Contact us

My IOPscience

Surface dominant photoresponse of multiferroic BiFeO3 nanowires under sub-bandgap illumination

This content has been downloaded from IOPscience. Please scroll down to see the full text. 2013 Nanotechnology 24 505710 (http://iopscience.iop.org/0957-4484/24/50/505710) View the table of contents for this issue, or go to the journal homepage for more

Download details: IP Address: This content was downloaded on 23/06/2014 at 09:11

Please note that terms and conditions apply.



Nanotechnology 24 (2013) 505710 (6pp)


Surface dominant photoresponse of multiferroic BiFeO3 nanowires under sub-bandgap illumination Kovur Prashanthi, Ravi Gaikwad and Thomas Thundat Department of Chemical and Materials Engineering, University of Alberta, Edmonton, AB, T6G 2V4, Canada E-mail: [email protected]

Received 21 September 2013, in final form 16 October 2013 Published 27 November 2013 Online at stacks.iop.org/Nano/24/505710 Abstract A surface dominant sub-bandgap photo-carrier generation has been observed in multiferroic BiFeO3 (BFO) nanowires, which is mainly attributed to the depopulation of surface states that exist within the bandgap. Mapping of surface potential using Kelvin probe force microscopy (KPFM) further supports the depopulation of surface states in BFO nanowires under sub-bandgap illumination. The mechanism of photovoltage generation in BFO nanowires is investigated by measuring the photoresponse with local illumination of visible laser pulses at different positions of the BFO nanowires. Interestingly, large photovoltage signals were observed when the laser spot was focused close to contact electrodes, showing a position dependent effect of photoresponse in the BFO nanowires. The sub-bandgap excitation of surface states in multiferroic nanowires offers potential new strategies for application in photovoltaic devices. S Online supplementary data available from stacks.iop.org/Nano/24/505710/mmedia (Some figures may appear in colour only in the online journal)

1. Introduction

is perpendicular to the in-plane component [14]. Electrostatic potentials which exist at the ferroelectric domain walls in BFO are attributed to their switchable photovoltaic effect [11]. Despite its many attractive properties, BFO as a material for solar energy conversion is undesirable due to its wide bandgap. It is well known that wide bandgap materials such as BFO have high density of surface states within the bandgap. Although surface effects do not influence the optical properties of bulk materials and thin films, they are dominant in 1D nanostructures because of their large surface-to-volume ratios. In this paper, we report the surface dominant photoresponse in BFO nanowires under sub-bandgap illumination. This sub-bandgap photoresponse in BFO nanowires is attributed to the surface photovoltage (SPV) phenomenon which originates from the depopulation of surface states that exist within the BFO bandgap. In addition to the presence of internal fields caused by the ferroelectric properties of BFO nanowires, the presence of a Schottky barrier at both ends of the nanowires, resulting from

One-dimensional (1D) semiconductor nanowires are emerging as promising candidates for developing next generation photovoltaic (PV) cells due to their strong light absorption properties, which are a result of large surface areas and direct charge transport paths along the geometry of the wires [1–5]. BiFeO3 (BFO) is an attractive material for photoconduction and photovoltaic applications because of its distinctive optical and electrical properties [6–9]. In ferroelectrics such as BFO, which lacks a center of symmetry, steady state photocurrents can exist in bulk as well as thin films under uniform illumination, a phenomenon called the bulk photovoltaic effect [6–9]. The photovoltaic effect in ferroelectric BFO is related to the light-induced generation of charge carriers and their separation under an internal electric field [10–13]. The photovoltaic effect is maximized when the electric field of the polarized light is parallel to the in-plane component of the ferroelectric polarization, and minimized when the field 0957-4484/13/505710+06$33.00


c 2013 IOP Publishing Ltd Printed in the UK & the USA

Nanotechnology 24 (2013) 505710

K Prashanthi et al

Figure 1. (a) FE-SEM image of a single BFO nanowire on Pt/Ti electrodes. (b) TEM image of an as-grown BFO nanowire. (c) High-resolution TEM image of an individual BFO nanowire showing the crystalline structure. Inset: selected area electron diffraction (SAED) pattern on the nanowire. (d) 3D morphology of an individual BFO nanowire device imaged with an AFM. Inset: the height profile of a BFO nanowire.

(Bruker Nano Inc., Santa Barbara, CA, USA) was used. A 75 W xenon arc lamp (Bentham TMc300) with an AM 1.5G filter was used to characterize the spectral response and the light intensity was calculated using a calibrated silicon photodiode. The dependence of the photocurrent on the wavelength was obtained by measuring the photocurrent generated in the device at 5 nm increments normalized by the photon flux of the source. A monochromatic probe based on a Bentham TMc300 single monochromator coupled with a 75 W xenon light source was introduced onto the sample during the test, giving coverage over the spectral range of 350–1100 nm. The output of the source was measured using a calibrated photodiode (Bentham).

inhomogeneous surface states and carrier concentration, may be responsible for the efficient separation of photogenerated carriers.

2. Experimental methods BFO nanowires were synthesized using a sol–gel based electrospinning method. Electrodes were first defined using photolithography and sputtering of titanium (Ti)/platinum (Pt) (10 nm/90 nm) on SiO2 grown Si substrates. BFO nanowires were directly deposited on an electrode patterned substrate using a sol–gel based electrospinning technique. A voltage of 15 kV was applied to the solution by means of a 25 A series DC–DC high voltage module through a stainless steel needle. A distance of 7.5 cm was maintained between the metallic jet and the rotating drum collector. The flow rate of the solution was fixed at 0.3 ml h−1 on the syringe pump and the electrospinning experiments were conducted at ambient temperature. The electrospun fibers were dried in a vacuum oven at 120 ◦ C for 1 h and further calcinated at 600 ◦ C for a duration of 2 h in an Argon gas atmosphere in order to create single phase BFO nanowires. A JAMP 9500F (JEOL) scanning electron microscope (SEM) in high vacuum mode was used for SEM imaging. An MFP-3D atomic force microscope (AFM) (Asylum Research, Santa Barbara, CA, USA) with tapping mode (in air) was used for recording the topography of the BFO nanowires. A silicon probe with Al reflex coating (spring constant: 42 N m−1 , resonant frequency: 300 kHz) was used for AFM imaging. For the KPFM and simultaneous topography measurements, a Dimension Icon AFM and Pt–Ir coated conductive probe (SCM-PIT) with spring constant of 2.8 N m−1 , resonant frequency of 75 kHz and radius of curvature about 20 nm

3. Results and discussion Field emission scanning electron microscopy (FE-SEM) images of a single BFO nanowire (figure 1(a)) revealed the diameter to be approximately 100 nm. Crystalline morphology and lattice constants of BFO nanowires were investigated using transmission electron microscopy (TEM) and are shown in figures 1(b) and (c). The size of BFO crystallites observed in figure 1(b) is approximately 25 nm. The crystallinity and phase purity of BFO nanowires were studied by x-ray diffraction (XRD) (supplementary data, figure S1 available at stacks.iop.org/Nano/24/505710/ mmedia). The spacing of the lattice fringes measured from high-resolution TEM (HRTEM) shown in figure 1(c) is ˚ which is also in agreement with the XRD pattern 2.78 A, ((110) plane). The selected area electron diffraction (SAED) pattern is presented in the inset of figure 1(c). The 3D AFM topographic image of the BFO nanowire device is shown in figure 1(d). The height and diameter of the BFO nanowire 2

Nanotechnology 24 (2013) 505710

K Prashanthi et al

observed from a cross-sectional profile (inset of figure 1(d)) are 90 nm and 100 nm respectively. The spectral response of single BFO nanowires for three scans is shown in figure 2. It is evident from figure 2 that the BFO nanowires show very strong photoresponse over the sub-bandgap region (visible to near infrared (IR)) as compared to the bandgap regions (Eg ∼ 2.55 eV). In the case of bulk semiconductor materials, photons are absorbed when their energy is higher than the bandgap. This makes it difficult to use semiconductors with a wide bandgap for light absorption when incident photon energy is lower than its bandgap. Although BFO is a wide bandgap (∼2.55 eV) material and shows significant photovoltaic effect within its bandgap [7, 9, 11], the BFO nanowires show a strong spectral response between visible (1.9 eV) and near IR (1.2 eV), which indicates a sub-bandgap optical absorption in these nanowires. A similar trend is observed for various sets of BFO nanowires synthesized under similar conditions (supplementary data, figure S2 available at stacks.iop.org/ Nano/24/505710/mmedia). In the case of many wide bandgap materials, a distribution of surface states exists in the forbidden energy gap which is filled up to the Fermi level. Hence photons having energy smaller than the bandgap can be absorbed by the surface states as a result of readily available electronic transitions. Since the nanowires have a high surface-to-volume ratio, sub-band transitions become dominant for optical absorption. Furthermore, surface states are bound and localized in positions which do not have well defined momentums. Decreasing the uncertainty of the carrier position leads to increased uncertainty in the momentum [15]. According to the k-selection rule for direct bandgap semiconductors, an electron at the bottom of the conduction band will recombine with the hole at the top of the valence band and may not hold strictly in sub-bandgap transitions at the surface, which may lead to an increased absorption in the sub-bandgap region of BFO nanowires. Figure 3(a) shows the transient photoresponse of the BFO nanowires illuminated at 635 nm (power: 5 mW) using a visible diode laser. This development of a photovoltage signal in BFO is determined by the fundamental properties of light absorption and surface built-in field within the BFO nanostructures. In bulk or thin film BFO, the fundamental mechanism behind the photovoltaic effect is the internally generated electric field which originates from ferroelectric domains. This internal electric field is responsible for the separation of photogenerated carriers. However, in nanostructured BFO, in addition to ferroelectric polarization, there is a significant contribution of surface potential. The energy bands at the surface of the BFO nanowire are bent due to trapped charges or surface defects, and the resulting surface potential can separate the photogenerated charge carriers. Significantly high photocurrent gains have been reported in SnO2 nanowires due to the built-in electric field induced by surface states which is responsible for spatial separation of photogenerated carriers [16]. The photogenerated carrier life times in BFO nanowire are extracted from the transient photoresponse curve and are shown in figures 3(b) and (c). Both rise and fall times

Figure 2. Photovoltage spectral response of a BFO nanowire under continuous illumination for three different scans where wavelength varied from 516 nm (2.4 eV) to 1033 nm (1.2 eV). The reported bandgap of BFO nanowires is 2.5 eV [7, 9, 11]. However, the BFO nanowire shows a strong spectral response over a wide range of wavelengths/energy between visible (1.9 eV) and near IR (1.2 eV), which indicates a strong optical absorption in the BFO nanowire by surface states. The inset shows the schematic of the BFO nanowire under illumination.

of the surface photovoltage pulse were found to proceed exponentially with time, with characteristic time constants obtained by fitting (figures 3(b) and (c)) of about 72 µs and 192 µs, respectively. For an ideal semiconducting device, the generation and recombination lifetimes should be the same. However, these lifetimes could be different when the generation and recombination mechanisms involve one or more energy levels in the bandgap [13]. This further supports the presence of surface states within the bandgap of BFO nanowires. The photoluminescence (PL) experiments on BFO nanowires also confirm the presence of surface states (supplementary data, figure S3 available at stacks.iop.org/Nano/24/505710/mmedia). In conventional photovoltaic cells, a barrier is set up by a p–n junction which separates photogenerated electrons and holes in opposite directions. The electrons and holes separated in this way are less likely to recombine with each other. However, in nanowire devices, the surface built-in potential separates the photogenerated electron–hole pairs making electron–hole recombination difficult. Nanowires offer charge separation mechanisms without needing a junction. Using density functional theory, it has been shown that tapered nanowires exhibit different degrees of quantum confinement along their length which can separate electrons and holes without any dopants [17]. The nanowire geometry is particularly beneficial for this step because it enables rapid radial or axial charge separation and efficient carrier collection through band conduction [18, 19]. Nanowires exhibit much higher carrier collection efficiency compared to nanoparticles as a result 3

Nanotechnology 24 (2013) 505710

K Prashanthi et al

Figure 3. (a) Transient surface photovoltage of the BFO device under visible pulsed laser irradiation. Photovoltage is measured under 0 V bias using pulsed illumination from a diode laser (wavelength: 635 nm; power: 5 mW). (b) The generation pulse and (c) the decay pulse with rise and fall times of 72 µs and 192 µs respectively. The black line shows the exponential fitting.

of faster band conduction rather than a trap-limited diffusion transport mechanism [20]. The sub-bandgap excitation of surface states can be explained using the surface photovoltage (SPV) approach [21, 22]. In the sub-bandgap SPV, variation of the surface charge takes place which changes the surface potential by depopulation and population of charge carriers. This situation is illustrated in figure 4(a). In surface state depopulation, illumination by photons with energy hν > EC − ET produces electron transitions from the surface state located at an energy ET into the conduction band, where they are swept quickly to the bulk because of the electric field due to the band bending. Therefore, negative charge on the surface is reduced and the band bending is decreased. In surface state population, illumination by photons with energy hν > ET − EV produces electron transitions from the valence band into a surface state located at an energy ET above the valence band maximum EV . Such transitions increase the negative charge on the surface and therefore the surface band bending is also increased (illustrated in figure 4(b)). Further experiments on BFO nanowires using ultraviolet photoemission spectroscopy (UPS) revealed the band edge position at 1.78 eV (supplementary data, figure S4 available at stacks.iop.org/Nano/24/505710/mmedia). The depopulation mechanism in BFO nanowires under sub-bandgap illumination is further investigated by mapping the surface potential of nanowires using Kelvin probe force microscopy (KPFM). The amplitude-modulated single pass KPFM technique in the lift mode was used to simultaneously image the topography and local contact potential distribution

Figure 4. (a) Energy band diagram showing depopulation surface states in BFO nanowires under sub-bandgap illumination. (b) Energy band diagram for population of surface states under sub-bandgap illumination.

of individual BFO nanowires. The topography and surface potential images for single BFO nanowires are shown in figures 5(a) and (b) respectively. The topography and the 4

Nanotechnology 24 (2013) 505710

K Prashanthi et al

Figure 5. (a) AFM topography of BFO nanowire used for surface potential mapping. (b) Surface potential along BFO nanowire with no sample bias (0 V) under the absence of illumination recorded with KPFM. (c) Surface potential mapping along the nanowire in the presence and absence of illumination with no sample bias (0 V). The surface potential decreased by around 80 mV under illumination which is consistent with depopulation of surface states resulting in decreased barrier height.

corresponding surface potential image of a single BFO nanowire is captured in ambient light conditions. As shown in figure 5(c), an overall decrease in the surface potential along the BFO nanowire was observed under sub-bandgap illumination (635 nm red laser) when no bias was applied to the metal contacts. In the dark, the surface potential of the BFO nanowire is about 180 mV whereas the local surface potential decreases significantly under illumination by 100 mV. This decrease in surface potential under sub-bandgap illumination is attributed to the depopulation of surface states. To investigate the type of surface states in BFO nanowires, an x-ray photoelectron spectroscopy (XPS) study was carried out (supplementary data, figure S5 available at stacks.iop.org/Nano/24/505710/mmedia). The XPS data supports the presence of mixed oxidation states (+2 and +3) of Fe in BFO nanowires thereby suggesting the presence of oxygen vacancy defects. The mechanism of photovoltage generation in BFO nanowires is investigated by measuring the photoresponse with local illumination of laser pulses at different positions

of the BFO nanowires. Figure 6(a) shows a schematic of the BFO nanowire device with the focused laser beam (635 nm) irradiated at different positions along the nanowire device. When the laser beam was focused at the two ends of the contacts, a photovoltage signal of opposite polarity was generated as shown in figure 6(b). On the other hand, a very small photovoltage signal was generated when the nanowire was irradiated near its midpoint. The focused laser inducing localized heating is attributed to the photothermoelectric effect and may contribute to the photovoltage [23, 24]. When the laser is illuminated near the midpoint of the nanowire, the photogenerated carriers are diffused randomly resulting in a negligible photovoltage signal. When the BFO nanowires are illuminated, photon energy is absorbed and charge carriers (electron–hole pairs) are generated. These carriers would diffuse randomly when there is a small electrical field in the nanowire. However, when the focused laser is illuminated near the metal–nanowire interface, the hot electrons might have enough energy to cross the Schottky barrier via tunneling 5

Nanotechnology 24 (2013) 505710

K Prashanthi et al

Figure 6. (a) Schematic of the BFO nanowire device with focused laser beam irradiated at its different sections. (b) Photoresponse upon pulsed irradiation of focused laser beam on three different portions of the nanowire device at zero bias.

or thermal emission resulting in significant photovoltage generation.

[2] Liao Z-M, Xu J, Zhang J-M and Yu D-P 2008 Appl. Phys. Lett. 93 023111 [3] Stelzner T, Pietsch M, Andr¨a G, Falk F, Ose E and Christiansen S 2008 Nanotechnology 19 295203 [4] Wallentin J et al 2013 Science 339 1057 [5] Peng K-Q and Lee S-T 2011 Adv. Mater. 23 198 [6] Belinicher V I and Sturman B I 1980 Sov. Phys.—Usp. 23 199 [7] Ji W, Yao K and Liang Y C 2010 Adv. Mater. 22 1763 [8] Ji W, Yao K and Liang Y C 2011 Phys. Rev. B 84 094115 [9] Yang S Y et al 2009 Appl. Phys. Lett. 95 062909 [10] Choi T, Lee S, Choi Y J, Kiryukhin V and Cheong S-W 2009 Science 324 63 [11] Yi H T, Choi T, Choi S G, Oh Y S and Cheong S-W 2011 Adv. Mater. 23 3403 [12] Yang S Y et al 2010 Nature Nanotechnol. 5 143 [13] Alexe M and Hesse D 2011 Nature Commun. 2 256 [14] Ho Sung J, Lee W-M, Hong Lee J, Chu K, Lee D, Moya X, Mathur N D, Yang C-H, Park J-H and Jo M-H 2013 NPG Asia Mater. 5 e38 [15] Hofmann P 2008 Solid State Physics: An Introduction (New York: Wiley–VCH) [16] Lin C-H, Chen R-S, Chen T-T, Chen H-Y, Chen Y-F, Chen K-H and Chen L-C 2008 Appl. Phys. Lett. 93 112115 [17] Wu Z, Neaton J B and Grossman J C 2008 Phys. Rev. Lett. 100 246804 [18] Garnett E C, Brongersma M L, Cui Y and Mcgehee M D 2011 Annu. Rev. Mater. Sci. 41 269 [19] Kayes B M, Atwater H A and Lewis N S 2005 J. Appl. Phys. 97 114302 [20] Law M, Greene L E, Johnson J C, Saykally R and Yang P 2005 Nature Materials 4 455 [21] Gatos H C 1973 J. Vac. Sci. Technol. 10 130 [22] Kronik L and Shapira Y 1999 Surf. Sci. Rep. 37 1–206 [23] St-Antoine B C, M´enard D and Martel R 2012 Nano Res. 5 73 [24] Xu X, Gabor N M, Alden J S, van der Zande A M and McEuen P L 2010 Nano Lett. 10 562

4. Conclusions We observed a remarkable sub-bandgap photovoltage phenomenon with BiFeO3 (BFO) nanowires due to the depopulation of surfaces states within the forbidden bandgap. The significant photoresponse in the visible and near IR wavelengths can be attributed to the dominant nature of surface states as a result of the high surface-to-volume ratio of the nanowires. The local heating induced by the illumination close to the nanowire–metal contact generates a significant photovoltage. The sub-bandgap voltage generation in BFO nanowire can be the basis for novel photovoltaic (PV) devices. An array of BFO nanowires should provide a key to the development of new generation solar cells, capable of cost-effective energy conversion on a large scale.

Acknowledgment This work was supported by the Canada Excellence Research Chairs (CERC) Program.

References [1] Tang J, Huo Z, Brittman S, Gao H and Yang P 2011 Nature Nanotechnol. 6 568 6

Surface dominant photoresponse of multiferroic BiFeO3 nanowires under sub-bandgap illumination.

A surface dominant sub-bandgap photo-carrier generation has been observed in multiferroic BiFeO3 (BFO) nanowires, which is mainly attributed to the de...
2MB Sizes 0 Downloads 0 Views