Sucrose-templated nanoporous BiFeO₃ for promising magnetically recoverable multifunctional environment-purifying applications: adsorption and photocatalysis

Abstract

Bismuth ferrite (BiFeO₃), which acts as a significant multiferroic material, exhibits unique magnetic and ferroelectric properties. Here, we adopted a sol–gel method to synthesize disordered nanoporous BFO with a sucrose template. By different characterization methods such as powder X-ray diffraction (XRD), transmission electron microscopy (TEM) and N₂ physisorption measurements, the large BET (Brunauer–Emmett–Teller) surface area (34.25 m² g⁻¹) and disordered pores (20–40 nm) of the sucrose-templated nanoporous BFO (SN-BFO) materials were determined. These SN-BFO materials exhibited high magnetic performance (M_s = 6.37 emu g⁻¹) and a favorable magnetic recycling ratio. In addition, SN-BFO displayed appropriate adsorption characteristics and good photocatalytic stability, which demonstrates that SN-BFO is a candidate for promising recoverable multifunctional environment-purifying applications in the future.

---

Introduction

In recent years, BiFeO₃ (BFO) has been widely studied owing to its unique room-temperature (RT) multiferroic properties with a high Curie temperature (T_c ∼ 1103 K) and Néel temperature (T_N ∼ 643 K), which make it a promising candidate in practical applications such as data storage, spintronics, sensors, etc.^1–3 However, on account of its modulated spiral spin structure with a period of 62 nm, weak ferromagnetism can be observed in bulk BFO,^3,4 which can be attributed to the defects and anisotropy in BFO⁴ and thus weakens the ME coupling effects in BFO to some extent. Therefore, various attempts have been made to improve the magnetic properties of BFO in order to not only improve the ME coupling effects but also expand new fields of application of BFO such as magnetic photocatalysis.^5,34 Treated as a potential photocatalyst, BFO possesses an appropriate band gap energy (∼2.2 eV), good chemical stability and intrinsic ferroelectric polarization, which can accurately separate photoinduced charge carriers.^5–7 Therefore, various BFO nanostructures have appeared recently with different catalytic behaviors such as oxidation of organic compounds, degradation of organic pollutants,⁸ etc. Besides photocatalytic activity, suitable magnetic performance is also essential for photocatalysts so they can be recovered, which is necessary for photocatalysts. However, as mentioned above, BFO is antiferromagnetic, which hinders its application in recoverable photocatalysts. To overcome these problems, various BFO nanostructures such as nanoparticles,⁵ nanofibers,⁹ nanotubes,¹⁰ and other Bi-based nanomaterials^31,33,35,36 have been reported and demonstrated not only to disrupt the 62 nm spin structure to enhance the magnetic properties but also to provide a larger surface area and shorter charge carrier diffusion length to enhance the photocatalytic activity. However, although enhanced magnetic and photocatalytic behaviors are observed in BFO nanostructures, the magnetism of these BFO nanostructures is still so weak that they cannot be fully or partially recovered after the degradation of pollutants. With their ultra-large surface area and favorable magnetization properties, mesoporous and nanoporous BFO nanomaterials exhibit excellent catalytic^11,12 or water oxidation¹³ performance and display considerable recoverability.^11,14 Therefore, it is expected that mesoporous or nanoporous BFO would exhibit recoverable photocatalytic behavior, which would make it promising for future practical photocatalytic applications. Furthermore, the adsorption characteristics of BFO have rarely been reported and we have investigated the adsorption properties of BFO NPs and SN-BFO in this work.

In this paper, non-surfactant sucrose was used as a non-solid template to synthesize BFO nanoporous structures with disordered pores by a one-pot sol–gel method. These SN-BFO structures exhibited greatly enhanced magnetic properties with a saturation magnetization value of about 6.0 emu g⁻¹, which is attributed to an increase in the Fe²⁺ content in the BFO lattice arising from incomplete combustion of the sucrose template during calcination. Moreover, SN-BFO also displayed favorable recoverable photocatalytic performance and adsorption properties in the degradation of rhodamine B (RhB), which make it a promising candidate for future practical multifunctional environment-purifying applications.

Experimental section

Synthesis of sucrose-templated nanoporous BiFeO₃

SN-BFO was synthesized by a soft chemical method.¹⁵ In order to prepare the BFO precursor, appropriate amounts of Bi(NO₃)₃·5H₂O and Fe(NO₃)₃·9H₂O (the molar ratio of Bi to Fe was 1 : 1) were dissolved in a 3 mol L⁻¹ aqueous solution of HNO₃. Tartaric acid was then added as a complexing agent in a 1 : 1 molar ratio to the metal ions. The obtained solution was kept under constant stirring for 1 h at RT until the gel was clear and uniform. Appropriate amounts of the sucrose template (the molar ratio of sucrose to metal ions was 1 : 1) were dropped into the solution with further vigorous stirring until the sucrose was completely dissolved. The mixed gel was then dried at 90 °C and the viscous mixture was annealed at 150 °C for 5 h to nucleate the BFO. The annealed yellow-brown solid was ground in a quartz vessel and annealed at 550 °C for 2 h in an air atmosphere in order to remove the sucrose template. Finally, the obtained red-brown powder was SN-BFO. For the sake of removing the secondary phases and obtaining a high-purity BFO phase, the red-brown powder was immersed in acetic acid for over 1 h and then washed with deionized water and isopropanol 3 times.

Physical characterization

The phase formation of the crystalline structures and the purity of the sample were determined by an X-ray powder diffractometer (XRD, Bruker AXS D8 Advance, Germany) with Cu Kα (I = 0.154178 nm) radiation. The nanoporous microstructure was investigated by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The field-dependent magnetization characteristics of all samples were measured by a superconducting quantum interference device (SQUID, Quantum Design) at RT. For analyzing the chemical components of the samples, X-ray photoelectron spectroscopy (XPS) was performed with an SSX-100 ESCA spectrometer. Nitrogen adsorption–desorption isotherms were measured at 77 K on a TriStar II 3020 sorption analyzer. Before measurements, all samples were first degassed at 150 °C under vacuum for 12 h. The specific surface areas were calculated by applying the Brunauer–Emmett–Teller (BET) methods to the adsorption isotherms in the relative pressure (P/P₀) range of 0.05–0.25.

Photocatalytic reactions

The photocatalytic degradation of rhodamine B (RhB) by SN-BFO was carried out at RT. A RhB solution (2.5 × 10⁻⁵ mol L⁻¹, 50 mL) with SN-BFO (1 mg mL⁻¹) was ultrasonicated for 10 minutes and stirred for 60 min in dark conditions. The solution was then irradiated by a simulated solar light (Microsolar 300C, Perfectlight, Beijing, China) with a light intensity of 0.2 W cm⁻² at the surface of the solutions. The concentrations of RhB before and after the catalytic reactions were determined from the UV absorption spectra (Lambda 35, PerkinElmer) with a time interval of 30 minutes.

Results and discussion

A disordered BFO nanoporous structure was prepared by a one-pot sol–gel method using non-surfactant sucrose as a non-solid template. Before calcination, a sucrose micella acted as a template, which made the BFO precursor assemble around it, as shown in Scheme 1, and BFO nucleated after calcination. When the temperature was over 200 °C, sucrose thermally decomposed into CO, CO₂ and acetone accompanying the formation of nanopores. The formation of a wide distribution of nanopores can mainly be attributed to the disordered accumulation and arrangement of the sucrose template in the precursor.¹⁶

Scheme 1 Overview of the synthesis process of SN-BFO.

Fig. 1(a) shows the XRD pattern recorded for the SN-BFO sample at RT. All the diffraction peaks can be indexed according to the standard rhombohedral distorted perovskite crystal structure of BFO with the R3c space group. The sharp and well-defined XRD peaks reveal that the sample is highly polycrystalline with few impurity phases (if the sample had not been immersed in acetic acid, diffraction peaks of Bi₂Fe₄O₉ and Bi₂₄Fe₂O₃₉ phases would have appeared in the XRD patterns). Fig. 1(b) shows the Raman shift of the SN-BFO sample, and all the recorded peaks can be identified according to the standard Raman modes of BFO with the rhombohedral R3c space group. As shown in Fig. 1(b), the Raman mode located at 274 cm⁻¹, which is related to the stretching and bending of the FeO₆ octahedra, is slightly weakened compared with that of the pure BFO NPs in our previous work.¹⁷ This indicates a reduction in the amount of FeO₆ octahedra and an increase in distortion caused by the introduction of Fe²⁺, which is demonstrated by XPS, as discussed later. Fig. 1(c) shows a typical TEM image of SN-BFO nanopowder, which indicates a disordered nanoporous structure with a three-dimensional interconnecting configuration. Fig. 1(d) shows the lattice fringes of the (012) and (014̄) planes with interplanar distances of 0.40 nm and 0.29 nm. In addition, the components of the expected composition of SN-BFO, which were quantitatively analyzed from energy-dispersive X-ray spectroscopy (EDS) data, indicate a Bi : Fe atomic ratio of 1 : 1.

Fig. 1 (a) X-ray diffraction pattern of SN-BFO powder; (b) Raman shift of SN-BFO powder; (c) typical TEM image of SN-BFO nanopowder; (d) HRTEM image of SN-BFO powder.

The N₂ adsorption–desorption isotherm of SN-BFO shown in Fig. 2(a) reveals its nanoporous structure, which is consistent with Fig. 1. The BET surface area of SN-BFO was calculated to be 34.25 m² g⁻¹, which is mainly due to nanoscale pores formed by the disappearance of the sucrose template during calcination. The pore size distribution of SN-BFO can also be determined from the desorption branch by the Barrett–Joyner–Halenda (BJH) method, as shown in Fig. 2(a) (inset). The obvious hysteresis loop that was obtained also demonstrates the disordered nanoporous structure,²⁰ and the irregular porosity was also determined from the values of dV/dR for the BJH pore size distribution. The pore diameter of SN-BFO was thus calculated to be between 20 and 50 nm. At low relative pressures of less than 0.6, the isotherm exhibits no obvious loops, which is possibly due to small-sized pores ranging from 3 to 5 nm, and the hysteresis loop that was formed at high relative pressures of greater than 0.6 can be assigned to pores with larger sizes above the 10 nm range. In this work, the hysteresis loop in the range from 0.7 to 1 is caused by disordered pores with sizes from 20 to 50 nm.^20,21 Fig. 2(b) shows the RT curve of magnetization versus magnetic field (M–H) for SN-BFO with a maximum applied magnetic field of 50 kOe. In our previous work,¹⁷ BFO NPs prepared in the same conditions without any templates exhibited antiferromagnetic characteristics (shown in Fig. 2(b) (inset)). The magnetization at 50 kOe and the coercive field (2E_c) of BFO NPs were 0.305579 emu g⁻¹ and 493.5 Oe, respectively, as a result of the space-modulated spiral spin structure. However, the saturation magnetization of SN-BFO derived from the M–H curve shown in Fig. 2(b) was 6.37316 emu g⁻¹ at 50 kOe, whereas the coercive field was about 40 Oe for SN-BFO (shown in Fig. 2(b) (inset)), which was even weaker than that of BFO NPs, indicating paramagnetism rather than ferromagnetism or antiferromagnetism in the SN-BFO samples. Compared with those of BFO thin films and other nanostructures, the coercive field of SN-BFO is much weaker and more insignificant, which reveals its soft characteristics and better adaptability for device applications. To determine the oxidation states of magnetic Fe ions in SN-BFO, X-ray photoelectron spectroscopy (XPS) was carried out and the recorded XPS spectra of the Fe 2p region of SN-BFO are plotted in Fig. 2(c). In particular, the Fe 2p_3/2 peak is centered at 710.1 eV for SN-BFO. In general, the binding energy of Fe 2p_3/2 is about 709.5 eV for Fe²⁺ and 711.0 eV for Fe³⁺.²² The broadening of the Fe 2p peak could be attributed to the nanoscale size of the SN-BFO powder and the coexistence of Fe²⁺ and Fe³⁺ oxidation states in the Fe 2p region with different binding energies. Data fitting of the Fe²⁺ and Fe³⁺ peaks demonstrates that the molar ratio of Fe²⁺ to Fe³⁺ is about 4 : 6, as for other natural polymer-templated BFO,²³ whereas it is usually lower than 1 : 2.²⁴ Such an enhancement in magnetization and transformation of the magnetic characteristics are mainly due to the increased Fe²⁺ content in SN-BFO, which results from the reducing atmosphere during calcination caused by incomplete combustion of the sucrose templates. The improvement in magnetization can therefore be explained as follows: first, the magnetic moment of Fe²⁺ is smaller than that of Fe³⁺, so an appropriate proportion of Fe²⁺ that is present in SN-BFO may seriously disturb the modulated spiral spin structure along the [110] direction with a period of 62 nm by forming a staggered arrangement of Fe²⁺ and Fe³⁺ in some regions of the BFO crystal lattice.¹⁸ In perfect BFO lattices, the total magnetic moment of Fe³⁺ in oxygen octahedra is almost zero, which is shown in Fig. 2(d). However, an increase in the Fe²⁺ content in the BFO lattice results in a change in the total magnetic moment, and a schematic is shown in Fig. 2(d). Second, the volume of Fe²⁺ is much larger than that of Fe³⁺, which causes a distortion of the crystal lattice.¹⁹ Moreover, the introduction of Fe²⁺ changes the Fe–O–Fe double-exchange interaction and further influences the electron cloud distribution of the oxygen octahedra to some extent. Thirdly, the ferromagnetic behavior of BFO may also be affected by the change in oxygen vacancies resulting from the large amount of Fe²⁺.²⁵

Fig. 2 (a) N₂ adsorption–desorption isotherm of SN-BFO at 77 K and the corresponding NLDFT pore size distribution (inset) of SN-BFO. (b) M–H curves of SN-BFO and BFO NPs (inset) with a maximum magnetic field of 50 kOe measured at RT. (c) XPS spectra of the Fe 2p_3/2 state in SN-BFO containing the contributions of Fe²⁺ and Fe³⁺. (d) Magnetic moments of Fe³⁺ in oxygen octahedra in a perfect BFO lattice (left) and the role of Fe²⁺ in the SN-BFO lattice (right).

Adsorption data for the pollutant rhodamine B (RhB) on SN-BFO and BFO NPs are shown in Fig. 3. Obviously, SN-BFO is more effective for the adsorption of RhB from water than BFO NPs. The adsorption kinetics, which reflect the uptake rate of the solute dyes, are an important feature of adsorption. Fig. 3(a) shows the adsorption kinetic curves of SN-BFO and BFO NPs for removing RhB. Under the test conditions (pH = 3 and a concentration of RhB of 12 mg L⁻¹), the adsorption capability for RhB increased rapidly in the first 10 minutes and then remained unchanged. After 60 min, the adsorption capacities of SN-BFO and BFO NPs reached 30.0 mg g⁻¹ and 10.2 mg g⁻¹, respectively. The adsorption isotherms of RhB on SN-BFO and BFO NPs at various initial concentrations have also been studied (Fig. 3(b)).

Fig. 3 (a) Adsorption kinetic curves of RhB on SN-BFO and BFO NPs. (b) Adsorption isotherms of RhB on SN-BFO and BFO NPs. (c) Langmuir isotherms for the adsorption of RhB onto SN-BFO and BFO NPs. (d) Plot of Q_e vs. pH for SN-BFO and BFO NPs.

With an increase in the RhB concentration from 0 mg L⁻¹ to 100 mg L⁻¹, the adsorption capacities of SN-BFO and BFO NPs were increased to 34.7 mg g⁻¹ and 12.1 mg g⁻¹, respectively. Apparently, SN-BFO exhibits a higher adsorption capacity for RhB than that of BFO NPs. A possible reason is the presence of more basic sites in SN-BFO, which is caused by its surface area being larger (34.25 m² g⁻¹) than that of BFO NPs (1.9 m² g⁻¹).

To further study the adsorption behavior of RhB on SN-BFO and BFO NPs, the Langmuir isotherm models have been applied. The Langmuir isotherm is expressed by eqn (1):

where C_e (mg L⁻¹) is the equilibrium concentration in the liquid phase, Q_e (mg g⁻¹) is the amount of adsorbate adsorbed per unit mass of adsorbent, Q_m is the maximum monolayer adsorption capacity (mg g⁻¹), and K_L is the Langmuir adsorption constant (L mg⁻¹), respectively. As shown in Fig. 3(c), C_e/Q_e was plotted against C_e and the slope of the fitted line was obtained. The Langmuir constants K_L and Q_m of SN-BFO and BFO NPs were calculated from the isotherm and their values are noted in Fig. 3(c). Fig. 3(d) shows the relationship between Q_e and pH, which indicates that when the pH value decreases, the value of Q_e of SN-BFO and BFO NPs increases owing to H⁺ activating the surface sites of BFO.

The photocatalytic behaviors of SN-BFO and BFO NPs were estimated via the degradation of the typical organic pollutant RhB (rhodamine B) under visible light at a pH value of 7. To eliminate the influence of the adsorption of RhB on SN-BFO, all the solutions were stirred in dark conditions for 60 min. Because the photocatalytic reaction of the degradation of RhB follows first-order kinetics, the curve of ln(C₀/C) vs. time exhibits a linear relationship, as shown in Fig. 4(a), and the photocatalytic reaction was repeated 3 times with the same catalysts separately. After illumination with SN-BFO for 180 min, the RhB concentration was reduced to approximately 0.2C₀ and the value of the rate constant k reached an average of about 0.0112 min⁻¹. The first-order rate constants (k) of the three photocatalytic reactions were 0.0113 min⁻¹, 0.0109 min⁻¹ and 0.0115 min⁻¹, respectively, which were greater than that of BFO NPs. The relatively high degradation efficiency may be attributed to the small particle size and large specific surface area of SN-BFO. The concentration of RhB after different periods of illumination was confirmed via UV-visible spectra and the absorption spectra of RhB over 180 min are shown in Fig. 4(b). The intensity of the characteristic peak of RhB at 553 nm decreased as the illumination time continued. The absorption peaks decreased considerably throughout the irradiation time and switched to lower wavelengths based on a N-de-ethylation mechanism.²⁶ The inset of Fig. 4(b) shows a shift from 556 nm to 534 nm for the solution under solar-light illumination. To acquire knowledge of the actual degradation process of RhB, the change in the TOC of the RhB solution was measured and is shown in Fig. 4(c). In 240 min of illumination the TOC decreased by 80% and, with an increase in the illumination time to 420 min, all organic species disappeared from the solution. From a comparison of TOC data and the remaining relative concentration of the RhB solution, we found that although after 300 min the relative concentration of RhB in solution was almost zero, TOC still remained in the solution, which was probably because some RhB molecules were adsorbed on the surface of SN-BFO. The TOC in the solution was mainly derived from phenol derivatives or aromatic esters, which originated from the breakdown products of RhB.^27,28 To demonstrate the recovery properties of SN-BFO and BFO NPs, a series of magnetic recycling experiments were carried out. Fig. 4(d) shows the relationship between the mass of SN-BFO/BFO NPs and the recycling times after recovery by magnetic separation by pressing a magnetic stirrer close to the reaction beaker and pouring out the liquid phase slowly. From Fig. 4(d), it can be seen that the recycled mass of BFO NPs decreased from 51.8 mg to 26.2 mg, which was a reduction of about 49.4%, after 3 recycling steps. Compared with this, the recycled mass of SN-BFO decreased from 51.8 to 50.5 mg, which was a decrease of only 2.5%, even in the low-intensity magnetic field of the magnetic stirrer.

Fig. 4 (a) Time-dependent degradation of RhB photocatalyzed by SN-BFO and BFO NPs under UV-visible light. (b) UV absorption spectra of RhB solution in photocatalytic reactions over 180 min. The inset shows the change in the maximum absorption wavelength of RhB solution over time. (c) Changes in TOC and the remaining relative concentration of RhB solution over time. (d) Magnetic recycling performances of SN-BFO and BFO NPs.

As a photocatalyst is irradiated by light with energy that is greater than the band gap energy, electrons are excited from the valence band (VB) to the conduction band (CB), producing holes in E_V. The holes may attract electrons from adsorbed pollutants or react with H₂O to form hydroxyl radicals, whereas electrons can reduce absorbed molecular oxygen to yield superoxide anion radicals (O₂⁻˙), which further form hydroxyl radicals via chain reactions.³² Both types of radicals may initiate redox reactions with organic molecules adsorbed on the surface of the photocatalyst, which leads to the destruction of pollutants. Furthermore, the dye-sensitized effect, namely, that most of the electrons in the E_C of SN-BFO may be derived from RhB*, should be taken into consideration. In fact, different BFO materials prepared by different methods may have diverse energy band structures, especially band gaps (E_g ∼ 2.0–2.9 eV). Because of this, a UV-vis absorbance experiment on SN-BFO was first carried out. As shown in Fig. 5(a), SN-BFO exhibits strong photoabsorption properties in the visible-light region. The inset of Fig. 5(a) shows that the estimated band gap of SN-BFO is about 2.0 eV, which was calculated by using Tauc's equation.²⁹ Based on the estimated energy band structure of SN-BFO, another experiment was designed to test whether RhB* can inject electrons into the conduction band of SN-BFO. Monochromatic light at 620 nm and 0.2 W cm⁻² was used to illuminate SN-BFO, as shown in Fig. 5(b). It was easily found that the degradation rate of RhB was much slower than when illuminated by solar light with the same power. From the UV absorption spectra of RhB solution, it can be discovered that only light in the range from 460 to 590 nm can excite RhB from the ground state to the excited state (RhB*). On the basis of the above facts, the degradation of RhB under monochromatic light at 620 nm is mainly due to photogenerated radicals in SN-BFO. However, the obvious decrease in the degradation rate of RhB also reflects the existence of the dye-sensitized effect, which is equally important in SN-BFO photocatalysis. For a fundamental investigation of which radical plays a key role in the photocatalytic process of SN-BFO, KI and t-BuOH were used as scavengers of different radicals. After the addition of KI (a quencher of positive holes and hydroxyl radicals on the catalyst surface³⁰), the inhibition of photocatalysis in Fig. 5(b) indicates that the degradation of RhB was mainly achieved by positive holes and hydroxyl radicals on the surface of SN-BFO. Moreover, t-BuOH was used as a scavenger of hydroxyl radicals and the degradation rate changed little, which indicates that positive holes may play a major role in the photocatalysis process, as shown schematically in Fig. 5(c). On the one hand, the photogenerated holes combine with ground-state RhB to form RhB⁺, which first undergoes a process of N-de-ethylation and fragments into phenol derivatives or aromatic esters,²⁷ finally degrading to CO₂ and H₂O. On the other hand, when illuminated by visible light, RhB may enter into the excited state and inject electrons into the conduction band of SN-BFO, being transformed into RhB⁺, which proceeds via the same degradation mechanism as mentioned above.

Fig. 5 (a) UV-vis absorption spectrum of SN-BFO. The inset shows the calculation of the band gap of SN-BFO. (b) Effect of different scavengers and specific monochromatic light on the photocatalytic activity of SN-BFO. (c) Schematic of the photocatalytic reaction of RhB on SN-BFO.

From the perspective of environmental purification, SN-BFO may play a dual role, namely, when the concentration of an organic pollutant is relatively low SN-BFO can degrade it by a photocatalytic method and when the concentration of an organic pollutant is high SN-BFO can act as an adsorbent to adsorb it and then be magnetically recycled.

Conclusions

In summary, a nanoporous BFO material with a large surface area was prepared by a sol–gel method with sucrose as a non-surfactant template. XRD confirmed the standard rhombohedral distorted perovskite crystal structure of the SN-BFO material with the R3c space group. TEM showed that SN-BFO displays a series of disordered pores (20–50 nm) and a three-dimensional interconnecting nanoporous structure. Single-phase highly crystalline BFO appeared to exhibit high magnetic performance (M_s = 6.373 emu g⁻¹), which was mainly due to the different ratio of Fe²⁺ to Fe³⁺ (4 : 6), which was determined by XPS. Because of its large BET surface area (34.25 m² g⁻¹), SN-BFO exhibits favorable adsorption and photocatalytic performances, which have been demonstrated by the degradation of RhB. SN-BFO also displays magnetically recoverable characteristics (mass loss of about 2.5% after 3 recycling steps) and stable photocatalytic activity, which can contribute to the dye-sensitized effect and intrinsic hole transfer. Owing to its excellent properties such as nanoporous structure, high magnetic performance, remarkable adsorption properties and appropriate photocatalytic activity, SN-BFO could be a promising candidate for future recoverable multifunctional environment-purifying applications.

Acknowledgements

The authors are grateful for the financial support from the National Natural Science Foundation of China (51372130 and 61401251), Tsinghua National Laboratory for Information Science and Technology (TNList) Cross-discipline Foundation, and the Open Foundation of State Key Laboratory of Electronic Thin Films and Integrated Devices (KFJJ201402).

Notes and references

1. W. Eerenstein, N. D. Mathur and J. F. Scott, Nature, 2006, 442, 759.
2. R. Ramesh and N. A. Spaldin, Nat. Mater., 2007, 6, 21.
3. J. Wang, J. B. Neaton, H. Zheng, V. Nagarajan, S. B. Ogale, B. Liu, D. Viehland, V. Vaithyanathan, D. G. Schlom, U. V. Waghmare, N. A. Spaldin, K. N. Rabe, M. Wutig and R. Ramesh, Science, 2003, 299, 1719.
4. C. Ederer and N. A. Spaldin, Phys. Rev. B: Condens. Matter Mater. Phys., 2005, 71, 060401(R).
5. F. Gao, X. Chen, K. Yin, S. Dong, Z. Ren, F. Yuan, T. Yu, Z. Zou and J.-M. Liu, Adv. Mater., 2007, 19, 2889.
6. L. Li, P. A. Salvador and G. S. Rohrer, Nanoscale, 2014, 6, 24.
7. Y. Zhang, A. M. Schultz, P. A. Salvador and G. S. Rohrer, J. Mater. Chem., 2011, 21, 4168.
8. (a) S. M. Sun, W. Z. Wang, L. Zhang and M. Shang, J. Phys. Chem. C, 2009, 113, 12826; (b) T. Soltani and M. H. Entezari, J. Mol. Catal. A: Chem., 2013, 377, 197; (c) Y. Liu, R. Zuo and S. Qi, J. Mol. Catal. A: Chem., 2013, 376, 1; (d) Z. Li, Y. Shen, C. Yang, Y. Lei, Y. Guan, Y. Lin, D. Liu and C.-W. Nan, J. Mater. Chem. A, 2013, 1, 823; (e) Z. Li, Y. Shen, Y. Guan, Y. Hu, Y. Lin and C.-W. Nan, J. Mater. Chem. A, 2014, 2, 1967.
9. S. Bharathkumar, M. Sakar, K. Rohith Vinod and S. Balakumar, Phys. Chem. Chem. Phys., 2015, 17, 17745.
10. (a) X. Zhu, Z. Liu and N. Ming, J. Mater. Chem., 2010, 20, 4015; (b) S. Mohan and B. Subramanian, RSC Adv., 2013, 3, 23737.
11. I. T. Papadas, K. S. Subrahmanyam, M. G. Kanatzidisb and G. S. Armatas, Nanoscale, 2015, 7, 5737.
12. Y. Huo, M. Miao, Y. Zhang, J. Zhu and H. Li, Chem. Commun., 2011, 47, 2089.
13. I. Papadas, J. A. Christodoulides, G. Kioseoglou and G. S. Armatas, J. Mater. Chem. A, 2015, 3, 1587.
14. R. Dhanalakshmi, M. Muneeswaran, P. Reddy Vanga, M. Ashok and N. V. Giridharan, Appl. Phys. A, 2016, 122, 13.
15. S. Ghosh, S. Dasgupta, A. Sen and H. S. Maiti, J. Am. Ceram. Soc., 2005, 88(5), 1349.
16. D.-W. Lee and K.-H. Lee, Microporous Mesoporous Mater., 2011, 142, 98.
17. J. Xu, D. Xie, C. Teng, X. Zhang, C. Zhang, Y. Sun, T.-L. Ren, M. Zeng, X. Gao and Y. Zhao, J. Appl. Phys., 2015, 117, 224101.
18. J. Liu, L. Fang, F. Zheng, S. Ju and M. R. Shen, Appl. Phys. Lett., 2009, 95, 022511.
19. S. M. Selbach, T. Tybell, M. A. Einarsrud and T. Grande, Chem. Mater., 2007, 19, 6478.
20. M. Humayun, A. Zada, Z. Li, M. Xie, X. Zhang, Y. Qu, F. Raziq and L. Jing, Appl. Catal., B, 2016, 180, 219.
21. H. Su, L. Jing, K. Shi, C. Yao and H. Fu, J. Nanopart. Res., 2010, 12, 967.
22. T. Schedel-Niedrig, W. Weiss and R. Schlögl, Phys. Rev. B: Condens. Matter Mater. Phys., 1995, 52, 17449.
23. Q. Xu, X. Zheng, Z. Wen, Y. Yang, D. Wub and M. Xu, Solid State Commun., 2011, 8, 624.
24. K. Chakrabarti, K. Das, B. Sarkar, S. Ghosh and S. K. De, Appl. Phys. Lett., 2012, 101, 042401.
25. Y. Wang, Q.-H. Jiang, H.-C. He and C.-W. Nan, Appl. Phys. Lett., 2006, 88, 142503.
26. S. Horikoshi, A. Saitou, H. Hidaka and N. Serpone, Environ. Sci. Technol., 2003, 37, 5813–5822.
27. T. Soltani and M. H. Entezari, Chem. Eng. J., 2013, 223, 145.
28. G. M. Liu and C. J. Zhao, New J. Chem., 2000, 24, 411.
29. D. A. Chang, P. Lin and T. Y. Tseng, J. Appl. Phys., 1995, 77, 4445.
30. G. T. Li, J. H. Qu, X. W. Zhang, J. H. Liu and H. N. Liu, J. Mol. Catal. A: Chem., 2006, 259, 238.
31. F. Qin, G. Li and R. Wang, et al., Chem.–Eur. J., 2012, 18, 16491–16497.
32. N. Wang, L. Zhu and M. Lei, et al., ACS Catal., 2011, 1, 1193–1202.
33. Z. Dai, F. Qin and H. Zhao, et al., Nanoscale, 2015, 7, 11991–11999.
34. M. S. Bernardo, et al., Chem. Mater., 2013, 25, 1533–1541.
35. F. Qin, et al., Nanoscale, 2014, 6, 5402–5409.
36. Z. Dai, et al., ACS Catal., 2016, 6, 3180–3192.

---