vs hυ plot to y-axis = 0, for n = 1/2 and 2 gives different Eg values, as shown in Table 1. As a conclusion, the band gaps for rutile phase TiO₂/clinoptilolite composites (2.85 eV for RTZ6), as shown in Table 1, are reduced considerably as compared to that with anatase phase (3.19 eV for ATZ1).

The PL spectra were determined to understand the rate of photo-generated electron-hole pair recombination in different TiO₂/clinoptilolite catalysts, as shown in Fig. S5. As can be seen in Fig. S5-i, pure rutile TiO₂ (R-TiO₂) exhibits the highest emission intensity, while ATZ1, RTZ6 and AR-TiO₂ have the least emission intensity. These PL signals result from oxygen vacancies and defects of TiO₂, in which, clinoptilolite as substrate is useful to decrease oxygen vacancies and intrinsic defects of TiO₂. In this case, small particles have large oxygen vacancy content and thus have higher probability of excitation, leading to the strong PL signals (Wang and Li, 2014; Zhang et al., 2014).

The textural parameters of all samples originating from N₂ adsorption-desorption isotherms are listed in Table 1, in which, the surface areas firstly decreased from 134 m²/g for RTZ1 to 21 m²/g for RTZ3, and then increased gradually to 51 m²/g for RTZ6, while the surface area (45.4 m²/g) for ATZ1 was low, but higher for ATZ3 (68.2 m²/g). These observations may be based on the fact that increasing acidity of the media used for hydrolysis-polycondensation duration of TiCl₄ could prevent the TiO₂ particles from agglomeration during hydrothermal synthesis, which in turn are beneficial to uniform and independent dispersion on the clinoptilolite.

Moreover, as listed in Table 1, Si/Al ratio of RTZ1-RTZ9 samples with TiO₂-loaded contents of around 6.55 to 31% exhibited 20-41, which became higher than that of pure clinoptilolite (around 10.6), due to the dealumination under high acidic conditions.

Fig. 2a and c shows rod-like particles with uniform dispersion in size of around 30 nm of RTZ6, which is different from that of the very well defined smooth surface of pure clinoptilolite (Fig. S3-a). ATZ1 (Fig. 2b and d) presents the deposition of spherical-like particles with comparatively less agglomeration. These observations are found in close agreement with crystal phase loaded TiO₂ calculated on the basis of XRD patterns (Table 1). The SEM images of ATZ1 and RTZ6 after photocatalytic activity, as shown Fig. S6, exhibit almost the same morphology to that of pristine catalysts which reflect the long term photo-stability of the composite catalysts. The SEM images of other samples (RTZ1-RTZ9 and ATZ2-ATZ3), as shown in Fig. S3 of SI section, also exhibit similar phenomena. Additionally, the TEM mapping of Ti in RTZ6 (Fig. 2e) and ATZ1 (Fig. 2f) illustrate uniform distribution on the surface of clinoptilolite, rather than going inside the micropore channels. The HRTEM images of RTZ6 (Fig. 2g) and ATZ1 (Fig. 2h) further reconfirm the presence of rutile and anatase phases in RTZ6 and ATZ1, respectively, which are aslo found in good agreement with that results depicted by XRD patterns (Fig. 1A). Selected area electron diffraction (SAED) images were shown in Fig. 2i, and j. As can be seen, the SAED patterns in RTZ6 (Fig. 2i) and ATZ1 (Fig. 2j) exhibit spotty ring pattern without any additional diffraction spots, corresponding to different crystal planes of rutile and anatase phases of TiO₂, respectively, which strongly support XRD analysis in Fig. 1A.

Close

Fig. 2

(a and b) SEM images, (c and d) TEM images, (e and f) Mapping of Ti, (g and h) HRTEM images, and (i and j) SAED images of rutile and anatase phases of TiO₂ in RTZ6 (a, c, e, g, and i) and ATZ1 (b, d, f, h, and j), respectively.

Export to PPT

3.2 Photo-catalytic degradation

Prior to photo-catalytic degradation, the solution was kept under continuous magnetic stirring for one hour in dark to attain adsorption-desorption equilibrium. The amount of adsorbed CV dye was found to vary between 5 and 15% for synthesized photo-catalysts. In order to elucidate the effect of UV-light on the degradation properties of CV dye, an experiment was carried out as a blank under UV-irradiation in the absence of photo-catalyst. Only about 5% degradation of CV dye was obtained under UV-irradiation within 60 min. Therefore, the influences of various parameters on the catalytic activity are thoroughly investigated and given in the following results and discussion section.

3.2.1

3.2.1 Effect of type and dose of TiO₂/clinoptilolites

As shown in Fig. 3, ATZ1 is the most photo-catalytically active (up to 67.6% degradation) when using anatase-containing TiO₂/clinoptilolites as catalysts, however, the rutile-contained catalysts are relatively less photo-catalytically active due to its fast electron-hole recombination characteristics (Allen et al., 2018, Behnajady et al., 2008). While RTZ6 is the most photo-catalytically active (58.8% degradation), which may be due to their relatively large surface area (51 m²/g) and bigger crystal size (26 nm) of rutile phase. Furthermore, large particle size and better dispersion of TiO₂ in RTZ6 as compare to other rutile contained catalysts leads to considerable PL quenching as clear form Fig. S5, indicating effective inhibition of recombination of generated electron-hole pair. Almquist and Biswas (2002) synthesized TiO₂ containing different particle sizes (5–165 nm) and concluded that an optimum particle sizes of 25–40 nm is the best for the photo-catalytic activity. They demonstrated that the increase in particle size of rutile TiO₂ tends to decrease the electron-hole recombination rate, in other words, the rate of electron-hole recombination increases on the particle surface with the decreasing particle size due to the enhanced proximity of the charges. Presently, we found that the degradation efficiencies of the pure rutile TiO₂ and anatase-rutile mixed TiO₂ (89.68% anatase + 10.32% rutile) were about 32.26 and 55.50%, respectively. Therefore, RTZ6 and ATZ1 were selected as the optimum photo-catalysts for further studies.

Close

Fig. 3

Effect of type of photo-catalyst on the photo-catalytic degradation of crystal violet dye present in aqueous solution under following conditions: catalyst dose = 0.5 g/L, pH = 5.8, CV dye concentration = 12 ppm, Time = 100 min.

Export to PPT

To determine the optimum amount of the obtained TiO₂/clinoptilolites for degradation of CV dye, five different doses i.e. 0.12, 0.25, 0.5, 0.75 and 1 g/L of TiO₂/clinoptilolite were used, and their degradation performances are shown in Fig. 4A. As can be seen, the degradation rate of CV dye firstly increased to 67% for ATZ1 (Fig. 4A-a) and 58% for RTZ6 (Fig. 4A-b) with increasing dose up to 0.5 g/L, then decreased to 53% for ATZ1 and 50% for RTZ6, while using larger doses (around 1.0 g/L). The relatively low degradation rate of CV dye with lower dose, i.e. 0.1 and 0.25 g/L was quite, because less number of catalyst particles interact with UV light to produce the reactive radicals i.e. ^•OH, O₂^•− etc. (Nezamzadeh-Ejhieh and Khorsandi, 2010). While high dose, i.e. 0.75 and 1 g/L, may scatter the light and causes difficulty for UV-light to pass through the solution easily, resulting in the decrease of degradation of CV dye (Nezamzadeh-Ejhieh and Khorsandi, 2010, Shafaei et al., 2010). In these cases, 0.5 g/L is selected as the optimum dose of used catalyst. Furthermore, ATZ1 (Fig. 4A-a) exhibits higher photo-catalytic activity as compared to RTZ6 (Fig. 4A-b) for all the doses of the used catalyst. The low degradation efficiency of RTZ6 is attributed to the relatively fast electron-hole recombination taking place in rutile phase of TiO₂ after UV-irradiation. However, the photo-catalytic activity using pure clinoptilolites as catalysts with different doses showed less than 6% degradation of the CV dye (Fig. 4A-c), suggesting poor photo-catalytic behavior of pure clinoptilolite.

Close

Fig. 4

Effect of various parameters on the % degradation of CV dye: (A) amount of photo-catalysts. Conditions: Initial pH = 5.8, dye concentration = 12 ppm, Time = 100 min, Room temperature. (B) pH of media. Conditions: dose of catalyst = 0.5 g/L, dye concentration = 12 ppm, Time = 100 min, Room temperature. (C) Irradiation time interval. Conditions: Initial pH = 8, dose of catalyst = 0.5 g/L, dye concentration = 12 ppm, Time = 100. (D) UV–Vis absorbance spectra of CV catalyzed by ATZ1. (a): ATZ1, (b): RTZ6, and (c): pure clinoptilolite.

Export to PPT

3.2.2

3.2.2 Effect of pH values, irradiation time, and initial concentration of dye

The pH value is an important factor, affecting the photo-catalytic degradation of organic contaminants in wastewater due to its strong influences on the production of ^•OH radicals and also the charge of the catalyst surface (Nezamzadeh-Ejhieh and Amiri, 2013). The effect of pH value was investigated by performing experiments at various pH values, and Fig. 4B shows that the photo-catalytic degradation rate is around 37% for ATZ1, 7% for RTZ6, and 5% for pure clinoptilolite at low pH value of around 2.0, but gradually increases to 89% for ATZ1 (Fig. 4B-a), 76% for RTZ6 (Fig. 4B-b), and 7% for pure clinoptilolite (Fig. 4B-c) with increasing initial pH (up to 8) of the solution. As shown in Fig. 1C, the pH_pzc values of 6.15 and 6.66 determined for RTZ6 and ATZ1, respectively, have direct influence on the photo-catalytic degradation. Therefore, the catalyst surface at pH < pH_pzc is positively charged, which hinders the adsorption of cationic CV, leading to the low degradation rate (Nezamzadeh-Ejhieh and Shirzadi, 2014). Both the composites, i.e. ATZ1 and RTZ6 present highest degradation efficiency at pH of 8 and then a little decrease in degradation efficiency occurs at pH 10. These results were similar to that reported by Nezamzadeh-Ejhieh and Khorsandi, (2014). The decreased degradation efficiency at pH > 8 is due to the appearance of repulsive forces between the highly negatively charged catalyst surface and free electron pairs of the CV dye. Therefore, relatively less molecules of CV dye can reach the catalyst surface where the highly reactive ^•OH radicals are generated. Furthermore, there might occurs two types of reactions such as (i) reaction of ^•OH with ^•OH due to the presence of greater amount of ^•OH radicals (Liu et al., 2012), (ii) reaction of ^•OH with ⁻OH, at high pH values producing relatively less reactive species (Nezamzadeh-Ejhieh and Moeinirad, 2011), and consequently reducing the % degradation.

As seen in Fig. 4C, the degradation of CV dye increases up to 89.5% for ATZ1 (Fig. 4C-a), 78.7% for RTZ6 (Fig. 4C-b), and 7% for pure clinoptilolite (Fig. 4C-c) with prolong time interval (around 100 min). The corresponding UV–vis spectra of CV dye at different irradiation time intervals using ATZ1 as a photocatalyst is also shown in Fig. 4D, which clearly illustrates a decrease in absorbance of the characteristic peak with increasing time interval.

Photo-catalytic degradation process using different initial concentrations of CV dye is shown in Fig. 5. As can be seen in Fig. 5A and B, the degradation rate using both RTZ6 and ATZ1 as catalysts decreases with increasing initial concentration of CV dye. It may be due to the fact that fix number of ^•OH radicals produced from a given amount of photo-catalyst and UV-light are reacting with different initial concentrations of CV dye. Furthermore, the opacity of solution increases with increasing concentration of CV dye and penetration of UV-light through the solution becomes difficult, producing relatively less number of the highly reactive ^•OH radicals.

Close

Fig. 5

Degradation of CV dye at different initial concentration of CV dye using RTZ6 (A and C) and ATZ1 (B and D). Conditions: 0.5 g/L of catalyst, pH = 8, RT, Time = 50 min.

Export to PPT

In general, the kinetic behavior of heterogeneous photocatalytic reactions is described by following Langmuir Hinshelwood equation:

(2)

$$r=-dc/dt=k\theta =k(KC/1+KC)$$ where r is the rate of reaction, k the rate constant of photocatalysis (mg/L.min), K the Langmuir constant, C the concentration of dye solution, and c the concentration of dye solution at any time (min). If the concentration C > 5.0 mmol/L, then KC >> 1 and the reaction becomes zero order kinetics. For dilute solutions where C < 1.0 mmol/L, KC << 1 and the reaction becomes pseudo-first order.

Hence, the photo-catalytic degradation of CV dye is found to fit the pseudo-first order kinetic expression (3):

(3)

$$\ln \frac{C_0}{C}=K_{\mathit{app}}·t$$ where C₀ and C are the concentration of CV dyes at time 0 and t respectively, k_app is the apparent rate constant and t is the time interval. A linear co-relation was observed between ln(C₀/C) vs irradiation time using different initial concentrations of CV dye, as shown in Fig. 5C and D. A decrease of k_app was observed with the increasing C₀ of CV dye (seen in Table 2), suggesting low degradation rate for high initial concentration of the dye.

Table 2 Observed rate constants and degradation (%) using RTZ6 and ATZ1 as photocatalysts at different initial concentration of CV dye.

Catalyst	Dye concentration (ppm)	Degradation (%)	kapp (min−1)	R2
RTZ6	4	71.97	0.023	0.987
	8	62.61	0.019	0.999
	12	52.28	0.012	0.969
	16	41.71	0.009	0.934
	20	18.48	0.003	0.971
ATZ1	4	79.21	0.030	0.985
	8	70.88	0.023	0.977
	12	64.58	0.019	0.985
	16	44.71	0.011	0.985
	20	22.60	0.005	0.964