Dual Extraction of Photogenerated Electrons and Holes from a

Subscriber access provided by Karolinska Institutet, University Library

Article

Dual extraction of photogenerated electron and hole from ferroelectric Sr Ba NbO semiconductor 0.5

0.5

2

6

Dayong Fan, Jian Zhu, Xiuli Wang, Shengyang Wang, Yong Liu, Ruotian Chen, Zhaochi Feng, Fengtao Fan, and Can Li ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b00809 • Publication Date (Web): 16 May 2016 Downloaded from http://pubs.acs.org on May 17, 2016

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 free 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 accessible to all readers and 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.

ACS Applied Materials & Interfaces 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 27

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

ACS Applied Materials & Interfaces

Dual extraction of photogenerated electron and hole from ferroelectric Sr0.5Ba0.5Nb2O6 semiconductor Dayong Fan,†,‡ Jian Zhu,†,‡ Xiuli Wang,† Shengyang Wang,†,‡ Yong Liu,†,‡ Ruotian Chen,†,‡ Zhaochi Feng,† Fengtao Fan,∗,† Can Li∗,† †State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian National Laboratory for Clean Energy, Dalian 116023, China ‡Graduate University of Chinese Academy of Sciences, Beijing 100049, China.

KEYWORDS : ferroelectric semiconductor, SBN, APV, photocatalysis, charge separation, KPFM.

ACS Paragon Plus Environment

1


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

Page 2 of 27

Abstract

The separation of photogenerated charges is a critical factor in the photocatalysis. Recently, anomalous photovoltaic (APV) field effects (Voc~103 V/cm) in ferroelectrics with their strong driving force for charge separation have attracted much attention in photocatalysis and photoelectrocatalysis. However, it is still unknown whether photogenerated electrons and holes can be simultaneously extracted by the strong driving force towards the surface of ferroelectrics and become available for surface reactions. This issue becomes critically important in photocatalysis, because the surface reaction utilizes both the electrons and holes which reach the surface. In this work, a model lateral symmetric structure metal/Sr0.5Ba0.5Nb2O6/metal with metal = (Ag or Pt) as electrodes was fabricated. The dual extractions of photogenerated electron and hole on the two opposite metal electrodes were achieved as revealed by photovoltaic and ferroelectrical hysteresis measurements and photo-assisted Kelvin Probe Force Microscopy (KPFM). It was found that the high Schottky barriers of the two opposite Sr0.5Ba0.5Nb2O6-Pt electrodes are key factors that alter the two Space Charge Regions (SCRs) by poling effect. The resulting built-in electrical fields with parallel direction near both electrodes significantly enhance the charge separation ability. Our model unravels the “driving force” for the charge separation in ferroelectric semiconductors, thus demonstrating the potential for high efficient charge separation in photocatalysis.


2

Page 3 of 27

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


Introduction Photogenerated charge separation is one of the most important steps in photocatalysis and photoelectrocatalysis. In most artificial photocatalyst designs, built-in electrical fields in p-n junction, phase junction, and space charge regions (SCR) offer the major driving force for the separation of photogenerated charges. However, in most of the cases, these driving forces are extremely limited by the band gap energy. Moreover, efficiency in a photocatalyst requires spatially separated electron and hole to prevent recombination, as well as spatially separated catalytic sites to prevent backward reactions. Ferroelectric materials are attracting much attention due to their unique photovoltaic effects created by the existence of ferroelectric domains and associated internal fields. Unlike conventional semiconductor materials, ferroelectric materials can provide high open-circuit voltages (Voc) up to 5 kV/cm, which are much larger than their band gap energy under light irradiation and as such are termed as anomalous photovoltaic (APV) effect.1-7 Recent years, the APV effect based ferroelectric materials were introduced for photocatalyst system to improve the charge separation efficiency. Photoreactivity enhanced by ferroelectric polarization was observed for TiO2–BaTiO3 core–shell nanowire.8 As well, single domain ferroelectric PbTiO3 particles were synthesized and used for selective photo-deposition of redox co-catalysts for water splitting.9-10 Moreover, the strong charge separation ability of ferroelectric semiconductors was utilized to enhance the activity of conventional semiconductor photocatalysts. For example by combining Ag2O with BaTiO3, it was found that electron and hole pairs were generated in an Ag2O nanoparticle and separated by the electric field created inside the BaTiO3 nanocube along its spontaneous polarization direction. Rohrer et al.11-12 confirmed through a series of experiments based on the TiO2/BaTiO3 model systems that dipolar


3


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

Page 4 of 27

fields from a ferroelectric substrate can enhance the reactivity of a supported rutile film. Rosei et al.13-14 demonstrated that tailoring the Fe/Cr ratio in Bi2FeCrO6 active layers to reduce its band gap without altering its ferroelectric properties is a promising route for the efficient utilization of broad band solar spectrum, without affecting the photogenerated electron–hole pairs separation ability. Although hydrogen evolution reaction (HER) and oxygen evolution reaction (OER) may be thermodynamically possible with some semiconductor photocatalysts, their performance in overall water splitting is far from expected. One possible reason is that the distinctive band bending at the surface of photocatalyst particles is either upward or downward. Thus band bending induced by electric fields help the extraction of only one kind of photogenerated charge carriers.15 In contrast, dual cocatalysts loading,16 phase junctions17 and Z-schemes18 have been exploited in selected photocatalysis systems with high efficiency design due to their welldesigned built-in electrical fields. Despite these efforts, the quantum efficiency (QE) is still restricted by the internal electron and hole recombination. In this regard then, ferroelectric domains with polar structure extended toward the surface offer the opportunity for concurrent spatial separation of photogenerated electron and hole.19 Overall, a fundamental understanding of the mechanism of extraction of photogenerated electrons and holes in ferroelectric material is still lacking. Strontium barium niobate SrxBa1-xNb2O6 (SBN) is a conventional ferroelectric material whose ferroelectricity was studied in 1960 by Francombe.20 Ferroelectric SBN has a tetragonal tungsten-bronze (TTB) structure with P4bm space group. The ferroelectricity comes from the displacement of all metal ions from the oxygen planes along the c axis.21 Therefore, only 180° domain can be formed in SBN and this property makes it a good candidate for the separation and


4

Page 5 of 27

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


transport of electrons and holes in opposite directions. Accordingly we expected that fabricating Sr0.5Ba0.5Nb2O6 (SBN-50) pellets and inserting between symmetric electrodes in a model device had the potential of revealing the dual extraction of photogenerated electrons and holes in poled SBN-50 pellets. We report here that SBN-50 exhibits an impressive enhancement of charge extraction due to the APV effect. We discuss the important roles of Schottky Barriers Height (SBH) between metal electrodes and SBN-50 in the APV effects. The resulting structures with their unique properties show great potential for application in photocatalysis due to their outstanding ability to separate photogenerated electrons and holes toward opposite poled electrodes.

Experimental Section A solid state reaction (SSR) method was used to prepare Sr0.5Ba0.5Nb2O6 polycrystalline samples using high purity BaCO3 (99%), SrCO3 (99.8%) and Nb2O5 (99.9%) (Alfa Aesar) mixed with Sr : Ba = 1 and ball-milled for 12 h, using ethanol as medium. After drying, the mixture was calcined at 1300℃ for 10 hours in an alumina crucible. For the electrical experiments, the powders were reground and mixed with polyvinyl alcohol (PVA) as a binder and pressed into disk shaped pellets by uniaxial pressing with 100 MPa for 3 minutes. The pellets were then sintered at 1350℃ for 2 h after pre-sintering at 1250℃ for 4 h at the heating rate of 5℃/min to avoid the generation of abnormal grain growth. The density of sintered SBN-50 pellet is around 95% of its theoretical density (TD). Ag and Pt pastes (Sino-Platinum Metal Co. Ltd. China) were used to fabricate electrodes. For the dielectric and ferroelectric measurements, the traditional vertical structure was adopted: the sintered pellets were polished carefully and then coated with Ag or Pt paste on both sides. The


5


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

Page 6 of 27

electrodes were fired under different conditions (Ag at 550℃ for 30 minutes, Pt at 1100℃ for 10 minutes). For photoelectric measurement, lateral symmetric structures metal/SBN-50/metal were adopted: the sintered pellets were polished carefully, and then thermally etched in 1300℃ for 30 minutes. On the top side of the pellets, two symmetric Ag or Pt electrodes were screen-printed, stripe shaped (5 mm × 2 mm) with inter-electrode distances of 2 mm; they were then fired in the same condition as mentioned above. Before testing, all samples were ultrasonically cleaned in ethanol. The traditional contact method performed at room temperature was used for poling the pellets and the holding process usually lasted 30 minutes unless specially mentioned in the text. The dark and UV-induced contact potential differences (CPD) on each poled electrode were detected by Kelvin Probe Force Microscopy (KPFM) equipped with a Pt-Ir tip. The details of this setup and scan mode were described elsewhere22. The time dependence of local contact potential profile was acquired along scan lines within an area of 10 nm × 10 nm for 512 seconds. Two-point dc transport properties were measured both in dark and under irradiation by I–V measurements (Keithley model 4200-SCS Semiconductor Characterization System, input impedance: > 1013 Ω, input leakage current: < 30 pA). As the source of UV irradiation, a LED lamp 365 nm (3.4 eV) was used to excite the sample in the absorption region of SBN-50 (Fig. S1) with a flux intensity of UV irradiation of ~2.72 × 1017 photons/cm2·s. The UV light was uniformly irradiated on the 2 mm gap between two electrodes, while avoiding the irradiation on the metal electrodes. The temperature dependence of dielectric properties was measured using a computer controlled Agilent 4294A impedance analyzer. The polarization hysteresis loops were recorded using a Radiant Precision LC ferroelectric testing system.


6

Page 7 of 27

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


Results and discussion Polarization switching for aligning the random domains of as-sintered SBN-50 (details about crystal structure and composition are given in Fig. S2 and Table S1) was done by setting up a large voltage bias between the two electrodes at room temperature. The grains of SBN-50 pellet are on the order of micron (Fig. S3), much larger than the ferroelectric nano-size domains in asgrown samples.23 As the poling voltage increase, the domains would grow up and result in the reduction of the number of domain walls.24 The diagrams of poling/I-V test circuits and ferroelectric domain orientations on the ideally poled SBN-50 pellet are shown in Figure 1a. Two kinds of metal were chosen as the contacting electrodes: Ag with work function of 4.3 eV25, which is usually used in conventional tests26-27, and Pt with work function of 5.3 eV28. These samples are noted for SBN-50-Ag and SBN-50-Pt, respectively. Ferroelectric domains between the electrodes in SBN-50 pellet would be aligned with the poling electric field. As shown in Figure 1b, the photovoltaic measurement of un-poled SBN-50-Pt exhibited a straight line passing through the origin. After poling of an SBN-50-Ag pellet with 30 kV/cm electric field, an opencircuit potential (Voc) of 2.2 V was obtained while, under the same poling conditions, the Voc of an SBN-50-Pt pellet reached up to 6.6 V (The obtained Voc from first time measurement of the sample was even larger, as shown in Fig. S4), which is much higher than that of an SBN-50-Ag pellet and much larger than the bandgap of SBN-50. We note that he output Voc remained stable even after the poled sample had been preserved under ambient condition for 33 hours (see Fig. S5b). The photovoltaic current always flowed in the opposite direction from the poling field. Furthermore, on the as-sintered SBN-50-Pt pellet after I-V measurement with Vmax = 200 V (E = 1 kV/cm), a Voc of 2.3 V was obtained. In contrast, no similar observations could be made for SBN-50-Ag. After heating to 165℃, the Voc decreased rapidly and diminished after heating at


7


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

Page 8 of 27

195℃ for 15 minutes (Fig. 1c). From the dielectric measurements (Fig. S6), SBN-50 showed a diffuse peak around 171℃, which corresponds to the nominal Curie temperature (Tc). These results indicate that the SBN-50 pellet prepared as described exhibits the APV effects only in ferroelectric phase, and that those effects vanish after thermal depoling to the para-electric phase around T = Tc + 25℃. Figure 2a shows the hysteresis P-E loops of a SBN-50-Pt pellet with different poling electric fields and the corresponding I-E loop with Emax = 20 kV/cm at 10 Hz frequency. We extracted a coercive field (Ec) of about 9.5 kV/cm from the domain switching current peak (Isw) although an observable loss in current is observed. It was found that the nominal remnant polarization (Pr, not saturated when the poling field is below Ec) increases with the amplitude of the poling field, a finding similar to that of a SBN single crystal.29 Theses result indicate that domains in the pellet are activated and oriented in the direction of the stronger applied electric field. Additionally, increasing of the poling electric field resulted in increases in photoconductivity and output Voc as shown in Figure 2b. In Figure 2c, the values of Pr and the output Voc of SBN-50-Pt pellets are plotted as functions of poling electric fields. Both Pr and Voc show a linear relationship with the applied electric fields, although the Voc with 30 kV/cm poling falls out of the fitting line due to current leakage. The polarization gets nearly saturated and forms a single-domain state with poling under E = 30 kV/cm (Fig. S7). A simple linear relationship is found between Pr and Voc: Voc (V) =1.5 × Pr (µC/cm2) + 0.19 (E ≤ 20 kV/cm). This equation clearly reflects that the APV properties of SBN-50 pellets strongly depend on Pr, which is proportional to volume fraction of domains oriented with the applied electric field.30 From the above experimental results, it is clear that SBN-50 pellets present ferroelectric properties and can give APV upon poling with electrical field. To demonstrate whether the built-


8

Page 9 of 27

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


in electrical field can help in the separation and transport of the photogenerated charge carriers to the two electrodes, photo-assisted KPFM was employed to study the photogenerated charge separation behaviors on the two electrodes, as shown in Figure 3a. The two electrodes connected to positive and negative poles of the home-made high DC voltage source during the poling process are labeled by superscript “+” and “-”, respectively. KPFM provides the contact potential difference (CPD) between sample surface and the Kelvin tip and can be calculated as CPD = (Φtip – Φsurface) / -e, where Φtip and Φsurface are the work functions of the kelvin tip and the sample surface, and e is the electronic charge. Typical KPFM images obtained at the two electrodes before and after light illumination are shown in Figure 3b. The cross sections of the images reflect the CPD changes of the two electrodes during light modulation. Figure 3c shows the CPD changes of the two electrodes on SBN-50-Ag modulated with light irradiation (365 nm). It is found that the CPD of the Ag+ electrode shows a sharp increase followed by a relative slow decrease upon light irradiation, and that the steady state CPD under illumination is higher than that in the dark by 0.12 V. Upon cutting off the illumination, the CPD undergoes a sharp decrease and returns to its original CPD. The increase in CPD on the Ag+ electrode can be assigned to the lowered Fermi level and thus indicates that the photogenerated holes are successfully extracted at the Ag+ electrode. As for the Ag- electrode, the CPD is less affected by the light modulation. These results suggest that the photogenerated holes can be collected on the Ag+ electrode, while no photogenerated charge carrier is collected on the Ag- electrode. In contrast the CPD changes of the two electrodes on SBN-50-Pt are very different from those of SBN-50-Ag, as shown in Figure 3d. The CPD of the Pt+ electrode shows a sharp increase followed by a slow decrease upon UV irradiation. Similar to that of the Ag+ electrode, the steady state CPD of the Pt+ electrode under illumination is higher than that in the dark state by 1.17 V.


9


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

Page 10 of 27

It is interesting to note that the CPD of the Pt- electrode shows a sharp decrease followed by a slow increase upon light irradiation. The steady state CPD for the Pt- electrode under illumination is lower than that of the dark state, by 1.08 V. Additionally the shape of the CPD profile from the Pt- electrode is mirror-symmetric to that of the Pt+ electrode. These results indicate that photogenerated holes can be collected on Pt+ electrode while photogenerated electrons can be collected on Pt- electrode. The high Voc shown in Figure 3e can be a result of the dual extraction of photogenerated electrons and holes to the opposite electrodes of SBN-50-Pt. It was reported that by heating the poled ferroelectrics to T = Tc + 25℃ and cooling at zero field, the ferroelectric-related effects are eliminated, a phenomenon called “thermally depoled unpolar phase”.31-32 In order to understand the role of this “thermally depoling process” played in photogenerated charge separation, the poled SBN-50-Pt pellet was heated to 195℃ and then cooled down to room temperature. From the CPD profiles plotted by red and blue dashed lines (Fig. 3d), a small CPD changes can be still observed for the Pt+ and Pt- electrodes. However, no output Voc is observed in the photovoltaic measurement of the SBN-50-Pt pellet after heated at 195℃. Thus, it is reasonable to assume that Pr in the SBN-50-Pt pellet approaches zero after heating at 195℃. However, the CPD changes on the two electrodes do not approach zero, an observation that suggests that some other factors are affecting the charge separation on both Pt electrodes. SBN-50-Pt samples after poling show abnormal increases of their dielectric constant with the increased temperature, as shown in Fig. S6. This effect can be assigned to the short-range migration of doubly ionized oxygen vacancies described in the literature.33-35 These doubleionized oxygen vacancies may accumulate near the cathode and migrate away from the anode, respectively, due to the applied high electric field (See details in Fig. S6 in Supporting


10

Page 11 of 27

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


Information). These inhomogeneous distributions of oxygen vacancies might be a factor that affects charge separation. After heating the sample to 250℃, the difference in light-induced CPD on both of the electrodes vanish (Fig. 3f). It should be mentioned that the abnormal jump of dielectric loss of SBN-50-Pt at T > 200oC (Fig. S6b) after high electric field poling was generally regarded as due to the migration of oxygen vacancies. After depoling the sample at 250oC for 30 min, the interfacial internal field induced by inhomogeneous distribution of the doubly oxygen vacancies disappears, due to the migration of thermal excited oxygen vacancies into the bulk. Table 1 summarizes the CPD values of the SBN-50-Pt pellets under different situations. It can be deduced that, for the dual charge extraction, the contribution of oxygen vacancies is limited (< 15%), while the internal unscreened electric field induced by ferroelectric polarization dominates this process.36 The flat band position (ϕ ) of SBN-50, as inferred from the Mott-Schottky (M-S) analysis, is calculated around -0.70 V vs. NHE at pH = 7 ( as shown in Fig. 4a). This value agrees well with the onset potential of anodic photocurrent when using methanol aqueous solution as sacrificial agent,37 as shown in Figure 4b. The flat band position (ϕ ) of unpoled SBN-50 is calculated to be 4.16 eV vs. absolute vacuum scale (AVS). The Tauc plots of SBN-50 polycrystalline particles show that the direct and indirect bandgaps are 3.42 eV and 3.17 eV, respectively (Fig. S8). Taking EVB (SBN-50) as 7.38 eV (vs. AVS) from the literature,38 the energy diagram of unpoled SBN-50 is depicted in Scheme 1a. From the diagram, we can infer that SBN-50 is an n-type ferroelectric semiconductor and has potential alignement for both HER and OER. This is further shown in Figure S9. From the above analysis, we can infer that a Schottky barrier for electrons with a barrier height of 0.34 eV is formed between the Ag and SBN-50, as shown in Scheme 1b. This barrier


11


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

Page 12 of 27

height increases to 1.34 eV in the SBN-50-Pt pellet, as shown in Scheme 1c. These values are consistent with the fact that the dark conductivity of SBN-50-Pt is 2.5 orders of magnitude lower than that of SBN-50-Ag (shown in Fig. S10). The strong dependence of the I-V character on the electrode materials implies that the high work function of the metal electrode suppresses the leakage current effectively. Thus, in the case of SBN-50-Pt pellet, deeply n-depleted regions with strong upward band bending near both contacts of Pt/SBN-50/Pt strongly prevent the free charges injection. In contrast the leakage current cannot be suppressed for conventional Ag/SBN-50/Ag structures. It is reported that the free charge injection will result in the screening of the external electric field during poling and further decreases the efficiency of ferroelectric polarization.39 When a high bias voltage is applied between the two electrodes, voltage drops occur in both SCRs. Because of the higher barrier height at SBN-50-Pt interfaces, one can expect the voltage drops for SBN-50-Pt contacts to be much higher than that for SBN-50-Ag and to result in a significant modification of the SCRs at SBN-50-Pt interface, as shown in Scheme 1d. We note that the Debye screening length of unpoled SBN is only 3.8 nm (See the details below Fig. S10 in Supporting Information) and higher photoconductivity can be induced by higher poling electric field (as shown in Fig. 2b). These findings suggest that the extent of the upward band bending will become larger on the interface of SBN-50-Pt+, while the SCR at SBN-50-Pt- side changes from a depletion layer to an accumulation layer, and that the direction of the intrinsic internal electric field changes to align with SBN-50-Pt+ side. As a consequence, two extended SCRs with the same direction of internal electric fields are obtained near both SBN-50-Pt- and SBN-50-Pt+ interfaces. These internal electric fields, which are induced mainly by the polar ferroelectric domain, are the driving forces for the separation and dual extraction of


12

Page 13 of 27

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


photogenerated electron and hole on both electrodes. As for the SBN-50-Ag pellet, the small voltage drops can only lead to the formation of a depleted SCR with upward bend bending on the anode side and very weak depleted SCR on the cathode side. This explains the fact that the photogenerated hole extraction was observed only for SBN-50-Ag+ while only a trace signal was observed for SBN-50-Ag-. The results imply that it is possible to simultaneously extract photogenerated electrons and holes toward spatially separated surface sites of a homogeneous ferroelectric material. Additionally, we notice the low photocatalysis activity of SBN-50, which may be ascribed to its low surface area (1.66 m2/g from BET measurement) due to the high calcination temperature. The Pt loading induces built-in Schottky barriers on the SBN-50 particle surface and prevents the recombination of photogenerated charge carriers formed by light excitation. However, the SCR width (w0 ≈ 38 nm for unpoled SBN-50-Pt) is much smaller when compared with the indirect optical absorption length at 365 nm ( >1 µm, Fig. S10). Our results indicate the poling effect may extend the SCR throughout the ferroelectric micro-size particles. The resulted SCR will facilitate the photo-excited electron and hole separation and extraction for higher photocatalysis efficiency.40

Conclusions In summary, model structures of lateral symmetric metal/SBN-50/metal with metal = Ag or Pt as electrodes were fabricated and both of them exhibited APV effects. It was found that desired APV effects on SBN-50-Pt can be ascribed to the simultaneous dual extraction of separated electrons and holes. High SBH in SBN-50-Pt junctions prevents the free charge carrier injection during the high electric field poling process, and thus significantly modifies the SCRs


13


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

Page 14 of 27

on both Pt+-SBN-50 and Pt--SBN-50 contacts. Two deeply extended SCRs with the same internal electric field direction facilitate the simultaneous dual extraction of separated electrons and holes to the two opposite directions. Most importantly in conclusion, the ferroelectric semiconductor SBN-50 with symmetric metal electrodes promises excellent potential applications in photocatalysis due to its unique ability to drive the photogenerated electron and hole to the opposite polar surfaces with clear spatial separation. Further improvements in design for effective poling of SBN-50 photocatalyst particulates are still under investigation for possible practical applications.


14

Page 15 of 27

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. (a) The circuits for the poling and I-V measurement on the lateral symmetric structure metal/SBN-50/metal (SMU: Source-Measure Unit) and the schematic of domain orientations in SBN-50 pellet between electrodes after an ideal poling with a high bias; I-V curves under 365 nm LED irradiation: (b) the poled SBN-50 with symmetrical Ag or Pt electrodes vs. un-poled SBN-50-Pt. The curves show that the output parameters largely depend on the electrode material, with SBN-50-Pt giving a Voc up to 6.6 V and Isc ~ -261 pA. (c) as-sintered SBN-50-Pt after dark swept I-V test with a max applied voltage of 200 V (~ 1 kV/cm)


15


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

Page 16 of 27

Figure 2. a) Hysteresis P-E loops of a SBN-50-Pt pellet with different Emax at frequency of 10 Hz and the corresponding I-E loop with Emax = 20 kV/cm. b) I-V curves under 365 nm LED irradiation: SBN-50-Pt after poling with various poling electric fields, with the black dots corresponding to the sample having undergone a depoling process after poled with E = 10 kV/cm; c) summarized relationships between poling electric field, nominal remnant polarization and photovoltage (see the details in Fig. S7);


16

Page 17 of 27

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 3. a) Photo-assisted KPFM was employed to study the photogenerated charge separation behaviors on the two electrodes; b) Typical KPFM images obtained at the Pt+ and Pt- electrodes before and after light illumination (scanning time for one picture: 512 s); The transient CPD signal under 365 nm LED irradiation recorded on the two poled electrode surfaces of c) SBN-50Ag poled with 30 kV/cm; d) SBN-50-Pt after poling with 30 kV/cm(solid line) and after thermal depoled at 195℃ for 15 mins(dashed line); e) Photovoltage vs. time for poled and unpoled SBN50-Pt; f) same as (d) but with sample after thermal depoling at 250℃ for 30 mins.


17


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

Page 18 of 27

Figure 4. a) Mott–Schottky plots for a SBN-50 electrode according to impedance measurements (Frequency: 1000 Hz; Counter electrode: Pt; Reference electrode: Ag/AgCl; ϕ(NHE) = ϕ(Ag/AgCl) + 0.199 V); b) linear sweep voltammetry of a SBN-50 electrode under chopped XeHg lamp irradiation measured in 0.1 M Na2SO4 + 10 vol% methanol solution (scan rate: 10 mV/s).


18

Page 19 of 27

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


SCHEMES

Scheme 1. Band alignment of SBN-50 with Pt or Ag as electrodes vs. absolute vacuum scale (AVS) a) before contact; b) after contact with Ag electrodes; c) after contact with Pt electrodes； d) schematic of poling effects on band structure (origin ： after poling, ∆ CPDs induced by irradiation are labeled by red areas)


19


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

Page 20 of 27

TABLES. Table 1. The analysis of the contribution to ∆CPD induced by UV irradiation (Pt electrodes situation) CPD

Pr/overallc

∆CPD

Pt (+)

Pt (-)

0.67 V

-1.46 V

Pt(+)

Pt(-)

1.17 V

-1.08 V

Pt(+)

Pt(-)

77%

96%

UV (RT) Darka (RT) UV (195℃b)

-0.50 V 0.09 V

-0.38 V -0.30 V 0.27 V

Dark (195℃)

-0.16 V

-0.04 V

-0.26 V

a

The dark values refer to the steady value after cutting off irradiation; bthe sample was heated to 195℃ for 15 mins and cooled down to RT for test; cthe ratio refers to UV-induced ∆CPD after depoled at 195℃ divided by original ∆CPD (all test at room temperature, assuming the complete vanishing of ferroelectric polarization in SBN-50 pellet after depoled at 195℃).

ASSOCIATED CONTENT Supporting Information Details about photoelectric test and photoelectrochemical test methods, poling effects on SBN50, flat-band position calculation, photocatalysis characterization, Debye screening length calculation, original KPFM images. This material is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Corresponding Authors


20

Page 21 of 27

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


*E-mail: [email protected].& [email protected]. Tel: 86-411-84379070. Fax: 86-411-84694447. Homepage: http://www.canli.dicp.ac.cn. Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes The authors declare no competing financial interest. ACKNOWLEDGMENT We thank Michel Dupuis for his careful revision of the manuscript. This work was supported by the National Natural Science Foundation of China (Grant No. 21090340) and the National Key Basic Research Program of China (973 Program, Grant No. 2014CB239403).


21


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

Page 22 of 27

REFERENCES

1.

Brody, P., Large Polarization-Dependent Photovoltages in Ceramic BaTiO3 + 5wt.%

CaTiO3. Solid State Commun. 1973, 12, 673-676. 2.

Glass, A. M.; von der Linde, D.; Negran, T. J., High‐Voltage Bulk Photovoltaic Effect

and the Photorefractive Process in LiNbO3. Appl. Phys. Lett. 1974, 25, 233-235. 3.

Brody, P. S.; Crowne, F., Mechanism for the High Voltage Photovoltaic Effect in

Ceramic Ferroelectrics. J. Electron. Mater. 1975, 4, 955-971. 4.

Brody, P. S., High Voltage Photovoltaic Effect in Barium Titanate and Lead Titanate-

Lead Zirconate Ceramics. J. Solid State Chem. 1975, 12, 193-200. 5.

Koch, W.; Munser, R.; Ruppel, W.; Würfel, P., Bulk Photovoltaic Effect in BaTiO3. Solid

State Commun. 1975, 17, 847-850. 6.

Heyszenau, H., Electron Transport in the Bulk Photovoltaic Effect. Phys. Rev. B 1978,

18, 1586-1592. 7.

Inoue, Y.; Sato, K.; Sato, K.; Miyama, H., Photoassisted Water Decomposition by

Ferroelectric Lead Zirconate Titanate Ceramics with Anomalous Photovoltaic Effects. J. Phys. Chem. 1986, 90, 2809-2810. 8.

Yang, W.; Yu, Y.; Starr, M. B.; Yin, X.; Li, Z.; Kvit, A.; Wang, S.; Zhao, P.; Wang, X.,

Ferroelectric Polarization-Enhanced Photoelectrochemical Water Splitting in TiO2–BaTiO3 Core–Shell Nanowire Photoanodes. Nano Lett. 2015, 15, 7574-7580.


22

Page 23 of 27

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


9.

Chao, C.; Ren, Z.; Zhu, Y.; Xiao, Z.; Liu, Z.; Xu, G.; Mai, J.; Li, X.; Shen, G.; Han, G.,

Self‐Templated Synthesis of Single‐Crystal and Single‐Domain Ferroelectric Nanoplates. Angew. Chem. Int. Ed. 2012, 51, 9283-9287. 10. Jimmy, C., Selective Deposition of Redox Co-catalyst(s) to Improve the Photocatalytic Activity of Single-domain Ferroelectric PbTiO3 Nanoplates. Chem. Commun. 2014, 50, 1041610419. 11. Burbure, N. V.; Salvador, P. A.; Rohrer, G. S., Photochemical Reactivity of Titania Films on BaTiO3 Substrates: Influence of Titania Phase and Orientation. Chem. Mater. 2010, 22, 58315837. 12. Burbure, N. V.; Salvador, P. A.; Rohrer, G. S., Photochemical Reactivity of Titania Films on BaTiO3 Substrates: Origin of Spatial Selectivity. Chem. Mater. 2010, 22, 5823-5830. 13. Nechache, R.; Harnagea, C.; Li, S.; Cardenas, L.; Huang, W.; Chakrabartty, J.; Rosei, F., Bandgap Tuning of Multiferroic Oxide Solar Cells. Nat. Photonics. 2015, 9, 61-67. 14. Li, S.; AlOtaibi, B.; Huang, W.; Mi, Z.; Serpone, N.; Nechache, R.; Rosei, F., Epitaxial Bi2FeCrO6 Multiferroic Thin Film as a New Visible Light Absorbing Photocathode Material. Small 2015, 11, 4018-4026 15. Goodman, A. M.; Rose, A., Double Extraction of Uniformly Generated Electron‐Hole Pairs from Insulators with Noninjecting Contacts. J. Appl. Phys. 1971, 42, 2823-2830. 16. Yang, J.; Yan, H.; Wang, X.; Wen, F.; Wang, Z.; Fan, D.; Shi, J.; Li, C., Roles of Cocatalysts in Pt–PdS/CdS with Exceptionally High Quantum Efficiency for Photocatalytic Hydrogen Production. J. Catal. 2012, 290, 151-157.


23


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

Page 24 of 27

17. Wang, X.; Xu, Q.; Li, M.; Shen, S.; Wang, X.; Wang, Y.; Feng, Z.; Shi, J.; Han, H.; Li, C., Photocatalytic Overall Water Splitting Promoted by an α–β Phase Junction on Ga2O3. Angew. Chem. Int. Ed. 2012, 51, 13089-13092. 18. Chen, S.; Qi, Y.; Hisatomi, T.; Ding, Q.; Asai, T.; Li, Z.; Ma, S. S. K.; Zhang, F.; Domen, K.; Li, C., Efficient Visible‐Light‐Driven Z‐Scheme Overall Water Splitting Using a MgTa2O6−xNy/TaON Heterostructure Photocatalyst for H2 Evolution. Angew. Chem. Int. Ed. 2015, 54, 8498-8501. 19. Giocondi, J. L.; Rohrer, G. S., Spatial Separation of Photochemical Oxidation and Reduction Reactions on the Surface of Ferroelectric BaTiO3. J. Phys. Chem. B 2001, 105, 82758277. 20. Francombe, M., The Relation between Structure and Ferroelectricity in Lead Barium and Barium Strontium Niobates. Acta Crystallogr. 1960, 13, 131-140. 21. Jamieson, P.; Abrahams, S.; Bernstein, J., Ferroelectric Tungsten Bronze‐Type Crystal Structures. I. Barium Strontium Niobate Ba0.27Sr0.75Nb2O5.78. J. Phys. Chem. 1968, 48, 50485057. 22. Zhu, J.; Fan, F.; Chen, R.; An, H.; Feng, Z.; Li, C., Direct Imaging of Highly Anisotropic Photogenerated Charge Separations on Different Facets of a Single BiVO4 Photocatalyst. Angew. Chem. Int. Ed. 2015, 54, 9111-9114. 23. Shvartsman, V. V.; Kleemann, W.; Łukasiewicz, T.; Dec, J., Nanopolar Structure in SrxBa1−xNb2O6 Single Crystals Tuned by Sr∕Ba Ratio and Investigated by Piezoelectric Force Microscopy. Phys. Rev. B 2008, 77, 054105.


24

Page 25 of 27

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


24. Nelson, C. T.; Gao, P.; Jokisaari, J. R.; Heikes, C.; Adamo, C.; Melville, A.; Baek, S.-H.; Folkman, C. M.; Winchester, B.; Gu, Y., Domain Dynamics During Ferroelectric Switching. Science 2011, 334, 968-971. 25.

Fomenko, V. S., Handbook of Thermionic Properties. Springer US: 1966.

26. Qu, Y.-Q.; Li, A.-D.; Shao, Q.-Y.; Tang, Y.-F.; Wu, D.; Mak, C.; Wong, K.; Ming, N.B., Structure and Electrical Properties of Strontium Barium Niobate Ceramics. Mater. Res. Bull. 2002, 37, 503-513. 27. Venet, M.; Garcia, D.; Eiras, J. A.; M'Peko, J. C.; Amorín, H., Ferroelectric Properties of Lanthanum and Titanium Modified SBN Ceramic System. Phys. Status Solidi. (b) 2003, 238, 198-203. 28. Scott, J., High-Dielectric Constant Thin Films for Dynamic Random Access Memories (DRAM). Annu. Rev. Mater. Sci. 1998, 28, 79-100. 29. Matyjasek, K.; Kaczmarek, S. M.; Ivleva, L. I., Temperature Dependence of Domain Switching in Cr Doped Sr0.61Ba0.39Nb2O6 Single Crystals. Ferroelectrics 2012, 426, 97-102. 30. Akdogğan, E. K.; Safari, A., Thermodynamics of Ferroelectricity. Springer US: 2008; p 3-16. 31. Granzow, T.; Woike, T.; Wöhlecke, M.; Imlau, M.; Kleemann, W., Polarization-Based Adjustable Memory Behavior in Relaxor Ferroelectrics. Phys. Rev. Lett. 2002, 89, 127601.


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

Page 26 of 27

32. Schmidt, G.; Arndt, H.; Borchhardt, G.; von Cieminski, J.; Petzsche, T.; Borman, K.; Sternberg, A.; Zirnite, A.; Isupov, V. A., Induced Phase Transitions in Ferroelectrics with Diffuse Phase Transition. Phys. Status Solidi. (b) 1981, 63, 501-510. 33. Li, W.; Chen, A.; Lu, X.; Zhu, J., Collective Domain-Wall Pinning of Oxygen Vacancies in Bismuth Titanate Ceramics. J. Appl. Phys. 2005, 98, 024109. 34. Ang, C.; Yu, Z.; Cross, L., Oxygen-Vacancy-Related Low-Frequency Dielectric Relaxation and Electrical Conduction in Bi: SrTiO3. Phys. Rev. B 2000, 62, 228. 35. Li, W.; Su, D.; Zhu, J.; Wang, Y., Mechanical and Dielectric Relaxation in NeodymiumModified Bismuth Titanate Ceramics. Solid State Common. 2004, 131, 189-193. 36. Zenkevich, A.; Matveyev, Y.; Maksimova, K.; Gaynutdinov, R.; Tolstikhina, A.; Fridkin, V., Giant Bulk Photovoltaic Effect in Thin Ferroelectric BaTiO3 Films. Phys. Rev. B 2014, 90, 161409. 37. Hong, S. J.; Lee, S.; Jang, J. S.; Lee, J. S., Heterojunction BiVO4/WO3 Electrodes for Enhanced Photoactivity of Water Oxidation. Energy Environ. Sci. 2011, 4, 1781-1787. 38. Scaife, D., Oxide Semiconductors in Photoelectrochemical Conversion of Solar Energy. Sol. Energy 1980, 25, 41-54. 39. Rabe, K. M.; Ahn, C. H.; Triscone, J.-M., Physics of Ferroelectrics: a Modern Perspective. Springer Science & Business Media: 2007; Vol. 105. 40. Park, S.; Lee, C. W.; Kang, M.-G.; Kim, S.; Kim, H. J.; Kwon, J. E.; Park, S. Y.; Kang, C.-Y.; Hong, K. S.; Nam, K. T., A Ferroelectric Photocatalyst for Enhancing Hydrogen Evolution: Polarized Particulate Suspension. Phys. Chem. Chem. Phys. 2014, 16, 10408-10413.


26

Page 27 of 27

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


Table of Contents Graphic

A lateral symmetric Pt-/Sr0.5Ba0.5Nb2O6/Pt+ photocatalyst pellet was fabricated. It was found that this model catalyst exhibits anomalous photovoltaic (APV) effects and the photogenerated electrons and holes can be respectively collected on the opposite electrodes, as revealed by photo-assisted KPFM. This unique feature is ascribed to the formation of the parallel built-in electric field, which was induced by the poling process.


27

Dual Extraction of Photogenerated Electrons and Holes from a

Recommend Documents