Size Effect on Optical and Photocatalytic Properties in BiFeO3

Feb 2, 2016 - (5, 6, 8, 9) BFO could potentially becom an alternative to the widely ... BFO nanoparticles with average particle sizes of 30, 50, 120, ...
0 downloads 0 Views 4MB Size
Subscriber access provided by ORTA DOGU TEKNIK UNIVERSITESI KUTUPHANESI

Article 3

Size Effect on Optical and Photocatalytic Properties in BiFeO Nanoparticles Xiaofei Bai, Jie Wei, Bobo Tian, Yang Liu, Thomas Reiss, Nicolas Guiblin, Pascale Gemeiner, Brahim Dkhil, and Ingrid Canero-Infante J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.5b09945 • Publication Date (Web): 02 Feb 2016 Downloaded from http://pubs.acs.org on February 15, 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.

The Journal of Physical Chemistry C 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 33

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

Size Effect on Optical and Photocatalytic Properties in BiFeO3 Nanoparticles Xiaofei Bai1, Jie Wei1,2, Bobo Tian1,3, Yang Liu1, Thomas Reiss4, Nicolas Guiblin1, Pascale Gemeiner1, Brahim Dkhil1, Ingrid C. Infante1,5∗ 1

Laboratoire Structures, Propriétés et Modélisation des Solides, CentraleSupélec,

CNRS-UMR8580, Université Paris-Saclay, 92295 Chatenay-Malabry Cedex, France 2

Electronic Materials Research Laboratory, Key Laboratory of Ministry of Education

& International Center for Dielectric Research, Xi'an Jiaotong University, Xi'an 710049, P. R. China 3

National Laboratory for Infrared Physics, Shanghai Institute of Technical Physics,

Chinese Academy of Sciences, Shanghai 200083, China 4

Laboratoire Mécanique des Sols, Structures et Matériaux, CentraleSupélec, CNRS-

UMR8579, Université Paris-Saclay, 92295 Chatenay-Malabry Cedex, France 5

Materials Research and Technology Department, Luxembourg Institute of Science

and Technology, 4422 Belvaux, Luxembourg

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

ABSTRACT: In this work, we study the influence of grain and particle size on the structural and optical properties of BiFeO3 (BFO) nanoparticles and resulting photocatalytic activity. Unexpectedly, the photocatalytic activity is found to decrease while the expected surface reaction area increases by decreasing the particles size. We show that while the global structure, polarization, particle morphology and band gap are only weakly altered, if at all, some optical features namely the Urbach energy and low-energy bands in the absorption spectra are substantially changed. We argue that these optical modifications related to defects and local distortions are mainly affected at the skin-layer that is inherent to oxides like BFO. By reducing the particle size of BFO nanoparticles, the skin-layer is thus altered which in turn changes the photocatalytic properties.

2 ACS Paragon Plus Environment

Page 2 of 33

Page 3 of 33

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

INTRODUCTION The use of ferroelectric materials to convert light into electrical1,2, mechanical3,4 or chemical5,6 energy has generated a broad interest these recent years for fundamental understanding reasons as well as application goals as they may be promising in photovoltaic, phototransduction or photocatalytic devices7. Ferroelectrics exhibit an intrinsic spontaneous polarization that is at the heart of the aforementioned photoinduced phenomena. Indeed, once the light excitation generates electrons-hole pairs, the polarization acts as an internal electric field favouring the charge carrier separation and thus reducing their recombination. Interestingly, the photogenerated charges can migrate to the material surface and serve as redox sources for degradation of contaminant molecules5,8 in wastewater treatment as well as for water splitting6,9 in sustainable hydrogen fuel cells. Among useful ferroelectrics for photo-induced applications, the so-called multiferroic BiFeO3 (BFO) is the most promising candidate because of its relatively small band gap (Eg ∼ 2.6-2.8 eV) in comparison to other classical ferroelectric oxides like BaTiO3, PbZrTiO3 or LiNbO3 (Eg > 3 eV) allowing to benefit from a wider part of the sunlight spectrum and its larger polarization value (P ∼ 100 µC/cm2)10, ensuring a more efficient separation of the photogenerated charge carriers. While intensively studied in thin film form for its ferroelectric, magnetic and magnetoelectric properties10 as well as most recently for its photovoltaicity11, there is an increasing interest for photocatalytic and photolysis processes under visible light irradiation using this material5,6,8,9. BFO could potentially becom an alternative to the widely investigated photocatalytic material TiO2 which suffers from its large band gap allowing to use only ∼ 4 % of the solar spectrum12. However, to make BFO a good photocatalyst requires enhancing the surface area and reactivity by simultaneously keeping or even increasing its absorption capability 3 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

in order to guarantee an efficient charge transfer to surrounding foreign molecules and participate in oxidation-reduction processes leading to molecule decomposition. It has been demonstrated that using different BFO particles, from nanosized to core shells structures, can indeed be beneficial for enhancing the photocatalytic activity. The nanosize shape lowers the band gap13 which might cause a larger absorption of sunlight. Moreover, the particle morphology is also a contributing parameter as it has been shown on micron-sized BFO particles14 because of the more efficient photoabsorption of {111}-cubic like facets15. Core-shell nanostructures based on BFO coated with TiO2 have been also used to enhance surface reactivity16. However, despite growing research in this field, the coupling between nanosize shaping and optical and photocatalytic responses in BFO has been not demonstrated yet. Here, by investigating at room temperature BFO nanopowders with sizes ranging from 30 to 190 nm, we show that, in contrast to common belief, the photocatalytic activity decreases with the particle size reduction. Such unexpected behaviour is caused by the decrease of the photoabsorption efficiency of BFO particle when their size is reduced although the global properties (morphology, polarization, average strain and energy gap) are only weakly altered. The depressed optical properties are explained by trapped defects and local distortions that are mainly altered within the skin-layer of BFO which is known to have its own properties and to be an intrinsic property of many other oxides with ABO3 perovskite structure17,18.

EXPERIMENTAL SECTION

BFO nanoparticles with average particle size of 30, 50, 120 and 190 nm have been synthesized using conventional wet chemical methods. In this process, due to esterification between tartaric acid and ethylene glycol used as chelating agents, a 3D

4 ACS Paragon Plus Environment

Page 4 of 33

Page 5 of 33

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

network structure is formed allowing to obtain high quality powder samples. The asobtained dried precursor solution was then calcined under controlled heating conditions at temperatures ranging from 450 °C to 600 °C to tune the particle sizes. The phase purity and crystal structure of BFO powders were characterized by X-ray diffraction (XRD) on a Bruker D2 Phaser diffractometer using Cu Kα radiation (λ = 1.5406 Å), theta-theta configuration. Data collected with angular step size of 0.01° and scanning speed being 0.01°/s, with 5 sec at every step point, for structural refinement procedure to determine crystallite size. The size and morphology of the BFO particles were determined from the imaging of disperse nanopowders on carbon grids using a Zeiss Leo 1530 Gemini field-emission scanning electron microscopy (FE-SEM). To access the optical response and its features, room temperature UltraViolet-Visible (UV-Vis) diffuse reflectance spectra on BFO powders were measured using a PerkinElmer spectrometer (Lambda 850) equipped with a Harrick Praying MantisTM diffuse reflectance accessory. The non-absorbing standard BaSO4 powder has been used as the total reflectance reference and the focused beam spot size was approximately of 2 mm2. The photocatalytic activity was evaluated by the degradation of Rhodamine-B dye (RhB) in aqueous solution under visible-light irradiation (500 W Xe-lamp with a cut-off filter allowing λ > 420 nm). The temperature reaction was kept at room temperature to avoid any thermal catalytic effect. The degradation efficiency of RhB was evaluated recording the transmittance spectra (a PerkinElmer spectrometer, Lambda 950) of the retrieved centrifuged samples. The intensity of the absorption peak of RhB (at around 553 nm) is expected to be directly related to the molecule concentration through Beer-Lambert equation19. Practically, we extract the maximum intensity of the absorption peak (A) after different illumination processes and compare this value to the corresponding one for

5 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

its initial state (A0), which is expected to be proportional to the RhB concentration in the solution (C after different exposures and C0 for the initial state), thus from these results, we obtain A/A0 ~ C/C0, leading us to determine the molecule concentration ratio and corresponding photodegradation efficiency (C0-C)/C0.

RESULT AND DISCUSSION

Figure 1 depicts the electronic micrographs obtained by FE-SEM showing the size distribution and morphology of BFO particles depending on the synthesis temperature.

Figure 1 FE-SEM images of BFO powders with average nanoparticles of (a) ∼ 50 nm (b) ∼ 190 nm. The corresponding inset figures are selected focused areas depicting that similar surface morphology is observed for different particle sizes.

6 ACS Paragon Plus Environment

Page 6 of 33

Page 7 of 33

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

The particle size changes on average from ∼ 30 nm to ∼ 190 nm when the synthesis temperature is enhanced from 450 °C to 600 °C, as reported in Table 1.

Table 1 Sample name of BFO nanoparticles and their corresponding synthesis temperature, particle size, crystallite size, lattice parameters, cc/ac ratio, cubic-like distortion, local inhomogeneous distortion and inhomogeneous strain

ε values.

The particle distribution is similar from one sample to the other and varies between +/- ∼ 20 % with respect to the average particle size (see Table 1). A closer inspection of each particle (see the inset in Figure 1) also indicates that the particles have similar morphology with a faceted shape whatever the synthesis temperature was. Thus, here the morphology as an influent parameter for the optical properties can be ruled out. Figure 2 displays the XRD patterns for BFO powders with different average particle sizes at room temperature showing pure phase BFO free of any parasitic phase.

7 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

Figure 2. Fine scanning XRD patterns of different BFO particle size powders: (a) 30 nm, (b) 50 nm, (c) 120 nm, and (d) 190 nm.

Using JANA2006 software20, Rietveld refinements from the X-ray diffraction patterns indicate that the rhombohedral symmetry with R3c space group is maintained down to the lower particle size. The analysis reveals that the cubic-like lattice parameters ac and cc are only weakly affected, if any, with a cc/ac ratio and local inhomogeneous distortion (see Table 1) varying from 1.0148 to 1.0149 and from 1.46% to 1.47% respectively when the particle size is decreased from 190 nm to 30 nm, respectively. Therefore, the polarization which is directly proportional to the squareroot of the distortion is believed to remain almost unchanged in the studied size range. Using the Bragg peak widths extracted from the Rietveld analysis and applying the Williamson-Hall method21, we can access the crystallite size as well as the so-called inhomogeneous strains or local distortions arising from defects including vacancies, 8 ACS Paragon Plus Environment

Page 8 of 33

Page 9 of 33

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

dislocations or interfaces, among others. Inspecting Figure 3, we observe a clear broadening of (104) and (110) peaks as size decreases13. Other families of planes (and peaks) confirm the existence of similar size and strain effects. On the different intensity of (104) and (110) peaks, the coalescence of crystals related to the crystalline growth mechanism for larger nanoparticles is at the origin of the observation of a preferential plane orientation. It is worth mentioning that the crystallite size coincides with the size of the smallest particles while the bigger particles are constituted of some crystallites (Table 1). By decreasing the crystallite/particle size, the inhomogeneous strain corresponding to local distortions related to defects rises up by three times from the biggest to the smallest particles as reported in Table 1. Downscaling BFO particles results then in an enhancement of the amount and/or strength of defects and their corresponding local strains.

Figure 3. Magnified XRD patterns Figure 2 around (104) and (110) peaks for different size BFO particles (bottom to top, particle size is 30 nm, 50 nm, 120 nm and 190 nm).

9 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

Let us now focus on the time dependent photo-degradation of the RhB dye under visible light irradiation while in solution with BFO nanoparticles of different size. Figure 4a shows the typical intensity variation of the absorbance peak of RhB under visible-light irradiation that decreases with increasing the time indicating that RhB has been decomposed by BFO with reaction time. In order to evaluate the degradation efficiency of RhB, the maximum intensity ratio C/C0 is plotted in Figure 4b, in which C0 and C are the maximum intensity in the initial state (0 h) and later (1-4 h) absorption spectra of RhB under visible light irradiation.

Figure 4 (a) Absorption spectra of RhB under visible-light irradiation in solution with ∼ 190 nm BFO nanopowders. (b) Photodegradation of RhB as a function of the illuminated time when in solution with BFO nanopowders of different particle sizes. The curves are fitting results using the first order kinetic equation C/C0 = exp (-kt)

10 ACS Paragon Plus Environment

Page 10 of 33

Page 11 of 33

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

Compared with the blank test for equivalent time dependences obtained from the light induced degradation of pure RhB in aqueous solution, it can be seen that the RhB degradation becomes efficiently accelerated by up to 50 % by using BFO particles. However, unexpectedly, the degradation is found to be less efficient once the particle size is reduced. This is clearly seen in Table 2 through the continuous decrease of (C0-C)/C0 ratio, i.e. the degradation efficiency, with particle size decreasing.

Table 2 Sample name and corresponding values of photodegradation efficiency (C0-C)/C0 (deduced after 4 h of irradiation), kinetic rate constant k, absorption onset energy EAbs, Urbach energy EU, and absorption area below 2.95 eV.

The degradation efficiency is strongly lowered by over 30 % from the 190 nm sample to the 30 nm one. A first order kinetic reaction law i.e. C/C0 = exp (-kt) where k is the kinetic rate and t the time, is used to fit the data in Fig. 3b. Table 2 reports the constant k obtained for the different particle size. It appears that the faster reaction occurs for the bigger particles despite a lower surface area is exposed to the RhB dye. Reducing the particle size is generally beneficial for photocatalysis, as it leads to quadratic growth of the specific surface area and thus enhances the surface reaction. 11 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

However, it is not always the case that the smaller the particle size, the higher the efficiency. Nevertheless, here, in contrast to previous reports14, the shape of the particles is similar and the lowering of the degradation efficiency cannot be attributed to better light absorption from some specific facets with higher surface energy. Besides, the polarization, which is expected to only vary slightly, if at all, with the change of the nanocrystallite size, cannot be considered as a parameter that could affect charge carrier separation. In contrast, some defects and their corresponding local strains revealed by the broadening of the Bragg peaks do exist and their amount and/or strength increases with decreasing particles size. These defects, if localized at the surface, may indeed favor the recombination of electron-hole charge carriers reducing the reaction activity between BFO and the RhB molecules. Moreover, the defects localized in the inner part of the particle can trap the charge carriers created within the particle, limiting their mean free path and thus their migration towards the surface. In addition, the local distortions surrounding/associated to these defects can also affect the electronic structure, by shifting or lifting the degeneracy of existing energy levels or by creating new extra ones that can be localized within the band gap. Additionally, from Figure 5b, although the total absorption is similar for different size samples, a great number of defects in smaller size sample could lead to greater scattering of carriers. This reduces the photoefficiency, which will lead to a change on the light absorption properties of BFO particles. It is worth mentioning that their nature cannot be straightforwardly determined and further efforts are required to answer to this question. Here, we note that the more plausible origin of these defects are the oxygen vacancies13, as they are usually expected in oxides as point defects, other possibilities include like the so-called skin-layer22,23. This skin-layer having its own phase transitions is believed to be the site of trapping centers for charge carriers

12 ACS Paragon Plus Environment

Page 12 of 33

Page 13 of 33

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

and interestingly it is associated to strong local distortions. Increasing the surface/volume ratio by decreasing the particle size might make the skin-layer have an increased role on (1) the average increase of the local distortions, becoming more relevant in smaller particles (Table 1), and (2) the development of more local defects, i.e. more efficient trapping centers for the photogenerated charge carriers, as observed from the decrease of the degradation efficiency and the rate constant in such particles (Table 2). In order to get further insights into the possible mechanisms involved in the decrease of the photocatalytic activity with particle size reduction, we studied the modifications of the optical properties related to particle size changes in BFO. Figure 5a shows the reflectance spectra of different size BFO at room temperature, in which a highly noticeable variation of reflectance is observed at ∼600 nm for all the samples, thus indicating light absorption within the visible range. The absorption spectra of BFO nanoparticles are analyzed through the corresponding Kubelka-Munk function F(R), which is calculated using F(R)=(1-R)2/2R, where R is the experimental reflectance referred to the standard BaSO4. The different F(R) curves are presented in Figure 5b as a function of the photon energy. Different low energy onsets can be seen in the inset of Figure 5b.

13 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

Figure 5 (a) Room temperature UV-Vis reflectance spectra of different size BFO samples. (b) Room temperature UV-Vis absorption spectra of different size BFO samples derived from the diffuse reflectance (R) spectra using Kubelka-Munk functions F(R). The inset shows the low energy crystal-field transition bands for samples with different average particle sizes. (c) Schematic representation of the crystal-field effect lowering the symmetry from cubic octahedral Oh to rhombohedral C3v environment and the energy levels involved in BFO. Ec and Ev indicate the expected bottom conduction and top valence band, respectively. EU represents the energy level of defects.

Because of the spin-charge-lattice couplings, BFO absorption spectrum reveals a complex and puzzling electronic structure24-30 due to superimposition of charge transfer excitation bands associated with interatomic transitions between O-2p, Fe-3d, Bi-6s and Bi-5p levels and absorption bands (d-d bands transitions) originating from on-site Fe3+ (3d5 high-spin HS) crystal-field transitions linked to its rhombohedral symmetry environment (Figure 5c). The typical t2g3eg2 high spin configuration of Fe3+ ions in the cubic octahedral environment transforms into a11e2e2 electronic configuration in the rhombohedral environment. In addition, the lowering of the local symmetry by defects/strains can lead to shift of the energy levels, lift of degeneracy (e 14 ACS Paragon Plus Environment

Page 14 of 33

Page 15 of 33

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

orbitals can split) and/or insertion of localized in-gap energy levels and thus adds some extra absorption features. Figure 5b shows at a first glance that the global shape of the absorption does not change whatever the particle size. Above ∼ 2.2 eV, the absorption jumps and shows two broad bands at ∼ 3.4 eV and ∼ 5.2 eV. On its low-energy side, the p-d charge-transfer band at ∼ 3.4 eV is affected by the 4T2g crystal-field electronic level localized at ∼ 3.3 eV as well as additional bands previously reported at ∼ 2.8 eV, 2.9 eV and 3.1 eV by second-harmonicgeneration measurements29 and explained by electronic states and multimagnon couplings. The band at ∼ 5.2 eV can be assigned to (p+d)-p charge-transfer31 and the tail on the left-side energy at ~4.5eV is explained by p-d charge-transfer excitations27. While the charge band gap is expected at ∼ 2.6-2.8 eV25,27, a narrow d-d band absorption at ∼ 2.55 eV corresponding to (4Eg-4Ag) crystal field levels and a shoulder at ∼ 2.4 eV in the linear spectra above ∼ 2.2 eV due to d-d charge-transfer (and possibly to double-excitons) can be observed29. Actually in these conditions i.e. absence of a sharp absorption edge, the determination of the band gap is not straightforward. Typically, the band gap is determined using Tauc plot according to the relation (F(R)hν)n=A(hν-Eg), using the experimentally derived F(R) curves with R the diffuse reflectance, h the Plank’s constant, ν the photon frequency, A being a constant value, Eg the band gap and n the exponential coefficient related to the band gap transition nature (n = 2 for direct band gap). Using the direct band gap approach with our set of data provides an absorption onset energy of EAbs∼ 2.2 eV (here, EAbs substitutes Eg in general expression of the Kubelka Munk relation), in agreement with previous reports13 but lower than the usual Eg values obtained from single crystals or thin films29. Even though this absorption onset energy is lower than that seen in crystalline materials, it is relatively close to the absorption onset noted there. Our 15 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

results show a very weak decrease, if any, with the particle size reduction from 190 nm to 30 nm, as summarized in Table 2. From the different Kubelka-Munk curves, we can determine the Urbach energy EU related to the variation of the energy levels close to the conduction and/or valence bands, being affected by different sources of disorder, e.g. point defects or strains. Using Urbach analysis through lnF(R)=hν/EU relationship, EU is found to vary from 0.11 eV to 0.20 eV for 190 nm to 30 nm particle size, respectively (Table 2). These latter EU values are in very good agreement with x-ray photoelectron spectroscopy measurements30 on BFO revealing the presence of defect trapped states at 0.2 eV below the conduction band that have been attributed to defects at the particle boundaries. Below ∼ 2.2 eV, two other absorption onsets can be seen at ∼ 1.4 eV (although localized just at the limit of our measurement range) and ∼ 1.9 eV (see inset Figure 5b), being attributed to 6A1g to 4T1g and 6A1g to 4T2g crystal-field transitions, respectively (Figure 5c). Strikingly and in contrast to the other absorption bands, these low-energy crystal-field bands are obviously affected by the decrease of the particle size; the positions are red-shifted and the intensity is also decreased (within the detection limits), as shown in the inset of Figure 5b. These bands have been demonstrated to be magnetically sensitive and, more interestingly, by using an oscillator strength analysis involving electron-phonon coupling28, it has been also shown that their temperature dependent behaviour changes at 150 K, which remarkably corresponds to the temperature value of one of the transitions reported for the skin-layer of BFO23. We note that the magnetic changes have been previously attributed to the skin-layer at this peculiar temperature23. Moreover, as such a skinlayer is also known to be the site of trapped states for charge carriers, we propose here that the defect states evidenced from our Urbach analysis could be mainly located in the skin-layer. Therefore, both that the Urbach energy and low-energy bands changes 16 ACS Paragon Plus Environment

Page 16 of 33

Page 17 of 33

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

when the particle size is varied could suggest that the skin-layer and the defects trapped in it plays a major role in modifying the optical response of BFO particles. However, we should consider other mechanisms accounting for these different surface effects. As a matter of fact, the variation of the polarization state near the surface or through the nanoparticles, related to the ferroelectric domain configuration, will lead to a polarization gradient. Secondly, these nanoparticles should present Stern layers developed in the solution during the photocatalytic experiments that should influence the path and recombination of carriers within the solution. To unveil the contribution of these effects, further investigations are in process. Figure 6 shows the absorption area above 420 nm (i.e. energies below 2.95 eV) determined using F(R) (Figure 5b) and corresponding to the total optical energy absorbed by BFO particles in our photocatalysis experiment in comparison to the degradation efficiency.

Figure 6 The absorption area extracted from F(R) above 420 nm (i.e. below 2.95 eV) and photodegradation efficiency (C0-C)/C0 as a function of the different average particle size.

17 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

It appears that the absorption by BFO particles decreases with the reduction of the particle size. Therefore, the decrease of degradation efficiency of RhB molecules with particle size reduction can be simply explained by the decrease of BFO capability to absorb the light once the particle size is reduced. We believe that the ability of BFO to absorb is also related to its skin-layer. Such skin-layer which is the site of defects and local distortions can indeed affect the optical absorption of BFO by shifting and/or splitting energy levels because of local symmetry reduction and inserting extra in-gap levels that can trap the charge carriers. By decreasing the particle size, the average shape, polarization, distortion and band gap are only weakly altered, if any, while the skin-layer is believed to vary substantially (Urbach energy and low-energy bands) leading to the observed decrease of degradation capabilities of BFO nanoparticles.

CONCLUSIONS

In conclusion, a detailed structural, microstructural, optical and photocatalytic investigation on different particle sized BFO particles is conducted. While the particle size is reduced and thus the surface reaction is increased, the photocatalytic activity of BFO is found to decrease in the investigated particle size range. Such unexpected behaviour is explained by changes occurring in the optical response of BFO due to the existence of trapped defects and local distortions mainly altered at the skin-layer. Decreasing the particle size affects the skin-layer response while the global properties (polarization, strain, band gap) are only weakly affected. The skin-layer, being an inherent property of many other compounds, should be then taken into account for further investigation and use in future photocatalysis applications of other functional oxides with the ABO3 perovskite structure like BiFeO3.

AUTHOR INFORMATION

18 ACS Paragon Plus Environment

Page 18 of 33

Page 19 of 33

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

Corresponding Author

Tel: +33 1 41 13 16 23 E-mail: [email protected] Notes The authors declare no competing financial interest.

ACKNOWLEDGEMENTS This work has been supported by the French Ile-de-France region through the DIM OxyMORE program funding, the French ANR program NOMILOPS (ANR-11BS10-016-02) project and the National Science Foundation of China (NSFC No. 51272204), Specialized Research Fund for the Doctoral Program of Higher Education (No. 20110201120003). X.F. Bai, J. Wei, B. B. Tian and Y. Liu also wish to thank the China Scholarship Council (CSC) for funding their stay in France.

REFERENCES

1.

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, J. W. et al. Above-Bandgap Voltages from Ferroelectric Photovoltaic Devices. Nature Nanotech. 2010, 5, 143-147. 2.

Alexe, M. Local Mapping of Generation and Recombination Lifetime in

BiFeO3 Single Crystals by Scanning Probe Photoinduced Transient Spectroscopy. Nano Lett. 2012, 12, 2193-2198. 3.

Kundys, B.; Viret, M.; Colson, D.; Kundys, D. O. Light-Induced Size Changes

in BiFeO3 Crystals. Nature. Mater. 2010, 9, 803-805.

19 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

4.

Lejman, M.; Vaudel, G.; Infante, I. C.; Gemeiner, P.; Gusev, V. E.; Dkhil, B.;

Ruello, P. Giant Ultrafast Photo-Induced Shear Strain in Ferroelectric BiFeO3. Nature. Comm. 2014, 5, 4301. 5.

Gao, F.; Chen, X.; Yin, K.; Dong, S.; Ren, Z.; Yuan, F.; Yu, T.; Zou, Z.; Liu, J.

M. Visible‐Light Photocatalytic Properties of Weak Magnetic BiFeO3 Nanoparticles. Adv. Mater. 2007, 19, 2889-2892. 6.

Deng, J.; Banerjee, S.; Mohapatra, S. K.; Smith, Y. R.; Misra, M. Bismuth

Iron Oxide Nanoparticles as Photocatalyst for Solar Hydrogen Generation from Water. J. Fundam. Renew. Energy. Appl. 2011, 1, 1-10. 7.

Kreisel, J.; Alexe, M.; Thomas, P. A. A Photoferroelectric Material Is More

Than the Sum of Its Parts. Nature. Mater. 2012, 11, 260. 8.

Zhang, Y.; Schultz, A. M.; Salvador, P. A.; Rohrer, G. S. Spatially Selective

Visible Light Photocatalytic Activity of TiO2/BiFeO3 Heterostructures. J. Mater. Chem. 2011, 21, 4168-4174. 9.

Lou, X. W.; Wang, Y.; Yuan. C.; Lee, J. Y.; Archer, L. A. Template‐Free

Synthesis of SnO2 Hollow Nanostructures with High Lithium Storage Capacity. Adv. Mater. 2006, 18, 2325-2329. 10.

Yang, Y.; Infante, I. C.; Dkhil, B.; Bellaiche, L. Strain Effects on Multiferroic

BiFeO3 Films. C. R. Phys. 2015, 16, 193-203. 11.

Choi, T.; Lee, S.; Choi, Y.; Kiryukhin, V.; Cheong, S. W. Switchable

Ferroelectric Diode and Photovoltaic Effect in BiFeO3. Science. 2009, 324, 63-66. 12.

Schneider, J.; Matsuoka, M.; Takeuchi, M.; Zhang, J.; Horiuchi, Y.; Anpo, M.;

Bahnemann, D. W. Understanding TiO2 Photocatalysis: Mechanisms and Materials. Chem. Rev. 2014, 114, 9919-9986.

20 ACS Paragon Plus Environment

Page 20 of 33

Page 21 of 33

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

13.

Mocherla, P. S. V.; Karthik, C.; Ubic, R.; Rao, M. S. R.; Sudakar, C. Tunable

Bandgap in BiFeO3 Nanoparticles: The Role of Microstrain and Oxygen Defects. Appl. Phys. Lett. 2013, 103, 022910. 14.

Li, S.; Lin, Y. H.; Zhang, B. P.; Wang, Y.; Nan, C. W. Controlled Fabrication

of BiFeO3 Uniform Microcrystals and Their Magnetic and Photocatalytic Behaviors. J. Phys. Chem. C. 2010, 114, 2903-2908. 15.

Fei, L.; Yuan, J.; Hu, Y.; Wu, C.; Wang, J.; Wang, Y. Visible Light

Responsive Perovskite BiFeO3 Pills and Rods with Dominant {111} C Facets. Cryst. Growth Des. 2011, 11, 1049-1053. 16.

Li, S.; Lin, Y. H.; Zhang, B. P.; Li, J. F.; Nan, C. W. BiFeO3/TiO2 Core-Shell

Structured Nanocomposites as Visible-Active Photocatalysts and Their Optical Response Mechanism. J. Appl. Phys. 2009, 105, 054310. 17.

Ravy, S.; Bolloc’h, D. L.; Currat, R.; Fluerasu, A.; Mocuta, C.; Dkhil, B.

SrTiO3 Displacive Transition Revisited Via Coherent X-Ray Diffraction. Phys. Rev. Lett. 2007, 98, 105501. 18.

Xu, G.; Gehring, P.; Stock, C.; Conlon, K. The Anomalous Skin Effect in

Single Crystal Relaxor Ferroelectric PZN-x PT and PMN-x PT. Phase Transitions

2006, 79, 135-152. 19.

A=ε⋅C⋅l, where A is the absorbance of the used solution, ε is the molar

attenuation coefficient, C is the corresponding concentration and l is the pathlength on the solution. 20.

Petricek, V.; Dusek, M.; Palatinus, L. Crystallographic Computing System

Jana2006: General Features. Z. Kristall. 2014, 229, 345-352. 21.

Williamson, G. K.; Hall, W. H. X-Ray Line Broadening from Filed

Aluminium and Wolfram. Acta. Metall. 1953, 1, 22-31.

21 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

22.

Martí, X.; Ferrer, P.; Herrero. Albillos, J.; Narvaez, J.; Holy, V.; Barrett, N.;

Alexe, M.; Catalan, G. Skin Layer of BiFeO3 Single Crystals. Phys. Rev. Lett. 2011, 106, 236101. 23.

Jarrier, R.; Marti, X.; Herrero-Albillos, J.; Ferrer, P.; Haumont, R.; Gemeiner,

P.; Geneste, G.; Berthet, P.; Schülli, T.; Cevc, P.et al. Surface Phase Transitions in BiFeO3 Below Room Temperature. Phys. Rev. B 2012, 85, 184104. 24.

Neaton, J. B.; Ederer, C.; Waghmare, U. V.; Spaldin, N. A.; Rabe, K. M. First-

Principles Study of Spontaneous Polarization in Multiferroic BiFeO3. Phys. Rev. B

2005, 71, 014113. 25.

Clark, S. J.; Robertson, J. Band Gap and Schottky Barrier Heights of

Multiferroic BiFeO3. Appl. Phys. Lett. 2007, 90, 132903. 26.

Basu, S.; Martin, L.; Chu, Y.; Gajek, M.; Ramesh, R.; Rai, R.; Xu, X.;

Musfeldt, J. Photoconductivity in BiFeO3 Thin Films. Appl. Phys. Lett. 2008, 92, 091905. 27.

Pisarev, R.; Moskvin, A.; Kalashnikova, A.; Rasing, T. Charge Transfer

Transitions in Multiferroic BiFeO3 and Related Ferrite Insulators. Phys. Rev. B 2009, 79, 235128. 28.

Xu, X.; Brinzari, T.; Lee, S.; Chu, Y.; Martin, L.; Kumar, A.; McGill, S.; Rai,

R.; Ramesh, R.; Gopalan, V. Optical Properties and Magnetochromism in Multiferroic BiFeO3. Phys. Rev. B 2009, 79, 134425. 29.

Ramirez, M.; Kumar, A.; Denev, S.; Podraza, N.; Xu, X.; Rai, R.; Chu, Y.;

Seidel, J.; Martin, L.; Yang, S. Y. Magnon Sidebands and Spin-Charge Coupling in Bismuth Ferrite Probed by Nonlinear Optical Spectroscopy. Phys. Rev. B 2009, 79, 224106.

22 ACS Paragon Plus Environment

Page 22 of 33

Page 23 of 33

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

30.

Ramachandran, B.; Dixit, A.; Naik, R.; Lawes, G.; Rao, M. R. Charge

Transfer and Electronic Transitions in Polycrystalline BiFeO3. Phys. Rev. B 2010, 82, 012102. 31.

Chen, P.; Podraza, N.; Xu, X.; Melville, A.; Vlahos, E.; Gopalan, V.; Ramesh,

R.; Schlom, D.; Musfeldt, J. Optical Properties of Quasi-Tetragonal BiFeO3 Thin Films. Appl. Phys. Lett. 2010, 96, 131907.

23 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

Table of Contents (TOC) Image

24 ACS Paragon Plus Environment

Page 24 of 33

Page 25 of 33

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

The Journal of Physical Chemistry

ACS Paragon Plus Environment

10 (202)

20 30 40 50 60

2 Theta (deg) 70 80

120 nm

90

(d)

10

ACS Paragon Plus Environment

20

30

2 Theta (deg) 40

30 40

50

50

(208) (220) (10 10) (036) (128) (134) (026) (042) (404)

20

(211) (116) (018) (214)

10

(024)

90

(202)

80

(202)

(208) (220) (10 10) (036) (128) (134) (026) (042) (404)

(211) (116) (018) (214)

(024)

(113)(006)

(012)

Intensity (arb.units)

(208) (220) (10 10) (036) (128) (134) (026) (042) (404)

(211) (116) (018) (214)

(110)

(110)

(104)

(b)

(110)

70

(113)(006)

60

(104)

50

(012)

(202) (024)

(113)(006)

(104)

30 nm

Intensity (arb.units)

40

(208) (220) (10 10) (036) (128) (134) (026) (042) (404)

(c) 30

(211) (116) (018) (214)

20

(110)

10

(104)

(012)

(a)

(024)

(113)(006)

(012)

Intensity (arb.units)

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

Intensity (arb.units)

The Journal of Physical Chemistry Page 26 of 33

50 nm

60 70

60

2 Theta (deg)

70

80

80

90

2 Theta (deg) 190 nm

90

30

31

190 nm 120 nm 50 nm 30 nm

(110)

(104)

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

The Journal of Physical Chemistry

Intensity (arb. units)

Page 27 of 33

32

33

2 Theta (deg) ACS Paragon Plus Environment

34

1.0

Page 28 of 33

0h 1h 2h 3h 4h 190 nm

(a)

0.8 0.6 0.4 0.2 0.0 400

450

500

550

600

650

Wavelength (nm) 1.0

(b)

0.9 0.8

C/C0

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

Absorbance (arb.units)

The Journal of Physical Chemistry

RhB 30nm 50nm 120nm 190nm

0.7 0.6 0.5 0.4

0

1

2

Time (h)

ACS Paragon Plus Environment

3

4

Page 29 of 33

Energy (eV) 3

2

(a)

(b)

50 40 30 nm 50 nm 120 nm 190 nm

30 20

F(R) (arb.units)

60

4

F(R) (arb.units)

6 5

R (%)

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

The Journal of Physical Chemistry

10

1.4

1.5

1.6

1.7

1.8

1.9

2.0

2.1

30 nm 50 nm 120 nm 190 nm

2.2

Energy (eV)

200

300

400

500

600

700

800

900

2

Wavelength (nm)

(c)

ACS Paragon Plus Environment

3

4

Energy (eV)

5

6

The Journal of Physical Chemistry

0.50

Absorption Area

3.0 0.45 2.8 0.40 2.6 0.35 2.4 40

80

120

160

Particle Size (nm) ACS Paragon Plus Environment

200

0.30 240

(C0-C)/C0

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

Page 30 of 33

Page 31 of 33

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

The Journal of Physical Chemistry

Sample name

Synthesis  Mean  Crystalline  Temperature  Particle Size  Size (XRD) (°C) (SEM) (nm) (nm)

ac (Å)

cc (Å)

cc/ac

Local  inhomogeneous  Inhomogeneous  distortion strain  (cc‐ac)/cc

30 nm

450

327

28+/‐5

3.9452(5) 4.0040(1)

1.0148(9)

1.46(8)%

(24.74.8)10‐4

50 nm

500

488

39+/‐5

3.9443(5) 4.0030(4)

1.0149(7)

1.47(5)%

(14.62.5) 10‐4

120 nm

550

12022

46+/‐6

3.9450(4) 4.0042(7)

1.0150(1)

1.47(9)%

(8.43.0) 10‐4

190 nm

600

18938

61+/‐2

3.9443(6) 4.0034(3)

1.0149(8)

1.47(6)%

(8.22.8) 10‐4

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

Page 32 of 33

Sample  name

Photodegradation (C0‐C)/C0 (%)

Rate constant k (h‐1)

Absorption onset energy Eabs (eV)

Urbach Energy

30 nm

33%

0.12

2.18(1)

0.19(1)

2.40(1)

50 nm

41%

0.14

2.21(1)

0.16(1)

2.47(1)

120 nm

44%

0.16

2.22(1)

0.11(1)

2.58(1)

190 nm

46%

0.17

2.24(1)

0.11(1)

2.94(1)

ACS Paragon Plus Environment

EU (eV)

Absorption  area

Page 33 of 33

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

TOC image 63x47mm (300 x 300 DPI)

ACS Paragon Plus Environment