Improving the broadband photocatalytic performance of TiO₂ through a highly efficient optical converter

Abstract

Developing efficient photocatalysts that respond to near-infrared (NIR) light holds significant scientific importance for improving the utilization of solar energy. In this study, the photocatalyst β-NaLuF₄:Yb³⁺/Tm³⁺@TiO₂ that simultaneously responds to both ultraviolet (UV) and NIR light is successfully synthesized. Instead of the conventional use of β-NaYF₄:Yb³⁺/Tm³⁺, β-NaLuF₄:Yb³⁺/Tm³⁺ upconversion (UC) nanoparticles (NPs) are employed as the NIR-UV light converter to achieve stronger UV UC luminescence. The data indicate that the photocatalytic degradation efficiency and rate of β-NaLuF₄:Yb³⁺/Tm³⁺@TiO₂ synthesized in this work are superior to those of NIR-driven photocatalysts based on β-NaYF₄:Yb³⁺/Tm³⁺ under both NIR and sunlight irradiation. Further analysis of the UC emission spectra and the decay curves of the samples reveals an effective fluorescence resonance energy transfer (FRET) process between β-NaLuF₄:Yb³⁺/Tm³⁺ and TiO₂ after the combination, which makes the NIR-driven photocatalytic reactions happen. All results confirm that the present β-NaLuF₄:Yb³⁺/Tm³⁺@TiO₂ is an exceptionally efficient photocatalyst, owing to its responsiveness to both UV and NIR light.

---

Introduction

Environmental pollution and energy shortages have become two major challenges that urgently need to be addressed in today's world. Under solar irradiation, photocatalytic materials can not only decompose pollutants but also achieve water splitting for hydrogen production, making it one of the best solutions to tackle the above two problems simultaneously.^1–6 Therefore, research on high-performance photocatalysts holds significant practical value for the sustainable development of humanity.^7–11 Currently, the most widely used photocatalyst, TiO₂, possesses advantages such as strong chemical stability, non-toxicity, low cost and so on. However, the bandgap width of TiO₂ (3.2 eV) restricts its ability to utilize near-infrared (NIR) light for efficient photocatalytic reactions, which accounts for approximately 46% of solar radiation, resulting in a huge loss of solar power.^12–17

Rare earth ion-doped upconversion (UC) nanoparticles (NPs) possess the unique ability to transform NIR light into either ultraviolet (UV) or visible (Vis) light. This transforming ability enables their integration with photocatalysts to realize photocatalytic reactions under NIR illumination.^18,19 For instance, Qin et al. reported a composite material of NaYF₄:Yb³⁺/Tm³⁺ and TiO₂, in which an effective fluorescence resonance energy transfer (FRET) process occurs between the UV emission level ¹I₆ and ¹D₂ of Tm³⁺ and TiO₂.²⁰ Then, the NIR-driven photocatalysis process is carried out based on this FRET process. Similar reports include β-NaYF₄:Yb³⁺/Tm³⁺@ZnO, β-NaYF₄:Yb³⁺/Tm³⁺@CdSe, β-NaYF₄:Yb³⁺/Tm³⁺@Ag₃PO₄ and so on.^21–27

As mentioned above, in the composite structures of UC materials and photocatalysts reported so far, β-NaYF₄:Yb³⁺/Tm³⁺ is usually the preferred converter for NIR to UV light due to the high efficiency of β-NaYF₄ in the UC process. However, β-NaLuF₄, which shares a similar crystal cell structure and phonon energy with β-NaYF₄, is also an extremely efficient UC matrix. Moreover, according to the intensity borrowing mechanism proposed by O. Guillot Noel et al., the unique electronic states at the valence band top of Lu-based compounds increase the vibrational intensity of the doped rare earth ions in the crystal.²⁸ Therefore, the UC transition probability of the luminescent centers in Lu-based compounds is often higher than that of the corresponding Y-based compounds, giving rise to much stronger UC intensity appearing in Lu-based compounds. The validity of this theory has been substantiated by numerous reports.^29,30 Nevertheless, currently, there are few reports on using β-NaLuF₄:Yb³⁺/Tm³⁺ as a conversion agent for the NIR photocatalytic reaction.

Herein, β-NaLuF₄:Yb³⁺/Tm³⁺ is selected as a NIR-UV light converter to combine with TiO₂ for the aim of developing a high-performance NIR-driven photocatalyst. The FRET processes between β-NaLuF₄:Yb³⁺/Tm³⁺ and TiO₂ are thoroughly investigated using both steady-state and transient spectroscopy techniques. Additionally, rhodamine B (RhB) is employed as the mimical pollutant to evaluate the photocatalytic performance of β-NaLuF₄:Yb³⁺/Tm³⁺@TiO₂ under NIR light and sunlight irradiation, in which the results indicate that the sample developed in our work exhibits a much higher photodegradation efficiency and rate compared to the reported NIR-driven photocatalysts formed by β-NaYF₄:Yb³⁺/Tm³⁺. Furthermore, a detailed study of the photoelectrochemical properties of the samples is conducted to elucidate the photocatalytic mechanism of β-NaLuF₄:Yb³⁺/Tm³⁺@TiO₂. The findings reveal that β-NaLuF₄:Yb³⁺/Tm³⁺@TiO₂ is an efficient photocatalytic material that can be driven by both UV and NIR light.

Experimental

Chemicals

LuCl₃·6H₂O, YbCl₃·6H₂O and TmCl₃·6H₂O are supplied by Beijing Founde Star Science & Technology Co, Ltd. Cetyltrimethylammonium bromide (CTAB), titanium diisopropoxide bis (acetylacetonate) (TDAA), isopropanol (IPA), ethylene diamine tetraacetic acid (EDTA) and benzoquinone (BQ) are provided by Shanghai Aladdin Biochemical Technology Co., Ltd. Cyclohexane, methanol, ethanol, NH₄F, NH₃·H₂O (30 wt%) and NaOH are obtained from Chongqing Chuandong Chemical (Group) Co, Ltd. 1-Octadecene (ODE, 90%) and Oleic acid (OA, 90%) are acquired from Alfa Aesar. All the reagents are utilized directly in the experiments without additional purification.

Preparation of β-NaLuF₄:y% Yb³⁺/x% Tm³⁺ (UCNPs), x = 0.5, 1, 1.5, 2, y = 10, 20, 30, 40, 50, 60, 70

The UCNPs are synthesized by the solvothermal method. Specifically, the rare-earth chlorides are weighed according to their stoichiometric ratio using an electronic balance for the preparation of 1 mmol UCNPs and then transferred into a 100 mL three-neck flask containing 6 mL oleic acid (OA) and 15 mL octadecene (ODE). Nitrogen is employed as the shielding gas to prevent the occurrence of oxidation phenomena during the reaction process. Initially, the mixture in the three-neck flask is heated to 140 °C and maintained for 30 minutes, resulting in the formation of rare-earth oleate complexes. Following this, the mixture is allowed to gradually cool down to 50 °C, at which juncture 5 mL of a methanol solution containing 2.5 mmol of NaOH and 4 mmol of NH₄F is introduced. Thereafter, the solution is heated up to 80 °C and sustained at this temperature for 30 minutes to facilitate the complete evaporation of methanol. Subsequently, the reactants are subjected to a temperature of 300 °C for a duration of 1.5 hours. After cooling naturally to ambient temperature, the end product is successfully synthesized. Finally, the UCNPs are washed with ethanol and cyclohexane and collected by centrifugation. The resulting centrifuged product is dispersed in 2 mL of cyclohexane for the subsequent coating process.

Preparation of UCNPs@TiO₂ composite structure

The preparation of the UCNPs@TiO₂ composite structure involved the following steps: initially, UCNPs dissolved in cyclohexane are introduced into 20 mL of deionized water containing 0.1 g of CTAB, which is subsequently agitated until it achieved a white milky appearance. Then, the mixture is heated to 80 °C within a water bath, aiming to evaporate the cyclohexane and thus render the solution transparent. Once the solution is cooled to room temperature, the solution is washed using deionized water and centrifugated twice to yield CTAB-modified UCNPs. The sample is subsequently dispersed in 10 mL of isopropanol followed by mixing with another solution consisting of 10 mL isopropanol and 144 μL TDAA. Under continuous magnetic stirring, 0.3 mL NH₃·H₂O and 2.5 mL deionized water are added into the above solution. The mixture is maintained under magnetic stirring at room temperature for a period of 12 hours, and then collected by centrifugation and washed with deionized water and ethanol. After the mixture is dried in an oven at 70 °C for 12 hours, the mixture is annealed at 500 °C for 3 hours to form UCNPs@TiO₂.

Photocatalytic experiment

The photocatalytic efficacy of UCNPs@TiO₂ is assessed through the degradation of RhB under the irradiation of diverse photosources, including a 300 W Xe lamp assembled with various optical filters and a 980 nm semiconductor laser. To be specific, 10 mg of UCNPs@TiO₂ solid powder is dispersed in a 30 mL RhB solution with a concentration of 10 mg L⁻¹. Then, the previously described mixture is transferred into a reaction vessel, where it is maintained under dark conditions for 1 hour to realize the adsorption–desorption equilibrium between UCNPs@TiO₂ and RhB before illumination. During the process of photocatalytic experimentation, 1 mL of the liquid is extracted from the solution at the same time interval and then centrifuged to remove UCNPs@TiO₂ from the solution. The UV-vis absorbance spectrum of the remaining liquid is tested to determine the degradation status of RhB. The photocatalytic performances of pure UCNPs, anatase TiO₂ and the physical mixture of UCNPs and TiO₂ (UCNPs/TiO₂) are also evaluated using the above method. In addition, the UCNPs@TiO₂ is recovered through centrifugation at the end of every cycle to evaluate its performance for repeated use.

Characterization

The X-ray diffraction (XRD) data of the samples are obtained using an XD-2 X-ray diffractometer provided by Beijing Persee. The morphology and elemental analysis data are collected using a JEOL JEM 2100 transmission electron microscope (TEM) equipped with an energy-dispersive spectroscopy (EDS) analyzer. The absorption spectra are measured using a Cary 5000 UV-vis-NIR spectrophotometer. An FLS 1000 spectrometer equipped with a 980 nm laser as the excitation source is employed to measure the spectral data.

Photoelectrochemical characterization is performed through a CHI760E-typed electrochemical workstation equipped with a conventional three-electrode setup in the 0.2 M Na₂SO₄ electrolyte. Here, a Pt plate electrode and saturated Ag/AgCl electrode serve as the counter electrode and reference electrode, respectively. Meanwhile, the prepared photocatalysts are layered onto FTO glass with an active area close to 1 cm² to form the working electrode. A 300 W Xe lamp with an NIR filter (800–2500 nm) is employed as the light source for the experiments.

Results and discussion

Structure and morphology

A two-step wet chemical method is used to synthesize the UCNPs@TiO₂ composite structure photocatalyst, as presented in Fig. 1(a). First, CTAB serves as a surfactant to endow the as-prepared UCNPs with hydrophilic properties for utilization in the following stage. Next, TDAA is employed as the Ti source for the formation of UCNPs@TiO₂, which is mixed with the CTAB-modified UCNPs in isopropanol along with the immission of NH₃·H₂O and deionized water under magnetic stirring at room temperature for 12 hours. The successful synthesis of the UCNPs@TiO₂ composite material is demonstrated using XRD patterns. As shown in Fig. 1(b), each peak presented in the UCNPs and TiO₂ corresponds precisely to β-NaLuF₄ and anatase TiO₂, referenced by JCPDS no. 27-0726 and JCPDS no. 21-1272, respectively. Furthermore, the XRD patterns of UCNPs@TiO₂ distinctly reveal the coexistence of both UCNPs and TiO₂ phases without the appearance of obvious extraneous peaks, verifying the purity of the composite NPs. The weak diffraction peak marked with an asterisk belongs to NaCl, which has no influence on the optical properties of the samples.

Fig. 1 (a) Schematic representation of the preparation process of UCNPs@TiO₂ composite structure. (b) XRD patterns of UCNPs, UCNPs@TiO₂ and anatase TiO₂.

TEM characterization is performed to examine the morphology and structure of the as-synthesized NPs. The TEM image illustrated in Fig. 2(a) reveals that the UCNPs are uniformly dispersed with an average diameter of approximately 200 nm. Meanwhile, the size and morphology of the UCNPs remain almost unchanged after being modified by the hydrophilic CTAB because of the ultra-thin layer of CTAB molecules, as depicted in Fig. 2(b) and (c). Consequently, the CTAB-modified can be readily mixed with TDAA in isopropanol to form the UCNPs@TiO₂ composite structure, as displayed in Fig. 2(d). Ulteriorly, the high-resolution TEM (HR-TEM) image of UCNPs@TiO₂ is examined to shed light on the structural information of the composite NPs. As delineated in Fig. 2(e), the lattice fringe of the UCNPs is identified to be 2.90 Å, aligning with the (101) crystal plane of β-NaLuF₄ (JCPDS No. 27-0726). Additionally, the interplanar spacing of TiO₂ was found to be 3.41 Å, in perfect agreement with the (101) lattice plane of anatase TiO₂ (JCPDS no. 21-1272). Element mapping of an individual UCNPs@TiO₂ particle, shown in Fig. 2(f), vividly displays the distribution of the whole elements, encompassing Na, Lu, F, Yb, Tm, Ti and O. Notably, Na, Lu, F, Yb, and Tm predominantly occupy the inner of the particle, whereas Ti and O are primarily situated on the exterior. This situation of element distribution further confirms the successful formation of the UCNPs@TiO₂ composite structure.

Fig. 2 TEM images of (a) UCNPs, (b) CTAB-modified UCNPs, (c) a single UCNP modified by CTAB and (d) UCNPs@TiO₂. (e) HR-TEM and (f) elemental mapping image of UCNPs@TiO₂. (g) Absorption spectra of UCNPs, TiO₂ and UCNPs@TiO₂ as well as (h) the plots of (αhν)²versus hν of TiO₂ and UCNPs@TiO₂.

In order to gain insight into the absorption ability of the synthesized samples, UV-vis absorption spectra of UCNPs, TiO₂ and UCNPs@TiO₂ are measured. As presented in Fig. 2(g), the UCNPs only reveal a pronounced NIR absorption band peaked at about 980 nm, denoting the transition from ²F_7/2 to ²F_5/2 of Yb³⁺ ions. In contrast, anatase TiO₂ is characterized by its strong absorption in the UV region with an edge near 400 nm, belonging to the band-band transition of TiO₂. The UCNPs@TiO₂ composite NPs integrate the absorption capacity in both UV and NIR regions mentioned above. Moreover, the UV absorption edge of the composite NPs experiences a slight red shift to a long wavelength direction, resulting from the composite structure between UCNPs and semiconductor.^31,32 The bandgap (E_g) can be deduced by Tauc's equation:

αhν = A·(hν − E_g)^n/2 (1)

here, α denotes the absorption coefficient, h symbolizes Planck's constant and ν represents the vibrational frequency. A is a proportional constant. The n value is set to 1 for direct bandgap semiconductors and 4 for indirect bandgap semiconductors. As can be seen from Fig. 2(h), the calculated bandgap energies of anatase TiO₂ and UCNPs@TiO₂ are 3.16 eV and 3.06 eV respectively, which means that more light with low energy can be utilized by UCNPs@TiO₂ for photocatalytic reaction.

Optical properties

As illustrated in Fig. 3(a) and (b), under the excitation of 980 nm NIR light, Tm³⁺ can produce UC luminescence in the UV and blue light regions through the sensitization of Yb³⁺, including UV transition ¹I₆ → ³H₆ at 289 nm, ¹I₆ → ³F₄ at 345 nm and ¹D₂ → ³H₆ at 362 nm as well as blue transition ¹D₂ → ³F₄ at 450 nm and ¹G₄ → ³H₆ at 477 nm. Notably, the UC emission in the UV region of UCNPs@TiO₂ plays a crucial role in achieving NIR-driven photocatalytic reaction due to the bandgap width of TiO₂. To obtain strong UV UC luminescence in Yb³⁺ and Tm³⁺ doped β-NaLuF₄, a series of UCNPs doped with various concentrations of Yb³⁺ and Tm³⁺ are prepared, of which the corresponding XRD data are presented in Fig. S1 (ESI†). As shown in Fig. 3(a), with the doping concentration of Yb³⁺ fixed at 20%, a dependence of the luminescence intensity of UCNPs on the concentration of Tm³⁺ is detected, finding that the strongest UV UC emission is obtained at the Tm³⁺ doping concentration of 0.5%. Subsequently, through adjusting the doping concentration of Yb³⁺ in UCNPs with a fixed Tm³⁺ doping concentration of 0.5%, the strongest UV UC luminescence corresponds with the Yb³⁺ doping concentration of 50%, as shown in Fig. 3(b). Therefore, it can be determined that the optimal doping concentration of Yb³⁺ and Tm³⁺ is 50% and 0.5% respectively, which will be used in subsequent studies of UCNPs@TiO₂ composite NPs.

Fig. 3 UC spectra of the UCNPs doped with (a) 20% Yb³⁺ and x% Tm³⁺ and (b) y% Yb³⁺ and 0.5% Tm³⁺. (c) UC spectra of UCNPs and UCNPs@TiO₂. Decay curves of (d) ¹I₆ level, (e) ¹D₂ level and (f) ¹G₄ level of Tm³⁺ in UCNPs and UCNPs@TiO₂ respectively.

Fig. 3(c) reveals that the UV emissions of Tm³⁺ at 289 nm, 345 nm and 362 nm nearly vanish after integrating UCNPs with TiO₂, while the intensity of blue UC emissions at 450 nm and 477 nm remains almost unchanged, indicating a significant alteration in UC luminescence of Tm³⁺ due to the introduction of TiO₂. In order to explore the energy transfer (ET) mechanism between Tm³⁺ and TiO₂, the decay curves of the relevant energy levels of Tm³⁺ in UCNPs and UCNPs@TiO₂ are measured, including the ¹I₆, ¹D₂ and ¹G₄ levels. As can be seen from Fig. 3(d)–(f), after coating with TiO₂, the lifetimes of Tm³⁺: ¹I₆ and ¹D₂ levels are significantly shortened, whereas the lifetime of the ¹G₄ level stays nearly constant. Here, the lifetime values are obtained from the integrated area of the corresponding normalized decay curves.³³ This phenomenon suggests that the ET between the ¹I₆ and ¹D₂ levels of Tm³⁺ and TiO₂ primarily occurs through a FRET process, rather than a radiation-reabsorption process, because of the unavailability of the reabsorption process on changing the lifetimes of the energy levels. Moreover, referring to the position of the ¹I₆ and ¹D₂ states in Tm³⁺ ions, the two energy levels reveal a good match with the bandgap of TiO₂ indeed, allowing direct ET from the Tm³⁺ ions at the excited state of ¹I₆ and ¹D₂ to the semiconductor TiO₂. This provides a good precondition for realizing an efficient photocatalytic process stimulated by NIR light. In the present work, the ET efficiencies from Tm³⁺: ¹I₆ and ¹D₂ level to TiO₂ are calculated to be 11% and 13%, respectively, by using the following equation:³⁴

η = 1 − τ_T/τ_U (2)

where τ_U and τ_T are the lifetime values of the corresponding levels in UCNPs and UCNPs@TiO₂.

Photocatalytic properties

The photocatalytic performances of UCNPs@TiO₂ can be evaluated through the photodegradation of RhB under various light irradiations, which can be regarded as the values of C/C₀. Here, C₀ represents the initial concentration of the RhB solution, while C denotes the concentration of the RhB solution following a period of illumination. Both of the two concentrations can be calibrated through the absorption spectra of the corresponding RhB solutions. As shown in Fig. 4(a), pristine TiO₂ exhibits a quite poor photocatalytic activity under the irradiation of 980 nm laser, which is close to the performance of the control group without any photocatalysts. This inefficacy stems from the mismatch between the 980 nm wavelength and the bandgap width of anatase TiO₂. The observed minor degradation of RhB in both experimental groups should be attributed to the thermal decomposition caused by the slight increase of solution temperature under the irradiation of a 980 nm laser. Similarly, the pure UNCPs show no ability to degrade RhB. Nevertheless, RhB is significantly degraded in the presence of UCNPs@TiO₂ with a 97% reduction within 10 hours, benefiting from the efficient NIR-UV energy conversion through UCNPs followed by the effective FRET process from UCNPs to TiO₂.

Fig. 4 Photocatalytic performance of various samples under the irradiation of (a) 980 nm wavelength, (b) NIR light and (c) simulated sunlight. Reaction kinetics of RhB degradation with various samples under the irradiation of (d) 980 nm wavelength, (e) NIR light and (f) simulated sunlight.

The degradation rate k of RhB can be calculated using the Langmuir–Hinshelwood kinetics model as follows:

−ln(C/C₀) = kT (3)

where T represents the duration of light exposure. As illustrated in Fig. 4(d), the photo-degradation rate constant of UCNPs@TiO₂ is determined to be 0.2356 h⁻¹. Further testing on the degradation of RhB under the illumination of NIR light (>800 nm) reveals that only the UCNPs@TiO₂ composite material owns significant photocatalytic activity, degrading about 66% of RhB in 10 hours, with a degradation rate of 0.1096 h⁻¹, which is similar to the situation irradiated by 980 nm laser, as plotted in Fig. 4(b) and (e).

The ability of a photocatalyst to harness sunlight for photocatalytic processes stands as a critical benchmark for its practical application. Therefore, the photocatalytic performance testing of the samples irradiated by simulated sunlight is conducted. Fig. 4(c) shows that the degradation effect of the pure UCNPs is virtually indistinguishable from that of the control group, suggesting that the degradation of RhB in the two groups is primarily due to thermal decomposition triggered by light exposure. Compared to anatase TiO₂, the UCNPs@TiO₂ composite material exhibited much stronger catalytic performance, degrading nearly 80% of RhB within 90 minutes, of which the degradation efficiency is 1.6 times higher than that of anatase TiO₂. Moreover, the photocatalytic degradation rate constant of UCNPs@TiO₂ is calculated to be 0.7572 h⁻¹, significantly surpassing the value 0.4446 h⁻¹ of anatase TiO₂, as depicted in Fig. 4(f).

Additionally, to verify the pivotal role of the UCNPs@TiO₂ composite structure in enhancing the photocatalytic performance of TiO₂, the photocatalytic activity of the physically mixed UCNPs/TiO₂ is detected (see Fig. 4(c)), showing only a negligible improvement in degradation of RhB compared to anatase TiO₂ and is significantly lower than that of UCNPs@TiO₂. The huge discrepancy in photocatalytic properties indicates that an effective FRET process is hard to be established between UCNPs and TiO₂ in the physically mixed sample, in which only a small amount of NIR light can be utilized for photocatalytic reactions via the radiation-reabsorption process, demonstrating the significant advantages of the composite structure in markedly boosting the catalytic effectiveness of TiO₂ through facilitating efficient FRET between UCNPs and TiO₂.

The stability of the photocatalysts is another important criterion for its practical application. Thus, the cyclic degradation properties of the UCNPs@TiO₂ are carried out. As shown in Fig. S2 (ESI†), the degradation effect of UCNPs@TiO₂ on RhB only slightly decreases after three experiment cycles, evidencing its outstanding recyclability. Table S1 (ESI†) lists a series of typical NIR-responsive photocatalysts along with their associated parameters. Under the irradiation of 980 nm wavelength, the photocatalytic degradation efficiency and rate of UCNPs@TiO₂ significantly surpass those of the composite materials based on β-NaYF₄:Yb³⁺/Tm³⁺ and TiO₂. Similarly, when exposed to simulated sunlight, the photocatalytic performance of UCNPs@TiO₂ also outperforms other NIR-responsive photocatalytic materials, underscoring its potential value in photocatalytic applications.

Photoelectrochemical properties

To delve deeper into the properties of photo-generated carrier separation and transport in the prepared samples induced by NIR light irradiation, the transient photocurrent response (TPR) and electrochemical impedance spectroscopy (EIS) are conducted. First, as illustrated in Fig. 5(a), a notably higher photocurrent density is observed in UCNPs@TiO₂ with the periodic switching on and off of NIR light. In contrast, almost no photocurrent is detected in UCNPs and TiO₂ individually, suggesting that TiO₂ can effectively harness NIR light to facilitate the separation of electrons and holes for photocatalytic reactions after being integrated with UCNPs. Subsequently, as depicted in Fig. 5(b), the arc radii in EIS Nyquist plots of UCNPs, TiO₂ and UCNPs@TiO₂ decrease progressively, indicating that the lowest charge transfer resistance belongs to UCNPs@TiO₂. These data collectively demonstrate that UCNPs@TiO₂ exhibits superior photocatalytic activity under NIR light irradiation.

Fig. 5 (a) TPR and (b) EIS Nyquist plots of the samples excited by NIR light. (c) Mott–Schottky plots of anatase TiO₂ and UCNPs@TiO₂vs. NHE. (d) RhB degradation efficiency of UCNPs@TiO₂ mixed with various scavengers under the irradiation of simulated sunlight.

Generally speaking, the conduction band potential (E_CB) of n-type semiconductors closely aligns with their flat band (E_fb). Therefore, the Mott–Schottky plots of the samples are detected under dark conditions to ascertain the position of the conduction band edge. The curves in Fig. 5(c) and Fig. S3 (ESI†) exhibit positive slopes, confirming that both anatase TiO₂ and UCNPs@TiO₂ are n-type characteristics. The E_fb values of the two samples obtained from the intercept of the potential axis are −0.35 V and −0.37 V vs. the normal hydrogen electrode (NHE), respectively, which can be considered as their E_CB values. Following this, using eqn (3):

E_VB = E_CB + E_g (4)

the valence band potentials (E_VB) of anatase TiO₂ and UCNPs@TiO₂ are calculated to be +2.81 V and +2.69 V, respectively.

Beyond that, it is widely recognized that the photocatalytic degradation of organic pollutants is primarily accomplished through active species such as photogenerated holes (h⁺), hydroxyl radicals (˙OH), and superoxide radicals (˙O₂⁻). To determine the predominant active species produced by UCNPs@TiO₂ under light illumination, EDTA, BQ and IPA are employed as the specific scavengers for h⁺, ˙O₂⁻ and ˙OH, respectively, in the photodegradation tests of RhB. As illustrated in Fig. 5(d) and Fig. S4 (ESI†), the addition of either EDTA or BQ results in only a minor decrease in the RhB degradation rate compared to the control group without any quenchers. However, the introduction of IPA leads to a significant reduction in the degradation rate from 78% to 38%, suggesting that ˙OH is the dominant reactive species produced by UCNPs@TiO₂ under light exposure.

Photocatalytic mechanism

According to the detected reactive species and the UC emission of UCNPs@TiO₂, the potential photocatalytic mechanism operating in UCNPs@TiO₂ is presented in Fig. 6. Under sunlight irradiation, UV light is directly employed by TiO₂ for photocatalytic reaction, while NIR light is converted to UV light via a series of UC processes and then utilized by TiO₂. Thereinto, the UC processes initiate with Yb³⁺ ions absorbing the NIR component of sunlight and pumping to the ²F_5/2 level. Subsequently, energy is transferred from the excited Yb³⁺ ions to the adjacent Tm³⁺ ions through ET1, ET2 and ET3 routes, enabling the population of Tm³⁺: ³H₅, ³F₂ and ¹G₄ levels. Due to the serious energy mismatch, Tm³⁺ ions at the ¹G₄ state are difficult to be excited to the ¹D₂ level through direct ET with Yb³⁺ ions. Instead, the population of the Tm³⁺: ¹D₂ level is primarily achieved through cross-relaxation processes, such as ³F₂ + ³H₄ → ³H₆ + ¹D₂. The Tm³⁺ ions at the ¹D₂ level can be then populated to the ³P₂ level involves an additional ET process (ET4), followed by multistep nonradiative relaxations to realize the population of ¹I₆ level. Thanks to the excellent matching of Tm³⁺: ¹I₆ and ¹D₂ levels with E_g of TiO₂, an extremely efficient FRET process can occur from the Tm³⁺ ions at the above two states to the TiO₂ adsorbed on the surface of UCNPs, allowing TiO₂ to generate electrons in the CB and holes in the VB, respectively. More importantly, the suitable potentials of CB and VB in TiO₂ compared with the standard redox potential of O₂/˙O₂⁻ and OH⁻/˙OH make the electrons and holes easily to be reacted with O₂ and OH⁻/H₂O, respectively, resulting in the production of ˙O₂⁻ and ˙OH for photocatalysis.

Fig. 6 Possible photocatalytic mechanism in UCNPs@TiO₂.

Conclusions

In summary, β-NaLuF₄:Yb³⁺/Tm³⁺ NPs are selected as the efficient converters for NIR-UV light integrated with TiO₂ to endow this semiconductor with photocatalytic ability under irradiation of NIR light. By analyzing the steady state and transient spectral data of the samples, a significant FRET process between the UV emission energy levels of Tm³⁺ (including ¹I₆ and ¹D₂) and TiO₂ is identified after the successful combination of UNCPs and TiO₂, which enables TiO₂ to generate electrons and holes in the conduction and valence bands, respectively, for photocatalytic reaction under NIR irradiation. Compared to the reported NIR-responsive photocatalysts composed of the commonly used β-NaYF₄:Yb³⁺/Tm³⁺ NPs, the UNCPs@TiO₂ composite material exhibits much higher photodegradation efficiency and rate for pollutants under both NIR and simulated sunlight irradiation. The preparation of the UNCPs@TiO₂ composite material based on β-NaLuF₄:Yb³⁺/Tm³⁺ NPs not only paves the way for substantially raising the utilization efficiency of sunlight but also provides a new strategy for the design of NIR-driven photocatalysts.

Data availability

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work is financially supported by the National Natural Science Foundation of China (11704054, 52104392) and the Natural Science Foundation of Chongqing (CSTB2022NSCQ-MSX0366, csct2021jcyj-msxmX0578).

References

1. C. L. Torres-Martínez, R. Kho, O. I. Mian and R. K. Mehra, J. Colloid Interface Sci., 2001, 240(2), 525–532.
2. A. Di Paola, E. García-López, G. Marcì and L. Palmisano, J. Hazard. Mater., 2012, 211, 3–29.
3. D. Chen, Y. Cheng, N. Zhou, P. Chen, Y. Wang, K. Li, S. Huo, P. Cheng, P. Peng, R. Zhang, L. Wang, H. Liu, Y. Liu and R. Ruan, J. Cleaner Prod., 2020, 268, 121725.
4. S. Patial, P. Raizada, V. Hasija, P. Singh, V. K. Thakur and V. H. Nguyen, Mater. Today Energy, 2021, 19, 100589.
5. X. Zhang, J. Wang, X. X. Dong and Y. K. Lv, Chemosphere, 2020, 242, 125144.
6. P. Singh and A. Borthakur, J. Cleaner Prod., 2018, 196, 1669–1680.
7. A. Kubacka, M. Fernandez-Garcia and G. Colon, Chem. Rev., 2012, 112(3), 1555–1614.
8. S. N. Ahmed and W. Haider, Nanotechnology, 2018, 29(34), 342001.
9. D. Bahnemann, Sol. Energy, 2004, 77(5), 445–459.
10. X. Qiu, Y. Zhang, Y. Zhu, C. Long, L. Su, S. Liu and Z. Tang, Adv. Mater., 2021, 33(6), 2001731.
11. D. Spasiano, R. Marotta, S. Malato, P. Fernandez-Ibanez and I. Di Somma, Appl. Catal., B, 2015, 170, 90–123.
12. K. Nakata and A. Fujishima, J. Photochem. Photobiol., C, 2012, 13(3), 169–189.
13. Q. Guo, C. Zhou, Z. Ma and X. Yang, Adv. Mater., 2019, 31(50), 1901997.
14. J. Schneider, M. Matsuoka, M. Takeuchi, J. Zhang, Y. Horiuchi, M. Anpo and D. W. Bahnemann, Chem. Rev., 2014, 114(19), 9919–9986.
15. A. Meng, L. Zhang, B. Cheng and J. Yu, Adv. Mater., 2019, 31(30), 1807660.
16. A. Fujishima, X. Zhang and D. A. Tryk, Surf. Sci. Rep., 2008, 63(12), 515–582.
17. S. Peiris and H. B. De Silva, J. Chin. Chem. Soc., 2021, 68(5), 738–769.
18. Q. Wu, X. Lai, W. Li, L. Luo and P. Du, J. Alloys Compd., 2023, 967, 171803.
19. Q. Wu, Q. Zhang, W. Li, L. Luo and P. Du, Chem. Eng. J., 2023, 475, 146192.
20. Y. Tang, W. Di, X. Zhai, R. Yang and W. Qin, ACS Catal., 2013, 3, 405–412.
21. W. Wang, W. Huang, Y. Ni, C. Lu and Z. Xu, ACS Appl. Mater. Interface, 2014, 6(1), 340–348.
22. W. Wang, M. Ding, C. Lu, Y. Ni and Z. Xu, Appl. Catal., B, 2014, 144, 379–385.
23. S. Duo, J. Zhang, H. Zhang, Z. Chen, C. Zhong and T. Liu, Opt. Mater., 2016, 62, 240–249.
24. X. Guo, C. Chen, S. Yin, W. Song, F. Shi and W. Qin, J. Photochem. Photobiol., A, 2015, 297, 14–19.
25. X. Guo, W. Song, C. Chen, W. Di and W. Qin, Phys. Chem. Chem. Phys., 2013, 15(35), 14681–14688.
26. X. Bi, G. He, W. Di and W. Qin, Mater. Lett., 2016, 173, 187–190.
27. Y. Zhang, X. Lan, S. H. Park, L. Wang, D. Liu and J. Shi, Carbon, 2020, 169, 111–117.
28. Y. Li, J. Zhang, Y. Luo, X. Zhang, Z. Hao and X. Wang, J. Mater. Chem., 2011, 21, 2895–2900.
29. J. Ouyang, D. Yin, X. Cao, C. Wang, K. Song and B. Liu, Dalton Trans., 2014, 43(37), 14001–14008.
30. J. Ouyang, D. Yin, K. Song, C. Wang, B. Liu and M. Wu, J. Nanosci. Nanotechnol., 2015, 15(4), 2798–2803.
31. H. Anwer and J. W. Park, Appl. Catal., B, 2019, 243, 438–447.
32. C. Zhang, Z. Fu, F. Hong, J. Dou, T. Dong, Y. Zhang, D. Li, G. Liu, X. Dong and J. Wang, J. Solid State Chem., 2021, 300, 122248.
33. G. Xiang, M. Yang, Z. Liu, Y. Wang, S. Jiang, X. Zhou, L. Li, L. Ma, X. Wang and J. Zhang, Inorg. Chem., 2022, 61(13), 5425–5432.
34. C. Song, Z. Ye, G. Wang, J. Yuan and Y. Guan, ACS Nano, 2010, 4(9), 5389–5397.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4tc02695k

---