- Email: [email protected]
anatase phase junctions fabricated in coaxial nanorod arrays
Nano Energy (2015) 15, 406–412
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/nanoenergy
RAPID COMMUNICATION
Photovoltaic device based on TiO2 rutile/anatase phase junctions fabricated in coaxial nanorod arrays Pengli Yana,b, Xiang Wangb,c, Xiaojia Zhengb,c, Rengui Lib, Jingfeng Hanb,c, Jingying Shib, Ailong Lib,c, Yang Gana,n, Can Lib,nn a
School of Chemical Engineering & Technology, Harbin Institute of Technology, Harbin 150001, China State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences; Dalian National Laboratory for Clean Energy, Dalian 116023, China c University of Chinese Academy of Sciences, Beijing 100049, China b
Received 16 March 2015; received in revised form 4 May 2015; accepted 8 May 2015 Available online 16 May 2015
KEYWORDS
Abstract
Prototype PV device; Phase junction; Titania; Coaxial nanorod
Phase-junctions formed in mixed crystalline phases of semiconductor based photocatalysts (TiO2 and Ga2O3) shows enhanced photocatalytic activity owing to efficient separation of photogenerated charges. However, the phase junction effect on charge separation has not been directly verified yet. In this study, we fabricated a prototype photovoltaic device based on TiO2 rutile/anatase coaxial nanorod arrays (NRAs) to demonstrate charge separation at the interface of polymorphic crystal phases. The device—FTO/rutile NRAs/anatase/ITO shows an ordinary photovoltaic response (open-circuit voltage Voc: 154 mV, short-circuit current density Jsc: 1.76 mA/cm2), contrasting with photo resistor behavior of two TiO2 single phase devices of core–core (FTO/rutile NRAs/ITO) and shell–shell (FTO/anatase/ITO). Experimental evidences suggest that the built-in electric field at the interface of rutile/anatase phase junctions in the FTO/rutile NRAs/anatase/ITO device provides the direct driving force for efficient separation of photogenerated charges. The demonstrated strategy of fabricating phase-junction photovoltaic devices may inspire further investigations on new device converting solar into electrical energy and highlight the key role of the phase junction interface in the properties of the mix-phase photocatalysts. & 2015 Published by Elsevier Ltd.
n
Corresponding author. Tel.: +86 45186413708. Corresponding author. Tel.: +86 41184379070. E-mail addresses: [email protected] (Y. Gan), [email protected] (C. Li).
nn
http://dx.doi.org/10.1016/j.nanoen.2015.05.005 2211-2855/& 2015 Published by Elsevier Ltd.
Photovoltaic device based on TiO2 rutile/anatase phase junctions
407
Introduction
Experimental section
Semiconductor heterojunction plays an important role in photogenerated charge separation of photocatalytic [1–3] and photovoltaic [4,5] systems for solar energy conversion. A heterojunction, usually formed between two different semiconductors [6–8] or semiconductors with different doping levels [9,10], provides the potential driving force for separating photoinduced charge carriers at the interface. In a similar fashion, many semiconductor materials exist in various crystalline phases—namely polymorphs [11–13]. The junctions formed between different phases of the same semiconductor material with identical chemical composition can be defined as phase junctions [13]. Photocatalysts with phase junctions has aroused substantial interest since it could deliver much higher activity than that of single phase catalyst [12–17]. TiO2 nanoparticles with engineered surface anatase/rutile phase junctions give rise to enhanced photocatalytic properties, compared with single anatase or rutile phase [13,15–17]. Furthermore, Ga2O3 photocatalysts with tailored surface α–β phase junctions also show significantly enhanced activity for overall water splitting [12]. It was suggested that the enhanced photocatalytic activity is due to more efficient separation of photogenerated charges across the phase junction interface [12,14,15]. However, other studies indicate that the morphology of nanoclusters and the pore size distribution can be also contributed to the increased catalytic activity [16,17]. Clearly, the role of the phase junction in the photocatalytic systems has not been well understood yet. Previous attempts suggest that the differences in lattice structures of rutile and anatase TiO2 can cause different electronic densities and band structure, leading to slight difference in their band alignments [18]. EPR results indicate that anatase phase possess a higher conduction band level than rutile phase, in consistent with the mechanism that electrons migrate from the rutile phase to the shallow electron trapping sites in the anatase phase [19]. However, to date, there is no report on the charge transfer between the phase junctions detected from the external circuit. The energy levels of rutile and anatase TiO2 show slight difference (conduction band difference ΔEc =0.370.1 eV and valence band difference ΔEv 0.570.1 eV) [20–22], the potential barrier at the interface which provides the driving force for charge separation is thus limited. Nanorods facilitates directional motion of carriers due to reduced electron scattering or trapping at defects such as grain boundaries existed in nanoparticles, thus finding their way into frontiers of the photovoltaic (PV) field [23,24] and photoelectrochemical systems [25,26]. Core/shell architectures markedly increase the junction area and shortens the collection distance for carriers, which may be comparable to the diffusion length of minority carriers [10]. In order to collect enough charges, a structure of coaxial nanorods with large phase junction area and less grain boundaries would be highly desirable. In this study, we fabricated photovoltaic devices based on coaxial rutile and anatase phase junctions and demonstrate the feasibility of phase junction PV device. We found that the device based on anatase/rutile junction can result an apparent photovoltaic effect with Voc of 160 mV, Isc of 1.1 mA/cm2.
TiO2 nanorod arrays were prepared by a hydrothermal method. TiO2 anatase layer and ITO layer were fabricated by a physical vapor deposition (PVD) method. The fabrication process, general materials and instrumentation used in this work are described in more detail in Supplementary information.
Results and discussion Firstly, TiO2 nanorod arrays were prepared on fluorine doped tin oxide (FTO) coated glass substrates by a typical hydrothermal method [26] (see Supplementary information for details), followed by annealing in a muffle furnace at 500 1C for 2 h. The SEM image of pristine nanorod arrays (NRAs) shows that square-sectioned nanorods are formed (Figure 1a). The width of the nanorod is 75–150 nm, the width various with the qualities of the nanowires bundles (inset of Figure 1a). The cross-section SEM image (Figure 1c) of the NRAs displays a layer of roughly vertical nanorod arrays. The average length of nanorods is 3.8 μm and the cross sections of NRAs appear quite distinct and smooth (Figure 1c). The TiO2 shells were fabricated by a DC reactive physical vapor deposition (PVD) method and a subsequent annealing process (see Supporting information for details). Briefly, Ti atoms sputtered from the target reacted with oxygen, thereby TiO2 particles deposited on the NRAs. Due to the highly anisotropic nanorods, TiO2 particles aggregated along the axial and radial directions. In this way, coaxial core (TiO2 nanorod)–shell (TiO2 nanoparticles) structures were obtained. After deposition, both the outer diameter (ca. 500 nm) and length (ca. 4 μm) of the coated nanorods are larger than those of the pristine nanorods (Figure 1b and d). The cross-section view of coaxial NRAs shows that the morphology of nanorod turns to be a baseball bar with a mushroom heads up. The top view of anatase layer shows a cauliflower-like morphology (Figure 1b), which is induced by the shadowing effect during deposition. Because the sputtering pressure (O2 + Ar mixture) was 1.0 Pa, collision between sputtered particles and the inert/reactive gas would affect the deposition process of particles. The present target-substrate distance (140 mm) is much larger than the mean free path of the sputtered particles, so that the TiO2 particles are scattered by collision during traveling from the target to the substrate, thus resulting in restrained surface mobility. This causes a shadowing effect for TiO2 deposition, which can account for the observed capping of anatase columns on the top of NRAs (Figure S1). Some TiO2 particles enter the voids and crevices between nanorods and condense along the nanorods, giving rise to a gradient thickness of anatase coating. According to the SEM top and cross-sectional view images (Figure 1), the volume ratio between rutile NRAs and anatase in the mixed hybrids is about 4:1, roughly. The AFM 3D images of the coaxial NRAs indicates that the top surface show an inverted pyramid structure (Figure 1e), which is beneficial to facilitate light reflection and reabsorption inside the device. The XRD pattern of the as-prepared TiO2 nanorod arrays shows a single rutile phase (Figure 2a). The XRD pattern of
408
P. Yan et al.
Figure 1 SEM top view of (a) the rutile nanorods arrarys (NRAs) and (b) TiO2 rutile/anatase coaxial NRAs; SEM cross-sectional view images of (c) the rutile nanorod arrays and (d) TiO2 rutile/anatase coaxial nanorod arrays; (e) AFM image of TiO2 rutile/anatase coaxial NRAs.
the rutile NRAs/anatase coaxial structure gives (1) an additional peak corresponding to anatase (101) peak (Figure 2a), indicating the formation of the anatase phase after deposition, as well as (2) a significant drop of the intensity of the rutile (001) peak. The Raman spectra of the PVD TiO2 film over FTO substrate (Figure 2b) show vibrational modes at 143 cm 1, 395 cm 1, 515 cm 1 and 638 cm 1, which are in accordance with characteristics of
single anatase phase [27]. Both XRD and Raman results confirm that the TiO2 rutile/anatase coaxial structure is successfully prepared. TEM images in Figure 3a shows that the shell of the coaxial structure is about 20 nm thick. The highresolution transmission electron microscopy (HRTEM) image of the pristine nanorod clearly shows the continuous (101) atomic planes with a lattice spacing of 2.5 Å (Figure 3b).
Photovoltaic device based on TiO2 rutile/anatase phase junctions
409
Figure 2 (a) XRD patterns of TiO2 NRAs and TiO2 rutile/anatase coaxial nanorod arrays and (b) Raman spectra of PVD TiO2 film, TiO2 NRAs and TiO2 rutile/ anatase coaxial nanorod arrays.
Figure 3 (a) TEM image of a rutile/anatase coaxial nanorod; (b) HRTEM image of rutile nanorods and (c) HRTEM image of the circled part in (a).
rutile NRAs anatase
0.5 2.0
0.16
0.0
Abs (a.u.)
-0.5 -1.0 -1.5 FTO/NR/A/ITO-Light FTO/NR/A/ITO-dark
-2.0
1.5
0.14
1.0
0.12
0.5
0.10
Voc (V)
Current Density (mA/cm2)
0.18
2.5
1.0
-2.5 -3.0 -0.10
0.08
0.0
-0.05
0.00
0.05
0.10
0.15
0.20
Voltage (V)
360
380
400
420
440
Wavelength (nm)
Figure 4 Photocurrent density–voltage (J–V) curve of the TiO2 phase-junction photovoltaic device (FTO/rutile/anatase/ITO) (tested under UV illumination).
Figure 5 UV–vis spectra of TiO2 rutile and anatase phases and Voc of TiO2 phase-junction photovoltic device (FTO/rutile/ anatase/ITO) under UV illumination with different wavelengths.
In addition, the SAED indicates that the [110] axis is perpendicular to the nanorod side walls (inset in Figure 3b), and nanorods grow along the [001] direction, which is consistent with XRD results. The HRTEM recorded from the circle area in Figure 3a shows a lattice spacing of 3.5 Å, corresponding to the {101} planes of anatase layers (Figure 3c), which is confirmed by the new diffraction spots appear in the SAED (inset in Figure 3c).
The photovoltaic properties of the assembled phase junction device were evaluated under UV illumination (wavelength: 385 nm, light intensity: 25 mW/cm2, illuminated from the FTO side) with a LED light source. J–V data recorded from an optimized device yields an open-circuit voltage (Voc) of as high as 154 mV, a short circuit current density (Isc) of 1.76 mA/cm2 and a fill factor (FF) of 28.7% (Figure 4). To explore the effect of test direction, we
410
P. Yan et al.
Figure 6 (a) Illustrations of the proposed band alignment of the rutile/anatase coaxial solar cell and (b) schematic drawing of the TiO2 phase junction device and the charge carrier transport route.
swapped the positive and negative electrodes. The J–V test shows that the Voc turns to be 154 mV (Figure S2), indicating that the applied bias does not reverse the charge carrier transfer direction. Repeated J–V tests (details see Table S1) provides the mean values for Voc (160731 mV), Isc (1.0770.3 mA/cm2) and FF (2872%). Note that the rutile phase acts as the electron collector for all J–V tests. To further investigate the effect of illumination direction, J–V curves were acquired when illumination was swapped to ITO side. Again, PV response was observed with lower Voc and Isc values than illuminated from the FTO side (Figure S3). However, to achieve a positive Voc the photogenerated holes are supposed to accumulate at the side of rutile phase, the rutile phase still acts as the electron collector. To further clarify the effect of the phase junction, the FTO/rutile NRAs/anatase/ITO device was characterized in the 365–410 nm spectral range. Figure 5 shows that the Voc curve of the device is associated with the absorption band edges of the rutile and anatase phase. The device yields a maximum Voc when both the rutile and anatase phase are excited with 365 nm UV light. Nevertheless, the Voc of the device drop to a minimum as the anatase phase shows no light absorption, indicating the phase junction is indispensable for the photovoltaic performance. The PV response under 410 nm can be ascribed to the wide half peak width of the LED light (Figure S4). To understand the essential effect of the TiO2 phase junction on the photovoltaic performance, TiO2 single phase devices, including FTO/rutile NRAs/ITO device and FTO/ anatase/ITO device, were also fabricated and characterized. We found that the J–V curves of the single TiO2 phase devices are nearly zero crossing regardless of the light illumination (Figure S5). This experimental fact unambiguously verifies that the phase junction between anatase and rutile results in the photovoltaic effect, whereas a single phase (either anatase or rutile) fails to generate any photovoltaic effect. Based on our findings aforementioned, a plausible mechanism of the migration route of photogenerated charge carriers in the TiO2 phase junction device is illustrated in Figure 6. The Mott–Schottky plots of rutile NRAs and anatase indicated that the flat band position of the rutile NRAs electrode is about 0.05 V positive than that of anatase electrode and the Fermi level of the anatase is
more close to the CB position (Figure S6). Thus, the band structures of rutile and anatase are shown schematically in Figure 6a. The staggered band alignments allow holes to be accumulated at the rutile side, while electrons at the anatase side [20]. As shown in Figure 6b, the electron–hole carriers generated in the rutile and anatase are separated at the phase junction interface, whereby the electrons diffuse into anatase and are collected at the ITO electrode, subsequently transfer to the external circuit. Since the rutile nanorods grown on the FTO are single crystalline, the rutile/anatase core–shell structure renders shorter collection distance for holes, so that the accumulated holes can be effectively transferred and collected by the FTO electrode.
Conclusions We fabricated a prototype TiO2 phase-junction based photovoltaic device consisting of a coaxial core (rutile)–shell (anatase) architecture. This design takes the full advantage of both core/shell architectures and ordered onedimensional (1D) TiO2 nanorod arrays. The coaxial TiO2 nanorod arrays phase junction device shows PV response with a Voc of 154 mV and Isc of 1.76 mA/cm2, while the single TiO2 phase devices act as photo resistors. Moreover, our results highlight the key role of the polymorphic crystal phases in efficient separation of photogenerated charges. This work offers guidance to design various phase junctions or heterojunctions based devices, such as UV detectors or light sensor. Moreover, the phase junction device exhibit rapid response of about 4 μs, the detailed discussion will be reported in the near future.
Acknowledgments This work was financially supported by 973 National Basic Research Program of the Ministry of Science and Technology (2014CB239400), National Natural Science Foundation of China (No. 21373209), the Fundamental Research Funds for the Central Universities (No. HIT. KISTP. 201406).
Photovoltaic device based on TiO2 rutile/anatase phase junctions
Appendix A.
Supporting information
Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/ j.nanoen.2015.05.005.
References [1] Z. Li, W. Luo, M. Zhang, J. Feng, Z. Zou, Energy Environ. Sci. 6 (2013) 347–370. [2] M. Moriya, T. Minegishi, H. Kumagai, M. Katayama, J. Kubota, K. Domen, J. Am. Chem. Soc. 135 (2013) 3733–3735. [3] E.S. Kim, N. Nishimura, G. Magesh, J.Y. Kim, J.-W. Jang, H. Jun, J. Kubota, K. Domen, J.S. Lee, J. Am. Chem. Soc. 135 (2013) 5375–5383. [4] W. Qin, W. Yu, W. Zi, X. Liu, T. Yuan, D. Yang, S. Wang, G. Tu, J. Zhang, F.S. Liu, C. Li, J. Mater. Chem. A 2 (2014) 15303–15307. [5] S.V. Kershaw, A.S. Susha, A.L. Rogach, Chem. Soc. Rev. 42 (2013) 3033–3087. [6] G. Guerguerian, F. Elhordoy, C.J. Pereyra, R.E. Marotti, F. Martín, D. Leinen, J.R. Ramos-Barrado, E.A. Dalchiele, Nanotechnology 22 (2011) 5401–5409. [7] X. Liu, W. Zhou, Z. Yin, X. Hao, Y. Wu, X. Xu, J. Mater. Chem. 22 (2012) 3916–3921. [8] S.J. Hong, S. Lee, J.S. Jang, J.S. Lee, Energy Environ. Sci. 4 (2011) 1781–1787. [9] M. Gharghi, E. Fathi, B. Kante, S. Sivoththaman, X. Zhang, Nano Lett. 12 (2012) 6278–6282. [10] B. Tian, X. Zheng, T.J. Kempa, Y. Fang, N. Yu, G. Yu, J. Huang, C.M. Lieber, Nature 449 (2007) 885–889. [11] C. Li, M. Li, J. Raman Spectrosc. 33 (2002) 301–308. [12] X. Wang, Q. Xu, M. Li, S. Shen, X. Wang, Y. Wang, Z. Feng, J. Shi, H. Han, C. Li, Angew. Chem. 51 (2012) 13089–13092. [13] J. Zhang, Q. Xu, Z. Feng, M. Li, C. Li, Angew. Chem. Int. Ed. 47 (2008) 1766–1769. [14] T. Xia, N. Li, Y. Zhang, M.B. Kruger, J. Murowchick, A. Selloni, X. Chen, ACS Appl. Mater. Interfaces 5 (2013) 9883–9890. [15] T. Kawahara, Y. Konishi, H. Tada, N. Tohge, J. Nishii, S. Ito, Angew. Chem. 114 (2002) 2935–2937. [16] R.R. Bacsa, J. Kiwi, Appl. Catal. B-Environ. 16 (1998) 19–29. [17] C. Wang, X. Zhang, C. Shao, Y. Zhang, J. Yang, P. Sun, X. Liu, H. Liu, Y. Liu, T. Xie, D. Wang, J. Colloid Interface Sci. 363 (2011) 157–164. [18] Y. Ma, X. Wang, Y. Jia, X. Chen, H. Han, C. Li, Chem. Rev. 114 (2014) 9987–10043. [19] D.C. Hurum, A.G. Agrios, K.A. Gray, T. Rajh, M.C. Thurnauer, J. Phys. Chem. B 107 (2003) 4545–4549. [20] D.O. Scanlon, C.W. Dunnill, J. Buckeridge, S.A. Shevlin, A.J. Logsdail, S.M. Woodley, C.R. Catlow, M.J. Powell, R.G. Palgrave, I.P. Parkin, G.W. Watson, T.W. Keal, P. Sherwood, A. Walsh, A.A. Sokol, Nat. Mater. 12 (2013) 798–801. [21] P. Deák, B.l. Aradi, T. Frauenheim, J. Phys. Chem. C 115 (2011) 3443–3446. [22] V. Pfeifer, P. Erhart, S. Li, K. Rachut, J. Morasch, J. Brötz, P. Reckers, T. Mayer, S. Rühle, A. Zaban, I. Mora Seró, J. Bisquert, W. Jaegermann, A. Klein, J. Phys. Chem. L 4 (2013) 4182–4187. [23] Q. Wei, K. Hirota, K. Tajima, K. Hashimoto, Chem. Mater. 18 (2006) 5080–5087. [24] L. De Marco, M. Manca, R. Giannuzzi, F. Malara, G. Melcarne, G. Ciccarella, I. Zama, R. Cingolani, G. Gigli, J. Phys. Chem. C 114 (2010) 4228–4236. [25] G. Wang, H. Wang, Y. Ling, Y. Tang, X. Yang, R.C. Fitzmorris, C. Wang, J.Z. Zhang, Y. Li, Nano Lett. 11 (2011) 3026–3033. [26] B. Liu, E.S. Aydil, J. Am. Chem. Soc. 131 (2009) 3985–3990. [27] J. Zhang, M. Li, Z. Feng, J. Chen, C. Li, U.V. Raman, J. Phys. Chem. B 110 (2006) 927–935.
411 Ms. Pengli Yan received her bachelor degree in Chemical Engineering and Materials from Harbin Institue of Technology in 2010. She is currently a Ph.D. student of Harbin Institue of Technology and doing her research in State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics under the supervision of Prof. Can Li. Her research interest is the photovoltaic and photoelectrochemistry.
Mr. Xiang Wang received his bachelor degree in Chemistry from Northeastern University in 2007. In 2013, he received his Ph.D. degree from State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences. His research interest focused on phase junction, photocatalysis and photoelectrochemistry.
Mr. Xiaojia Zheng received his bachelor degree from Sichuan University in 2009. In 2014, he received his Ph.D. degree from State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian National Laboratory for Clean Energy. His research interest focusd on photovoltaic cell.
Mr. Rengui Li received his Ph.D. degree from State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian National Laboratory for Clean Energy. He is currently an associate researcher in Dalian Institute of Chemical Physics, Chinese Academy of Sciences. His research interest focused on photocatalysis.
Mr. Jingfeng Han received his bachelor degree from School of Chemistry and Chemical Engineering, China University of Petroleum in 2007. In 2015, he received his Ph.D. degree from State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian National Laboratory for Clean Energy. His research interest is the photocatalysis and photoelectrochemistry.
Ms. Jingying Shi received her Bachelor and Master degree in 1995 and 1998 respectively in Fujian Normal University. After the graduation, she worked as a teacher in Quanzhou Normal College for four and a half years. In 2006, she received her Ph.D. degree from Zhejiang University. She is currently an associate researcher in Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Her research interest focused on photoelectrochemistry.
412
P. Yan et al. Mr. Ailong Li received his bachelor degree from Ocean University of China in 2012. He is currently a Ph.D. candidate of State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics under the supervision of Prof. Can Li. His research interest is the photocatalysis and photoelectrochemistry.
Prof. Yang Gan received his Ph.D. degree in Institute of Metal Research, Chinese Academy of Sciences in 2001. Currently he is a Professor and Director of the Catalysis Science and Engineering in School of Chemical Engineering and Materials, Harbin Institute of Technology. His research focus on the relationship of Surface microstructure and surface chemical properties of ceramic materials and inorganic oxides.
Prof. Can Li is the director of State Key Laboratory of Catalysis and the director of Dalian National Laboratory for Clean Energy. His research interests include (1) UV Raman spectroscopy; (2) environmental catalysis and green catalysis; (3) heterogeneous asymmetric catalysis, and (4) solar energy ultilization based on photocatalysis, Photoelectrocatalysis, and photovoltaics. He has published more than 500 peer-reviewed papers with over 10,000 citations.
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/nanoenergy
RAPID COMMUNICATION
Photovoltaic device based on TiO2 rutile/anatase phase junctions fabricated in coaxial nanorod arrays Pengli Yana,b, Xiang Wangb,c, Xiaojia Zhengb,c, Rengui Lib, Jingfeng Hanb,c, Jingying Shib, Ailong Lib,c, Yang Gana,n, Can Lib,nn a
School of Chemical Engineering & Technology, Harbin Institute of Technology, Harbin 150001, China State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences; Dalian National Laboratory for Clean Energy, Dalian 116023, China c University of Chinese Academy of Sciences, Beijing 100049, China b
Received 16 March 2015; received in revised form 4 May 2015; accepted 8 May 2015 Available online 16 May 2015
KEYWORDS
Abstract
Prototype PV device; Phase junction; Titania; Coaxial nanorod
Phase-junctions formed in mixed crystalline phases of semiconductor based photocatalysts (TiO2 and Ga2O3) shows enhanced photocatalytic activity owing to efficient separation of photogenerated charges. However, the phase junction effect on charge separation has not been directly verified yet. In this study, we fabricated a prototype photovoltaic device based on TiO2 rutile/anatase coaxial nanorod arrays (NRAs) to demonstrate charge separation at the interface of polymorphic crystal phases. The device—FTO/rutile NRAs/anatase/ITO shows an ordinary photovoltaic response (open-circuit voltage Voc: 154 mV, short-circuit current density Jsc: 1.76 mA/cm2), contrasting with photo resistor behavior of two TiO2 single phase devices of core–core (FTO/rutile NRAs/ITO) and shell–shell (FTO/anatase/ITO). Experimental evidences suggest that the built-in electric field at the interface of rutile/anatase phase junctions in the FTO/rutile NRAs/anatase/ITO device provides the direct driving force for efficient separation of photogenerated charges. The demonstrated strategy of fabricating phase-junction photovoltaic devices may inspire further investigations on new device converting solar into electrical energy and highlight the key role of the phase junction interface in the properties of the mix-phase photocatalysts. & 2015 Published by Elsevier Ltd.
n
Corresponding author. Tel.: +86 45186413708. Corresponding author. Tel.: +86 41184379070. E-mail addresses: [email protected] (Y. Gan), [email protected] (C. Li).
nn
http://dx.doi.org/10.1016/j.nanoen.2015.05.005 2211-2855/& 2015 Published by Elsevier Ltd.
Photovoltaic device based on TiO2 rutile/anatase phase junctions
407
Introduction
Experimental section
Semiconductor heterojunction plays an important role in photogenerated charge separation of photocatalytic [1–3] and photovoltaic [4,5] systems for solar energy conversion. A heterojunction, usually formed between two different semiconductors [6–8] or semiconductors with different doping levels [9,10], provides the potential driving force for separating photoinduced charge carriers at the interface. In a similar fashion, many semiconductor materials exist in various crystalline phases—namely polymorphs [11–13]. The junctions formed between different phases of the same semiconductor material with identical chemical composition can be defined as phase junctions [13]. Photocatalysts with phase junctions has aroused substantial interest since it could deliver much higher activity than that of single phase catalyst [12–17]. TiO2 nanoparticles with engineered surface anatase/rutile phase junctions give rise to enhanced photocatalytic properties, compared with single anatase or rutile phase [13,15–17]. Furthermore, Ga2O3 photocatalysts with tailored surface α–β phase junctions also show significantly enhanced activity for overall water splitting [12]. It was suggested that the enhanced photocatalytic activity is due to more efficient separation of photogenerated charges across the phase junction interface [12,14,15]. However, other studies indicate that the morphology of nanoclusters and the pore size distribution can be also contributed to the increased catalytic activity [16,17]. Clearly, the role of the phase junction in the photocatalytic systems has not been well understood yet. Previous attempts suggest that the differences in lattice structures of rutile and anatase TiO2 can cause different electronic densities and band structure, leading to slight difference in their band alignments [18]. EPR results indicate that anatase phase possess a higher conduction band level than rutile phase, in consistent with the mechanism that electrons migrate from the rutile phase to the shallow electron trapping sites in the anatase phase [19]. However, to date, there is no report on the charge transfer between the phase junctions detected from the external circuit. The energy levels of rutile and anatase TiO2 show slight difference (conduction band difference ΔEc =0.370.1 eV and valence band difference ΔEv 0.570.1 eV) [20–22], the potential barrier at the interface which provides the driving force for charge separation is thus limited. Nanorods facilitates directional motion of carriers due to reduced electron scattering or trapping at defects such as grain boundaries existed in nanoparticles, thus finding their way into frontiers of the photovoltaic (PV) field [23,24] and photoelectrochemical systems [25,26]. Core/shell architectures markedly increase the junction area and shortens the collection distance for carriers, which may be comparable to the diffusion length of minority carriers [10]. In order to collect enough charges, a structure of coaxial nanorods with large phase junction area and less grain boundaries would be highly desirable. In this study, we fabricated photovoltaic devices based on coaxial rutile and anatase phase junctions and demonstrate the feasibility of phase junction PV device. We found that the device based on anatase/rutile junction can result an apparent photovoltaic effect with Voc of 160 mV, Isc of 1.1 mA/cm2.
TiO2 nanorod arrays were prepared by a hydrothermal method. TiO2 anatase layer and ITO layer were fabricated by a physical vapor deposition (PVD) method. The fabrication process, general materials and instrumentation used in this work are described in more detail in Supplementary information.
Results and discussion Firstly, TiO2 nanorod arrays were prepared on fluorine doped tin oxide (FTO) coated glass substrates by a typical hydrothermal method [26] (see Supplementary information for details), followed by annealing in a muffle furnace at 500 1C for 2 h. The SEM image of pristine nanorod arrays (NRAs) shows that square-sectioned nanorods are formed (Figure 1a). The width of the nanorod is 75–150 nm, the width various with the qualities of the nanowires bundles (inset of Figure 1a). The cross-section SEM image (Figure 1c) of the NRAs displays a layer of roughly vertical nanorod arrays. The average length of nanorods is 3.8 μm and the cross sections of NRAs appear quite distinct and smooth (Figure 1c). The TiO2 shells were fabricated by a DC reactive physical vapor deposition (PVD) method and a subsequent annealing process (see Supporting information for details). Briefly, Ti atoms sputtered from the target reacted with oxygen, thereby TiO2 particles deposited on the NRAs. Due to the highly anisotropic nanorods, TiO2 particles aggregated along the axial and radial directions. In this way, coaxial core (TiO2 nanorod)–shell (TiO2 nanoparticles) structures were obtained. After deposition, both the outer diameter (ca. 500 nm) and length (ca. 4 μm) of the coated nanorods are larger than those of the pristine nanorods (Figure 1b and d). The cross-section view of coaxial NRAs shows that the morphology of nanorod turns to be a baseball bar with a mushroom heads up. The top view of anatase layer shows a cauliflower-like morphology (Figure 1b), which is induced by the shadowing effect during deposition. Because the sputtering pressure (O2 + Ar mixture) was 1.0 Pa, collision between sputtered particles and the inert/reactive gas would affect the deposition process of particles. The present target-substrate distance (140 mm) is much larger than the mean free path of the sputtered particles, so that the TiO2 particles are scattered by collision during traveling from the target to the substrate, thus resulting in restrained surface mobility. This causes a shadowing effect for TiO2 deposition, which can account for the observed capping of anatase columns on the top of NRAs (Figure S1). Some TiO2 particles enter the voids and crevices between nanorods and condense along the nanorods, giving rise to a gradient thickness of anatase coating. According to the SEM top and cross-sectional view images (Figure 1), the volume ratio between rutile NRAs and anatase in the mixed hybrids is about 4:1, roughly. The AFM 3D images of the coaxial NRAs indicates that the top surface show an inverted pyramid structure (Figure 1e), which is beneficial to facilitate light reflection and reabsorption inside the device. The XRD pattern of the as-prepared TiO2 nanorod arrays shows a single rutile phase (Figure 2a). The XRD pattern of
408
P. Yan et al.
Figure 1 SEM top view of (a) the rutile nanorods arrarys (NRAs) and (b) TiO2 rutile/anatase coaxial NRAs; SEM cross-sectional view images of (c) the rutile nanorod arrays and (d) TiO2 rutile/anatase coaxial nanorod arrays; (e) AFM image of TiO2 rutile/anatase coaxial NRAs.
the rutile NRAs/anatase coaxial structure gives (1) an additional peak corresponding to anatase (101) peak (Figure 2a), indicating the formation of the anatase phase after deposition, as well as (2) a significant drop of the intensity of the rutile (001) peak. The Raman spectra of the PVD TiO2 film over FTO substrate (Figure 2b) show vibrational modes at 143 cm 1, 395 cm 1, 515 cm 1 and 638 cm 1, which are in accordance with characteristics of
single anatase phase [27]. Both XRD and Raman results confirm that the TiO2 rutile/anatase coaxial structure is successfully prepared. TEM images in Figure 3a shows that the shell of the coaxial structure is about 20 nm thick. The highresolution transmission electron microscopy (HRTEM) image of the pristine nanorod clearly shows the continuous (101) atomic planes with a lattice spacing of 2.5 Å (Figure 3b).
Photovoltaic device based on TiO2 rutile/anatase phase junctions
409
Figure 2 (a) XRD patterns of TiO2 NRAs and TiO2 rutile/anatase coaxial nanorod arrays and (b) Raman spectra of PVD TiO2 film, TiO2 NRAs and TiO2 rutile/ anatase coaxial nanorod arrays.
Figure 3 (a) TEM image of a rutile/anatase coaxial nanorod; (b) HRTEM image of rutile nanorods and (c) HRTEM image of the circled part in (a).
rutile NRAs anatase
0.5 2.0
0.16
0.0
Abs (a.u.)
-0.5 -1.0 -1.5 FTO/NR/A/ITO-Light FTO/NR/A/ITO-dark
-2.0
1.5
0.14
1.0
0.12
0.5
0.10
Voc (V)
Current Density (mA/cm2)
0.18
2.5
1.0
-2.5 -3.0 -0.10
0.08
0.0
-0.05
0.00
0.05
0.10
0.15
0.20
Voltage (V)
360
380
400
420
440
Wavelength (nm)
Figure 4 Photocurrent density–voltage (J–V) curve of the TiO2 phase-junction photovoltaic device (FTO/rutile/anatase/ITO) (tested under UV illumination).
Figure 5 UV–vis spectra of TiO2 rutile and anatase phases and Voc of TiO2 phase-junction photovoltic device (FTO/rutile/ anatase/ITO) under UV illumination with different wavelengths.
In addition, the SAED indicates that the [110] axis is perpendicular to the nanorod side walls (inset in Figure 3b), and nanorods grow along the [001] direction, which is consistent with XRD results. The HRTEM recorded from the circle area in Figure 3a shows a lattice spacing of 3.5 Å, corresponding to the {101} planes of anatase layers (Figure 3c), which is confirmed by the new diffraction spots appear in the SAED (inset in Figure 3c).
The photovoltaic properties of the assembled phase junction device were evaluated under UV illumination (wavelength: 385 nm, light intensity: 25 mW/cm2, illuminated from the FTO side) with a LED light source. J–V data recorded from an optimized device yields an open-circuit voltage (Voc) of as high as 154 mV, a short circuit current density (Isc) of 1.76 mA/cm2 and a fill factor (FF) of 28.7% (Figure 4). To explore the effect of test direction, we
410
P. Yan et al.
Figure 6 (a) Illustrations of the proposed band alignment of the rutile/anatase coaxial solar cell and (b) schematic drawing of the TiO2 phase junction device and the charge carrier transport route.
swapped the positive and negative electrodes. The J–V test shows that the Voc turns to be 154 mV (Figure S2), indicating that the applied bias does not reverse the charge carrier transfer direction. Repeated J–V tests (details see Table S1) provides the mean values for Voc (160731 mV), Isc (1.0770.3 mA/cm2) and FF (2872%). Note that the rutile phase acts as the electron collector for all J–V tests. To further investigate the effect of illumination direction, J–V curves were acquired when illumination was swapped to ITO side. Again, PV response was observed with lower Voc and Isc values than illuminated from the FTO side (Figure S3). However, to achieve a positive Voc the photogenerated holes are supposed to accumulate at the side of rutile phase, the rutile phase still acts as the electron collector. To further clarify the effect of the phase junction, the FTO/rutile NRAs/anatase/ITO device was characterized in the 365–410 nm spectral range. Figure 5 shows that the Voc curve of the device is associated with the absorption band edges of the rutile and anatase phase. The device yields a maximum Voc when both the rutile and anatase phase are excited with 365 nm UV light. Nevertheless, the Voc of the device drop to a minimum as the anatase phase shows no light absorption, indicating the phase junction is indispensable for the photovoltaic performance. The PV response under 410 nm can be ascribed to the wide half peak width of the LED light (Figure S4). To understand the essential effect of the TiO2 phase junction on the photovoltaic performance, TiO2 single phase devices, including FTO/rutile NRAs/ITO device and FTO/ anatase/ITO device, were also fabricated and characterized. We found that the J–V curves of the single TiO2 phase devices are nearly zero crossing regardless of the light illumination (Figure S5). This experimental fact unambiguously verifies that the phase junction between anatase and rutile results in the photovoltaic effect, whereas a single phase (either anatase or rutile) fails to generate any photovoltaic effect. Based on our findings aforementioned, a plausible mechanism of the migration route of photogenerated charge carriers in the TiO2 phase junction device is illustrated in Figure 6. The Mott–Schottky plots of rutile NRAs and anatase indicated that the flat band position of the rutile NRAs electrode is about 0.05 V positive than that of anatase electrode and the Fermi level of the anatase is
more close to the CB position (Figure S6). Thus, the band structures of rutile and anatase are shown schematically in Figure 6a. The staggered band alignments allow holes to be accumulated at the rutile side, while electrons at the anatase side [20]. As shown in Figure 6b, the electron–hole carriers generated in the rutile and anatase are separated at the phase junction interface, whereby the electrons diffuse into anatase and are collected at the ITO electrode, subsequently transfer to the external circuit. Since the rutile nanorods grown on the FTO are single crystalline, the rutile/anatase core–shell structure renders shorter collection distance for holes, so that the accumulated holes can be effectively transferred and collected by the FTO electrode.
Conclusions We fabricated a prototype TiO2 phase-junction based photovoltaic device consisting of a coaxial core (rutile)–shell (anatase) architecture. This design takes the full advantage of both core/shell architectures and ordered onedimensional (1D) TiO2 nanorod arrays. The coaxial TiO2 nanorod arrays phase junction device shows PV response with a Voc of 154 mV and Isc of 1.76 mA/cm2, while the single TiO2 phase devices act as photo resistors. Moreover, our results highlight the key role of the polymorphic crystal phases in efficient separation of photogenerated charges. This work offers guidance to design various phase junctions or heterojunctions based devices, such as UV detectors or light sensor. Moreover, the phase junction device exhibit rapid response of about 4 μs, the detailed discussion will be reported in the near future.
Acknowledgments This work was financially supported by 973 National Basic Research Program of the Ministry of Science and Technology (2014CB239400), National Natural Science Foundation of China (No. 21373209), the Fundamental Research Funds for the Central Universities (No. HIT. KISTP. 201406).
Photovoltaic device based on TiO2 rutile/anatase phase junctions
Appendix A.
Supporting information
Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/ j.nanoen.2015.05.005.
References [1] Z. Li, W. Luo, M. Zhang, J. Feng, Z. Zou, Energy Environ. Sci. 6 (2013) 347–370. [2] M. Moriya, T. Minegishi, H. Kumagai, M. Katayama, J. Kubota, K. Domen, J. Am. Chem. Soc. 135 (2013) 3733–3735. [3] E.S. Kim, N. Nishimura, G. Magesh, J.Y. Kim, J.-W. Jang, H. Jun, J. Kubota, K. Domen, J.S. Lee, J. Am. Chem. Soc. 135 (2013) 5375–5383. [4] W. Qin, W. Yu, W. Zi, X. Liu, T. Yuan, D. Yang, S. Wang, G. Tu, J. Zhang, F.S. Liu, C. Li, J. Mater. Chem. A 2 (2014) 15303–15307. [5] S.V. Kershaw, A.S. Susha, A.L. Rogach, Chem. Soc. Rev. 42 (2013) 3033–3087. [6] G. Guerguerian, F. Elhordoy, C.J. Pereyra, R.E. Marotti, F. Martín, D. Leinen, J.R. Ramos-Barrado, E.A. Dalchiele, Nanotechnology 22 (2011) 5401–5409. [7] X. Liu, W. Zhou, Z. Yin, X. Hao, Y. Wu, X. Xu, J. Mater. Chem. 22 (2012) 3916–3921. [8] S.J. Hong, S. Lee, J.S. Jang, J.S. Lee, Energy Environ. Sci. 4 (2011) 1781–1787. [9] M. Gharghi, E. Fathi, B. Kante, S. Sivoththaman, X. Zhang, Nano Lett. 12 (2012) 6278–6282. [10] B. Tian, X. Zheng, T.J. Kempa, Y. Fang, N. Yu, G. Yu, J. Huang, C.M. Lieber, Nature 449 (2007) 885–889. [11] C. Li, M. Li, J. Raman Spectrosc. 33 (2002) 301–308. [12] X. Wang, Q. Xu, M. Li, S. Shen, X. Wang, Y. Wang, Z. Feng, J. Shi, H. Han, C. Li, Angew. Chem. 51 (2012) 13089–13092. [13] J. Zhang, Q. Xu, Z. Feng, M. Li, C. Li, Angew. Chem. Int. Ed. 47 (2008) 1766–1769. [14] T. Xia, N. Li, Y. Zhang, M.B. Kruger, J. Murowchick, A. Selloni, X. Chen, ACS Appl. Mater. Interfaces 5 (2013) 9883–9890. [15] T. Kawahara, Y. Konishi, H. Tada, N. Tohge, J. Nishii, S. Ito, Angew. Chem. 114 (2002) 2935–2937. [16] R.R. Bacsa, J. Kiwi, Appl. Catal. B-Environ. 16 (1998) 19–29. [17] C. Wang, X. Zhang, C. Shao, Y. Zhang, J. Yang, P. Sun, X. Liu, H. Liu, Y. Liu, T. Xie, D. Wang, J. Colloid Interface Sci. 363 (2011) 157–164. [18] Y. Ma, X. Wang, Y. Jia, X. Chen, H. Han, C. Li, Chem. Rev. 114 (2014) 9987–10043. [19] D.C. Hurum, A.G. Agrios, K.A. Gray, T. Rajh, M.C. Thurnauer, J. Phys. Chem. B 107 (2003) 4545–4549. [20] D.O. Scanlon, C.W. Dunnill, J. Buckeridge, S.A. Shevlin, A.J. Logsdail, S.M. Woodley, C.R. Catlow, M.J. Powell, R.G. Palgrave, I.P. Parkin, G.W. Watson, T.W. Keal, P. Sherwood, A. Walsh, A.A. Sokol, Nat. Mater. 12 (2013) 798–801. [21] P. Deák, B.l. Aradi, T. Frauenheim, J. Phys. Chem. C 115 (2011) 3443–3446. [22] V. Pfeifer, P. Erhart, S. Li, K. Rachut, J. Morasch, J. Brötz, P. Reckers, T. Mayer, S. Rühle, A. Zaban, I. Mora Seró, J. Bisquert, W. Jaegermann, A. Klein, J. Phys. Chem. L 4 (2013) 4182–4187. [23] Q. Wei, K. Hirota, K. Tajima, K. Hashimoto, Chem. Mater. 18 (2006) 5080–5087. [24] L. De Marco, M. Manca, R. Giannuzzi, F. Malara, G. Melcarne, G. Ciccarella, I. Zama, R. Cingolani, G. Gigli, J. Phys. Chem. C 114 (2010) 4228–4236. [25] G. Wang, H. Wang, Y. Ling, Y. Tang, X. Yang, R.C. Fitzmorris, C. Wang, J.Z. Zhang, Y. Li, Nano Lett. 11 (2011) 3026–3033. [26] B. Liu, E.S. Aydil, J. Am. Chem. Soc. 131 (2009) 3985–3990. [27] J. Zhang, M. Li, Z. Feng, J. Chen, C. Li, U.V. Raman, J. Phys. Chem. B 110 (2006) 927–935.
411 Ms. Pengli Yan received her bachelor degree in Chemical Engineering and Materials from Harbin Institue of Technology in 2010. She is currently a Ph.D. student of Harbin Institue of Technology and doing her research in State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics under the supervision of Prof. Can Li. Her research interest is the photovoltaic and photoelectrochemistry.
Mr. Xiang Wang received his bachelor degree in Chemistry from Northeastern University in 2007. In 2013, he received his Ph.D. degree from State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences. His research interest focused on phase junction, photocatalysis and photoelectrochemistry.
Mr. Xiaojia Zheng received his bachelor degree from Sichuan University in 2009. In 2014, he received his Ph.D. degree from State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian National Laboratory for Clean Energy. His research interest focusd on photovoltaic cell.
Mr. Rengui Li received his Ph.D. degree from State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian National Laboratory for Clean Energy. He is currently an associate researcher in Dalian Institute of Chemical Physics, Chinese Academy of Sciences. His research interest focused on photocatalysis.
Mr. Jingfeng Han received his bachelor degree from School of Chemistry and Chemical Engineering, China University of Petroleum in 2007. In 2015, he received his Ph.D. degree from State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian National Laboratory for Clean Energy. His research interest is the photocatalysis and photoelectrochemistry.
Ms. Jingying Shi received her Bachelor and Master degree in 1995 and 1998 respectively in Fujian Normal University. After the graduation, she worked as a teacher in Quanzhou Normal College for four and a half years. In 2006, she received her Ph.D. degree from Zhejiang University. She is currently an associate researcher in Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Her research interest focused on photoelectrochemistry.
412
P. Yan et al. Mr. Ailong Li received his bachelor degree from Ocean University of China in 2012. He is currently a Ph.D. candidate of State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics under the supervision of Prof. Can Li. His research interest is the photocatalysis and photoelectrochemistry.
Prof. Yang Gan received his Ph.D. degree in Institute of Metal Research, Chinese Academy of Sciences in 2001. Currently he is a Professor and Director of the Catalysis Science and Engineering in School of Chemical Engineering and Materials, Harbin Institute of Technology. His research focus on the relationship of Surface microstructure and surface chemical properties of ceramic materials and inorganic oxides.
Prof. Can Li is the director of State Key Laboratory of Catalysis and the director of Dalian National Laboratory for Clean Energy. His research interests include (1) UV Raman spectroscopy; (2) environmental catalysis and green catalysis; (3) heterogeneous asymmetric catalysis, and (4) solar energy ultilization based on photocatalysis, Photoelectrocatalysis, and photovoltaics. He has published more than 500 peer-reviewed papers with over 10,000 citations.