UiO-66/Palygorskite/TiO₂ Ternary Composites as Adsorbents and Photocatalysts for Methyl Orange Removal

Abstract

Metal–organic frameworks are recognized as a new generation of emerging porous materials in a variety of applications including adsorption and photocatalysis. The present study presents the development of ternary composite materials made through the coupling of UiO-66 with palygorskite (Pal) clay mineral and titanium dioxide (TiO₂) applied as adsorbent and photocatalyst for the removal of methyl orange (MO) from aqueous solutions as a typical anionic dye. The prepared materials were characterized using XRD, ATR, DR UV/Vis, and TGA analysis. Detailed kinetic experiments revealed that the presence of the clay at low amounts in the composite outperformed the adsorption efficiency of pure UiO-66, increasing MO adsorption by ca. 8%. In addition, coupling Pal/UiO-66 with TiO₂ for the production of ternary composites provided photocatalytic properties that resulted in complete removal of MO. This was not observed in the pure UiO-66, the Pal/UiO-66 composite, or the pure TiO₂ material. This study presents the first example of clay mineral/MOF/TiO₂ composites with improved performance in removing dyes from aqueous solutions and highlights the importance of coupling MOFs with low-cost clay minerals and photocatalysts for the development of multifunctional advanced composites.

1. Introduction

The decrease in environmental quality made clear the need to fill the void between development and sustainability. High amounts of pollutants end up in the aquatic environment [1]. Among the different xenobiotics that enter into the environment, synthetic dyes are a common class of organic pollutants with high concentrations in wastewater since they are used in many different industries (paper, printing, textiles, food, and pharmaceutical industry) [2,3,4,5]. Different technologies have been suggested for dye removal from effluents, among which adsorption [6,7,8,9] and photocatalysis [10,11,12] are considered suitable mostly due to the environmentally friendly character of the process. Adsorption offers the benefit of simplicity, high efficiency, and cost-effectiveness. On the other hand, photocatalysis is highly favorable when solid materials are applied as photocatalysts and when solar light is used for the decomposition of organic pollutants.

In both processes, efficiency is driven by the properties of the material used. Regarding adsorption, many different types of materials have been developed and applied in dye removal from the aquatic phase [13,14,15]. Metal–organic frameworks (MOFs) have been effectively applied as solid absorbing media including dye removal [2,16,17,18,19,20,21]. MOFs are highly porous synthetic materials that offer the benefit of chemical, textural, and compositional tunability [2]. In addition, they have also been applied as photocatalysts for the degradation of organic dyes [11,22] and have been coupled efficiently with traditional photocatalysts to improve performance [16,17,23,24] or the development of multifunctional composites [17,25]. Clay minerals have been also extensively studied as adsorbents for dyes removal [19,26,27]. They are low-cost, nontoxic materials with significant cation exchange capacity. However, their negative charge in neutral conditions limits their application as adsorbents of anionic chemical species [28,29]; therefore, chemical modification is required [30,31].

TiO₂ is the most used photocatalyst owing to its inherent properties such as high photocatalytic activity and nontoxicity [32,33,34]. Among the different polymorphs of TiO₂, anatase is preferred mostly due to the better charge handling properties [33]. Numerous studies have applied TiO₂-based materials for the degradation of dyes [35]. Formation of composites through the coupling of TiO₂ with MOF structures has also been exploited in photocatalytic reactions [11,16,17,22,24,25,36]. Clay mineral/TiO₂ composites have also proven to be more active than the individual counter parts in a variety of photocatalytic reactions [37,38,39]. Since adsorption is a prerequisite for photocatalytic degradation reactions, adsorption of dyes on TiO₂ [40,41] or MOF/TiO₂ composites [42] has also been investigated. As expected, pure TiO₂ materials present much lower adsorption capacity than MOF structures. Recently, Wu et al. showed that coupling MIL-101 (Cr) with TiO₂ increased the adsorption performance against methyl orange (MO) (Scheme 1) [42]. In addition, the adsorption of cationic dyes onto UiO-66/TiO₂ composites was not affected by the presence of TiO₂, while the composites presented superior photocatalytic activity [25]. Clay/MOF composites have also been developed and studied as adsorbents. However, the effect of the presence of both clay minerals and TiO₂ has never been studied.

Herein, novel ternary composites bearing UiO-66, palygorskite, and TiO₂ were developed using wet chemical processes. The materials were characterized and tested as adsorbents and photocatalysts for the removal of MO from aqueous solutions. The presence of palygorskite in the composite increased the adsorption efficiency while TiO₂ provided photocatalytic properties able to decompose MO. To the best of our knowledge, such composites made of these particular phases have never been reported before.

2. Materials and Methods

All chemicals were used as received without further purification. Palygorskite (Pal, from Ventzia continental basin, Western Macedonia, Greece) was obtained as follows: particles <2 μm were collected by gravity sedimentation, and the clay fraction was separated using centrifugation methods [38,39]. Only the clay fraction of the sample was used for the preparation of the composites in order to avoid other mineral impurities.

2.1. Synthesis of UiO-66

Synthesis of UiO-66 (U) was based on a previous reported process [17,43]. Briefly, 1.25 g of zirconium (IV) chloride (ZrCl₄, 99.5%, Aldrich, Germany) was dissolved in 50 mL of N,N-dimethyl formamide (DMF, 99.9%, Aldrich) followed by the addition of 10 mL of acetic acid, forming solution A. On the other hand, 1.23 g of terephthalic acid (Aldrich) was mixed in 100 mL of DMF, forming solution B. The two solutions were mixed and heated under stirring to 150 °C for 24 h in an oil bath. The material formed was collected by centrifugation and washed three times with ethanol and with acetone. Solvent exchange with acetone was performed over a period of 4 days. The solvent was changed every second day. Finally, the material was dried at 100 °C over night and stored in a desiccator.

2.2. Synthesis of TiO₂

Anatase TiO₂ (T) was synthesized using the microemulsion process [44]. Titanium(IV) isopropoxide (Aldrich, Germany) was added to an inverse emulsion of 50 mL of water and n-heptane (85/10 v/v vs. H₂O). The water/Ti molar ratio was equal to 100. Triton X-100 was used as a surfactant in 1-hexanol (105/100 v/v vs. surfactant). The solution was stirred for 24 h at room temperature. The material was collected and washed thoroughly with methanol by centrifugation and dried at 110 °C overnight. Anatase TiO₂ was obtained by calcination for 2 h at static air and 400 °C using a ramp of 1 °C·min⁻¹.

2.3. Synthesis of Binary Palygorskite/UiO-66 Composite

For the development of the palygorskite/UiO-66 composite (PU), a similar synthesis process was used with a slide modification. Firstly, 0.1 g of palygorskite was added to the ZrCl₄ solution, and the amounts of ZrCl₄ and terephthalic acid were adjusted, keeping the same ratio, such that the final composite contained a theoretical nominal amount of 10% by weight of palygorskite. The same washing and activation process was used as in the U sample.

2.4. Synthesis of Palygorskite/UiO-66/TiO₂ Composite

The Pal/UiO-66/TiO₂ ternary composite (PUT) was developed using a facile wet impregnation method. Preformed PU and TiO₂ were mixed at a mass ratio equal to 6/4 in 100 mL of double-distilled water. The mixture was first sonicated for 30 min and then stirred overnight. Mixing was followed by removing the solvent using rotary evaporation, drying overnight at 100 °C, and a thermal treatment step at 200 °C for 4 h (ramp 1 °C·min⁻¹).

2.5. Materials Characterization

X-ray diffraction (XRD) measurements were performed using a Bruker D8-Advance diffractometer equipped with a Lynx Eye detector operating at 40 kV and 40 mA, using K_α radiation of Cu at 1.5418°. A Thermo Fisher Nicolet iS10 spectrometer was used to obtain the attenuated total reflectance (ATR) IR spectrum in the 400–4000 cm⁻¹ range. Thermogravimetric analysis (TGA) was performed using a TA Instrument Q5000IR. Each sample was heated under air flow (25 mL·min⁻¹) up to 650 °C with a dynamic ramping rate depending on the weight loss. Diffuse reflectance UV/Vis absorption spectra were recorded in the range of 200–800 nm using a PerkinElmer Lambda 950 Scan spectrophotometer equipped with integrating sphere. For this analysis, BaSO₄ was used as the reference. An X-band Bruker ER200D electron paramagnetic resonance (EPR) spectrometer operating at the X-band frequencies was used to perform spin trap experiments. The 5,5-dimethyl-1-pyrroline N-oxide (DMPO) was used as spit trap for the detection of hydroxyl radicals (${\mathrm{OH}}^{•}$) at room temperature. A suspension of 1 mg·mL⁻¹ of the catalyst in water was sonicated for 10 min. Then, 100 μL of the above suspension was mixed with 100 μL of 0.005 M DMPO directly in the EPR flat cell. A quartz EPR flat cell was used.

2.6. Adsorption Experiments

The material (20 mg) was placed in an Erlenmeyer flask containing 50 mL of 90 mg·L⁻¹ MO aqueous solution. The pH of the suspension was adjusted to 5. Adsorption experiments were performed in a temperature-controlled incubator. The temperature was maintained at 25 °C, and the adsorption process was performed under dark conditions throughout the whole experiment. For the preparation of all solutions, Milli-Q water was used. Adsorption of MO was quantified through the following process; aliquots of 1.5 mL were periodically taken, filtered using a 0.45 μm PTFE Millipore disc, and measured using a Varian Cary 100 Scan spectrophotometer. Quantification of MO was based on comparison with standards. The adsorption capacity evaluation was performed using the following equation:

$${\mathrm{q}}_{\mathrm{e}}=\frac{\left({\mathrm{C}}_0-{\mathrm{C}}_{\mathrm{e}}\right)\times \mathrm{V}}{\mathrm{m}},$$

(1)

where q_e (mg·g⁻¹) is the adsorbed amount at equilibrium, C₀ and C_e (mg·L⁻¹) are the initial dye concentrations and at equilibrium, respectively, V is the volume of the mixture (mL), and m is the mass of the adsorbent (mg). The removal efficiency was calculated using the following equation:

$$\mathrm{RE}=\frac{\left({\mathrm{C}}_0-{\mathrm{C}}_{\mathrm{e}}\right)\times 100}{{\mathrm{C}}_0}.$$

(2)

2.7. Photocatalytic Reactions

Photocatalytic tests were carried out after the adsorption process reached equilibrium. In brief, after incubating the material with the MO solution under dark conditions for 1.5 h, the mixture was irradiated using six UV lamps (Philips 24W/10/4P lamps, λmax = 365 nm) symmetrically positioned in the periphery of the photoreactor. The temperature of the mixture was maintained at 25 °C with water circulation by means of a cryostat. Reactions were performed at ambient pressure. Samples were periodically taken and filtered using a 0.45 μm PTFE Millipore disc. The UV/Vis spectra were then collected using a Varian Cary 100 Scan spectrophotometer.

2.8. Recycling of the Materials

Recycling was performed by redispersing the materials in water at pH 10 to accelerate the desorption process followed by sonication for 10 min and washing thoroughly via centrifugation until pH was neutral. Then, the material was washed ethanol three times and finally dried at 100 °C under vacuum for 5 h.

3. Results and Discussion

3.1. Materials Characterization

Structural information of the prepared materials was first obtained by means of X-ray diffraction (XRD) analysis. Figure 1 presents the diffraction patters of all materials. A typical diffraction pattern for the pure UiO-66 was observed [25,45]. The PU composite presented the same diffraction pattern as the U sample. This observation verified that the presence of Pal did not affect the synthesis of UiO-66. No diffraction peak attributed to Pal was detected in the PU sample. Two reasons are probably responsible for this. Firstly, the amount of Pal in the composite was small. Secondly, there was an intense diffraction peak from UiO-66 in the ca. 8.5° region that overlapped with the main diffraction peak of Pal. TiO₂ presented the typical diffraction pattern of anatase. The PUT composite had a diffraction pattern that was a combination of the pure UiO-66 and TiO₂ sample, without any obvious differences compared with the patterns of their pure counterparts.

The materials were further characterized using attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy, and the data are given in Figure 2a. The characteristic peaks of UiO-66 were observed in the materials containing this specific MOF (U, PU, and PUT) [46,47,48]. The symmetric and asymmetric stretch vibrations in the COO⁻ group of the organic linker were observed at 1392 cm⁻¹ and 1582 cm⁻¹, while the vibrational mode of the C=C bond in the benzene ring was centered at 1510 cm⁻¹. Vibrational modes of the C–H, C=C, O–H, and OCO bonds were detected in the region 810–710 cm⁻¹. The broad peak at the beginning of the spectra (400–800 cm⁻¹) was attributed to the presence of TiO₂ [44,49]. Regarding the Pal sample, an intense peak was observed in the region 97–1060 cm⁻¹. This was not observed in the pure UiO-66 sample but was present in all composites bearing Pal (i.e., PU and PUT). This peak was attributed to the Si–O–Mg asymmetric stretching vibration [39,46,50] and verified the presence of Pal in the composite materials. It should be noted that the peak centered at 1582 cm⁻¹ in the UiO-66 sample was shifted to higher wavenumbers in the composite samples bearing Pal (Figure 2b). This is probably linked with the in-situ development of UiO-66 in the presence of Pal and suggests an interaction of the clay mineral with the carboxylate groups of the organic linker in the MOF structure. Overall, ATR verified the presence of UiO-66, Pal, and TiO₂ in the composite materials.

In order to get information regarding the amount of the different parts in the composites, the prepared materials were subjected to thermogravimetric (TGA) analysis. The TGA profiles of all materials are illustrated in Figure 3. A smooth mass decrease was observed in the pure TiO₂ that ended at ca. 7% total mass loss at 550 °C. The mass loss was greater in the Pal sample (ca. 18% at 550 °C). In this temperature range, the mass loss in the Pal sample was attributed to both superficially adsorbed and loosely bound zeolitic water, as well as structural OH₂ [51,52]. In the materials containing UiO-66, the small weight loss in the temperature range 50–400 °C was due to the loss of guest molecules (DMF). The fast drop in mass loss at temperatures higher than 450 °C was attributed to the decomposition of the organic framework of UiO-66. This second drop in mass was observed at lower temperatures in the composite bearing TiO₂. Taking into account the mass loss in the composites at temperatures higher than 400 °C (indicated by Δmass in Figure 3 for sample U) and the corresponding weight loss in the pure Pal and TiO₂, the Pal content was calculated equal to 7 wt.% in the PU and PUT samples, and the TiO₂ content was calculated equal to 39 wt.% in the PUT composite.

Diffuse reflectance UV/Vis spectroscopy was applied to study the electronic properties related with light absorption. Figure 4a presents the UV/Vis absorption spectra of the materials. The corresponding Tauc plots [38,39,53] used to estimate the bandgap energies (E_g) are given in Figure 4b. Materials containing TiO₂ were treated as direct, while the remaining materials were treated as indirect semiconductors [17]. Pure UiO-66 absorbed light at wavelengths lower than 310 nm in accordance with the literature [24]. The same absorption onset was observed for the PU sample. This corresponded to E_g values of ca. 4 eV. The absorption edge of TiO₂ was observed at ca. 390 nm, while Pal absorbed light in the whole visible region [38,39]. Light absorption of the composites was a combination of the absorption of the individual counter parts. The absorption onset of the PUT material was observed at ca. 390 nm, as for the pure TiO₂. This corresponded to an E_g value of 3.18 eV, typical for pure anatase TiO₂ [16]. A broad absorption was observed in visible region (400–800 nm) for both PU and PUT samples. This was strictly attributed to the presence of Pal. Overall, it was verified through DR UV/Vis spectroscopy that all parts in the composite contributed to the light absorption properties of the material.

3.2. Dye Adsorption Evaluation

The effect of contact time on MO adsorption over the prepared materials is given in Figure 5a. Adsorption equilibrium was achieved within less than 5 min for all materials containing UiO-66. On the contrary, equilibrium was reached at longer reaction times for pure Pal and TiO₂. Significant differences were also detected in the experimentally observed adsorption capacity at equilibrium (q_e,exp). The q_e,exp together with the removal efficiency (RE %) are given in

Table 1. Experimentally obtained and calculated values for q_e, removal efficiency (RE %), and correlation coefficient (R²) for the pseudo-second-order kinetic model used to simulate the adsorption process and photocatalytic rate constant (k).

Sample	qe,exp (mg·g−1)	RE (%) 1	qe,cal (mg·g−1)	k2 (g·mg−1·min−1) 2	R2	k (min−1) 3
U	195.7	87.1	196.1	0.012	0.999
PU	213.3	94.9	212.8	0.074	0.999
PUT	174.7	77.5	175.4	0.108	0.999	0.0153
TiO2	41.3	18.3				0.0043
Pal	20.9	9.3

¹ Removal efficiency; ² k₂: rate constant of the second-order adsorption kinetics; ³ k: first-order rate constant of the photocatalytic removal of MO.

. Both TiO₂ and Pal presented low q_e and removal efficiency. On the contrary, the pure U and PU composite presented high adsorption capacity, with the PU sample being the most efficient (q_e = 213.3 mg·g⁻¹). This suggests that the presence of Pal at low amounts in the composite was beneficial for MO adsorption. An increase in the adsorption capacity of dyes was previously shown in oxide/MOFs or clay mineral/MOF composites including UiO-66 [25,42,54,55], as well as in other MOF-based composites [20,21]. In the specific case of Pal/UiO-66 composites, formation of microporosity at the interface of the two parts was shown to increase the specific surface area that may affect adsorption efficiency [56]. A decrease in the q_e and the removal efficiency was observed in the composite bearing TiO₂ (i.e., PUT). This was probably related to the relatively high amount of TiO₂ in the PUT composite that presented low q_e. However, the removal efficiency in these composites was not a linear combination of the individual counter parts considering their content in the composites. Taking into account the fast kinetic profile of MO adsorption, the high q_e, and the removal efficiency, the adsorption behavior of the PUT composite was dominated by UiO-66. Lastly, in order to verify the stability of the composites, recycling adsorption experiments using the most efficient composite were performed (Figure S1). The stability was evidenced by the small loss of adsorption efficiency after three consecutive cycles [56].

The pseudo-second-order kinetic model [54,55,57,58,59] was used to fit the experimental data as represented by Equation (3).

$$\frac{\mathrm{t}}{{\mathrm{q}}_{\mathrm{t}}}=\frac{1}{{\mathrm{k}}_2{\mathrm{q}}_{\mathrm{e}}^2}+\frac{\mathrm{t}}{{\mathrm{q}}_{\mathrm{e}}},$$

(3)

where q_e is the amount of MO adsorbed at equilibrium, q_t the MO adsorbed at time t, and k₂ is the rate constant (g·mg⁻¹·min⁻¹) of the second-order adsorption kinetics. The fitted adsorption kinetic curves for the materials containing UiO-66 as a function of time are given in Figure 5b. The correlation coefficient R² together with the calculated adsorption at equilibrium (q_e,cal) is given in Table 1. High correlation coefficients were obtained for all materials. In addition, q_e,cal matched the experimental observation. These findings indicate that MO adsorption onto the pure U sample and all composites containing UiO-66 followed the pseudo-second-order kinetic model [54,55,57,58]. Overall, the adsorption experiments showed that Pal induced an increase in MO adsorption, while the presence of ca. 40 wt.% TiO₂ did not significantly affect the adsorption given its relatively high content in the composite and the low adsorption capacity.

3.3. Photocatalytic Activity Evaluation

The materials were further tested as photocatalysts for the degradation of MO. Before starting the catalytic process, the mixture containing 90 mg·L⁻¹ MO and 20 mg of the material was stirred in dark conditions for 1.5 h, as indicated by the adsorption experiments. The shaded area in Figure 6a corresponds to the adsorption process. After adsorption/desorption equilibrium was achieved, the sample was irradiated with UV light. The kinetics of MO photocatalytic degradation was monitored for 2 h. The data are presented in Figure 6a. No obvious degrease in MO concentration was detected by the Pal, pure UiO-66, and PU composite. Clay minerals do not possess any photocatalytic activity. On the contrary, UiO-66 is photoactive [37,53]. However, on the basis of the U and PU light absorption properties (Figure 4), light irradiation at wavelengths lower than 310 nm was required for their activation. Therefore, it is not surprising that U and PU were not active under the conditions of our experiment. MO concentration was reduced when materials containing TiO₂ were used. This implies that PUT is photoactive and able to degrade MO. UiO-66/TiO₂ composites were previously proven to be active in photocatalytic degradation of dyes [25,60]. Pure TiO₂ removed approximately 50% of the initial MO. Comparing the materials presenting photoactivity, complete decolorization was achieved only in the PUT composite within less than 2 h irradiation. The composite bearing UiO-66 and TiO₂ (UT) also presented photoactivity, but the removal efficiency was lower than that of the PUT composite. The high removal efficiency of PUT against MO was linked to both functions of the PUT composite (i.e., adsorbent and photocatalyst). Most of the MO (77.5%) was adsorbed, while the remainder was decomposed via photocatalysis.

In order to further evaluate the photocatalytic efficiency of the prepared materials, the photocatalytic degradation rates were calculated. The data were well fitted with a fist-order kinetic model. Rate constants (k) were estimated by fitting the experimental data with the following equation:

$$\ln \frac{{\mathrm{C}}_0}{\mathrm{C}}=\mathrm{kt},$$

(4)

where k is the apparent first-order rate constant (min⁻¹), t is the irradiation time (min), C₀ is the initial concentration (i.e., after the adsorption/desorption process reached equilibrium), and C is the concentration of MO at reaction time t. The data are presented in Figure 6b, and the photocatalytic degradation rates (k) are summarized in Table 1. The PUT composite presented higher degradation rates than the pristine TiO₂. This has previously been seen in UiO-66/TiO₂ [24,60] composites, as well as in clay/TiO₂ or TiO₂-based composites containing other natural porous minerals [37,38,39,61].

The improved photoactivity in TiO₂ composites has been linked with synergistic effects related to the increase in charge availability [11,25,45], as well as the deposition of TiO₂ nanoparticles on the support, which decreases agglomeration phenomena [37,39]. Both factors are important for photocatalytic reactions in the liquid phase [12]. An additional contributor to the improved photoactivity was the increase in the contact between TiO₂ and the substrate [17,53]. This could be achieved through the increase in MO concentration close to TiO₂ due to the high adsorption efficiency of UiO-66. The large size of MO molecular structure and the fast adsorption kinetics suggest that MO was most likely adsorbed onto the external surface of UiO-66 where TiO₂ particles were present. This potentially increased the contact between TiO₂ and MO, improving photoactivity.

To further verify the photocatalytic activity of the composites, in situ spin-trap electron paramagnetic resonance (EPR) spectroscopy was performed using DMPO to study the reactive oxygen species formed under irradiation. Figure 7 presents the EPR spectrum obtained at room temperature after 15 min irradiation of a PUT aqueous suspension under light irradiation conditions identical to the photocatalytic experiments. An EPR signal with a 1:2:2:1 intensity ratio corresponding to the DMPO–OH^• spin adduct was observed. These data clearly indicate that the PUT composite was able to produce OH^• under irradiation. A similar analysis was performed for the PU composite. However, no DMPO–OH^• was detected (Figure 7). This suggests that TiO₂ was required in the composite to produce reactive oxygen species. These data may explain the inability of PU and pure U to decompose MO under the conditions of our experiment.

4. Conclusions

A facile two-step synthesis process was applied for the synthesis of clay mineral/MOF/TiO₂ ternary composites as adsorbents and photocatalysts for the removal of MO from aqueous solutions. Pure Pal and TiO₂ presented low adsorption capacities. Low amounts of Pal in the composite increased adsorption by ca. 8%. The presence of TiO₂ at 40 wt.% reduced the adsorption efficiency but not significantly. In addition, it induced photocatalytic properties in the composite. Materials made of Pal, UiO-66, and TiO₂ completely removed MO, while pure TiO₂ removed ca. 50% of the initial dye concentration. This study highlights the importance of composites for the development of efficient materials presenting multifunctionality and presents the first example of coupling UiO-66 with low-cost clay minerals and TiO₂ for the development of dual-function advanced materials.