Halide Perovskite Nanocrystals Inks for

14 downloads 0 Views 2MB Size Report
Oct 25, 2016 - ligand-free, lead halides-based perovskites inks by laser abla- tion synthesis assisted in solution (LASiS). LASiS is a technique known as an ...
www.advancedsciencenews.com

www.advenergymat.de

Francesco Lamberti, Lucio Litti, Michele De Bastiani, Roberto Sorrentino, Marina Gandini, Moreno Meneghetti,* and Annamaria Petrozza* After the publication of the two landmark papers which demonstrate the embodiment of perovskite semiconductors in efficient solid-state solar cells,[1] such devices have reached power conversion efficiencies higher than 20%.[2] Moreover, metal halide perovskites have also demonstrated great potential for lighting and lasing applications.[3,4] These materials not only show exceptional primary optoelectronic properties such as a direct bandgap,[5] small exciton binding energy,[6] low carrier recombination rates,[7] ambipolar transport,[8] and tunability of the bandgap covering a wavelength range from the near-infrared (NIR)[9] to the ultraviolet,[10] they are also very attractive for their ease of processability for mass production (e.g., printing from solution)[11] and for a large availability of their chemical components.[12] For the most efficient optoelectronic devices, the semiconductor films are currently processed from precursor solutions, such as PbI2 and CH3NH3I dissolved in organic solvents (e.g., dimethylformamide, DMF) for CH3NH3PbI3.[13] After deposition, the constituent ions self-assemble during crystallization directly upon the selected substrate when treated at temperatures below 120 °C. Such process is very versatile, and being accessible to a wide range of scientists it has allowed an unprecedented number of research laboratories to enter the field and push it quickly to reach exceptional advancements. However, it presents relevant drawbacks. First of all, it makes very difficult the reproducibility of the thin film morphology and the related optoelectronic properties.[14] In fact, the thin films are polycrystalline but can exhibit varying morphologies determined by different factors including precursor ratio, solvent, processing additives, substrate roughness and surface energy, atmospheric/environmental conditions, annealing temperature, and treatment time.[15] It also makes the thin films highly subject to structural and chemical defects. Importantly, such defects have been recently linked to the presence of hysteretic behaviors in current density–voltage (J–V) characteristics of perovskite-based diodes[16] and to optical

Dr. F. Lamberti, Dr. M. De Bastiani, R. Sorrentino, M. Gandini, Dr. A. Petrozza Center for Nano Science and Technology @Polimi Istituto Italiano di Tecnologia via Giovanni Pascoli 70/3, 20133 Milan, Italy E-mail: [email protected] Dr. L. Litti, Dr. M. De Bastiani, Prof. M. Meneghetti Department of Chemical Sciences Università degli Studi di Padova Padova 35131, Italy E-mail: [email protected]

DOI: 10.1002/aenm.201601703

Adv. Energy Mater. 2017, 1601703

instability of the materials upon light soaking.[17] In the first case, it has been demonstrated that an electric field applied to a perovskite film deposited between symmetric contacts, induces the formation of reversible p-i-n structures upon ion migration.[18,19] In the second case, although the absorption edge of the semiconductor can be easily tuned between 1.6 and 2.3 eV by mixing the halides composition, photo-induced ion migration causes the formation of a sub-bandgap emissive state, tentatively assigned to the formation of stable phases upon ion segregation.[17,20] This phenomenon hampers the use of the semiconductor in multijunction solar cells and in tunable light-emitting devices.[21] Importantly, it has been proven that when the perovskite semiconductors are synthetized through highly controlled synthetic processes, such as solution synthesis approaches to produce colloidal nanocrystals, such issue is solved.[22] However, these nanocrystals always have bulky organic ligands which passivate their surface and at the same time hamper the fabrication of conductive thin films. In this Communication, we present the synthesis of high-quality, ligand-free, lead halides-based perovskites inks by laser ablation synthesis assisted in solution (LASiS). LASiS is a technique known as an easy methodology to produce nanomaterials with a plasma plume confined by a solution.[23,24] Among different methodologies for preparing perovskite NCs, like sequential deposition,[25] colloidal quantum dot inspired synthesis,[26] and the re-precipitation method,[27] the present technique is unique for obtaining stable colloidal solutions without the use of any surfactants or templates and with a small distribution of dimensions (on the scale of some tens of nm) and shapes of the nanocrystals. Surface charges of the nanoparticles usually determine a coulomb stabilization of the colloidal solutions without the need of stabilizing ligands.[28] We show that stable colloidal dispersions of PbI2 and PbBr2 in organic solvents can be prepared and that they can be fully converted into mixedhalide perovskite inks. Here we demonstrate, starting from the obtained colloidal solutions, the fabrication of thin films which do not show the formation of a rectifying junction when polarized between two symmetric contacts and are photo-stable when we tune the semiconductor bandgap. The experimental setup for the LASiS technique is shown in Figure 1a. A pulsed beam of 9 ns at 2.33 eV (Nd:YAG laser) is focused on the surface of PbI2 or PbBr2 compressed powder tablet immersed in iodo-benzene or bromo-benzene, respectively (see the Experimental Section for the ablation conditions and the ablated material production). The ablation with ns pulses is thought to produce nanoparticles based on thermal processes following the absorption of laser pulses.[29] Figure 1b,c shows the scanning electron microscopy (SEM)

© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

wileyonlinelibrary.com

(1 of 5)  1601703

Communication

High-Quality, Ligands-Free, Mixed-Halide Perovskite Nanocrystals Inks for Optoelectronic applications

Communication

www.advenergymat.de

www.advancedsciencenews.com

by EDX analysis. In particular we have: full iodine perovskite, MAPbI3.16 (PVK-A); mixed halide perovskite rich in iodine content MAPbI1.7Br1.3 (PVK-B); mixed halide perovskite rich in bromine content MAPbI0.52Br2.48 (PVK-C); and full bromine perovskite MAPbBr2.94 (PVK-D). Full halide perovskite samples (Figure 3a,b) have a cubic shape, whereas different shapes are obtained for the mixed ones (Figure 3c,d). Dimensions of the iodine PVK-A (50–100 nm) and bromine PVK-D (100–200 nm) crystals are found larger than those of the precursor NCs (see above). Even larger crystals (up to 1 µm) were observed for the PVK-D sample. Figure 3e shows a zoom-in of the full X-ray diffraction (XRD) pattern (see Figures S3 and S4, Supporting Information) for the four different types of nanocrystals. As expected, for reduced lattice constant values, the peak at 14.1° related to the (100) plane, Figure 1.  a) Schematic description of the laser ablation synthesis setup; SEM images showing shifts toward larger 2ϑ values going from the the b) PbI2 and c) PbBr2 ablated crystals N. The inset pictures show the obtained colloidal pure iodine PVK-A[31] to crystals with higher solutions. bromine contents. One can also note that the FWHM of the peaks gets smaller and smaller with the increasing of the bromine content. By using images of the ablated crystals of PbI2 NCs and PbBr2 NCs, the Scherrer equation[32] (with 0.9 as shape factor, as usual for respectively. The iodine NCs are almost spherical with a diameter of about 30–50 nm, while the bromine-based NCs have cubic grains),[33] we estimate the grain size of the different PVK larger diameter and present different shapes from almost and find that the nanocrystalline domains are about 25 nm for spherical to elongated nanocrystals. As a result, while the PbI2 PVK-A increasing to 60 nm for PVK-D (Figure 3f). The full XRD pattern (Figures S3 and S4, Supporting Information) also shows NCs show a very sharp absorption edge, the PbBr2 absorpthat a full conversion of the PbI2 and PbBr2 NCs has occurred tion line is affected by strong scattering effects (see Figure S1, Supporting Information). The Energy Dispersive X-Ray (EDX) since peaks related to the precursor materials are not observed. analysis confirms the stoichiometry of the crystals for both No preferential orientation can be observed by comparing the cases (Figure S2, Supporting Information): PbI2.02 and PbBr2.30. patterns with those reported in literature.[3] A top-view SEM The observed little excess of halogens probably derives from a image of a drop-casted film (Figure S6, Supporting Informacontamination of the nanocrystals by the solvent used for the tion) confirms the XRD analysis concerning the dimensions of synthesis of the materials. the grains for PVK-A sample and the relatively high roughness Perovskite inks can be prepared by mixing the NC preof the deposited film due to the drop-casting technique.[34] cursor solutions with MAI/MABr salts dissolved in isopropanol as described in the Experimental Section and presented in Figure 2. In Figure 2, the absorption spectra of thin films made from the four different inks are shown. As expected for mixed-halide perovskites, the absorption edge blue-shifts with increasing bromine content.[30] Importantly, upon photo-excitation, the emission wavelength, for all the thin films, follows the absorption edge, contrary to what has been observed for the same type of perovskite films obtained from other synthetic techniques. Prolonged excitation, as reported in Figure S5 (Supporting Information), does not influence the result, supporting the conclusion that ion migration and consequent formation of other phases are hindered. The stability of the emission properties of the synthesized PVKs shows that an improved quality of the synthetized semiconductors is obtained. This allows us to synthetize nanocrystals with different stoichiometry, which exhibit different bandgaps. Figure 2.  UV–vis absorption spectra of quartz-coated substrates with all In Figure 3 the SEM images of the perovskite nanocrystals the perovskite inks. In the inset picture, perovskite inks with increased are shown together with their exact stoichiometry obtained bromine content (from A to D). 1601703  (2 of 5)

wileyonlinelibrary.com

© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Adv. Energy Mater. 2017, 1601703

www.advancedsciencenews.com

www.advenergymat.de

Communication Figure 3.  Morphological characterization of the perovskites. SEM images of a) full iodine perovskite (PVK A), b) mixed halides perovskite with higher iodine content (PVK B), c) mixed halide perovskite with higher bromine content (PVK C), and d) full bromine sample (PVK D). e) (100) XRD peak of the four synthesized PVK samples, f) nanocrystal grain size calculated following the Scherrer equation and iodine content obtained from qualitative EDX analysis as a function of XRD (100) peak position. Error bars are calculated as reported in the Experimental Section.

To further verify the hypothesis of ion migration hindrance, we performed electrical measurements to observe the (photo) current response upon polarization, which can be correlated to ion migration within the perovskite thin film.[16,19] Figure 4a shows the SEM images of the interdigitated gold electrodes coated with a homogeneous film of PVK-C (mixed halide perovskite with excess content in Br) used for the

measurements. Electrical measurements of the device were performed in glovebox, under a nitrogen atmosphere. Applying an electric field of ±1.4 V μm−1, both in dark and under illumination (Figure 4c,d) we find that: (a) almost no hysteresis is observed during forward and backward polarization and (b) a symmetrical shape between forward and reverse bias application (black line—without polarization curves) is found. Recent

Figure 4.  Electrical characterization of lateral symmetric devices realized using interdigitated gold electrodes coated by PVK C ink. a) SEM images showing the interdigitate electrodes with a 10 µm channel and b) a zoom of the channel, c) dark and d) light current/voltage (I/V) curves with and without pre-polarization.

Adv. Energy Mater. 2017, 1601703

© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

wileyonlinelibrary.com

(3 of 5)  1601703

www.advenergymat.de

www.advancedsciencenews.com

Communication

still maintaining the possibility of fabricating conductive thin films which are electrically photo-stable. First of all, this will guarantee the possibility of fabricating solar cells with optimal optical band for multijunction applications. In addition, the clean tunability of the semiconductor bandgap along the vis– NIR spectrum, has immediate impact on the color tunability and optimization of perovskites for light-emitting applications.

Experimental Section

Figure 5.  Photocurrent time response for positive and negative applied electric field of 1.4 V μm−1 using gold contacts coated by mixed perovskite film with Br excess (PVK-C).

works[16,19] have shown that a p-n junction can be formed at the interface between perovskite and metal contact due to ion migration and accumulation, therefore, also giving the usual rectifying diode characteristic in “dark” conditions. On the contrary, we do not observe any rectifying features. In addition, the photocurrent time response with an applied positive electric field has instantaneous and fully symmetric response when the device polarization is alternated (Figure 5). These data confirm that that ion migration is strongly forbidden. In order to explain such results one should recall the fabrication methodology of the presented nanocrystals. The fundamental steps of the laser ablation assisted in solution, namely, the plasma formation, followed by the evolutions of shock waves and of cavitation bubbles, are intensively studied and many aspects of the formation of nanoparticles are understood.[35] However, other aspects of the nanoparticles formation are still not deeply explored, in particular the interaction of the plasma produced by the ablation (evaluated to be at higher than 2–3000 K)[36] with the solvent, which may decompose, leaving carbon materials on the nanoparticles surface as already proved, for example, for gold nanoparticles synthesized in toluene.[24] We may hypothesize that these carbon centers can prevent the ion motion although their concentration must be still low enough since the reaction with MAI or MABr occurs[37] and the Raman spectra of the films do not show evidence of the presence of carbon materials. Please note, that the prevalent role of the surface is in agreement with recent finding from Huang and co-workers who identify the grain boundaries as the main source and path for ion migration.[38] A parallel mechanism of ion motion hindrance can also be envisaged from the role of the solvent itself. In fact, it has been already demonstrated[39] that undercoordinated halogen ions on the perovskite surface can favor supramolecular halogen bond complexation which passivates defective sites, without affecting the connectivity and electrical conductivity between grains and layers. In conclusion, in this Communication we demonstrate the synthesis of ligands-free nanocrystal perovskite semiconductor inks, also suitable for large-area processing. They allow for a better control of the semiconductor quality prior deposition, 1601703  (4 of 5)

wileyonlinelibrary.com

Laser Ablation Synthesis and Perovskite Ink Production: PbI2 NCs and PbBr2 NCs were produced by LASiS. The second harmonic of a Nd:YAG laser (Quantel), at 532 nm, was focused, with a fluence of 1 J cm−2, on the surface of a PbI2 (or PbBr2) compressed powder tablet. The 9 ns laser pulses were used at 10 Hz repetition rate. Tablets were placed under the solvents used for the synthesis, namely, I-benzene and Br-benzene for PbI2 NCs and PbBr2 NCs, respectively. The ablation was performed for 20 min with aliquots of 5 mL, obtaining a final volume of 50 mL for each type of material. Concentration of lead halide contents was determined by centrifugation of the solutions (5 min at 30 000 RCF) and dissolution of the precipitate with dimethylformamide. The lead halide content of colloidal solutions was estimated by optical extinction. The molar extinction coefficients for PbI2 and PbBr2 were estimated by standard solutions of these salts in DMF: 7700 m−1 cm−1 for PbI2 at 322 nm and 3400 m−1 cm−1 for PbBr2 at 282 nm. The lead halide content for both the nanocrystal dispersions were in the range of 0.1 × 10−3 m. Perovskite inks are produced as follows: 999 µL of NCs colloidal solution are vigorously mixed under sonication with 1 µL of 0.2 m MAI/MABr isopropanol solution for 30 min in a sonication bath at 50 °C. The films are produced by drop-casting some µL of the ink on top of clean quartz substrates heated at 100 °C. Optical Characterization: UV–vis absorption was recorded with a Perkin Elmer Lambda 1050. Photoluminescence spectra were obtained with a Horiba JobinYvon spectrofluorimeter, exciting the sample with a xenon lamp source. Morphological Characterization: SEM images and EDX spectra were acquired with a Zeiss SUPRA40 Field Emission SEM, equipped with an Oxford EDX detector for qualitative EDX analysis. The measurement has an estimated error of 4%. X-Ray Analysis: The X-ray diffraction spectra were recorded with a BRUKER D8 ADVANCE diffractometer with Bragg–Brentano geometry and equipped with a Cu Kα1 (λ = 1.544060 Å) anode at operating voltage of 40 kV and operating current of 40 mA. All the diffraction patterns were recorded at room temperature over an angular range (2θ) between 10° and 60°, a step size of 0.020°, and an acquisition time of 1 s. Step size was used as error source for experimental data. The error of the grain size was calculated using the specific error propagation formula for Scherrer equation. Electrical Analysis: Interdigitated gold pads were used for electrical analysis. Gold contacts are 50 nm height on silica and channels between contacts are 2.5 and 10 µm wide. Perovskite inks were deposited by dropcasting a RT solution on hot substrates (100 °C) until the entire surface was homogeneously covered, as observed with an optical microscope. The electrical measurements were performed in a glovebox under a nitrogen atmosphere. An Agilent B1500 Semiconductor Parameter Analyzer (Agilent, USA) was used for electronic measurements scanning the voltage in order to apply an electric field from 0 V to ±1.4 V µm−1 both in dark and in light conditions. Polarization was obtained with the device applying an electric field of ±1.4 V µm−1 for 100 s before performing the measurement. Slow and fast scan (100 mV s−1 and 1 V s−1) was applied to all measurements.

Supporting Information Supporting Information is available from the Wiley Online Library or from the author.

© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Adv. Energy Mater. 2017, 1601703

www.advancedsciencenews.com

www.advenergymat.de

The research leading to these results received funding from Fondazione Cariplo (Project GREENS No. 2013-0656 and Project IPER-LUCE No. 2015-0080). Received: August 1, 2016 Revised: October 25, 2016 Published online:

[1] a) M. M. Lee, J. Teuscher, T. Miyasaka, T. N. Murakami, H. J. Snaith, Science 2012, 338, 643; b) H.-S. Kim, C.-R. Lee, J.-H. Im, K.-B. Lee, T. Moehl, A. Marchioro, S.-J. Moon, R. Humphry-Baker, J.-H. Yum, J. E. Moser, M. Grätzel, N.-G. Park, Sci. Rep. 2012, 2, 591. [2] M. Saliba, S. Orlandi, T. Matsui, S. Aghazada, M. Cavazzini, J.-P. Correa-Baena, P. Gao, R. Scopelliti, E. Mosconi, K.-H. Dahmen, F. De Angelis, A. Abate, A. Hagfeldt, G. Pozzi, M. Graetzel, M. K. Nazeeruddin, Nat. Energy 2016, 1, 15017. [3] H. Zhu, Y. Fu, F. Meng, X. Wu, Z. Gong, Q. Ding, M. V. Gustafsson, M. T. Trinh, S. Jin, X. Y. Zhu, Nat. Mater. 2015, 14, 636. [4] a) Z.-K. Tan, R. S. Moghaddam, M. L. Lai, P. Docampo, R. Higler, F. Deschler, M. Price, A. Sadhanala, L. M. Pazos, D. Credgington, F. Hanusch, T. Bein, H. J. Snaith, R. H. Friend, Nat. Nanotechnol. 2014, 9, 687; b) B. R. Sutherland, S. Hoogland, M. M. Adachi, C. T. O. Wong, E. H. Sargent, ACS Nano 2014, 8, 10947. [5] X. Ke, J. Yan, A. Zhang, B. Zhang, Y. Chen, Appl. Phys. Lett. 2015, 107, 091904. [6] A. Miyata, A. Mitioglu, P. Plochocka, O. Portugall, J. T.-W. Wang, S. D. Stranks, H. J. Snaith, R. J. Nicholas, Nat. Phys. 2015, 11, 582. [7] M. B. Johnston, L. M. Herz, Acc. Chem. Res. 2016, 49, 146. [8] G. Giorgi, J.-I. Fujisawa, H. Segawa, K. Yamashita, J. Phys. Chem. Lett. 2013, 4, 4213. [9] T. M. Koh, K. Fu, Y. Fang, S. Chen, T. C. Sum, N. Mathews, S. G. Mhaisalkar, P. P. Boix, T. Baikie, J. Phys. Chem. C 2014, 118, 16458. [10] S. D. Stranks, H. J. Snaith, Nat. Nanotechnol. 2015, 10, 391. [11] S.-G. Li, K.-J. Jiang, M.-J. Su, X.-P. Cui, J.-H. Huang, Q.-Q. Zhang, X.-Q. Zhou, L.-M. Yang, Y.-L. Song, J. Mater. Chem. A 2015, 3, 9092. [12] M. Ye, X. Hong, F. Zhang, X. Liu, J. Mater. Chem. A 2016, 4, 6755. [13] a) W. Zhang, S. Pathak, N. Sakai, T. Stergiopoulos, P. K. Nayak, N. K. Noel, A. A. Haghighirad, V. M. Burlakov, D. W. deQuilettes, A. Sadhanala, W. Li, L. Wang, D. S. Ginger, R. H. Friend, H. J. Snaith, Nat. Commun. 2015, 6, 10030; b) H. He, Q. Yu, H. Li, J. Li, J. Si, Y. Jin, N. Wang, J. Wang, J. He, X. Wang, Y. Zhang, Z. Ye, Nat. Commun. 2016, 7, 10896; c) W. Nie, H. Tsai, R. Asadpour, J.-C. Blancon, A. J. Neukirch, G. Gupta, J. J. Crochet, M. Chhowalla, S. Tretiak, M. A. Alam, H.-L. Wang, A. D. Mohite, Science 2015, 347, 522; d) F. Deschler, M. Price, S. Pathak, L. E. Klintberg, D.-D. Jarausch, R. Higler, S. Hüttner, T. Leijtens, S. D. Stranks, H. J. Snaith, M. Atatüre, R. T. Phillips, R. H. Friend, J. Phys. Chem. Lett. 2014, 5, 1421. [14] H. Cho, S.-H. Jeong, M.-H. Park, Y.-H. Kim, C. Wolf, C.-L. Lee, J. H. Heo, A. Sadhanala, N. Myoung, S. Yoo, S. H. Im, R. H. Friend, T.-W. Lee, Science 2015, 350, 1222. [15] a) G. Niu, X. Guo, L. Wang, J. Mater. Chem. A 2015, 3, 8970; b) B. Saparov, D. B. Mitzi, Chem. Rev. 2016, 116, 4558. [16] M. De Bastiani, G. Dell’Erba, M. Gandini, V. D’Innocenzo, S. Neutzner, A. R. S. Kandada, G. Grancini, M. Binda, M. Prato, J. M. Ball, M. Caironi, A. Petrozza, Adv. Energy Mater. 2016, 6, 1501453.

Adv. Energy Mater. 2017, 1601703

[17] E. T. Hoke, D. J. Slotcavage, E. R. Dohner, A. R. Bowring, H. I. Karunadasa, M. D. McGehee, Chem. Sci. 2015, 6, 613. [18] a) Y. Zhao, C. Liang, H. Zhang, D. Li, D. Tian, G. Li, X. Jing, W. Zhang, W. Xiao, Q. Liu, F. Zhang, Z. He, Energy Environ. Sci. 2015, 8, 1256; b) T. Leijtens, G. E. Eperon, N. K. Noel, S. N. Habisreutinger, A. Petrozza, H. J. Snaith, Adv. Energy Mater. 2015, 5, 1614. [19] Z. Xiao, Y. Yuan, Y. Shao, Q. Wang, Q. Dong, C. Bi, P. Sharma, A. Gruverman, J. Huang, Nat. Mater. 2015, 14, 193. [20] R. J. Sutton, G. E. Eperon, L. Miranda, E. S. Parrott, B. A. Kamino, J. B. Patel, M. T. Hörantner, M. B. Johnston, A. A. Haghighirad, D. T. Moore, Adv. Energy Mater. 2016, 6, 1502458. [21] G. Xing, N. Mathews, S. S. Lim, N. Yantara, X. Liu, D. Sabba, M. Grätzel, S. Mhaisalkar, T. C. Sum, Nat. Mater. 2014, 13, 476. [22] a) L. Protesescu, S. Yakunin, M. I. Bodnarchuk, F. Krieg, R. Caputo, C. H. Hendon, R. X. Yang, A. Walsh, M. V. Kovalenko, Nano Lett. 2015, 15, 3692; b) Q. A. Akkerman, S. G. Motti, A. R. S. Kandada, E. Mosconi, V. D’Innocenzo, G. Bertoni, S. Marras, B. A. Kamino, L. Miranda, F. De Angelis, A. Petrozza, M. Prato, L. Manna, J. Am. Chem. Soc. 2016, 138, 1010; c) Q. A. Akkerman, V. D’Innocenzo, S. Accornero, A. Scarpellini, A. Petrozza, M. Prato, L. Manna, J. Am. Chem. Soc. 2015, 137, 10276. [23] V. Amendola, M. Meneghetti, J. Mater. Chem. 2007, 17, 4705. [24] V. Amendola, S. Polizzi, M. Meneghetti, J. Phys. Chem. B 2006, 110, 7232. [25] J. Burschka, N. Pellet, S.-J. Moon, R. Humphry-Baker, P. Gao, M. K. Nazeeruddin, M. Gratzel, Nature 2013, 499, 316. [26] Y. Hassan, Y. Song, R. D. Pensack, A. I. Abdelrahman, Y. Kobayashi, M. A. Winnik, G. D. Scholes, Adv. Mater. 2016, 28, 566. [27] F. Zhang, H. Zhong, C. Chen, X.-G. Wu, X. Hu, H. Huang, J. Han, B. Zou, Y. Dong, ACS Nano 2015, 9, 4533. [28] a) J.-P. Sylvestre, S. Poulin, A. V. Kabashin, E. Sacher, M. Meunier, J. H. T. Luong, J. Phys. Chem. B 2004, 108, 16864; b) H. Muto, K. Yamada, K. Miyajima, F. Mafuné, J. Phys. Chem. C 2007, 111, 17221. [29] C. Chenard-Lemire, L. J. Lewis, M. Meunier, Appl. Surf. Sci. 2012, 258, 9404. [30] P. Gao, M. Gratzel, M. K. Nazeeruddin, En. Environ. Sci. 2014, 7, 2448. [31] a) A. Suzuki, H. Okada, T. Oku, AIP Conf. Proc. 2016, 1709, 020022; b) Y. Zhao, A. M. Nardes, K. Zhu, Faraday Discuss. 2014, 176, 301; c) S. A. Kulkarni, T. Baikie, P. P. Boix, N. Yantara, N. Mathews, S. Mhaisalkar, J. Mater. Chem. A 2014, 2, 9221; d) N. Pellet, J. Teuscher, J. Maier, M. Grätzel, Chem. Mater. 2015, 27, 2181. [32] A. L. Patterson, Phys. Rev. 1939, 56, 972. [33] P. Scardi, M. Leoni, R. Delhez, J. Appl. Crystallogr. 2004, 37, 381. [34] M. Eslamian, F. Zabihi, Nanoscale Res. Lett. 2015, 10, 462. [35] a) S. Ibrahimkutty, P. Wagener, T. d. S. Rolo, D. Karpov, A. Menzel, T. Baumbach, S. Barcikowski, A. Plech, Sci. Rep. 2015, 5, 16313; b) A. Tamura, A. Matsumoto, K. Fukami, N. Nishi, T. Sakka, J. Appl. Phys. 2015, 117, 173304. [36] S. Tetsuo, T. Kazuhiro, H. O. Yukio, M. Mahito, J. Phys. D: Appl. Phys. 2002, 35, 65. [37] F. Palazon, Q. A. Akkerman, M. Prato, L. Manna, ACS Nano 2016, 10, 1224. [38] Y. Shao, Y. Fang, T. Li, Q. Wang, Q. Dong, Y. Deng, Y. Yuan, H. Wei, M. Wang, A. Gruverman, J. Shield, J. Huang, Energy Environ. Sci. 2016, 9, 1752. [39] A. Abate, M. Saliba, D. J. Hollman, S. D. Stranks, K. Wojciechowski, R. Avolio, G. Grancini, A. Petrozza, H. J. Snaith, Nano Lett. 2014, 14, 3247.

© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

wileyonlinelibrary.com

(5 of 5)  1601703

Communication

Acknowledgements