Adsorption and photocatalytic degradation behaviors of rhodamine dyes on surface-fluorinated TiO₂ under visible irradiation

Abstract

Surface-fluorinated TiO₂ (F-TiO₂) particles was synthesized by a simple fluorosilanization method to serve as a visible-light photocatalyst for rapid degradation of Rhodamine B (RhB) dyes. The fluroalkylsilane (FAS-17) was covalently immobilized on the TiO₂ particles via robust Si–O bonds. The changes in surface properties of the F-TiO₂ particles were characterized by X-ray photoelectron spectroscopy (XPS), Fourier transform infrared spectroscopy (FTIR), X-ray diffraction (XRD), scanning electron microscope (SEM) and zeta potential measurements. The optical property of the F-TiO₂ particles was determined by UV-visible spectroscopy. Upon the fluorination modification, the zeta potential of TiO₂ particles switched from positive to negative, whilst the optical property of bulk TiO₂ particles remained almost unchanged. Photodegradation experimental results demonstrated that zwitterionic RhB dyes were more favorably adsorbed on F-TiO₂ rather than on the Ti(IV) sites of the pristine TiO₂, and that the photodegradation reaction of RhB on F-TiO₂ proceeded much faster than that on the pristine TiO₂. However, the adsorption and photodegradation rate of anionic methyl orange (MO) dyes did not show an obvious change on F-TiO₂. These results suggested that the molecular structure of dyes played a key role in visible-light photodegradation reactions on F-TiO₂ particles. Positive-charged diethylamine groups of RhB molecular structures were postulated to promote the adsorption of RhB dyes on the surface of F-TiO₂ particles, thus leading to easy electron injection and induction of rapid photodegradation under visible-light illumination.

---

1. Introduction

Industrial development is pervasively connected with a copious amount of various toxic pollutants, which are hazardous to the environment, human health, and difficult to degrade by natural means. Conventional physicochemical and biological treatment methods, such as coagulation, adsorption, filtration, and anaerobic granular sludge, have been the main-stream techniques for the purification of organic compound-contaminated water over the past decades. However, these approaches are less effective to cleanse the anthropogenic organic pollutants at very low concentrations.¹ Photocatalysis, as an environmentally-benign approach, has therefore been attractive to treat persistent low-concentration organic pollutants in recent years.²

TiO₂ is the most extensively-studied photocatalyst in the decontamination of organic pollutants.³ Under light illumination beyond its band-gap limits, TiO₂ shows strong oxidation ability to oxidize almost all organic compounds into CO₂. However, one obstacle toward its practical application is that TiO₂ can only be activated to induce photocatalytic reaction, by ultraviolet (UV) light, which accounts for only about 4% of the incoming sunlight. In view of practical applications, it is difficult to find an alternative photocatalyst as versatile, stable, abundant, environmentally-benign and economical as TiO₂. Consequently, modification of TiO₂ particles has been usually performed to achieve visible-light photosensitivity for effective photodegradation of organic pollutants under solar irradiation.

Various strategies have been proposed to modify TiO₂-based catalysts for enhanced visible light performance.^4–7 The doping (or incorporation) of metallic nano-materials, such as Cu₂O, WO₃, Ag and Au, into TiO₂ lattice has been found effective to improve the photocatalytic activities of TiO₂ in the visible light region. TiO₂ doped with nonmetallic elements, such as N,⁸ S,⁹ C,¹⁰ I,¹¹ Br¹² and Cl,¹³ has also been proven to shift the optical absorption edge of TiO₂ toward lower energy. A facile method to synthesize heterogeneous crystalline TiO₂–halloysite nanotubes with anatase/rutile mixed phase has been described recently. The as-synthesized catalyst was found to have higher photocatalytic degradation of rhodamine B (RhB) and gentian violet than the commercial titania P25.¹⁴ TiO₂ hybridized with a conducting polymer such as polyaniline has also verified to enhance photocatalytic activity towards the decomposition of organic molecules when irradiated with visible light.¹⁵ In a whole, the above doping approaches are aimed to alter the absorption band of bulk TiO₂ to visible region by either narrowing the band-gap of TiO₂ or introducing new absorption band. Most of them focus mainly on the improvement of the visible response of TiO₂ by changing band-gap and other thermodynamic properties, but the kinetic properties are seldom investigated. Heterogeneous photocatalytic reactions primarily take place on the TiO₂ surface, hence surface features of TiO₂ are critical to the photocatalytic efficiencies. However, surface modification of TiO₂ does not shift the band energy to the visible light regions, it usually alters the adsorption, surface acidity, surface charge, and surface functional groups of TiO₂ surfaces, thus leading the change of the photocatalytic activities and mechanisms.¹⁶

In recent years, photosensitization of dyes has been reported to be one of the most effective ways to extend the photoresponse of TiO₂ into the visible region.^17,18 In such a system, the dye pollutants rather than the TiO₂ photocatalyst are subject to the visible light excitation. The dye molecules are excited by visible-light illumination to transfer electrons into the conduction band of TiO₂, thus resulting in the formation of cationic radicals of the dyes. The injected electrons then react with O₂ molecules adsorbed on the surface of TiO₂ to generate a series of active oxygen radicals, such as O₂⁻˙, HO˙, and H₂O₂ species. The subsequent radical chain reactions lead to the degradation and mineralization of the dye pollutants.^19–21 Because the photosensitization of dyes under visible irradiation can be initiated by the interfacial electron injection from the excited dye molecule into the TiO₂ catalyst, more adsorption of dyes may enhance interfacial interactions between dyes and TiO₂ to improve the photosensitization ability. Zhao et al. reported that rapid and more adsorption of dyes on the TiO₂ particles caused the substantial enhancement of the extrinsic adsorption bands, thus leading to the increase in the degradation rate.²²

Among various strategies to modify TiO₂ particles, surface fluorination of TiO₂ has been widely investigated.^23,24 The fluorine-doped or surface fluorinated TiO₂ particles exhibited the enhanced photocatalytic activity under the UV and visible light illumination. Wang et al. reported that surface fluorination not only changed the adsorption modes of dyes on the fluorinated TiO₂, but also caused the dramatic change in the photocatalytic degradation kinetics and mechanisms of the adsorbed dyes.²¹ Surface fluorinated TiO₂ has been also reported to be more effective than pure TiO₂ for the photocatalytic oxidation of acid orange 7 and phenol,²³ as they could enhance both photocatalytic reactions and photoelectrochemical behaviors. In the previous studies, surface fluorinations were also achieved by adding NaF into the dispersions under acidic conditions (pH 3–4), thus resulting in the replacement of surface hydroxyl groups by fluoride anion (F⁻),^23,25 or the dissolution of anatase TiO₂ to etch the surface of rutile TiO₂ by HF.^21,26 The NaF-modified TiO₂ was found to significantly enhance the photocatalytic activities to dyes under UV light irradiation, but its photocatalytic ability was lower than the pristine TiO₂ under visible light irradiation.²³ However, to the best of our knowledge, few researches have been documented for the surface fluorination of TiO₂ particles by a simple fluorosilanization process.

Herein, the aim of this work is to fluorinate TiO₂ particles by fluorosilanization for enhanced photocatalytic degradation of dyes under visible light illumination. The fluorosilane FAS-17 was chosen due to its low surface energy, abundant fluorine atoms and big electronegativity. The surface-fluorinated TiO₂ (F-TiO₂) particles were characterized by X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), scanning electron microscope (SEM) and zeta potential measurements. The change in optical properties of the F-TiO₂ surface was determined by UV-visible spectroscopy. The adsorption of a zwitterionic RhB dye and an anionic MO dye on the F-TiO₂ surface were investigated, and the photodegradation behaviors of these two dyes were determined under visible light illumination. The postulated degradation mechanism of the RhB dye on the F-TiO₂ particles was proposed to interpret the photodegradation behaviors of dyes under visible light illumination.

2. Experimental section

2.1 Materials

Powder anatase TiO₂ was kindly provided by Taihai Co. (Panzhihua City, China). The chemicals, such as rhodamine B (RhB), and methyl orange (MO), rhodamine 101 inner salt (Rh 101 inner), rhodamine 6G (Rh 6G) and solvents, such as ethanol, were obtained from Aladdin Reagent Co. (Shanghai, China) and were used as received without further purifications. 1H,1H,2H,2H-Perfluorodecyltriethoxysilane (FAS-17) was purchased from Sicong Chemicals Co. (Xiamen, China). The deionized water used in the following experiments was purified using an Ultrapure reverse osmosis system (Ultrapure Technol. Co., Chengdu, China), and were used throughout the study. The molecular structures of RhB, Rh 101 inner, Rh 6G, and MO were schematically shown in ESI (Fig. S1†).

2.2 Preparation of surface-fluorinated TiO₂ particles

The surface fluorination of TiO₂ particles with FAS-17 was performed using a typical procedure similar to those described previously.²⁷ A 0.3 mL aliquot of FAS-17 was firstly hydrolyzed in a 5 mL ethanol solution at ambient temperature under continuous stirring for 2 h. A 0.7 g aliquot of TiO₂ particles was ultrasonically dispersed in a 150 mL round flask containing 55 mL of ethanol for 30 min. Subsequently, the FAS-17 mixture was added dropwise into the round flask under continuous stirring. The fluorosilanization of TiO₂ particles was allowed to proceed at 70 °C for 5 h with a reflux condenser. At the end of reaction, the surface-fluorinated TiO₂ (defined as F-TiO₂) particles were harvested by centrifugation at 6000 rpm and ultrasonically washed three times with copious amount of ethanol to remove the physically-adsorbed FAS-17, if any. The as-synthesized F-TiO₂ particles were finally dried in a vacuum oven at 90 °C overnight.

2.3 Surface characterization

The crystalline structure of TiO₂ particles before and after surface fluorination was characterized by X-ray powder diffraction (XRD) (DX-2700, Dandong, China). The voltage and anode current were 40 kV and 30 mA, respectively. The monochromatic Cu Kα (1.54056 Å) and scanning mode with interval of 0.03° and set time of 0.05 s were used to collect the XRD pattern of TiO₂ particles. The change in surface morphology of the F-TiO₂ particles was characterized by a scanning electron microscope (SEM) (S-4800, Hitachi, Japan). The particle size distribution and zeta potential of the TiO₂ particles suspensions (0.1 g L⁻¹) were determined respectively, by dynamic laser scanning Nanoparticle and Zeta potential analyzer (ZEN3690, Malvern, UK), and the pH was adjusted between pH 2.5 and 11.0 using a 1 mol L⁻¹ NaOH and HCl solutions. The optical properties of the pristine TiO₂ and F-TiO₂ particles were characterized by a UV-visible spectrophotometer with an integrating sphere (UV2100, Shimadzu, Japan). The surface composition of the F-TiO₂ was determined by X-ray photoelectron spectroscopy (XPS) (XSAM800, Kratos, UK) with monochromatized Al Kα radiation at constant dwell time of 100 ms and pass energy of 40 eV. The pressure in the analysis chamber was maintained at 10⁻⁷ to 10⁻⁹ Torr during each measurement. To compensate the surface charging effect, all the core-level spectra were referred to the C 1s hydrocarbon peak with binding energy (BE) at 284.6 eV. The photoluminescence spectra (PL) were recorded via a fluorescence spectrometer (F-7000, Hitachi, Japan).

The FTIR spectra of the RhB-adsorbed pristine TiO₂ and F-TiO₂ particles were obtained on a Fourier transform infrared spectrometer (Spectrum II L1600300, PerkinElmer) in the transmission mode. Prior to the measurement, the RhB adsorption process was as follows: the pristine TiO₂ and F-TiO₂ particles were dispersed separately in 16 mg L⁻¹ solution of RhB with highly-speed stirring, and then allowed to adsorb overnight to reach the adsorption equilibrium. The RhB-adsorbed particles were subsequently harvested by centrifugation, and then dried in air at 35 °C. All the operations were carried out in dark to prevent photodegradation of the adsorbed dyes. The TiO₂ particles dispersed in deionized water under the same conditions were used as the control for FTIR measurement.

2.4 Adsorption and visible light photocatalysis activities of modified F-TiO₂ particles

The photocatalytic degradation of RhB and MO was investigated to determine the photocatalytic activities of F-TiO₂ particles under visible light illumination. A photoreactor equipped with a 1000 W xenon lamp was used to perform the photocatalysis experiments (Institute of Electric Light Source, Kaifeng, China). Typically, the xenon lamp was positioned inside a cylindrical Pyrex vessel and surrounded by a circulating water Pyrex jacket, which served to cool the lamp for stable reaction temperature. A cutoff filter was placed outside the Pyrex jacket to secure visible-light irradiation only.

To eliminate the interference caused by adsorption on the photodegradation behaviors of dyes on the pristine TiO₂ and F-TiO₂ particles, the adsorption profiles of zwitterionic RhB and anionic MO dyes were determined prior to the photocatalytic experiments. Typically, a 100 mg aliquot of particles was well dispersed in 10 mL of 16 mg L⁻¹ dye aqueous solution by highly-speed stirring, and the adsorption was allowed to proceed for a predetermined time at ambient temperature. All the operations were performed in the dark to prevent the photocatalytic reactions. The dye concentration after adsorption was determined by a UV-visible spectrophotometer.

In the photocatalytic experiments, the TiO₂ suspension was prepared by adding 100 mg of TiO₂ particles to 10 mL of 16 mg L⁻¹ dye (RhB or MO) aqueous solution, or alternatively adding 150 mg of TiO₂ particles to 30 mL of 10 mg L⁻¹ dye (RhB, Rh 6G, or Rh 101 inner). Prior to visible light irradiation, the suspension was stirred in the dark for 1 h to reach the adsorption/desorption equilibrium on the TiO₂ particle surfaces under ambient air-equilibrated conditions. During the light irradiation process, the suspension was allowed to proceed under vigorous stirring for a predetermined time. At the end of photocatalytic reaction, the suspension was separated by centrifugation at 8000 rpm to remove the solid particles, and the supernatant was collected for further analyses. The concentration of the RhB, MO, Rh 6G and Rh 101 inner solution was determined by UV-visible spectrophotometry (TU-1810, Persee, China) at a wavelength of 553 nm, 465 nm, 526 nm and 575 nm, respectively.

3. Results and discussion

3.1 Surface properties and structure of the F-TiO₂ particles

The surface fluorination of TiO₂ particles was achieved by fluorosilanizing the TiO₂ particles with FAS-17, as schematically illustrated in Fig. 1. The reaction involves the hydrolysis of FAS-17 to convert siloxane groups (Si–OC₂H₅) to silanol groups (–Si–OH) and the fluorosilanization of TiO₂ particles. TiO₂ particles have been widely recognized to possess abundant hydroxyl groups (–OH) on the surface.²⁸ The ligand-exchanging reaction between the surface hydroxyl groups (–OH) of TiO₂ and Si–OH of FAS-17 results in the formation of robust Si–O–Ti bonds.^29,30 Meanwhile, the dehydration condensation reaction among the –Si–OH groups leads to the formation of polysiloxanes. The surface characteristics, crystalline structure and size distribution of the as-synthesized F-TiO₂ particles were characterized by SEM, XPS, XRD, UV-visible and zeta potential measurements. The detailed results are discussed as follows.

Fig. 1 Schematic illustration of the fluorosilanization process of FAS-17 onto the TiO₂ particles surfaces.

3.1.1 SEM imaging and size distribution. The change in surface morphologies of the TiO₂ particles before and after surface fluorination was characterized by SEM imaging. Fig. 2 shows the respective SEM image and size distribution diagrams of the pristine TiO₂ and the F-TiO₂ particles. For pristine TiO₂ particles, a disordered porous structure is observed and the particle surface is smooth and angular (Fig. 2a). No obvious change in surface morphology can be observed on the F-TiO₂ particles, albeit that the surface of TiO₂ particles becomes rough due to the deposition of silane layers (Fig. 2c). Moreover, some particles appear to stick together due to the silanization treatment, thus leading to the increase in TiO₂ particle size. This result is further ascertained by the size distribution of the pristine TiO₂ and F-TiO₂ particles as shown in Fig. 2b and d. The average diameter of pristine TiO₂ is 278.1 ± 3 nm, while the average diameter of F-TiO₂ increases to 318.1 ± 3 nm (Table 1) upon the surface fluorination. The noticeable increase in the average diameters of F-TiO₂ particles is caused by both the deposition of FAS-17 layer and the aggregation of some TiO₂ particles.

Fig. 2 The SEM images and size distribution diagrams of (a and b) the pristine anatase TiO₂ and (c and d) the F-TiO₂ particles.

Table 1 The average particle size and zeta potential of TiO₂ particles before and after surface fluorination

	Anatase	F-TiO2
a Zeta potentials of TiO2 particles were obtained at pH 6.5.
Zeta potential values^a (mV)	7.2 ± 1	−21.1 ± 2
Particle sizes (nm)	278.1 ± 3	318.1 ± 3

---

3.1.2 FTIR and XPS spectra. The changes in surface chemistry of the TiO₂ particles before and after surface fluorination were ascertained by the FTIR measurement. Fig. S2† (ESI) shows the respective FTIR spectra of the pristine TiO₂ and F-TiO₂ particles. The peak centered at a wavenumber of 3400 cm⁻¹ (ESI, Fig. S2a†), attributable to the stretching vibrations of hydroxyl groups (–OH), confirms the presence of reactive –OH on the pristine TiO₂ surfaces. The broad peak, centered at a wavenumber of around 1632 cm⁻¹ is associated with the molecularly-adsorbed water (H–O–H bending vibrations) (ESI, Fig. S2b†).^31–33 The successful fluorination of TiO₂ surfaces can be deduced from the appearance of three additional bands with wavenumber at 1153, 1206, and 1247 cm⁻¹, attributable to the stretching vibration of C–F species (ESI, Fig. S2b†). These C–F species are characteristic peaks of fluorosilane FAS-17.

The surface composition of the FAS-17-modified TiO₂ was also characterized by XPS. Fig. 3 shows the wide scan, C 1s and F 1s core-level XPS spectra of the pristine TiO₂ and F-TiO₂ particles, respectively. In comparison with the wide scan XPS spectrum of the pristine TiO₂ surface (Fig. 3a), four additional photoelectron lines with binding energies (BEs) at 99, 151, 685 and 832 eV, attributable to Si 2p, Si 2s, F 1s and F KLL species,³⁴ respectively, are discernible in the wide scan spectrum of the F-TiO₂ particle surface (Fig. 3a), indicative of the successfully immobilization of FAS-17 layer onto the TiO₂ surface. The presence of FAS-17 layer on the TiO₂ surface can also be deduced from the curve-fitted C 1s core-level spectrum, which consists of six peak components with BEs at 283.1, 284.8, 285.6, 289.4, 291.4 and 293.8 eV, attributable to C–Si, C–H, C–CF_x, CF–CF_x, C–F₂ and C–F₃ species,³⁵ respectively (Fig. 3b), as well as the appearance of the F 1s core-level spectrum with BE at 689 eV (inset of Fig. 3b). The fluorine peak components of C–CF_x, CF–CF_x, C–F₂ and C–F₃ species are characteristics of the FAS-17 molecules.²⁷ Thus, the surface fluorination of TiO₂ particles has been successfully synthesized by covalently immobilizing of FAS-17.

Fig. 3 (a) Wide scan XPS spectra of the pristine TiO₂ (red line) and F-TiO₂ (black line) surfaces (b) the C 1s and F 1s core-level XPS spectra of the F-TiO₂ particle surface.

3.1.3 XRD patterns. The change in crystalline structure of the TiO₂ particles before and after surface fluorination was characterized by XRD (ESI, Fig. S3†). Despite of the coverage of fluorosilane layers on the TiO₂ particle surfaces, no significant shift in peak position of TiO₂ phase is observed from the XRD pattern of the F-TiO₂. This result is consistent with the fact that the FAS-17 layers do not be embedded into the TiO₂ lattices during surface fluorination. The persistence of the crystalline structure of TiO₂ particles after fluorosilanization is in good agreement with the previous findings that the surface fluorination, unlike the fluorine doping of TiO₂, cause no significant alternation in crystalline structures of TiO₂ particles.^24,36

3.1.4 UV-visible spectra. The change in optical properties of the TiO₂ particles before and after modification was determined by UV-visible spectroscopy (ESI, Fig. S4†). It is evident that no significant shift in the fundamental absorption edge of TiO₂ (about 3.31 eV) is caused by the surface fluorination modification. This result is in good agreement with the previous conclusions that the surface fluorination modification has no effect the optical absorption edge of TiO₂.²³ Particularly, even the F-doped TiO₂ particles cannot cause the alternation of the optical absorption edge of TiO₂ particles, as reported previously by Yamaki et al.³⁷

3.1.5 Zeta potential. The surface charge of the TiO₂ particles before and after surface fluorination was characterized by zeta potential measurement, and the results are summarized in Table 1. The zeta potential of TiO₂ particles at different pH conditions was measured to determine the effect of pH values on the surface charge of TiO₂ particles. The change in zeta potential of the pristine TiO₂ and F-TiO₂ particles as a function of pH values is shown in ESI Fig. S5.† The zeta potentials of both pristine TiO₂ and F-TiO₂ particles show an evident decrease with the increase in pH values. The point of zero charge (pzc) of the pristine TiO₂ is estimated to be about pH 7.8, while the pzc value of the F-TiO₂ particles shifts to around pH 4.7. This result indicates that the surface charge of TiO₂ particles is substantially influenced by surface fluorination. As shown in Table 1, at pH 6.5, the zeta potential of TiO₂ particles undergoes a dramatic change from positive to negative values after the surface fluorinated modification. The pristine anatase TiO₂ particles show a positively-charged surface with the averaged zeta potential value of 7.2 ± 1 mV, while the F-TiO₂ particles have a negatively-charged surface with zeta potential values of −21.1 ± 2 mV. The opposite change in the surface charge of the F-TiO₂ particles is probably associated with the presence of abundant fluorine atoms of the FAS-17. It has been reported that fluorine can effectively lower the surface free energy due to its small atomic radius and being as the biggest electronegativity among atoms.²⁷

3.2 Photocatalytic activities of the modified F-TiO₂ under visible light irradiation

3.2.1 Adsorption behaviors of dyes on the modified F-TiO₂ particles. In general, the observed photocatalytic activity of a photocatalyst is a combined effect of many factors, such as surface area, phase structure, crystallinity, surface hydroxyl density, and surface charge.^38,39 Pre-adsorption is indispensable for dyes to be photodegraded on the TiO₂ particle surface, since rapid adsorption of dyes facilitates electron injection. To further elucidate the effect of surface fluorination on the dye adsorption, the adsorption behaviors of zwitterionic RhB and anionic MO dyes were investigated for the pristine TiO₂ and the F-TiO₂ particles under continuous stirring in the dark. Fig. 4 shows the adsorption profiles of RhB and MO on the pristine TiO₂ and the F-TiO₂ particles as a function of adsorption time, respectively. The pristine TiO₂ shows no adsorption to both dyes in the dark, as the concentration of the RhB and MO solution remained almost unchanged. However, the F-TiO₂ particles show different adsorption behaviors to RhB and MO dyes. The pre-adsorption of RhB on the F-TiO₂ particles increase rapidly to about 60%, and reach adsorption equilibrium afterwards (Fig. 4b), while no pre-adsorption of MO on the F-TiO₂ particles is observed (Fig. 4b). This phenomenon is probably associated with the electrostatic interaction between the F-TiO₂ and zwitterionic RhB dyes. After fluorination, the TiO₂ particles become more electronegative, and possess the negative surface charge at pH 6.5. Therefore, the negatively-charged F-TiO₂ particle facilitates the adsorption of positively-charged diethylamine group in the RhB molecules owing to electrostatic attraction, while it hinders the adsorption of anionic MO dye due to the electrostatic repulsion.

Fig. 4 The adsorption profiles of (a) zwitterionic RhB and anionic MO on the pristine TiO₂ and (b) F-TiO₂ particles as a function of time in the dark. Adsorption conditions: initial dye concentration (C₀) of 16 mg L⁻¹, temperature: at room temperature.

3.2.2 Photodegradation activities of dyes on the modified F-TiO₂ particles. After the adsorption equilibrium of RhB with TiO₂ and F-TiO₂ had been achieved, the photocatalytic decomposition experiment was performed under irradiation with visible light to evaluate the photocatalytic ability of F-TiO₂ particles. Fig. 5 shows the photodegradation curves of RhB without and with the presence of TiO₂ particles before and after subtracting the adsorbed amount. The RhB dyes remain unchanged under visible light illustration in the absence of TiO₂ particles over 120 min (Fig. 5a), while the degradation ratio of the RhB dyes by the pristine TiO₂ particles is about 65% (Fig. 5b). On the other hand, the enhancement in photocatalytic degradation of RhB under the visible-light irradiation can be observed on the F-TiO₂ particles, since the photodegradation ratio of the RhB dyes by the F-TiO₂ particles reach as high as more than 99% after 120 min (Fig. 5b). Moreover, no degradation of the RhB dye solution containing FAS-17 is observed under the visible light illumination, indicative of no visible-light photocatalytic properties of the FAS-17 molecules. Similarly, no degradation of RhB dyes takes place in the dark in the presence of the F-TiO₂ particles. The above results are consistent with the fact that the visible-light photocatalytic activities are significantly enhanced by the F-TiO₂ particles.

Fig. 5 (a) The photocatalytic degradation curves of RhB without and with the presence of TiO₂ particles as a function of time under visible light illumination, (b) the photodegradation curve of RhB after subtracting the adsorbed amount with the presence of TiO₂ and F-TiO₂ particles. Reaction conditions: (a) initial dye concentration (C₀) of 16 mg L⁻¹, catalyst dosage of 10 g L⁻¹, (b) initial dye concentration (C₀) of 10 mg L⁻¹, catalyst dosage of 5 g L⁻¹, light intensity: 1000 W xenon lights.

To investigate the role of surface fluorination in affecting the photodegradation kinetics and mechanisms of dyes, the photodegradation of anionic MO dyes were also examined, and the photocatalytic degradation behaviors of the anionic MO dyes for the pristine TiO₂ and F-TiO₂ particles under visible-light irradiation are show in Fig. 6. The MO dyes remains unchanged throughout under visible-light irradiation in the absence of TiO₂ particles. Particularly, the MO dyes also remain relatively stable in the presence of photocatalysts, since the photodegradation ratios of the MO dyes by the pristine TiO₂ and the F-TiO₂ are only about 8% and about 11%, respectively, after 120 min of visible-light irradiation. These results suggest that the surface fluorination of TiO₂ particles cannot improve the visible-light photocatalytic activity of the anionic MO dyes. Taking together, the surface fluorination of TiO₂ particles can only lead to the selectively-enhanced photocatalytic activity to the zwitterionic RhB dyes rather than to the anionic MO dyes. This phenomenon is possibly ascribed to the enhanced pre-adsorption of RhB and the intrinsic sensitivity of RhB to visible light illumination for the F-TiO₂ particles. It has been reported that the adsorption behaviors of cationic dyes on the catalyst surface play an important role in improving their photocatalytic degradation rate.²²

Fig. 6 The photocatalytic degradation curves of MO with or without the presence of TiO₂ particles as a function of time under visible light illumination. Reaction conditions: initial dye concentration (C₀) of 16 mg L⁻¹, catalyst dosage of 10 g L⁻¹, light intensity: 1000 W xenon lights.

The pre-adsorption of the zwitterionic RhB dyes on the F-TiO₂ particles rapidly reaches as high as 60% prior to the visible light irradiation (Fig. 4 and 5a), while the anionic MO dyes show almost no pre-adsorption on the F-TiO₂ particles. As a consequence, the enhancement in visible-light photocatalytic degradation of the zwitterionic RhB dyes is probably ascribed to, the enhanced pre-adsorption and the intrinsic visible-light sensitivity of zwitterionic RhB dyes, as compared to the anionic MO dyes. To further confirm the effect of dye types and pre-adsorption of dyes on the photocatalytic ability, other two cationic dyes such as Rh 6G and Rh 101 inner were chosen to perform the photocatalytic degradation experiments by the pristine TiO₂ and F-TiO₂ particles. Similar to that of the zwitterionic RhB, the photocatalytic degradation ratios of the two cationic dyes by F-TiO₂ particles show a significant increase under visible light irradiation, and both reach higher than 95% (ESI, Fig. S6†). These above results are in good agreement with the previous findings that the dye structure and their anchoring groups on the photocatalyst are key factors to determine the interfacial electron transfer and the degradation ratio of the dye.¹⁹

3.2.3 The degradation of adsorbed dyes characterized by FTIR analyses. As about 60% of the RhB is adsorbed on the F-TiO₂ particles, it is of great importance to verify the degradation of the RhB adsorbed on the F-TiO₂ surfaces. Despite that the pink-colored F-TiO₂ particles has been turned to white after light irradiation, the RhB-adsorbed F-TiO₂ were further characterized by FTIR and UV-visible spectra to ascertain the photodegradation of RhB dyes on the F-TiO₂ surface. Fig. 7 shows the respective FTIR spectra of the F-TiO₂, RhB-adsorbed F-TiO₂ before and after visible-light illumination. By comparing the FTIR spectra of the three samples, the RhB dyes adsorbed on the F-TiO₂ surface are confirmed to be completely photo-degraded. As shown in Fig. 7b, upon the adsorption of RhB dyes, two additional peaks with wavenumber at 1183 and 1340 cm⁻¹, attributable to the stretching vibrations of the C–O–C and the C-aryl groups, respectively,⁴⁰ appear on the FTIR spectra of the particles. The broad peak, centered at a wavenumber of around 1632 cm⁻¹, is attributed to the molecularly adsorbed water (H–O–H bending vibrations),³¹ which is overlap with the C–N bond at a wavenumber of 1649 cm⁻¹ (Fig. 7b).²¹ However, these characteristic bands of RhB disappear in the FTIR spectrum of the RhB-adsorbed F-TiO₂ particles after visible-light irradiation (Fig. 7c). The FTIR results reveal that the RhB dyes adsorbed on the F-TiO₂ particle surfaces can be completely photodegraded under light irradiation.

Fig. 7 FTIR spectra of the (a) as-synthesized F-TiO₂, (b) F-TiO₂ adsorbed with RhB, and (c) F-TiO₂ after degradation of the adsorbed RhB.

3.2.4 Determination of the stability of the F-TiO₂ particles. To determine whether the release of fluorine ions can release from the F-TiO₂ particles during the photocatalytic processes, the stability of the fluorosilane layers was evaluated by the FTIR and UV-vis spectra of the F-TiO₂ particles after visible-light radiation. As shown in Fig. 7c, the characteristic bands with wavenumber at 1153, 1206, and 1247 cm⁻¹, attributable to the stretching vibration of C–F species, are still available, and the peak intensity also appears to remain unchanged as compared to the F-TiO₂ particles before visible-light illustration. These results suggest the good stability of the F-TiO₂ particles after light irradiation, and probably no fluorine release from the fluorosilane layers. To further verify the good stability of F-TiO₂ particles, the optical properties of the F-TiO₂ particles after visible-light irradiation were also measured by the UV-vis spectra (ESI, Fig. S7†). No evident difference of the UV-vis spectra can be distinguished from the visible-light irradiated F-TiO₂ particles as compared to the as-synthesized F-TiO₂ particles (ESI, Fig. S4†). The fundamental absorption edge of F-TiO₂ after irradiation still remains at about 3.31 eV.

3.2.5 Recyclability of the F-TiO₂ particles. To determine the recyclability of the F-TiO₂ particles, the used F-TiO₂ particles were harvested and recycled to the photocatalytic degradation of RhB under visible light irradiation. The photocatalytic degradation curves of the RhB in the presence of recycled TiO₂ and F-TiO₂ particles are shown in ESI Fig. S8† (second cycle). Surprisingly, both the recycled TiO₂ and F-TiO₂ particles do not show any decrease in the photocatalytic ability to the RhB dyes, and even cause a slight increase in the photodegradation ratio of RhB dyes. The photocatalytic degradation ratio of the RhB dyes by the recycled TiO₂ particles is about 65%, while the photodegradation ratio of RhB by the recycled F-TiO₂ particles reaches as high as more than 90% after 80 min of visible-light irradiation (ESI, Fig. S8†). These results confirm the good recyclability of the F-TiO₂ particles, and are consistent with the above FTIR data.

3.3 The postulated visible-light photocatalytic mechanisms of RhB dyes on the F-TiO₂ particles

Under visible light irradiation, the degradation of zwitterionic RhB dyes is substantially enhanced in the presence of F-TiO₂ particles, while the degradation of anionic MO dyes shows no significant change. Therefore, it can be speculated that the enhanced photodegradation of RhB dyes is probably not caused by the intrinsic change of F-TiO₂ particles, but is more possibly associated with dye sensitization. As liquid–solid heterogeneous photocatalysis is a well-known interface-based process, the contact of dye molecules with the surface of catalyst is prerequisite to efficient electron transfer. The enhanced adsorption of dyes can also eliminate the outer diffusion control in surface reaction. Therefore, the enhanced adsorption of photosensitive dyes will partly contribute to the high photocatalytic activity of F-TiO₂. Take together, a postulated mechanism of the enhanced visible-light photocatalytic activity of RhB on the F-TiO₂ particles has been proposed, and is schematically illustrated in Fig. 8. The positively-charged TiO₂ particle surfaces show the electrostatic repulsion to resist the adsorption of RhB dyes (Fig. 8a), while the negatively-charged F-TiO₂ surfaces show strong affinity to attract RhB dyes (Fig. 8b). According to the adsorption results (Fig. 4), the RhB dyes can rapidly adsorb onto the F-TiO₂ surface to reach dynamic adsorption/desorption equilibrium. Thus, the pre-adsorption of RhB is an initial and important step to enhance visible-light photocatalytic degradation on the F-TiO₂ particle surfaces.

Fig. 8 Schematic illustration of (a and b) the adsorption model of the cationic RhB dyes on the pristine TiO₂ and F-TiO₂ particle surface, and (c) electron-transfer processes and subsequent excitation of RhB dye on the F-TiO₂ particle surfaces.

Fig. 8c schematically illustrates the electron-transfer processes and subsequent excitation of RhB dye on the F-TiO₂ particle surfaces. As a matter of fact, the TiO₂ particles may be not excited as its absorption threshold is 385 nm under visible-light illumination, however, the adsorbed RhB on the TiO₂ surface can be excited at a wavelength longer than 470 nm to produce singlet and triplet states (denoted as RhB^*_ads).²² The RhB^*_ads subsequently injects an electron into the conduction band (or to some surface states) of the TiO₂ particles to convert RhB to the radical cation RhB˙⁺. In turn, the injected electron on the TiO₂ particles, TiO₂ (e⁻), can react with adsorbed oxidants (usually O₂) to produce reactive oxygen radicals. The radical cation RhB˙⁺ ultimately reacts with reactive oxygen radicals and/or molecular oxygen to yield intermediate products or other radical species, for which might lead to mineralization if secondary radical processes occurred.⁴¹ As a consequence, the pre-adsorption of cationic RhB dyes and activation of the visible-light sensitive RhB dyes to inject electrons to TiO₂ particles are important steps to enhance the visible-light photodegradation activities of the TiO₂ particles.

To confirm the above postulated mechanism of the activation of pre-adsorbed RhB dyes to inject electron to TiO₂ particles, the photoluminescence (PL) signals was obtained on a fluorescence spectrometer. The PL signals of semiconductor materials result from the recombination of photo-induced charge carriers. Generally, the lower PL intensity, the lower the recombination rate of photo-induced electron–hole pairs, and the higher the photocatalytic activity of semiconductor photocatalysts.⁴² Fig. 9 shows the PL emission spectra of pure TiO₂, F-TiO₂ and F-TiO₂ adsorbed with RhB in the wavelength range 300–700 nm. The intensity of the PL spectrum of pure TiO₂ is much higher than that of the F-TiO₂ and F-TiO₂/RhB samples. The intensity of F-TiO₂/RhB is the lowest, it can be reasonably inferred that the enhanced adsorption of RhB dye should have a higher photocatalytic performance.

Fig. 9 The PL spectrum of the pristine anatase TiO₂ (black line), the F-TiO₂ (red line) and the RhB-adsorbed F-TiO₂ (blue line). The adsorption experimental conditions: initial dye concentration (C₀) of 100 mg L⁻¹, catalyst dosage of 10 g L⁻¹.

4. Conclusions

With the objective to enhance the visible-light photocatalytic degradation activities of TiO₂ to organic pollutants, a facile wet fluorosilanization approach was described to modify the anatase TiO₂ particles. Various surface characterization techniques, including XPS, XRD, zeta potential, SEM, FTIR and UV-vis spectroscopy, were utilized to determine the surface characteristics change after the surface fluorination. The fluorination modification of TiO₂ surfaces caused no significant changes in the surface morphology, crystalline phase, and optical properties, but the dramatic alternation of surface charges from positive to negative values. As a result, both the adsorption modes and degradation of the dyes were greatly altered. Adsorption experimental results revealed that the F-TiO₂ particles had a high affinity to attract zwitterionic RhB dyes instead of anionic MO dyes due to the electrostatic interaction. The postulate mechanisms were proposed to interpret the electron transfer from the excited dyes to TiO₂ particles rather than the intrinsic electron and holes of TiO₂ particles.

Acknowledgements

The authors would like to acknowledge the financial assistance of key project of National Natural Science Foundation of China (No. 21236004).

References

1. C. Chen, W. Ma and J. Zhao, Chem. Soc. Rev., 2010, 39, 4206–4219.
2. K. Demeestere, J. Dewulf and H. van Langenhove, Crit. Rev. Environ. Sci. Technol., 2007, 37, 489–538.
3. A. Fujishima, T. N. Rao and D. A. Tryk, J. Photochem. Photobiol., C, 2000, 1, 1–21.
4. D. Chatterjee and A. Mahata, J. Photochem. Photobiol., A, 2004, 165, 19–23.
5. J. Choi, H. Park and M. R. Hoffmann, J. Phys. Chem. C, 2009, 114, 783–792.
6. J. Lin, R. Zong, M. Zhou and Y. Zhu, Appl. Catal., B, 2009, 89, 425–431.
7. B. Muktha, D. Mahanta, S. Patil and G. Madras, J. Solid State Chem., 2007, 180, 2986–2989.
8. R. Nakamura, T. Tanaka and Y. Nakato, J. Phys. Chem. B, 2004, 108, 10617–10620.
9. T. Ohno, M. Akiyoshi, T. Umebayashi, K. Asai, T. Mitsui and M. Matsumura, Appl. Catal., A, 2004, 265, 115–121.
10. S. U. Khan, M. Al-Shahry and W. B. Ingler, Science, 2002, 297, 2243–2245.
11. X. Hong, Z. Wang, W. Cai, F. Lu, J. Zhang, Y. Yang, N. Ma and Y. Liu, Chem. Mater., 2005, 17, 1548–1552.
12. H. Irie, S. Washizuka, N. Yoshino and K. Hashimoto, Chem. Commun., 2003, 1298–1299.
13. S. Yamazaki, T. Tanimura, A. Yoshida and K. Hori, J. Phys. Chem. A, 2004, 108, 5183–5188.
14. C. Li, J. Wang, S. Feng, Z. Yang and S. Ding, J. Mater. Chem. A, 2013, 1, 8045–8054.
15. C. Li, J. Wang, H. Guo and S. Ding, J. Colloid Interface Sci., 2015, 458, 1–13.
16. X. Quan, J. Niu, S. Chen, J. Chen, Y. Zhao and F. Yang, Chemosphere, 2003, 52, 1749–1755.
17. Y. Cho, W. Choi, C.-H. Lee, T. Hyeon and H.-I. Lee, Environ. Sci. Technol., 2001, 35, 966–970.
18. T. Wu, T. Lin, J. Zhao, H. Hidaka and N. Serpone, Environ. Sci. Technol., 1999, 33, 1379–1387.
19. D. Zhao, C. Chen, Y. Wang, W. Ma, J. Zhao, T. Rajh and L. Zang, Environ. Sci. Technol., 2007, 42, 308–314.
20. P. Chowdhury, J. Moreira, H. Gomaa and A. K. Ray, Ind. Eng. Chem. Res., 2012, 51, 4523–4532.
21. Q. Wang, C. Chen, D. Zhao, W. Ma and J. Zhao, Langmuir, 2008, 24, 7338–7345.
22. J. Zhao, T. Wu, K. Wu, K. Oikawa, H. Hidaka and N. Serpone, Environ. Sci. Technol., 1998, 32, 2394–2400.
23. H. Park and W. Choi, J. Phys. Chem. B, 2004, 108, 4086–4093.
24. J. Tang, H. Quan and J. Ye, Chem. Mater., 2007, 19, 116–122.
25. J. Lee, W. Choi and J. Yoon, Environ. Sci. Technol., 2005, 39, 6800–6807.
26. R. Nakamura, T. Okamura, N. Ohashi, A. Imanishi and Y. Nakato, J. Am. Chem. Soc., 2005, 127, 12975–12983.
27. C. Kang, H. Lu, S. Yuan, D. Hong, K. Yan and B. Liang, Chem. Eng. J., 2012, 203, 1–8.
28. L. G. Bach, M. Islam, S. Y. Seo and K. T. Lim, J. Appl. Polym. Sci., 2013, 127, 261–269.
29. A. Y. Fadeev and T. J. McCarthy, Langmuir, 2000, 16, 7268–7274.
30. D. Schondelmaier, S. Cramm, R. Klingeler, J. Morenzin, C. Zilkens and W. Eberhardt, Langmuir, 2002, 18, 6242–6245.
31. W.-C. Hung, S.-H. Fu, J.-J. Tseng, H. Chu and T.-H. Ko, Chemosphere, 2007, 66, 2142–2151.
32. S. Sivakumar, P. K. Pillai, P. Mukundan and K. Warrier, Mater. Lett., 2002, 57, 330–335.
33. S. Musić, M. Gotić, M. Ivanda, S. Popović, A. Turković, R. Trojko, A. Sekulić and K. Furić, J. Mater. Sci. Eng. B, 1997, 47, 33–40.
34. J. F. M. C. D. Wagner, J. E. Davis and W. M. Riggs, Handbook of X-ray Photoelectron Spectroscopy, Perkin-Elmer Corp., Eden Prairie, MN, USA, 1992.
35. H. Liu, L. Pan, K. Shen, J. Lang, J. Shi, Q. Cui, H. Li and C. Liu, Fertil. Steril., 2009, 92, 1150–1152.
36. D. Li, H. Haneda, N. K. Labhsetwar, S. Hishita and N. Ohashi, Chem. Phys. Lett., 2005, 401, 579–584.
37. T. Yamaki, T. Umebayashi, T. Sumita, S. Yamamoto, M. Maekawa, A. Kawasuso and H. Itoh, Nucl. Instrum. Methods Phys. Res., Sect. B, 2003, 206, 254–258.
38. D. Li, H. Haneda, N. Ohashi, S. Hishita and Y. Yoshikawa, Catal. Today, 2004, 93, 895–901.
39. D. Li, H. Haneda, N. K. Labhsetwar, S. Hishita and N. Ohashi, Chem. Phys. Lett., 2005, 401, 579–584.
40. N. Mchedlov-Petrossyan, S. Shapovalov, S. Egorova, V. Kleshchevnikova and E. A. Cordova, Dyes Pigm., 1995, 28, 7–18.
41. R. W. Matthews, Water Res., 1991, 25, 1169–1176.
42. J.-G. Yu, H.-G. Yu, B. Cheng, X.-J. Zhao, J. C. Yu and W.-K. Ho, J. Phys. Chem. B, 2003, 107, 13871–13879.

---

Footnote

† Electronic supplementary information (ESI) available: Structures of all the dyes used (Fig. S1), FTIR spectra (Fig. S2), X-ray diffraction patterns (Fig. S3), the forbidden bandwidth of the pristine TiO₂ and F-TiO₂ particles (Fig. S4), the zeta potentials of the pristine TiO₂ and F-TiO₂ particles (Fig. S5), the photodegradation curves of Rh 6G and Rh 101 inner salt (Fig. S6), the forbidden bandwidth of the F-TiO₂ after photodegradation (Fig. S7) and the photodegradation curves of the RhB dyes in the presence of recycled TiO₂ and F-TiO₂ particles (Fig. S8). See DOI: 10.1039/c5ra14379a

---