Hydrothermal synthesis of TiO₂/carbon composites and their application for removal of organic pollutants

2.1 Sample preparation

Hydrothermal synthesis of TiO₂/carbon composites (TiO₂/HTC) was carried out using the following procedure: 37 ml solution of glucose was mixed with 3 ml 35% hydrochloric acid and 6 ml of titanium isopropoxide was added dropwise. Different concentrations of glucose solution: 5, 10, 15 and 30 g dm⁻³ were used to achieve following Ti/C molar ratio in suspension: 0.05; 0.1; 0.15 and 0.30; the obtained samples were marked as TiO₂/HTC₁, TiO₂/HTC₂, TiO₂/HTC₃ and TiO₂/HTC₄, respectively. The characteristics of the obtained TiO₂/HTC composites were compared with the characteristics of TiO₂ – hydrothermal, obtained by similar procedure, using 37 ml of water instead of glucose solution. The suspension was transferred into 50 ml Teflon lined stainless steel autoclave where hydrothermal carbonization is carried out at temperature at 160 °C and self-generated pressure for 12 h. After completion of the reaction, autoclave cooled to room temperature and suspension was centrifuged. The resulting precipitate was filtrated and washed with distilled water and ethanol several times, and finally dried at 60 °C overnight.

2.2 Sample characterization

The thermogravimetric (TG) analysis was performed on an SDT Q600 instrument (TA Instruments) in O₂ atmosphere (flow rate: 100 cm³ min⁻¹; heating rate: 20 °C min⁻¹), ranging from room temperature to 800 °C.

The morphology and the surface properties of the obtained materials were investigated by scanning electron microscopy (SEM) using a Mira Tescan 3X at 20 keV.

Nitrogen adsorption–desorption isotherms were determined using a Micromeritics ASAP 2020 instrument. The specific surface area of samples was calculated according to the Brunauer, Emmett, Teller (BET) method from the linear part of the nitrogen adsorption isotherms. The total pore volume (V_tot) was given at p/p₀ = 0.998. Pore size distribution and volume of the mesopores (V_meso) were estimated by applying the Barrett, Joyner and Halenda method from the desorption branch of isotherm. The volume of micropores (V_micro) was calculated from alpha-S plot.

In order to study the crystal modifications of the TiO₂ X-ray diffraction (XRD) was used. XRD spectra were recorded in the range of 2θ of 20–60° with a scan speed 1 °C min⁻¹ using a Philips PW1710 diffractometer with CuKα radiation.

UV–Vis diffuse reflectance spectroscopy (DRS) was used to examine the effect of precursor molar ratio on the shift of energy band gap of TiO₂ to the visible region. The spectra of the obtained materials were recorded on a Shimadzu 2600 UV–VIS spectrophotometer with an integrating sphere in the range of 220–750 nm, using BaSO₄ as a reference.

Fourier transform infrared spectroscopy (FT-IR) measurements were used for characterization of functional groups. FT-IR spectra were recorded in the range from 400 to 4000 cm⁻¹ on Bomem MB-Series, Hartmann Braun.

2.3 Measurement of photocatalytic activity

The photocatalytic activity of the TiO₂/HTC composites was examined in the degradation process of methylene blue (MB). All experiments were performed at room temperature with 1 g dm⁻³ of TiO₂/HTC composites and the initial concentration of MB was 10 mg dm⁻³. The suspension was magnetically stirred in the dark for 60 min to establish the adsorption/desorption equilibrium, and then exposed to the UV light irradiation using 125 W high-pressure mercury lamp (Philips, HPLN, emission bands in the UV region at 304, 314, 335 and 366 nm, with maximum emission at 366 nm). UV lamp was located at 10 cm above the solution. The entire process of adsorption and decomposition lasted 5 h under constant stirring. The possibility of reusing TiO₂/HTC composites was examined through the determination of photocatalytic activity of sample TiO₂/HTC₄ after each of five cycles. After finishing a cycle, sample TiO₂/HTC₄ was rinsed and dried in the atmosphere, without any other treatments, and then used in next cycle. Photocatalytic activity of sample TiO₂/HTC₄ in the visible light was tested through photocatalytical degradation of MB, under the same experimental conditions. For this purpose 150 W tungsten halogen lamp was used and short wavelength components (λ < 420 nm) of the light were cut off using a glass optical filter. During the process of photocatalytic degradation, at given time intervals, samples were taken and filtered through the 0.45 μm PVDF filters, and then concentration of MB was measured. In the case of sample TiO₂/HTC₄, additional desorption experiment was performed, after the adsorption and photocatalytic experiments were completed. MB, remained on the sample surface after the adsorption and photocatalytic experiments, was desorbed by methanol in the batch system with constant shaking. Concentration of MB was measured using visible spectrophotometer (Specol, Carl-Zeiss, Jena), by measuring absorbance at 675 nm.

Also, photocatalytic activity of the TiO₂/HTC composites was examined in the degradation process of selected pharmaceuticals from multicomponent solution. Initial concentration for each of selected pharmaceuticals: diclofenac, azithromycin, doxycycline, erythromycin, bromazepam, atorvastatin, amlodipine, cilazapril, simvastatin and clopidogrel were 100 μg dm⁻³. All experiments, photocatalytic degradation under UV irradiation and desorption experiments on TiO₂/HTC₄, were performed under the similar conditions, as described for MB experiments. Concentration of pharmaceuticals was measured by high performance liquid chromatography-tandem mass spectrometry (LC-MS/MS Thermo Scientific).

Surveyor HPLC system was used for the separation of the analytes on the reverse-phase Zorbax Eclipse XDB-C18 column, 75 mm long, 4.6 mm i.d. and 3.5 μm particle size (Agilent Technologies). The mobile phase consisted of water (A), methanol (B) and 10% acetic acid (C) and gradient changes are shown in Fig. 1. An aliquot of 10 μl of the aqueous solution was injected into HPLC system. Linear ion trap mass spectrometer, LTQ XL, was used for detection and quantification of pharmaceuticals. The electrospray ionization technique was used and all pharmaceuticals were analyzed in the positive ionization mode. From obtained MS² spectra of pharmaceuticals, most abundant fragment ions were selected. The selected reaction monitoring (SRM) mode was used for quantification of all pharmaceuticals. Mass chromatogram of selected pharmaceuticals is shown in Fig. 2.

Close

Figure 1

Diagram of the mobile phase composition (A – water, B – methanol and C – 10% acetic acid) and flow rate.

Export to PPT

Close

Figure 2

Mass chromatogram of selected pharmaceuticals.

Export to PPT

3.1 Characterization of photocatalysts

Thermogravimetric analysis (TG) is used to estimate the TiO₂ and carbon content of hydrothermally synthesized samples, and TG curves of TiO₂ – hydrothermal and TiO₂/HTC composites are shown in Fig. 3. The mass loss for TiO₂ – hydrothermal was 3.37 wt% due to the removal of physically and chemically adsorbed water. For TiO₂/HTC composites, the mass loss up to the temperature of 125 °C is a consequence of physically adsorbed water, while the substantial mass loss observed from 450 °C to 500 °C, depending on the sample, is the result of carbon release during the thermal treatment. Based on the thermal analysis results, TiO₂ and carbon content of TiO₂/HTC composites was calculated and is shown in Fig. 3.

Close

Figure 3

TG curves of TiO₂ – hydrothermal and TiO₂/HTC composites with embedded TiO₂ and carbon content.

Export to PPT

Morphology of obtained TiO₂/HTC composites is shown in Fig. 4. For the purpose of comparison, SEM photographs of TiO₂ – hydrothermal, along with hydrothermal carbon (HTC) derived from glucose are shown in Fig. 4a and f, respectively. The similarity between morphologies of TiO₂/HTC composites and TiO₂ – hydrothermal can be noted, with respect to decrease in particle sizes with the increase in carbon content. On the other hand, HTC poses quite different morphology (Fig. 4f); it is characterized by the presence of large and clearly defined carbon spheres with smooth surface that form cluster-like structure (Sun and Li, 2004; Titirici et al., 2008). Addition of titanium isopropoxide to the glucose solution in order to obtain TiO₂/HTC composites (Fig. 4a–e), induces the morphological changes, in the first place, the disappearance of HTC spherical structure.

Close

Figure 4

SEM photographs of samples: (a) TiO₂ – hydrothermal; (b) TiO₂/HTC₁; (c) TiO₂/HTC₂; (d) TiO₂/HTC₃; (e) TiO₂/HTC₄ and (f) hydrothermal carbon.

Export to PPT

Textural properties TiO₂ – hydrothermal and TiO₂/HTC composites are summarized in Table 1. Obtained TiO₂/HTC composites are mesoporous, with the average pore diameter from 3.78 nm, for TiO₂/HTC₄, to 17.22 nm, for TiO₂ – hydrothermal. The results summarized in Table 1 show that average pore diameter and V_meso decrease with the increase in carbon content. On the other hand, increase in carbon content leads to the increased S_BET and V_micro.

XRD analysis was performed in order to analyze crystal phases of TiO₂, present in TiO₂/HTC composites, and to determine phase content and grain size. Also, the influence of precursor ratios (Ti/C), used for hydrothermal synthesis of TiO₂/HTC composites, on the phase formation was examined. The anatase is active form of photocatalyst in the processes of oxidation, while the rutile form is suitable to operate in the near UV range (350–400 nm), because it has a lower energy gap (3.0 eV) than the anatase modification (3.2 eV) (Hanaor and Sorrell, 2011). The presence of TiO₂ in the obtained samples was confirmed by the XRD characteristic peaks for the anatase (1 0 1) (2θ = 25.6°) and rutile (1 1 0) (2θ = 27.7°) phases (Kordouli et al., 2015). The XRD pattern of TiO₂ – hydrothermal (Fig. 5), shows intense peak at 27.7°, which corresponds to the rutile crystalline phase, and the peak corresponding to the anatase phase (25.6°) of low intensity. XRD diffraction patterns of TiO₂/HTC₁, TiO₂/HTC₂, TiO₂/HTC₃ and TiO₂/HTC₄ (Fig. 5) show that with an increase in the molar ratio of glucose, i.e. increase of carbon content in the material, peak intensity of the anatase phase increases while peak intensity of the rutile phase decreases. This is in accordance with finding of Zhong et al. (2010), that the higher concentration of glucose may be responsible for the formation of anatase phase.

Close

Figure 5

XRD patterns of obtained samples.

Export to PPT

TiO₂ anatase over rutile ratio was calculated from (Zhao et al., 2010a): $$\begin{array}{cc}\hfill & \text{anatase content}:A(\text{wt}.\%)=\frac{100}{1+1.265·(I_R/I_A)}\hfill \\ \hfill & \text{rutile content}:R(\text{wt}.\%)=100-A(\text{wt}.\%)\hfill \end{array}$$ with I_A and I_R the diffraction intensities of the anatase (1 0 1) and rutile (1 1 0) crystalline phases at 2θ = 25.6° and 27.7°, respectively. The average grain size D (nm) was determined from the XRD pattern according to the Debye–Scherrer equation: $$D=\frac{k\lambda }{\beta \cos \theta }$$ where k is a constant (0.9), λ is the X-ray wavelength (0.15418 nm), β (rad) is the full width at half maximum of the diffraction line, and θ (rad) is the diffraction angle (Tian et al., 2006). The values of β and θ of anatase and rutile were taken from anatase (1 0 1) and rutile (2 1 1) diffraction line, respectively. Anatase and rutile phase content, and grain sizes of examined samples are summarized in Table 2.

The observed increase in anatase phase with carbon content is confirmed by the results presented in Table 2: anatase content increases from 7.6%, for TiO₂ – hydrothermal, to 76.0%, for TiO₂/HTC₄. It can be noted that ratio of anatase and rutile phase obtained for sample TiO₂/HTC₄ is nearly identical with the ratio of anatase and rutile in photocatalytically active Degussa P25. Therefore, it can be assumed that this sample will be the most efficient in photocatalysis. The average grain size (Table 2) decreases with increase in carbon content: from 22.14 nm (TiO₂ – hydrothermal) to 9.76 nm (TiO₂/HTC₄) which is in a good agreement with the results obtained by scanning electron microscopy. This grain size decrease can be the consequence of inhibitory effect of amorphous carbon on grain growth of TiO₂. In addition, transformation of rutile to anatase phase, with carbon content increase, leads to decrease in grain size. Grain size decrease, accompanied with increase in BET surface area (Table 1), is in very good agreement with the values of average pore diameters (Table 1), indicating that pores measured are most likely interparticle spaces.

UV–Vis DRS spectra show the optical properties of different TiO₂/HTC composites. For the purpose of comparison, the UV–Vis DRS spectra of commercial TiO₂ (anatase) and TiO₂ – hydrothermal were also recorded. TiO₂ – hydrothermal (Fig. 6), containing rutile phase, shows the absorption of radiation in the wavelength range up to 440 nm, while the commercial titanium dioxide, which contains anatase phase, absorbs radiation of wavelengths up to 410 nm. On the other hand, TiO₂/HTC composites absorb radiation into the wider area of wavelengths, up to 750 nm. For TiO₂/HTC composites (Fig. 6), sample TiO₂/HTC₄ shows the highest absorption of radiation in the visible light region, while TiO₂/HTC₁ shows the lowest absorption. Observed increase in absorption of radiation with the carbon content in the material (Fig. 6) may be the consequence of the absorption of radiation in the visible light region by the carbon present in the obtained samples (Dong et al., 2009; Ren et al., 2007). Also, the reduction in glucose employed in the hydrothermal process may lead to carbonaceous species embedded in the TiO₂ matrix. This substitutional carbon could narrow the band gap of TiO₂ and make the catalyst to absorb visible light efficiently (Lin et al., 2011; Zheng et al., 2015). Considering that the samples of TiO₂/HTC composites absorb radiation in the wavelength area of 220–750 nm, it can be expected that obtained TiO₂/HTC composites, especially TiO₂/HTC₄, may be photocatalytically active in the visible region.

Close

Figure 6

DRS spectra of samples: TiO₂ – hydrothermal, TiO₂ – commercial and TiO₂/HTC composites.

Export to PPT

FT-IR spectra of examined TiO₂/HTC composites, along with the FT-IR spectra of HTC, TiO₂ – commercial and TiO₂ – hydrothermal, are shown in Fig. 7. For hydrothermal carbon (Fig. 7a), the wide band between 3000 and 3400 cm⁻¹, is assigned to the stretching vibrations of O—H in hydroxyl or carboxyl groups. The bands around 2815–3000 cm⁻¹ are the characteristic stretching vibrations of aliphatic C—H, while the absorption band near 1710 cm⁻¹ is attributed to stretching vibration of —C⚌O from carbonyl and carboxyl groups. The out-of-plane bending vibration of aromatic C—H bands was observed in the region of 750–875 cm⁻¹ (Fig. 7a), and the band at 1620 cm⁻¹ is attributed to the C⚌C vibrations (Chen et al., 2012; Sevilla and Fuertes, 2009), suggesting the aromatization of the material during the hydrothermal treatment. For titanium dioxide (hydrothermal and commercial) (Fig. 7b), the strong intensity of the absorption peak at about 3400 cm⁻¹ may correspond to the stretching vibration of adsorbed water and hydroxyls on the surface of TiO₂ and the peak at about 1630 cm⁻¹ derives from the bending vibration of the O—H bond in hydroxyls and adsorbed water (Shen et al., 2015; Wang et al., 2012; Yu and Shi, 2010). For all TiO₂/HTC composites (Fig. 7c) FT-IR spectra are similar and show a wide hydroxyl stretching vibration mode in carboxyl, phenol and/or intercalated H₂O (ν O—H) at 3400 cm⁻¹, and skeletal vibrations assigned to the components from the oxygen parched carbon domains (δ C⚌C) at 1620 cm⁻¹ (Kalijadis et al., 2015). According to the literature (Dong et al., 2009; Huang et al., 2008), the surface hydroxyl groups can be trapped by the holes generated under UV light irradiation to form hydroxyl radicals which can suppress electron-hole recombination and increase photocatalytic efficiency. Therefore, adsorbed water and hydroxyls represent an important parameter for enhancing the photocatalytic activity (Wang et al., 2012). The highest intensity peaks at 1630 and 3400 cm⁻¹, observed for sample TiO₂/HTC₄ (Fig. 7c), make it the most promising photocatalyst in the process of photocatalytic degradation.

Close

Figure 7

FT-IR spectra of samples: (a) HTC, (b) TiO₂ – commercial, TiO₂ – hydrothermal and (c) TiO₂/HTC composites.

Export to PPT

3.2 Photocatalytic activity

The process of removing MB in the presence of TiO₂/HTC composites (shown in Fig. 8) takes place in two phases. The first phase involves the removal of MB by adsorption in the dark, while the second phase represents photocatalytic degradation of MB under UV irradiation. TiO₂ – commercial showed the highest efficiency in photocatalytic degradation of MB, due to the presence of photocatalytically active anatase phase, while TiO₂ – hydrothermal, which mainly contains photocatalytically inactive rutile phase (Fig. 5), showed the lowest efficiency in degradation of MB. As it can be observed (Fig. 8), for all TiO₂/HTC composites, sample TiO₂/HTC₄ showed the highest adsorption capacity and photocatalytic activity toward MB. In addition, for the shorter time of photocatalytic degradation, sample TiO₂/HTC₄ shows much higher photocatalytic activity than TiO₂ – commercial, and that is of great importance in the case of using these materials for the process of water purification. The increase in adsorption and photocatalytic activity (Fig. 8) with the increase in carbon content is observed for all TiO₂/HTC composites. Carbon species present in TiO₂/HTC composites can provide more active sites and adsorb more reactive species due to large surface area and pore volume, which causes the enhanced photocatalytic activity (Lin et al., 2011). It was reported (Kim et al., 2012; Maletić et al., 2015) that high level of MB adsorption on the carbon carrier surface, elevates the photocatalytic activity of TiO₂/HTC composites due to photocatalytic degradation of more accessible, adsorbed MB. After the degradation, carbon surface is again available for adsorption, which elevates the photocatalytic activity of TiO₂/HTC composites and therefore improves its efficiency in MB removal. As it was already mentioned, increased carbon content in TiO₂/HTC composites leads to higher specific surface area (Table 1) and the higher level of adsorption (Fig. 8), and a direct dependence between porous properties and amount of MB adsorbed in the dark can be demonstrated. Therefore, sample TiO₂/HTC₄ showed superior photocatalytic activity in the overall process of MB removal, due to the highest specific surface area and MB adsorption, as well as the presence of photocatalytically active anatase phase. Additionally, high photocatalytic activity of TiO₂/HTC₄ can be explained by the ratio of anatase and rutile phase (Table 2) which is nearly identical with the ratio of anatase and rutile in photocatalytically active Degussa P25.

Close

Figure 8

Photocatalytic degradation of MB monitored as the concentration decrease versus time in the presence of TiO₂ and TiO₂/HTC composites under UV light.

Export to PPT

Photocatalytic degradation of MB on TiO₂/HTC₄ was presented in UV–Vis absorption spectra of MB aqueous solution under different reaction times (Fig. 9). As shown in Fig. 9, the absorption spectrum of the original solution (t = 0 min) shows four peaks around 290, 323, 660 and 605 nm. During the photocatalytic degradation of MB, absorption peak around 660 nm decreased rapidly with a blueshift of the maximum absorbance peak from 660 to 620 nm. In addition, the absorbance peaks around 605, 323 and 290 nm declined obviously, which indicate that the whole conjugated chromophore structure of MB is destroyed (Yang et al., 2012; Xiong et al., 2012). This confirms the photodegradation of MB in the presence of TiO₂/HTC₄ composite.

Close

Figure 9

UV–Vis absorption spectra of MB aqueous solution under different reaction times.

Export to PPT

The kinetics study of photocatalytic degradation of MB was examined by Langmuir–Hinshelwood kinetic model (Hamad et al., 2016) given by equation: $$\ln \frac{C_0}{C_t}=k_{\mathit{app}}·t$$ where C₀ is the initial concentration of MB, C_t is the concentration of MB at time t and k_app is rate constant. A plot of ln $\frac{C_0}{C_t}$ versus time in Fig. 10 represents a straight line, with the slope of linear regression equals to the apparent first order rate constant k_app. The values of k_app obtained for all tested samples are given in Table 3. With increasing carbon content, the photocatalytic activity of TiO₂/HTC composites increases, and TiO₂/HTC₄ showed the highest activity with the apparent reaction rate constant (k_app, 0.0102 min⁻¹) higher than that of TiO₂ – commercial (k_app, 0.0070 min⁻¹). According to the literature (Dong et al., 2009;Jin et al., 2016; Wang et al., 2015), apparent reaction rate constant, obtained for TiO₂/HTC₄, is comparable, or even higher than that obtained for different carbon modified TiO₂ photocatalysts and Degussa P25.

Close

Figure 10

Photocatalytic activity of hydrothermally synthesized samples and TiO₂-commercial.

Export to PPT

Since adsorption and photocatalytic degradation occur simultaneously, the amount of MB degraded on TiO₂/HTC₄ can be determined by subtracting the amounts of not degraded MB, remained in solution and on the TiO₂/HTC₄ surface after photocatalytic experiment, from the initial amount of MB. For that purpose, MB desorption test was performed after photocatalytic experiment. After adsorption in the dark, sample TiO₂/HTC₄ adsorbed 54.4% of the initial MB amount in the solution (0.5 mg MB) (Fig. 8). After the photocatalytic experiment, 0.021 mg MB was desorbed from the sample surface, while 0.034 mg MB remained in the solution. These results indicate that nearly all of the adsorbed MB was subsequently photocatalytically degraded on the TiO₂/HTC₄ surface, simultaneously releasing the surface active cites for a new adsorption followed by photocatalytic degradation.

The regeneration and recycling of TiO₂ photocatalysts is one of key steps in practical applications of this heterogeneous photocatalysis in water purification. Therefore, an examination of the photocatalytic activity of the recycled TiO₂/HTC₄ was carried out under UV light irradiation. The results are shown in Fig. 11. The degradation rate in first cycle was 91.9%, while after five cycles degradation rate was decreased to 81.2%.

Close

Figure 11

Recycle of the TiO₂/HTC₄ composite for the degradation of MB (initial concentration of MB 10 mg dm⁻³, 50 ml).

Export to PPT

The possibility of using TiO₂/HTC₄ as a photocatalyst in the visible light was tested through the photocatalytic degradation of MB. Obtained results (Fig. 12) showed that UV irradiation induces a higher decrease in MB concentration than the visible light. Nevertheless, concentration of MB was decreased for about 70% in the presence of TiO₂/HTC₄ and visible irradiation. Therefore, it can be said that TiO₂/HTC₄ showed satisfactory photocatalytic activity under the visible light.

Close

Figure 12

MB removal in the presence of TiO₂/HTC₄ under the UV and visible light.

Export to PPT

Obtained results suggest that TiO₂/HTC composites with higher carbon content can be successfully applied in degradation of organic pollutants. Therefore, TiO₂/HTC composites were utilized as a photocatalysts for photocatalytic degradation of pharmaceuticals from the multicomponent solution. Percentage of pharmaceuticals adsorbed in the dark and subsequently photocatalytically degraded in the presence of TiO₂/HTC composites is presented in Table 4. All tested TiO₂/HTC composites showed both very high levels of adsorption and photocatalytic degradation of azithromycin, atorvastatin, amlodipine, simvastatin and clopidogrel. Generally, the level of adsorption and photocatalytic degradation of azithromycin, doxycycline, bromazepam, atorvastatin and clopidogrel increase with the carbon content. On the other hand, it seems that carbon content does not affect the adsorption and photocatalytic degradation of diclofenac, erythromycin and cilazapril. Also, for all tested composites, level of cilazapril adsorption was the lowest, but after UV irradiation about 75% of cilazapril was photocatalytically degraded. This behavior may be the consequence of low affinity toward composite surface and slow adsorption, followed by good photocatalytic degradation of cilazapril. Obtained results indicate superior photocatalytic activity of TiO₂/HTC₄ in pharmaceutical removal from multicomponent solution. The amounts of pharmaceuticals degraded on the surface of TiO₂/HTC₄ were determined in the similar manner as in MB desorption test. The lowest level of subsequent photocatalytic degradation was observed for erythromycin and bromazepam: about 87% of erythromycin and 40% of bromazepam adsorbed on the surface of TiO₂/HTC₄ were subsequently photocatalytically degraded. In the case of diclofenac, cilazapril, simvastatin and clopidogrel, more than 93% of adsorbed pharmaceuticals are photocatalytically degraded. Nevertheless, complete subsequent photocatalytic degradation of pharmaceuticals adsorbed on the surface of TiO₂/HTC₄ was observed for azithromycin, doxycycline, amlodipine and atorvastatin.