Bi4Ti3O12 Heterostructures for Visible

Subscriber access provided by YORK UNIV

C: Surfaces, Interfaces, Porous Materials, and Catalysis

Selective Ferroelectric BiOI/Bi4Ti3O12 Heterostructures For Visible-light-driven Photocatalysis Amar Al-keisy, Long Ren, Xun Xu, Weichang Hao, Shi Xue Dou, and Yi Du J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.8b09816 • Publication Date (Web): 13 Dec 2018 Downloaded from http://pubs.acs.org on December 13, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Selective Ferroelectric BiOI/Bi4Ti3O12 Heterostructures For Visible-light-driven Photocatalysis Amar Al-Keisya,c, Long Rena, Xun Xua, Weichang Haob*, Shi Xue Doua, and Yi Dua,b* a

Institute for Superconducting and Electronic Materials (ISEM), University of Wollongong, Wollongong, NSW 2525, Australia b

School of Physics and BUAA-UOW Joint Research Centre, Beihang University, Beijing 100191, China c

University of Technology, Baghdad, Iraq

ABSTRACT: Ferroelectric-photocatalyst/photocatalyst heterojunctions have very attractive photocatalytic activities. Beside enhanced charge carrier separation due to their internal electric fields, charge transfer could be even further enhanced by designing the heterojunction interface. In this work, the polarization-adsorption interaction that exists in ferroelectric materials was employed for successful deposition of BiOI on specific surfaces of Bi4Ti3O12 plates in the dark at room temperature, where the positively polarized region was found. The crystal structure, morphology, and composition of samples were confirmed by X-ray diffraction, field emission

1

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

scanning electron microscopy, and X-ray photoelectron spectroscopy, respectively. Higher photocatalytic activity was achieved by the use of heterojunctions, with the reason behind the enhancement of activity confirmed to be the modified band structure, which contributed to the transfer of photoelectrons from Bi4Ti3O12 to BiOI, the increased visible light absorption, the increased active site area of positively polarized Bi4Ti3O12, and the elimination of the screening layer, which contributes impedance in charge transfer.

1. Introduction

The high recombination rates of photogenerated electron-hole pairs inside the bulk of a photocatalyst, the back-reactions of intermediate species, and inactivity towards visible light are regarded as the key issues blocking light-conversion efficiency. The driving force of charge separation (DFCS) via an internal electric field (IEF) plays a key role in reducing recombination rates

1-5.

The IEF that is in a p-n junction, unfortunately, only exists in the space charge layer.

Overall, according to a review on random heterojunctions 6, the enhancement of photocatalytic activity is not high. In the case of ferroelectric materials with periodic ferroelectric domains, the DFCS that is provided by the ferroelectric photovoltaic effect (FPV) has demonstrated its ability to create high photovoltage due to generation of a series of positive/negative boundaries similar to p-n junctions with a ferroelectric response

7-11.

Unfortunately, this architecture shows low

photocatalyst activity due to low photocurrent density

12-13.

A ferroelectric-semiconductor (F-S)

with a band gap in the visible range, such as Bi4Ti3O12 (BTO), has been considered attractive because it combines optoelectronic and ferroelectric properties 14. Single crystal BTO nanoplates prepared from molten salts have shown much higher photocatalytic activity compared with polycrystalline powder 15-17. The single crystal plate provides faceted surfaces, a monodomain, or

2


Page 2 of 25

Page 3 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


longer disordered chains in the crystal structure for layered structures, whereas in a photocatalyst, it provides pathways for electronic charge carriers (including selective photodeposition and spatially separated redox sites on its surfaces)

18-20.

Because of the BTO-

based layered structure, anisotropic behaviour is present, along with the presence of higher polarity for the ab-plane than along the c-axis in BTO electrical conductivity and optical properties

23.

21-22.

There is also anisotropy in the

Nevertheless, in monodomains, on the

nanoscale, the polarization potential can be screened by charges that come either from the bulk of the ferroelectric-semiconductor (n- or p-type) or from the environment (usually a liquid, since liquids contain many ions, cations, and polar molecules), which then generates a Stern layer as an impedance layer for charge transfer or generates an isolator layer that prevents electrons from reacting with target materials, degrading the catalytic activity

24.

Few reports have studied

ferroelectric-semiconductor/semiconductor (F-S/S) heterojunctions in powder form with the growth controlled over the specific polarization potential. In this work, we explore a new concept to form a heterojunction with built-in DFCS that has positive polarization (right hand circular polarization, c+) from the interface of BTO with an n-type semiconductor such as BiOI. We devised a plan to build a passivation layer to release captured photoinduced electrons in the screening layer (c+) at the narrow edges of BTO plates. BiOI works as an electron transporter for the electrons coming from BTO due to its having a lower conduction band than BTO and both a layered crystal structure (including Bi2O2+2 slabs in both BiOI and BTO), generating continuous bonding from BTO to BiOI, and a narrow band gap, which can promote photocatalytic activity and easy epitaxial growth at room temperature. The polarization-adsorption interaction is important for the selective deposition of photocatalyst on a specific polar surface 25-26. Moreover, photodeposition can be functionalized to explore the pathways of electron-hole transport and

3



redox site surfaces. We propose that this work represents a new strategy to enhance photocatalytic activity and open up opportunities for further exploration of ferroelectricsemiconductor/semiconductor interfaces.

2. Methods

2.1 Synthesis of sample

The BTO plates were synthesised by the molten salt method. The BTO plates were prepared as in previous reports by mixing bismuth oxide and titanium oxide with salts, but here, we used little modification, and we prepared amorphous BTO powder by wet reaction of the raw materials and then mixing them with salts. It was believed that this modification would be easier than in previous reports because it does not need much mixing and requires a shorter time for the reaction. In a typical experimental process: Bismuth nitrate (Bi(NO3)3∙5H2O, Sigma-Aldrich) and titanium butoxide (C16H36O4Ti, Sigma-Aldrich) were used as raw materials. A typical synthesis for BTO plates is described as follows: 4 mmol of Bi(NO3)3∙5H2O was dissolved into a solution of 20 ml ethylene glycol and 40 ml absolute ethanol, and then 3 mmol of C16H36O4Ti was dissolved in the above solution drop by drop. The pH was adjusted by adding KOH up to pH = 10. The clear solution changed to the white precursor under stirring, and it was kept under stirring for 4 h before being filtered by centrifuge. The white powder was washed many times with water and ethanol. Amorphous BTO was obtained after drying for a day. The weight ratio was 10:1 g for the salts (NaCl: KCl = 1:1) in the powder. The mixture was mixed in a mortar for 5 min, then sintered for 1 h at 750 °C, and cooled naturally to room temperature. The product was washed many times with water and ethanol to remove the salts and then dried.

4


Page 4 of 25

Page 5 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


2.2 Characterization

X-ray diffraction (XRD, GBC, MMA) using Cu Kα radiation was used to test the purity and the crystallinity of the samples whereas field emission scanning electron microscopy (FESEM, JEOL-7500) include energy dispersive spectroscopy (EDS) instrument coupled to the FESEM was used to the morphology and elements composition. X-ray photoelectron pectroscopy (XPS) (XPS, PHI660) measurement to determine chemical states of element and surface chemical

composition were carried out by using a monochromatic Cu Kα X-ray source. The transmission electron microscopy (TEM) and selected-area electron diffraction (SAED) (JEOL, JEM2010) was used to further morphologies and crystal structures. An ultraviolet-visible spectrophotometer (UV-Vis, Shimadzu-3600) with an integrating sphere attached to the instrument was used to collect diffuse reflectance spectra (DRS), with BaSO4 providing a background between 200 nm and 800 nm.

2.3 Photocatalytic activity evaluation

Rhodamine B (RhB) and phenol (Sigma-Aldrich) were used as model organic pollutants in the photocatalytic degradation experiment at room temperature. the photocatalytic activity was performed by 300 W Xe lamp with a cut-off filter (λ > 420 nm) as the light source. In a typical process, 25 mL of RhB (10 mg L−1) (or phenol 20 mg∙L−1) and 0.025 g of as-prepared sample were added into a 100 mL beaker with a diameter of 6 cm. The suspension samples were continuously stirred in the dark for 30 min to ensure an adsorption-desorption equilibrium, and then the suspension samples were irradiated by the light source. After illumination, in certain time, the samples (3ml in volume) were collected from the reaction suspension and centrifuged to remove the powder and then analysed using a UV-Vis spectrophotometer by recording the variation of the peak absorption at 556 nm for RhB and 270 nm for phenol. The Data was collected to analyse photocatalytic performance for powders. For comparison with other dyes, Methylene blue (MB) and Methylene orange (MO) were used for comparison with RhB. 3. Results and Discussion

5



According to the above results and the literature, we planned to achieve enhanced photocatalytic activity by building systematic heterojunctions. It was expected that photocatalyst deposited on positive polar areas of BTO plates or edges of BTO plates, which have low reactivity, would enhance photoreduction activity. (Synthesis of pure BTO and verification of its polar nature are shown in the Supporting Information in Figure S1 and Figure S2, respectively.) Therefore, the selection of a photocatalyst with a lower conduction band and layered crystal structure, such as BiOI, could support our plan. With the assistance of previous reports on BTO and BiOI, it was possible to visualize the band structure and mechanism, which are shown in Figure 1. The photoinduced electron charges can be transferred from BTO to BiOI, which is useful for charge separation and enhanced reduction sites.

The growth of BiOI on the positive edge of BTO was successfully accomplished at room temperature, and the mechanism for this growth of BiOI on the edges of BTO could be explained as follows. A physisorption interaction exists between the terminal edge of BTO and the I1present with Bi3+ in solution. Because of the permanent polarization of BTO, electrostatic interaction (van der Waals forces) occurs between the positively polarized BTO and the negative ions in solution, resulting in screening of the positive polarization. Next, when a chemical force exists between the adsorbed substance and the terminal edge of BTO in the presence of Bi3+, chemical reactions occur, with the formation of chemical bonds attaching BiOI to the BTO edge; this process is called chemisorption.

Therefore, BiOI was deposited on positive polar areas of BTO plates by epitaxial growth at room temperature and the colours of the BTO-BI-NP (NP: nanoparticles), BTO-BI-PC (PC: partially covered), and BTO-BI-HS (HS: heterojunction nanosheets) heterojunction samples were

6


Page 6 of 25

Page 7 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


changed from milky to milky-yellow, yellow, and yellow-orange, respectively on increasing the BiOI content.

The morphology of the samples was observed by FESEM and TEM (Figures 2 and 3). The synthesis steps for the BTO-BI-NP, BTO-BI-PC, and BTO-BI-HS heterojunction samples are schematically illustrated as shown in Figure 2. It can be seen that the BiOI has only grown at the edges of the BTO plates, and the growth of BiOI increased with an increasing amount of NaI/Bi(NO3)3∙5H2O solution in the reaction. In the BTO-BI-HS, almost all the edges of BTO appear to be covered, whereas BTO-BI-PC is partially covered, and BTO-BI-NP resembles nanoparticles, so a single plate heterojunction was investigated by EDS to find the elemental contents and distributions. Figure 3a-e corresponds to BTO-BI-HS, Bi, I, Ti, and O respectively. The TEM image shown in Figure 3 (f) provides more confirmation of the morphology shown by the FESEM.

The single crystal nature of the Bi4Ti3O12 plates is confirmed in the SAED pattern shown in Figure 3 (g) of the selected area in the red zone in Figure 3f, and the top view surfaces of the BTO plates are exposed {001} facets, which were determined to have a theoretical interfacial angle of 90° between the (020) and (200) planes. High resolution TEM (HRTEM) (Figure 3h) was conducted to observe the yellow zone in Figure 3f, and the BiOI shows {001} facets. The results for both BTO and BiOI indicate exposed {001} facets. It should be noted that there is large spontaneous polarization along the a-axis due to distortion of the entire TiO6 octahedral layer with respect to the Bi ions 3.

Furthermore, the crystallinity was studied for both BTO and BiOI. XRD of the heterojunction samples was carried out, as shown in Figure 4, and that of BTO is shown in Figure S1a. The 7



BiOI diffraction peaks are clearly observed, which are at 29.7° and 31.7°, and assigned to the (012) and (110) planes of tetragonal BiOI (JCPDS: 90-901-1785), whereas peaks of BiOI were not observed in BTO-BI-NP and BTO-BI-PC because of their low content of BiOI. Because the diffraction peaks related to BiOI have not been demonstrated in XRD, except for BTO Bi3+, XPS was conducted to confirm the chemical states and chemical compositions of the samples. The survey spectrum for the BTO sample reveals that it only contains Bi, Ti, and O elements without any impurity (Figure 4b) whereas in the heterojunctions, the samples include I element. The high-resolution results demonstrate that the two peaks with binding energy at 164 and 158.6 eV correspond to Bi 4f5/2 and Bi 4f7/2 of Bi3+ in the crystal (Figure 4c), respectively, and there is a slight shift in the binding energies in the heterojunctions. The binding energy was shifted to higher energy towards that of pure BiOI with increasing BiOI content. The peak at 528.9 eV is assigned to O 1s (Figure 4d), and this peak could be deconvoluted into three peaks with binding energies of 528.9 eV, 529.2 eV, and 531.7 eV, which correspond to Bi-O, Ti-O, and adsorbed hydroxyl groups. The binding energies of the heterojunctions were shifted up. Moreover, a small peak at 155.5 eV corresponds to oxygen vacancy, which is seen in both pure BiOI and pure BTO. The peaks at 630.3 eV and 618.8 eV are assigned to I 3d3/2 and I 3d5/2, respectively, and the chemical state is I1- (Figure 4e). There was no shift in the binding energy with decreasing BiOI content. The binding energy peak at 457.1 eV is assigned to Ti 2p3/2, whereas the Ti 2p1/2 peak overlaps with Bi 4d3/2 at 462.8 eV with the chemical state of Ti4+. The Ti 2p peaks were not clearly shifted with increasing BiOI content (Figure 4f). On comparing the XPS spectrum of pure BTO with that of a BTO/BiOI heterojunction, the elements Bi and O were shifted up in binding energy with no contribution from Ti and I, which could indicate that the Bi-O bonds of BTO

8


Page 8 of 25

Page 9 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


interacted with the Bi-O bonds of BiOI. They possess different electronegativity values, and electron transfer between them results in band bending.

The photocatalytic activity of the samples was evaluated by decolouration of organic dyes under visible light. RhB dye was used as a colourful dye. BTO-BI-NP (nanoparticles) and BTO-BI-HS (heterostructure) samples were selected to compare with pure BiOI and BTO samples. Before the photocatalytic reaction, the adsorption performance of all samples was investigated in the dark after an interval of 30 min. The samples were compared, as shown in Figure 5a, and it was found that the adsorption performance of the samples reached saturation within 30 min and exhibited high adsorption efficiency, possibly due to the polar surface of BTO. In addition, there are no significantly different results between the samples in the adsorption process. The BiOI-BI-HS sample exhibited the best efficiency toward the photodegradation of RhB. A typical visible absorbance spectrum of RhB degradation by BTOBI-HS is shown in Figure 5a, where the concentration of RhB is relative to the intensity of peak absorption at 556 nm, which decreased rapidly with increasing irradiation time and nearly reached zero within 12 min (Figure 5a).

The colourless and toxic pollutant phenol was used to evaluate the photodegradation performance of the samples over 60 min, and the results demonstrated that 47% of the phenol was photodegraded during the 60 min by BTO-BI-HS, as shown in Figure 5b, which indicates that BTO-BI-HS has excellent performance towards photodegradation of phenol. In contrast, only 23% and 26% phenol degradation were achieved by BTO and BTO-BI-NP, respectively, under visible light irradiation for 60 min. A comparison of the photodecolouration rate was undertaken for other dyes with different cations and anions, MB and MO dyes, respectively. The

9



adsorption results showed (Figure S3) excellent adsorption of RhB with adsorption capacity (Qe) of about 5 mg/g compared with 0.5 mg/g for MO. The photodegradation rates of cationic dyes (such as RhB, Crystal Violet (CV), and MB) for the BTO BTO-BI-NP and BTO-BI-HS samples were significantly higher than those for the anionic dyes (such as MO, Orange G (OG), and Alizarin Yellow (AY)). The surface polarity of the sample could be behind the selective adsorption. The cationic dyes are highly adsorbed due to the negative potential of polarization in BTO because this group of dyes carries a positive charge in their molecules when they are dissolved in water. Therefore, zeta-potential measurements (ζ) were conducted in deionized water (pH = 7) at room temperature. The results revealed that BTO had a negative value (ζ = -12 mV), indicating that BTO has negative surface charge. It is believed that the negative polarization surface (i.e. photo-oxidation active sites) promotes higher photodegradation and has higher activity than the positive polarization (i.e. photoreduction active sites) due to significantly different photodegradation activity. The comparison is is presented in Figure 5c, showing the enhanced photocatalytic activity of BTO-BI-HS, compared with the other photocatalysts. Considering that the same amounts of photocatalyst (0.025 g) were used in characterizing their performances, the weight of BiOI in BTO-BI-HS is only 10.7 % of that of pure BiOI. Therefore, the higher photocatalytic activity of BTO-BI-HS could demonstrate that the heterojunction must be beneficial for the whole photocatalytic process. The optical properties of the samples were investigated by diffuse reflectance. According to the absorption spectra (Figure 6a), all the samples show visible light absorption, but there is a shift to red light spectra from BTO to BTOBI-HS on increasing the BiOI content. The absorption edges in the spectra for the BTO, BTOBI-NP, BTO-BI-PC, BTO-BI-HS, and BiOI samples were 406, 428, 500, 590 nm, and 640 nm, respectively. The enhanced absorption of BTO-BI-HS in the visible light region, compared with

10


Page 10 of 25

Page 11 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


BTO-BI-NP and BTO-BI-PC, is expected to have a benefit for its photocatalytic performances. This result can be explained by the effective hybridization between Ti 3d orbitals and Bi 6p orbitals at the conduction band minimum, and O 2p orbitals and I 5p orbitals at the valence band maximum, near the heterojunction area, which might narrow the band gap of BTO and concomitantly enhance the light absorption capability. The valence and conduction band positions were determined by Mott-Schottky measurements. The flat-band potential (Efp) value is important for determining the Fermi level, because the Efp has been considered nearly equal to the Fermi level. For n-type semiconductors, the Fermi level is considered to be lower than the conduction band by 0.3 V 0.1 Ve. The Efp values of the BTO and BiOI films were estimated, as shown in Figure S4, to be -0.66 V and -0.59 V vs. Ag/AgCl (0.1 M KCl), corresponding to 0.372 V and -0.302 V vs. normal hydrogen electrode (NHE), respectively. These positive slopes indicate the n-type semiconducting properties of these samples. Thus, the conduction bands are at -0.472 V and -0.402 V, and from the band-gap estimates, the valence bands are at 2.57 V and 1.4 V for BTO and BiOI, respectively.

A typical BTO/BiOI heterojunction is useless because the conduction band of BiOI is lower (negatively) than that of BTO, and the valence band of BiOI is lower (positively) than that of BTO, and therefore, electrons and holes will be accumulated in the BiOI, so that higher charge recombination is expected. Herein, a single type of charge transfer from BTO to BiOI is useful for charge separation, as shown in Figure 6b. The IEF that is provided by the FPV is the driving force that drives the electrons to the edges of the BTO plates. Because the BTO is n-type, however, it is believed that some majority charge carriers (electrons) are driven to the positively polarized regions of BTO, and because the edge of BTO, which is {110}, has a small surface area similar to a bottleneck (as shown in the scanning electron microscope (SEM) and HRTEM 11



images), the positive pole of the polarization is roughly compensated by the charges of these electrons as well as photogenerated electrons during photocatalysis, resulting in low effective positive polarization. It is thus expected to exhibit low effective efficiency compared with the large area of {001} facets of BiOI. Herein, BiOI provides a larger surface area with extra electron pathways and more visible light absorption. BiOI nanosheets with {001} facets are more beneficial for the separation of photoinduced carriers and their movement to the surfaces, whether from BTO or BiOI.

To further analyse the reason for the enhancement shown by BTO-BI-HS, a comparison has carried out between BTO-BI-HS that was grown by our method and BTO/BiOI prepared by the conventional loading and mechanical mixing method with the same content of BiOI. The result, as shown in Figure S5, is that BTO-BI-HS has much higher photodegradation performance than BTO-BI-CM (conventional loading method) and BTO-BI-MM (mechanical mixing), due to the random growth of BiOI on BTO, as shown in Figure S6, which shows the EDS mapping of the BTO-BI-CM sample. The random type of deposition cannot lead to enhancement of charge separation. Besides this, the growth of BiOI on the BTO surface blocks the visible spectrum for BTO, and therefore, BTO-BI-MM has slightly higher activity than BTO-BI-CM because BiOI only partially blocks the visible range for BTO in BTO-BI-MM. This indicates that interface engineering is very important for charge transfer and enhancement.

On-off photocurrent was used to investigate charge separation in the samples. The photocurrent density was multiplied by at least six times because of the high charge separation of BTO-BI-HS compared with BTO, as shown in Figure 7a, and the photocurrent result agrees with the photocatalytic activity. The electron recombination was further investigated by photovoltage

12


Page 12 of 25

Page 13 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


decay under on-off light conditions. The decay time is related to the charge lifetime and is considered as indicating the recombination rate. The results are consistent with the photocurrent results (Figure 7b), with BTO-BI-HS showing the slowest decay time.

Furthermore, to further investigate the interfacing positive/negative poles of BTO with BiOI, a BiOI/BTO photoelectrode (Figure S7 photograph of BiOI/BTO thin film) with different switching polarization was prepared to measure the photocurrent, open circuit voltage (Voc), and for electrochemical impedance spectroscopy (EIS) and Mott-Schottky (MS) measurements. The results are shown in Figure 8a and b, where there is higher photocurrent with positive poling than with the negative poling and no poling photoelectrodes. Also, the positive poling led to a higher Voc than for no poling or negative poling.

EIS is important to investigate the charge transfer in the dark and under light, and a smaller radius of arc in the spectrum indicates lower impedance and higher charge transfer. Thus, positive poling in the dark and under light results in lower impedance (Figure 8c). In the MS (Figure 8d) measurements, the flat band potential was demonstrated to be higher in the negative poling photoelectrodes than in the no poling and positive poling photoelectrodes. The reason behind the different properties could be that a barrier is naturally created in the interface for negative poling, so it is believed that there is no charge transfer, even though there is no perfect polarization switching in the ferroelectric material based layered structure. Therefore, the photocurrent in negative poling is nearly all in the BiOI without any contribution of BTO photocurrent. Also, there is a high photovoltage in the positive poling photoelectrodes, as shown in Figure 8b, because of the long lifetimes of photogenerated carriers due to charge separation. The higher impedance in the negative poling photoelectrodes is due to the barrier in the interface

13



region. It is believed that the barrier was generated due to the creation of a virtual p-n junction. The impedance between BiOI and the electrolyte is neglected because it is the same for all photoelectrodes which were used for positive and negative poling.

The virtual p-type is due to accumulated photogenerated holes from BTO at the interface, and this thus gives rise to the flat band potential, as shown in Figure 8d. This kind of virtual p-n junction with this band structure cannot contribute to charge transport, and the conduction band (CB) of BiOI still has lower potential than the CB of BTO, since otherwise, anodic photocurrent could be observed. A schematic diagram has been drawn, as shown in Figure 8(e), to illuminate the band structures for positive poling and negative poling. To test the claim that holes play a key role in the decolouration process for pure BTO, a scavenger effect was created by adding sodium oxalate as a hole scavenger while adding calcium iodate as an electron scavenger. The bar plot, as shown in Figure S9, demonstrates that the photocatalytic activity toward RhB photodegradation is affected by added scavengers. Firstly, with the electron scavenger, the photocatalytic activity is less affected with BTO compared to BTO-BI-HS demonstrating that holes are much more effective toward the photodegradation of RhB in BTO, whereas the electrons play this role in photodegradation with BTO-BI-HS. Secondly, with the hole scavenger, the photocatalytic activity is much more affected with BTO compared with BTO-BIHS, implying the role of electrons in BTO-BI-HS. The BTO result with the hole scavenger is consistent with reference 2. As a result, the enhancement in photocatalytic activity of BTO-BIHS occurs because of the contribution of the electrons in the reduction process, while in BTO, only holes play a role in photodegradation.

14


Page 14 of 25

Page 15 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The high-performance activity of the heterojunction has a considerable effect on the band structure at the interface, and because the BTO and BiOI have a layered crystal structure and close band structures with a little higher CB for BTO than BiOI and only connection in the abplane (BiOI) with exposed {001} facets, as shown in Figures 2 and 3, which means continuous Bi-O bonding, it is possible for the pathways of electrons to cross from the Bi2O2 layer of BTO to the Bi2O2 layer of BiOI. Thus, the majority charge carriers (electrons) which are accumulated at the edge of BTO due to the positive polarization are transferred from the edge of BTO to the CB of BiOI. The transfer will stop when the Fermi levels are aligned. Consequently, salvation of the trapped electrons in the positively polarized regions has been achieved, and thus pathways are created for the escape of the photoinduced electrons through BiOI. As can be seen in Figure S9, the deposited Ag on the edge of BTO-BI-HS was achieved by photodeposition processing and then detected by EDS. The fFigure demonstrates that Ag atoms continue to be deposited on the BTO edge, even though there is growth of BiOI on the BTO edge. This indicates that photoinduced electrons in BTO migrate to the conduction band of BiOI and then react with Ag+.

Although the enhancement was carried out with a single type of charge (electrons), the photocatalytic activity was significantly enhanced. More extensive research is needed, however, for example, spatially mapping of the surface potential and photoresponse measurement on the the surface flow of electric current.

4. Conclusion

We prepared single crystal plates of BTO from molten salts and then selectively deposited BiOI on the positive edges of BTO where there is positive polarization, so that a heterojunction was successfully constructed where electrons are transferred and transported into BiOI. Because band 15



bending is generated at the interface between BTO and BiOI, the charges are spatially separated by the transfer electrons from BTO to BiOI. Significant enhancement of photodegradation has been obtained by this method compared with the conventional loading method or mixing method. The complete photodegradation of RhB by the BTO-BI-HS sample requires 12 min under visible light, whereas phenol photodegradation amounts to 47% in 60 min. This work presents an effective strategy of enhancing photocatalytic activity by rationally designing ferroelectric-semiconductor/semiconductor interfaces.

ASSOCIATED CONTENT

Supporting Information: contains methods, characterizations by XRD, SEM, and TEM, adsorption performance, EDS spectra, and more. AUTHOR INFORMATION Corresponding Author * E-mail for (Y. D.): [email protected] & (W.H.): [email protected] ACKNOWLEDGMENTS (This work is supported by the Australian Research Council (ARC) through a Discovery Project (DP 140102581), the National Natural Science Foundation of China (Grant No. 51072012, No. 51272015), the Fundamental Research Funds for the Central Universities (Grant No. YWF-16-JCTD-B-03), and the Beijing Key Discipline Foundation of Condensed Matter Physics. The authors acknowledge the service supported by the Electron Microscopy Centre at

16


Page 16 of 25

Page 17 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the University of Wollongong. The authors thank Dr T. Silver for her valuable comments on this work. REFERENCES 1. Al-Keisy, A.; Ren, L.; Zheng, T.; Xu, X.; Higgins, M.; Hao, W.; Du, Y. Enhancement of charge separation in ferroelectric heterogeneous photocatalyst Bi4(SiO4)3/Bi2SiO5 nanostructures. Dalton Trans. 2017, 46, 15582-15588. 2. Huang, H.; Tu, S.; Zeng, C.; Zhang, T.; Reshak, A. H.; Zhang, Y. Macroscopic Polarization Enhancement Promoting Photo- and Piezoelectric-Induced Charge Separation and Molecular Oxygen Activation. Angew. Chem., Int. Ed. 2017, 56, 11860-11864. 3. Al-keisy, A.; Ren, L.; Cui, D.; Xu, Z.; Xu, X.; Su, X.; Hao, W.; Dou, S. X.; Du, Y. A ferroelectric photocatalyst Ag10Si4O13 with visible-light photooxidation properties. J. Mater. Chem. A 2016, 4, 10992-10999. 4. Zhang, R.; Dai, Y.; Lou, Z.; Li, Z.; Wang, Z.; Yang, Y.; Qin, X.; Zhang, X.; Huang, B. Layered photocatalyst Bi2O2[BO2(OH)] nanosheets with internal polar field enhanced photocatalytic activity. CrystEngComm. 2014, 16, 4931-4934. 5. Li, L.; Salvador, P. A.; Rohrer, G. S. Photocatalysts with internal electric fields. Nanoscale 2014, 6, 24-42. 6. Reza Gholipour, M.; Dinh, C.-T.; Beland, F.; Do, T.-O. Nanocomposite heterojunctions as sunlight-driven photocatalysts for hydrogen production from water splitting. Nanoscale 2015, 7, 8187-8208. 7. Yi, H. T.; Choi, T.; Choi, S. G.; Oh, Y. S.; Cheong, S.-W. Mechanism of the Switchable Photovoltaic Effect in Ferroelectric BiFeO3. Adv. Mater. 2011, 23, 3403-3407. 8. Ji, W.; Yao, K.; Liang, Y. C. Bulk Photovoltaic Effect at Visible Wavelength in Epitaxial Ferroelectric BiFeO3 Thin Films. Adv. Mater. 2010, 22, 1763-1766. 9. Matsuo, H.; Noguchi, Y.; Miyayama, M. Gap-state engineering of visible-light-active ferroelectrics for photovoltaic applications. Nat. Commun. 2017, 8, 207. 10. Yang, S. Y.; Seidel, J.; Byrnes, S. J.; Shafer, P.; Yang, C. H.; Rossell, M. D.; Yu, P.; Chu, Y. H.; Scott, J. F.; Ager Iii, J. W.; Martin, L. W.; Ramesh, R. Above-bandgap voltages from ferroelectric photovoltaic devices. Nat. Nanotechnol. 2010, 5, 143. 11. Inoue, R.; Ishikawa, S.; Imura, R.; Kitanaka, Y.; Oguchi, T.; Noguchi, Y.; Miyayama, M. Giant photovoltaic effect of ferroelectric domain walls in perovskite single crystals. Sci. Rep. 2015, 5, 14741. 12. Yang, X.; Su, X.; Shen, M.; Zheng, F.; Xin, Y.; Zhang, L.; Hua, M.; Chen, Y.; Harris, V. G. Enhancement of Photocurrent in Ferroelectric Films Via the Incorporation of Narrow Bandgap Nanoparticles. Adv. Mater. 2012, 24, 1202-1208.

17



13. Moubah, R.; Rousseau, O.; Colson, D.; Artemenko, A.; Maglione, M.; Viret, M. Photoelectric Effects in Single Domain BiFeO3 Crystals. Adv. Funct. Mater. 2012, 22, 48144818. 14. Noguchi, Y.; Goto, T.; Miyayama, M.; Hoshikawa, A.; Kamiyama, T. Ferroelectric distortion and electronic structure in Bi4Ti3O12. J. Electroceram. 2008, 21, 49-54. 15. Chen, Z.; Jiang, H.; Jin, W.; Shi, C. Enhanced photocatalytic performance over Bi4Ti3O12 nanosheets with controllable size and exposed {001} facets for Rhodamine B degradation. Appl. Catal., B 2016, 180, 698-706. 16. He, H.; Yin, J.; Li, Y.; Zhang, Y.; Qiu, H.; Xu, J.; Xu, T.; Wang, C. Size controllable synthesis of single-crystal ferroelectric Bi4Ti3O12 nanosheet dominated with {0 0 1} facets toward enhanced visible-light-driven photocatalytic activities. Appl. Catal., B 2014, 156–157, 35-43. 17. Liu, Y.; Zhu, G.; Gao, J.; Hojamberdiev, M.; Zhu, R.; Wei, X.; Guo, Q.; Liu, P. Enhanced photocatalytic activity of Bi4Ti3O12 nanosheets by Fe3+-doping and the addition of Au nanoparticles: Photodegradation of Phenol and bisphenol A. Appl. Catal., B 2017, 200, 72-82. 18. Wang, B.; Liu, M.; Zhou, Z.; Guo, L. Surface Activation of Faceted Photocatalyst: When Metal Cocatalyst Determines the Nature of the Facets. J. Adv. Sci. 2015, 2, 1500153. 19. Li, R.; Zhao, Y.; Li, C. Spatial distribution of active sites on a ferroelectric PbTiO3 photocatalyst for photocatalytic hydrogen production. Faraday Discuss. 2017, 198, 463-472. 20. Zhu, J.; Pang, S.; Dittrich, T.; Gao, Y.; Nie, W.; Cui, J.; Chen, R.; An, H.; Fan, F.; Li, C. Visualizing the Nano Cocatalyst Aligned Electric Fields on Single Photocatalyst Particles. Nano Lett. 2017, 17, 6735-6741. 21. Azodi, M.; Harnagea, C.; Buscaglia, V.; Buscaglia, M. T.; Nanni, P.; Rosei, F.; Pignolet, A. Ferroelectric switching in Bi4Ti3O12 nanorods. IEEE Trans. Sonics Ultrason. 2012, 59, 1903-1911. 22. Katayama, S.; Noguchi, Y.; Miyayama, M. 3D Domain Structure in Bi4Ti3O12 Crystals Observed by Using Piezoresponse Force Microscopy. Adv. Mater. 2007, 19, 2552-2555. 23. Salazar-Kuri, U.; Mendoza, M. E.; Damjanovic, D.; Setter, N. Conductivity and Ferroelectric Hysteresis in Bi4Ti3O12 Single Crystals Around Room Temperature. Ferroelectrics 2013, 448, 114-122. 24. Habicht, S.; Nemanich, R. J.; Gruverman, A. Physical adsorption on ferroelectric surfaces: photoinduced and thermal effects. J. Nanotechnol. 2008, 19, 495303. 25. Peng, Y.; Liu, T.; Xu, J.; Wang, K. K.; Mao, Y. G. Facet-selective interface design of a BiOI(110)/Br-Bi2O2CO3(110) p–n heterojunction photocatalyst. CrystEngComm. 2017, 19, 6837-6844. 26. Zhao, M. H.; Bonnell, D. A.; Vohs, J. M. Effect of ferroelectric polarization on the adsorption and reaction of ethanol on BaTiO3. Surf. Sci. 2008, 602, 2849-2855.

18


Page 18 of 25

Page 19 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figures

Figure 1. Schematic illustration growth of BiOI on the positive edge of BTO and possible mechanism for charge separation at the heterojunction.

Figure 2. Synthesis steps (top), corresponding schematic illustrations of the crystal structures, and the corresponding FESEM images (inset photogrphess of powders) of the samples: (1) BTO, (2) BTO-BI-NP, (3) BTO-BI-PC, and (4) BTO-BI-HS plates, respectively.

19



Figure 3. (a) Typical single plate of BTO/BiOI and the corresponding EDS mapping of (b) Bi, (c) I, (d) Ti, and (e) O; and (f) TEM image for a BTO/BiOI heterojunction; (g) SAED pattern of the red square in (f), and (h) HRTEM image of the yellow square in (f), corresponding to BTO and BiOI, respectively.

Figure 4. (a) XRD patterns, (b) XPS survey spectra, and (c, d, e, and f) high-resolution XPS spectra for Bi 4f, O 1s, I 3d, and Ti 2p, respectively, for the BTO, BiOI, BTO-BI-NP, BTO-BIPC, and BTO-BI-HS samples.

20


Page 20 of 25

Page 21 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. (a and b) Photodegradation of RhB and phenol, respectively, for pure samples of BTO and BiOI, and BTO-BI-NP, BTO-BI-PC and BTO-BI-HS heterojunctions, and (c) comparison of photodegradation activity for MB, MO, RhB, and phenol.

21



Figure 6. (a) UV-vis diffuse reflectance spectra of pure BiOI (the blue line is BiOI) and BTO, and BTO-BI-NP, BTO-BI-PC, and BTO-BI-HS heterojunctions, and (b) possible band structure and heterojunction system.

Figure 7. (a) On-off photocurrent and (b) photovoltage decay for samples under visible light for BTO, BTO-BI-NP, and BTO-BI-HS.

22


Page 22 of 25

Page 23 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 8. (a) On-off photocurrent and (b) open circuit photovoltage of samples. (c) Nyquist plots of open-circuit potential in the dark and under illumination. (d) Mott–Schottky plots for positive poling, no poling, and negative poling samples. (e) and (f) Schematic diagrams of BTO/BiOI

23



thin film and corresponding band structures for negative poling and positive poling photoelectrode.

24


Page 24 of 25

Page 25 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TOC Graphic

25


Bi4Ti3O12 Heterostructures for Visible

Recommend Documents