Ferroelectric Polarization-Enhanced Photocatalysis in BaTiO3-TiO2 Core-Shell Heterostructures

Suppressing charge recombination and improving carrier transport are key challenges for the enhancement of photocatalytic activity of heterostructured photocatalysts. Here, we report a ferroelectric polarization-enhanced photocatalysis on the basis of BaTiO3-TiO2 core-shell heterostructures synthesized via a hydrothermal process. With an optimal weight ratio of BaTiO3 to TiO2, the heterostructures exhibited the maximum photocatalytic performance of 1.8 times higher than pure TiO2 nanoparticles. The enhanced photocatalytic activity is attributed to the promotion of charge separation and transport based on the internal electric field originating from the spontaneous polarization of ferroelectric BaTiO3. High stability of polarization-enhanced photocatalysis is also confirmed from the BaTiO3-TiO2 core-shell heterostructures. This study provides evidence that ferroelectric polarization holds great promise for improving the performance of heterostructured photocatalysts.


Introduction
Organic synthetic dyes in wastewater has become a hard-to-solve source of pollution, causing serious environmental and health problems because of their high solubility and chemical stability in water. Several techniques, including physical adsorption and biological or chemical treatments, have been developed for dye degradation, but they are limited in terms of their practical applications due to their intrinsic disadvantages caused by harmful active byproducts [1,2]. Photocatalysis is a promising pathway for wastewater purification, in which photogenerated charges in catalysts are reacted with water to induce actives of hydroxyl radicals (OH • ) and superoxide anions ( • O 2 − ) for dye degradation [3].
To achieve an effective degradation, catalysts need to possess the following characteristics: broad light absorption, efficient charge separation and transport, as well as good chemical stability. A large amount of photocatalysts including oxide semiconductors [4,5] and sulfide semiconductors [6,7] with a comparatively high degradation efficiency have been reported. Among them, TiO 2 is one of the most widely used photocatalysts, benefiting from its excellent chemical stability, non-toxicity, and low cost. Mesoporous TiO 2 nanoparticles have a large surface area, which is favorable for the creation of numerous active sites for photocatalytic reactions [8,9]. However, the mesoporous TiO 2 cannot fully support space charge effects for the separation of photogenerated charges as a result of its nanoscale grain size [10]. Because the photogenerated charges and photocatalytic reaction products are produced in close proximity to each other, the relatively high rate of charge recombination in mesoporous

Material Preparation
A hydrothermal process [24] was employed to synthesis the BaTiO 3 -TiO 2 core-shell heterostructures, and the meroporous TiO 2 nanoparticles as a reference sample. Commercial BaTiO 3 crystallines (99%, KJ Group, Hefei, China) were ultrasonically dispersed in ethanol, followed by the addition of hexadecylamine (HDA, 90%, Sigma-Aldrich, St. Louis, MO, USA) and NH 3 ·H 2 O, and the mixture was ultrasonicated continuously at room temperature for 10 min to create a homogeneous solution. Subsequently, titanium isopropoxide (TIP, 95%, Sigma-Aldrich, St. Louis, MO, USA) as the precursor of TiO 2 was dropwise added into the above solution under stirring for a hydrolysis reaction of 12 min. NH 3 ·H 2 O is used as a catalyst to promote the hydrolysis reaction of TIP. HDA is used as a surfactant segregated to the surface of BaTiO 3 crystallines. Hydrogen-bonding interactions between HDA molecules and TIP oligomers occur to generate inorganic-organic composites that coat BaTiO 3 . The amount of TIP (0.12, 0.16, 0.2 mL) and NH 3 ·H 2 O (0, 0.1, 0.2 mL), and the stirring rate during the hydrolysis reaction (160, 200, 240 rpm) were adjusted to build core-shell geometries.
The resultants of BaTiO 3 -TiO 2 /HDA nanocomposites were collected by centrifugation. After washing with ethanol and deionized (DI) water, the BaTiO 3 -TiO 2 /HDA nanocomposites were dispersed in a mixture of 20 mL ethanol and 10 mL DI water, and then sealed in a teflon-lined autoclave. The hydrothermal reaction was proceeded at 160 • C for 16 h. The HDA molecules could be removed with the solvothemal treatment at 160 • C to generate mesoporous in TiO 2 that remained as the 3 of 9 nanoshells. The obtained BaTiO 3 -TiO 2 samples were washed with ethanol and DI water to remove residual organics, then dried at 80 • C for 12 h. BaTiO 3 -TiO 2 core-shell heterostructures with different weight ratios of BaTiO 3 to TiO 2 (1:1, 1.2:1, 1.4:1) were synthesized through the hydrolysis reaction followed by the hydrothermal treatment. The optimum synthesis condition was verified by checking photodegradation efficiency of the catalysts synthesized at the different amount of TIP and NH 3 ·H 2 O, and the varied stirring rate during the hydrolysis reaction, as shown in Figure S3. Meanwhile, mesoporous TiO 2 nanoparticles were synthesized at the absence of BaTiO 3 while the other conditions remained unchanged.

Characterization
X-ray diffraction (XRD, DX-2700 diffractometer) with a Cu Kα radiation (35 kV, 25 mA) was used to record phase structures of the as-synthesized BaTiO 3 -TiO 2 core-shell heterostructures and TiO 2 nanoparticles, and the commercial BaTiO 3 crystallines. The morphology and crystal structure of the samples were characterized by a high resolution transmission electron microscope (HRTEM, JEM-2100F, JEOL Ltd., Tokyo, Japan and a field-emission scanning electron microscopy (FESEM, JSM-7800F, JEOL Ltd., Tokyo, Japan). Brunauer-Emmett-Teller (BET) surface areas of the samples were analyzed by Quadrasorb 2MP using N 2 as the adsorption gas. Ferroelectricity of the BaTiO 3 crystallines following the hydrothermal process was validated using an atomic force microscopy (AFM, Cypher™, Oxford Instruments, Goleta, CA, USA) via recording phase-voltage hysteresis and amplitude-voltage butterfly loops with switching spectroscopy piezoresponse force microscopy (SSPFM) [25], which is an effective tool for demonstrating of local ferroelectric characteristics. The photoluminescence (PL) spectra were recorded using PL mapping system (iHR320, Horiba Ltd., Kyoto, Japan) at an excitation wavelength of 325 nm ranging 300-600 nm.

Photocatalytic Activity Measurements
The photocatalytic activity was evaluated by degradation of rhodamine-B dye (RhB) solution under UV light irradiation. Photocatalysts of 30 mg were mixed into 50 mL of 15 ppm RhB solution with pH value of 5.7 ± 0.5. A UV spotlight (L9566-02, Hamamatsu Photonics, Hamamatsu, Japan) equipped with a 200 W mercury-xenon lamp and a 280-400 nm filter was used as the UV light source. The total optical power impinging on the mixture solution was 10 mW/mL. Prior to UV light irradiation, the mixture was stirred in the dark for 30 min to reach desorption-absorption equilibrium. A suspension of 3 mL was drawn out at the given time and then centrifuged at 8000 rpm to remove the photocatalyst completely. The RhB degradation rates were calculated from the UV-vis absorption spectra of the supernatant measured by a UV-vis spectrophotometer (Persee T10, PERSEE Analytics, Auburn, CA, USA) over the range 300-700 nm. The calibration curve was run at the maximum peak of 554 nm, as seen in Figure S5.

Results and Discussion
BaTiO 3 is a typical perovskite ferroelectric with a Curie temperature around 120 • C. Below the Curie temperature, phase transition from a cubic paraelectric to a tetragonal ferroelectric occurs accompanied with lattice distortion due to Ti ions shift along the [001] axis, as seen in Figure S1, producing a spontaneous polarization (P s = 27 µC/cm 2 ) [26]. In this work, commercial BaTiO 3 crystallines were used as the core material. After being subjected to the hydrothermal process, the BaTiO 3 crystallines with a size of 100-300 nm showed excellent ferroelectricity, as seen in Figure S2, which coincides with the size-dependent ferroelectric property in BaTiO 3 [27].
As seen in Figure 1a, the XRD patterns confirm that all the BaTiO 3 -TiO 2 composites consist of tetragonal BaTiO 3 (JCPDS 05-0626) and anatase TiO 2 (JCPDS 21-1272). All the diffraction peaks can be indexed as well-crystallized BaTiO 3 and TiO 2 . No impurity peaks were found in the BaTiO 3 -TiO 2 composites, implying no impurity phases formed in the composites. The representative structure of an individual BaTiO 3 -TiO 2 composite (1.2:1) was characterized by a field emission TEM (FETEM). The low-resolution FETEM image shows a distinguishable core-shell structure that the outer shell consists of mesoporous TiO 2 nanoparticles at a thickness less than 100 nm, as seen in Figure 1b, which is comparable to the depletion width [28]. A high-magnification FETEM image revealed a good lattice match of the core-shell structure. The lattice spaces of about 0.283 nm and 0.192 nm measured from the core and shell portions are in an agreement with the (110) and (200) planes of tetragonal BaTiO 3 and anatase TiO 2, respectively, as seen in Figure 1c. The BaTiO 3 -TiO 2 core-shell configuration was further confirmed by the energy dispersive X-ray spectrometry (EDS) analysis, as seen in Figure 1d, and line-scan EDS profile crossing the core-shell structure, as seen in Figure S4b.  (110) and (200) planes of tetragonal BaTiO3 and anatase TiO2, respectively, as seen in Figure 1c. The BaTiO3-TiO2 core-shell configuration was further confirmed by the energy dispersive X-ray spectrometry (EDS) analysis, as seen in Figure 1d, and line-scan EDS profile crossing the core-shell structure, as seen in Figure S4b. Surface area of photocatalyst is an important factor influencing photocatalysis; a larger surface area can provide more active sites on the surface for photodegradation of organic dyes. BET surface areas of all the samples used in this work were examined and are listed in Table 1. The BaTiO3-TiO2 core-shell heterostructures with different weight ratios of BaTiO3 toTiO2 demonstrated smaller surface areas via a comparison with the pure TiO2 nanoparticles. Among the BaTiO3-TiO2 core-shell heterostructures, the surface area decreased slightly with the increase of the weight ratio. To demonstrate the effect of ferroelectric polarization on the photocatalytic activity of BaTiO3-TiO2 core-shell heterostructures, the degradation capability on RhB was evaluated under the UV light irradiation. For comparison, the photodegradation ability of the pure TiO2 nanoparticles and BaTiO3 crystallines was also evaluated at the same experimental conditions. As seen in Figure 2, the results show that slight photodegradation of RhB was detected using the pure BaTiO3 crystallines as the catalyst under UV light irradiation. However, all the BaTiO3-TiO2 core-shell heterostructures Surface area of photocatalyst is an important factor influencing photocatalysis; a larger surface area can provide more active sites on the surface for photodegradation of organic dyes. BET surface areas of all the samples used in this work were examined and are listed in Table 1. The BaTiO 3 -TiO 2 core-shell heterostructures with different weight ratios of BaTiO 3 toTiO 2 demonstrated smaller surface areas via a comparison with the pure TiO 2 nanoparticles. Among the BaTiO 3 -TiO 2 core-shell heterostructures, the surface area decreased slightly with the increase of the weight ratio. To demonstrate the effect of ferroelectric polarization on the photocatalytic activity of BaTiO 3 -TiO 2 core-shell heterostructures, the degradation capability on RhB was evaluated under the UV light irradiation. For comparison, the photodegradation ability of the pure TiO 2 nanoparticles and BaTiO 3 crystallines was also evaluated at the same experimental conditions. As seen in Figure 2, the results show that slight photodegradation of RhB was detected using the pure BaTiO 3 crystallines as the catalyst under UV light irradiation. However, all the BaTiO 3 -TiO 2 core-shell heterostructures exhibited enhanced photocatalytic activity despite a decrease in their surface areas in comparison with the pure TiO 2 nanoparticles. The BaTiO 3 -TiO 2 core-shell heterostructures (1.2:1) had the best photodegradation activity, which was 1.8 times stronger than that of the pure TiO 2 nanoparticles after a 120 min photocatalysis. The results illustrate that a BaTiO 3 -TiO 2 heterostructure is necessary for inducing an efficient photocatalytic activity. Moreover, the weight ratio of BaTiO 3 to TiO 2 in the heterostructures also plays an important role in photocatalytic activity. The room temperature band gap of both BaTiO 3 [28] and anatase TiO 2 [29] can be taken to be 3.2 eV on the basis of the available data; therefore, difference in the light absorption of these two materials is negligible. In addition, the large decrement in C/C 0 of the pure BaTiO 3 crystallines at 120 min might have been caused by pyroelectric effect of BaTiO 3 [30] because the temperature changed noticeably after 90 min irradiation of the UV light. exhibited enhanced photocatalytic activity despite a decrease in their surface areas in comparison with the pure TiO2 nanoparticles. The BaTiO3-TiO2 core-shell heterostructures (1.2:1) had the best photodegradation activity, which was 1.8 times stronger than that of the pure TiO2 nanoparticles after a 120 min photocatalysis. The results illustrate that a BaTiO3-TiO2 heterostructure is necessary for inducing an efficient photocatalytic activity. Moreover, the weight ratio of BaTiO3 to TiO2 in the heterostructures also plays an important role in photocatalytic activity. The room temperature band gap of both BaTiO3 [28] and anatase TiO2 [29] can be taken to be 3.2 eV on the basis of the available data; therefore, difference in the light absorption of these two materials is negligible. In addition, the large decrement in C/C0 of the pure BaTiO3 crystallines at 120 min might have been caused by pyroelectric effect of BaTiO3 [30] because the temperature changed noticeably after 90 min irradiation of the UV light. Suppressing charge recombination and improving carrier transport are key challenges for the enhancement of the photocatalytic activity of heterostructured photocatalysts. Mesoporous TiO2 nanoparticles with a large surface area can benefit from the creation of numerous active sites for photocatalytic reactions; however, this process can also cause charge recombination by decreasing electron diffusion length and impeding charge transport due to the highly random surfaces/boundaries and the absence of interfacial space charges. Our concept for this work was to use ferroelectric polarization-induced internal field from the BaTiO3 core to facilitate charge carrier separation and further promote charge transport to the surface of TiO2 nanoparticles for an efficient photocatalytic reaction. Figure 3 is the schematic illustration of photogenerated charge separation and transport facilitated by polarization-induced internal field and degradation of RhB based on the BaTiO3-TiO2 core-shell heterostructures. Polarization charges in the BaTiO3 core can attract charges with opposite signs in the TiO2, which move toward the core; meanwhile, the same sign charges are repelled to the outer shell. An internal electric field can be created in the heterostructure because the polarization charges with a high density in the BaTiO3 cannot be fully compensated by lowconcentration free charges in the adjacent TiO2. The average internal electric field formed from the BaTiO3 crystallines can be estimated by adopting the classical dipole field model [31] = = where ε is the relative permittivity of TiO2, σ is the polarization charge density, n is the unit normal to the surface, and f is the volume fraction occupied by the BaTiO3 nanocrystallines. Taking ε = 100 and f = 1 yield an estimated electric field of ~40 V/µm. This polarization-induced internal field generated locally in the heterostructure can penetrate the mesoporous TiO2 with a thickness less than Suppressing charge recombination and improving carrier transport are key challenges for the enhancement of the photocatalytic activity of heterostructured photocatalysts. Mesoporous TiO 2 nanoparticles with a large surface area can benefit from the creation of numerous active sites for photocatalytic reactions; however, this process can also cause charge recombination by decreasing electron diffusion length and impeding charge transport due to the highly random surfaces/boundaries and the absence of interfacial space charges. Our concept for this work was to use ferroelectric polarization-induced internal field from the BaTiO 3 core to facilitate charge carrier separation and further promote charge transport to the surface of TiO 2 nanoparticles for an efficient photocatalytic reaction. Figure 3 is the schematic illustration of photogenerated charge separation and transport facilitated by polarization-induced internal field and degradation of RhB based on the BaTiO 3 -TiO 2 core-shell heterostructures. Polarization charges in the BaTiO 3 core can attract charges with opposite signs in the TiO 2 , which move toward the core; meanwhile, the same sign charges are repelled to the outer shell. An internal electric field can be created in the heterostructure because the polarization charges with a high density in the BaTiO 3 cannot be fully compensated by low-concentration free charges in the adjacent TiO 2 . The average internal electric field formed from the BaTiO 3 crystallines can be estimated by adopting the classical dipole field model [31] where ε is the relative permittivity of TiO 2 , σ is the polarization charge density, n is the unit normal to the surface, and f is the volume fraction occupied by the BaTiO 3 nanocrystallines. Taking ε = 100 and f = 1 yield an estimated electric field of~40 V/µm. This polarization-induced internal field generated locally in the heterostructure can penetrate the mesoporous TiO 2 with a thickness less than the depletion width; it can also act as a driving force for separating electron-hole pairs and promoting them to transport toward to degradation reaction on the TiO 2 surfaces [32]. As a result, the BaTiO 3 -TiO 2 core-shell heterostructures with smaller surface areas present enhanced photodegradation activities compared with the pure TiO 2 nanoparticles used as the catalyst. The dependence of photocatalytic activities on weight ratio might be relevant to the configuration of the core-shell structures. Too few or too many of BaTiO 3 can lead to inharmonious configurations with the formation of either thicker TiO 2 shells (>100 nm) or BaTiO 3 aggregates with multiple crystallines, thus weakening the effect of polarization-induced internal field from the BaTiO 3 core.
Nanomaterials 2019, 9, x FOR PEER REVIEW 6 of 9 TiO2 core-shell heterostructures with smaller surface areas present enhanced photodegradation activities compared with the pure TiO2 nanoparticles used as the catalyst. The dependence of photocatalytic activities on weight ratio might be relevant to the configuration of the core-shell structures. Too few or too many of BaTiO3 can lead to inharmonious configurations with the formation of either thicker TiO2 shells (>100 nm) or BaTiO3 aggregates with multiple crystallines, thus weakening the effect of polarization-induced internal field from the BaTiO3 core. PL signals of semiconductor materials result from the recombination of photogenerated charge carriers. In general, the lower the PL intensity, the lower the recombination rate of photogenerated electron-hole pairs and the higher the photocatalytic activity of semiconductor photocatalysts. In this work, PL spectra are employed to monitor the influence of ferroelectric polarization-induced internal field on the dynamic behaviors of photogenerated charge carriers in the BaTiO3-TiO2 core-shell heterostructures. As shown in Figure 4, PL intensities of all the BaTiO3-TiO2 core-shell heterostructures decreased in comparison with those of the pure TiO2 nanoparticles, which confirms the decrease in recombination rate of photogenerated charge carriers in the presence of light illumination. Among the heterostructures, the weight ratio of 1.2:1 exhibits the lowest PL intensity due to the lower recombination rate of photogenerated electron-hole pairs than the others. This explains the improved photodegradation activity, which is in agreement with the photocatalysis characteristics, as shown in PL signals of semiconductor materials result from the recombination of photogenerated charge carriers. In general, the lower the PL intensity, the lower the recombination rate of photogenerated electron-hole pairs and the higher the photocatalytic activity of semiconductor photocatalysts. In this work, PL spectra are employed to monitor the influence of ferroelectric polarization-induced internal field on the dynamic behaviors of photogenerated charge carriers in the BaTiO 3 -TiO 2 core-shell heterostructures. As shown in Figure 4, PL intensities of all the BaTiO 3 -TiO 2 core-shell heterostructures decreased in comparison with those of the pure TiO 2 nanoparticles, which confirms the decrease in recombination rate of photogenerated charge carriers in the presence of light illumination. Among the heterostructures, the weight ratio of 1.2:1 exhibits the lowest PL intensity due to the lower recombination rate of photogenerated electron-hole pairs than the others. This explains the improved photodegradation activity, which is in agreement with the photocatalysis characteristics, as shown in Figure 2.
The BaTiO 3 -TiO 2 core-shell heterostructures (1.2:1) that has the best photodegradation activity was used to evaluate the stability of photcatalysis by checking their cyclic degradation ability. As shown in Figure 5, a small reduction (~3%) in the photodegradation activity was detected after four consecutive cycles of photocatalysis. The results demonstrate that the BaTiO 3 -TiO 2 core-shell heterostructures possess good stability of photogenerated charge separation and transport during long-term operation which can be attributed to the polarization-induced internal field.
work, PL spectra are employed to monitor the influence of ferroelectric polarization-induced internal field on the dynamic behaviors of photogenerated charge carriers in the BaTiO3-TiO2 core-shell heterostructures. As shown in Figure 4, PL intensities of all the BaTiO3-TiO2 core-shell heterostructures decreased in comparison with those of the pure TiO2 nanoparticles, which confirms the decrease in recombination rate of photogenerated charge carriers in the presence of light illumination. Among the heterostructures, the weight ratio of 1.2:1 exhibits the lowest PL intensity due to the lower recombination rate of photogenerated electron-hole pairs than the others. This explains the improved photodegradation activity, which is in agreement with the photocatalysis characteristics, as shown in Figure 2. The BaTiO3-TiO2 core-shell heterostructures (1.2:1) that has the best photodegradation activity was used to evaluate the stability of photcatalysis by checking their cyclic degradation ability. As shown in Figure 5, a small reduction (~3%) in the photodegradation activity was detected after four consecutive cycles of photocatalysis. The results demonstrate that the BaTiO3-TiO2 core-shell heterostructures possess good stability of photogenerated charge separation and transport during long-term operation which can be attributed to the polarization-induced internal field.

Conclusions
In summary, we fabricated BaTiO3-TiO2 core-shell heterostructures with a varied weight ratio of BaTiO3 to TiO2 for photodegradation of RhB. The mesoporous TiO2 nanoshells were formed on the BaTiO3 core by optimizing the hydrolysis reaction followed by the hydrothermal process. The BaTiO3-TiO2 core-shell heterostructures exhibited considerably higher photocatalytic activity than the pure TiO2 nanoparticles, with a maximum 1.8 times enhancement obtained from the heterostructures with the optimal weight ratio of 1.2:1. Such an enhanced photocatalytic activity is attributed to the facilitation of charge separation and transport based on the internal electric field originated from the spontaneous polarization of ferroelectric BaTiO3. No remarkable reduction of photocatalytic activity was observed after four consecutive cycles, indicating good stability of polarization-enhanced photocatalysis for the BaTiO3-TiO2 core-shell heterostructures. This research demonstrated that the ferroelectric polarization could be an effective approach to enhance performance of heterostructured photocatalysts.
Supplementary Materials: The following are available online at www.mdpi.com/xxx/s1, Figure S1. Crystal structure of (a) cubic (paraelectric) and (b) tetragonal (ferroelectric) BaTiO3. Figure S2. FESEM image (a) and (b) representative phase-voltage hysteresis loop (red) and amplitude-voltage butterfly loop (black) of BaTiO3 after subjected to the hydrothermal process. Figure S3. Photodegradation of RhB in the presence of catalysts synthesized at the different amount of TIP and NH3·H2O, and the varied stirring rate during the hydrolysis reaction Figure S4. Line-scan EDS profile (b) along the red dot dash line crossing the BaTiO3-TiO2 core-shell heterostructure (a). Figure S5. Absorption spectra of RhB solution in the presence of BaTiO3-TiO2 core-shell

Conclusions
In summary, we fabricated BaTiO 3 -TiO 2 core-shell heterostructures with a varied weight ratio of BaTiO 3 to TiO 2 for photodegradation of RhB. The mesoporous TiO 2 nanoshells were formed on the BaTiO 3 core by optimizing the hydrolysis reaction followed by the hydrothermal process. The BaTiO 3 -TiO 2 core-shell heterostructures exhibited considerably higher photocatalytic activity than the pure TiO 2 nanoparticles, with a maximum 1.8 times enhancement obtained from the heterostructures with the optimal weight ratio of 1.2:1. Such an enhanced photocatalytic activity is attributed to the facilitation of charge separation and transport based on the internal electric field originated from the spontaneous polarization of ferroelectric BaTiO 3. No remarkable reduction of photocatalytic activity was observed after four consecutive cycles, indicating good stability of polarization-enhanced photocatalysis for the BaTiO 3 -TiO 2 core-shell heterostructures. This research demonstrated that the ferroelectric polarization could be an effective approach to enhance performance of heterostructured photocatalysts.