Symposium EP3 : Perovskite-Based Photovoltaics and Optoelectronic Devices

The remarkable optoelectronic properties of organohalide-lead perovskite materials show resilience to defects and grain boundaries which inspires hope for their use in low-cost, next generation photovoltaic technology. However, little is known about the thermomechanical properties of perovskite properties and the effect of these defects and grain boundaries on their resistance to fracture. In our preliminary mechanical evaluations of perovskite solar cells, we have consistently measured fracture energies so low so as to prohibit the realistic utilization of perovskites in PV devices. In addition to the compromising mechanical integrity of the perovskite active layer, the auxiliary hole and electron transporting materials critical for cell performance also exhibit very poor mechanical strength and adhesion to the perovskite layer. It is therefore imperative that the fundamental thermomechanical properties of perovskite solar cells are better understood and that the advances are made in both engineering the mechanical integrity of perovskite materials and in developing robust hole and electron transporting layers that adhere well to perovskite active layers in PV devices.
After identifying that PC₆₀BM thin films present in several perovskite architectures as electron transporting layers exhibit the lowest resistance to fracture amongst any other component present, we designed a cross-linkable fullerene alternative. This fullerene derivative features a styrene substituent that is designed to both imbue solubility to the fullerene facilitating the deposition of uniform thin films and provide an avenue for cross-linking following deposition through thermal curing. Mechanical analysis of solution deposited thin films of this fullerene derivative exhibit improvements in the fracture resistance in comparison to PC₆₀BM. Interestingly, even as-deposited films exhibit a ten-fold enhancement in fracture resistance while films cured by thermal annealing can be transformed into solvent resistant films. Perovskite devices fabricated with this fullerene electron transporting layer exhibit high photovoltaic conversion efficiencies, comparable to those achieved when PC₆₀BM is utilized.

Achieving a power conversion efficiency (PCE) of 20% for perovskite solar cells would attract the attention of investors and the solar PV industry who are interested in the commercialisation of printed solar films. Using a turnkey roll-to-roll printer is recognised as a cost-effective method for producing solar films. As a principal member of the Victorian Organic Solar Cell Consortium (VICOSC), the Commonwealth Scientific and Industrial Research Organisation (CSIRO) has developed the capability to print organic solar modules on 10cm and 30cm wide plastic films. The CSIRO has also devoted its efforts to perovskite research and developed a process for deposition of methylammonium lead triiodide (MAPbI₃) by scalable lab-to-fab slot die coating. The resulting 15% PCE achieved on glass substrates is comparable with those made by non-scalable spin coating processes. However, MAPbI₃ is inherently unstable and undergoes a reversible phase transition between tetragonal and cubic phase at operating temperatures. To address this instability, we have substituted MAPbI₃ with the relatively photo-stable formamidinium lead trihalide (FAPbX₃) that is insensitive to phase transition. Using the same slot die coating conditions, FAPbX₃ yielded equivalent or improved efficiency indicating that large scale printing of FAPbX₃ films via roll-to-roll manufacturing is feasible. A comparison of stability between the two types of perovskite cells under various conditions will be discussed and further improvement of perovskite device structure will be presented.

The rapid rise in power conversion efficiency of perovskite solar cells, crossing 20% in just few years, shows the huge potential of this new technogy to emerge as an alternative to the various existing photovoltaic technologies. However, there are several issues, such as hysteresis, performance stability, materials durability, which need to be addressed prior to commercialization of the perovskite based solar cells. Although decomposition of perovskite in ambient operating conditions is considered as a major practical problem at present, we observed even a more serious problem in the perovskite based devices. In a recent study, we found that perovskite solar cells show photo-induced performance degradation; continuous drop in photocurrent density (J_sc) uder illumination (1 sun), both in close and open circuit conditions of the devices. Interestingly, such deterioration in the performance was found not to be due to decomposition of perovskite as no photobleaching of the absorber (perovskite) was observed after cyclic photocurrent-dark current or light soaking measurements. Besides, the performacne could recover partially or fully by relaxing the devices in dark for hours. This recovery in performance on storing the devices in dark was found to depend strongly on the device structure; planar structure cells showed partial recovery while TiO₂ mesoporous devices showed complete recovery. Hence, the photo-induced performance degradation is believed to be due to a temporary electronic/ionic change (reversible) occuring in the device under illumination. Such change in the perovskite film or at the interfaces, which prevents carrier extraction, must be influenced by compositon of perovskite and configurtion of the device. Here, we report how composition of perovskite, which was varied by changing the composition of the precursors, and device configuration (thickness and morphology of TiO₂ compact layer and perovskite layer) influence the photo-induced performance degradation behaviour of perobskite solar cells.

Over the last few years the power conversion efficiency of organometal-halide perovskite (such as CH₃NH₃PbI₃) based solar cells has skyrocketed at an unprecedented pace to values around 20%. However, device progress overshadows our understanding of physical/chemical processes in the material. For instance, the susceptibility of CH₃NH₃PbI₃-based devices to water, despite being well documented, is still not well understood and as such remains controversial. Infrared (IR) spectroscopy is a powerful method to investigate physical and chemical properties of hybrid perovskites [1]. Herein we use IR spectroscopy and electrical characterization in controlled atmospheres to demonstrate that water infiltration into CH₃NH₃PbI₃ occurs much faster and at much lower humidity than previously thought. We propose a molecular model where water molecules affect the hydrogen bonding between the methylammonium cations and the Pb-I cage. Furthermore, the exposure of CH₃NH₃PbI₃ to ambient environment leads to an increase of the photocurrent by more than one order of magnitude in lateral devices. Based on transient photocurrent measurements we speculate that the effect is associated with enhanced proton conduction when light is combined with water and oxygen exposure. Our results suggest that water infiltration plays an important role in the opto-electronic properties of CH₃NH₃PbI₃ based devices, well beyond the known water triggered degradation processes.

References:
[1] The Journal Of Physical Chemistry Letters, 2015, Volume 6, p. 2913–2918, T. Glaser et al.

Solar cells based on CH₃NH₃Pb(I,Br)₃ have attracted substantial recent interest, because of the unprecedented rise in power conversion efficiency to values above 20%, which in turn is made possible by the near ideal band gap, strong optical absorption, high carrier mobilities, long minority carrier lifetimes, and relatively benign defects and grain boundaries for the perovskite absorber. Despite the high power conversion efficiencies, the incorporation of a soluble (leachable) form of the heavy metal lead, coupled with issues of device stability and electrical hysteresis, pose barriers to commercializing this exciting technology. One fundamental question for the perovskite PV field is to understand how important Pb is to the high device performance, as well as to the issues of stability and hysteresis. In this talk, we will explore prospective solar cell absorber materials beyond CH₃NH₃Pb(I,Br)₃ and address what the prospects are for finding alternative perovskite-related systems that can offer high performance, while mitigating some of the other issues mentioned above.

Organic-inorganic hybrid perovskites are a special class of semiconductors that have revolutionized the prospects of emerging photovoltaic technologies, in forms of both light harvesters and hole transport materials. The inorganic chemistry of this class of materials is fascinating. These compounds adopt the ABX₃ perovskite structure, which consists of a network of corner-sharing BX₆ octahedra, where the B atom is a divalent metal cation (typically Ge²⁺, Sn²⁺ or Pb²⁺) and X is a monovalent anion (typically Cl⁻, Br⁻, I⁻); the A cation is selected to balance the total charge and it can be a Cs⁺ or a small molecular species. Such perovskites afford several important features including excellent optical properties that are tunable by controlling the chemical compositions, they exhibit ambipolar charge transport with high mobilities. Some members exhibit long electron and hole diffusion lengths. The fundamental similarities and differences between MeNH₃PbI₃, MeNH₃SnI₃ and MeNH₃GeI₃ perovskites as well as other low dimensional materials will be discussed.

We report Sn/Pb perovskite solar cells with high short circuit current (Jsc:30mA/cm²) by photoconverting the near IR light up to 1050nm. Conventional MAPbI3 perovskite solar cells can harvest light up to 800nm and 20-23mA/cm2 of Jsc has been reported. In order to increase the Jsc more, we have alrady reported Sn/Pb binary perovskite solar cells photoconverting the near IR light up to 1050nm with 20mA/cm2 of Jsc. The crystal disordering was compared with each other by using Urbach energy (steepness factor) which was measured by Photoacoustic spectra. The steepness factor was initially 0.5-0.6 which was extremely lower than 1.2 of MAPbI3 (MA:methylammonium cation), suggesting that the binary perovskite solar cells had much more disordering than MAPbI3. The fabrication process was improved and finally the steepness factor was increased to 1.3. By reducing the disordering density in the crystal, we achieved 30mA/cm2 of Jsc. The point of the improvement was the starting materials and the process. Previously we used mixture of SnI2 and PbI2. This time for a new method, we used mixture of SnI2 (DMSO)2 complex and PbI2(DMSO)2 complex instead of SnI2 and PbI2. In addition, anti-solvent method was used during spincoating process. For Sn/Pb perovskite solar cells with 20mA/cm2 of Jsc, the charge separation ratio calculated by charge carrier dynamics analysis obtained by transient spectroscopy measurement was 90% and charge collection ratio was 67%. The low charge collection ration was improved to 90 % after the new process was used (30mA/cm2 of Jsc). Therefore, the increase in the Jsc is explained by the increase in the charge collection ratio. From the analysis of the optical loss in the range from 400nm to 1100nm, reflection loss from glass surface was large in the range from 800nm to 1000nm, reaching 20% to 60 % loss. Scattering loss was 20-30% in the range from 900nm to 1000nm. In order to enhance the Jsc more, anti-reflective coating on the glass and light confinement structures are needed. We analyzed Sn/Pb ratio from the top to the bottom in the Sn/Pb binary perovskite layer. It was found that Pb was rich on the top and Sn was rich in the bottom layer. Since conduction band and valence band levels vary by the Sn/Pb ratio, graded band structure in Sn/Pb binary layer was expected. The graded band structure was supported by very long charge recombination time reaching several ms in TiO2/MASnPbI3 binary perovskite layer, where the recombination time of TiO2/MAPbI3 layer was shorter than micro second. The graded band structure is also observed in CIGS compound type solar cells. The similarity to CIGS is discussed from the view point of graded band structure.

Metal-halide perovskites have been actively researched in the past few years as an efficient light harvester material, that can be solution processed at low temperatures. The highest certified power conversion efficiency of these materials has reached 20%. This class of materials can also be incorporated into other devices such as light emitting diodes and lasers. In this presentation we will present results of computational efforts, supported by experimental synthesis and characterization studies, aimed at identifying lead-free halides with desirable properties for replacing the lead-based halide perovskites that have been a main focus of research in this field. We present results for band gaps and thermodynamic stability derived by a computational screening study based on density functional theory methods. In total, 360 AMX₃ (A = K, Rb, Cs; X = Cl, Br, I; M = divalent cation) perovskite compounds [1] and 297 A₂MX₆ (A = K, Rb, Cs; X = Cl, Br, I; M = tetravalent cation) perovskite-derived compounds were considered. Their thermodynamic stability was evaluated through the Materials Project database with calculated energies to determine whether they would decompose to simpler binary phases. Six energetically stable compounds with desirable band gaps were identified as promising light harvesters: RbSnBr₃, CsSnBr₃, CsGeI₃, K₂MnBr₆, K₂PdBr₆, and Rb₂NiBr₆. The second goal of this study is based on calculations of the so-called branch point energy, or charge neutrality level, derived directly from the band structures of these semiconductor compounds. The position of branch point energy relative to the conduction band minimum and valence band maximum can be used to predict whether the intrinsic defects are n-type, p-type or self-compensating. Using this simple criteria, CsSnI₃ was predicted to be intrinsically p-type, consistent with experiment, while CsGeI₃ and CsPbI₃ were predicted to have self-compensating intrinsic defects. In addition, the assumption that the position of branch point energy displays a near-universal value relative to the vacuum level [2, 3], can be used to align the band positions of different semiconductors in a high-throughput manner, which provides useful guidance in tandem solar cell design.

[1] T. Krishnamoorthy, H. Ding, C. Yan, W. L. Leong, T. Baikie, Z. Zhang, M. Sherburne, S. Li, M. Asta, N. Matthews, and S. B. Mhaisalkar, “Lead-free Germanium Iodide Perovskite Materials for Photovoltaic Application,” J. Mater. Chem. A (accepted)
[2] Chris G. Van de Walle and J. Neugebauer. Universal alignment of hydrogen levels in semiconductors, insulators and solutions. Nature 423, 626-628 (2003).
[3] W. Walukiewicz. Intrinsic limitations to the doping of wide-gap semiconductors, Physica B 302–303, 123–134 (2001).

Acknowledgement: This work was supported by the Singapore Berkeley Research Initiative for Sustainable Energy(SinBeRISE) Program. We thank Chen Yan, Wei Lin Leong, Tom Baikie and Shuzhou Li for discussions and support.

Hybrid organic-inorganic perovskites (such as methylammonium-lead-iodide, MAPbI₃) have proven to be excellent light-harvesting materials for photovoltaic applications as they yielded unprecedented high power conversion efficiencies compared to conventional dye-sensitized solar cells. However, these materials exhibit limitations (such as long-term stability) as well as anomalies that have been matter of intense investigations: large apparent dielectric constant ε^/ at low frequencies and pronounced hysteresis during voltage-current sweeps. Common explanations in literature invoke the occurrence of ferroelectricity or trapping/detrapping of mobile charge carriers at grain boundaries or interfaces.
In this contribution, we present the results of d.c. polarization measurements as well as a.c. impedance spectroscopy (performed in dark and under illumination) proving that both ion and electron conductivities contribute to the overall electrical transport properties of MAPbI₃ [1]. In particular, we demonstrate that the ionic conduction, which is due mostly to iodine migration, is significant and can even exceed the electronic conduction. Moreover, the d.c. experiments allow for following the kinetics of stoichiometric changes, from which the characteristic relaxation time τ^δ and chemical diffusion coefficient D^δ of iodine was determined (2.4x10^-8 cm²s^-1).
The ionic conductivity of such compounds naturally explains the anomalies mentioned above. If MAPbI₃ is enclosed between two electrodes that block the ions - as it is the case during photoelectrochemical operation - and is charged with a d.c. current, a stoichiometric polarization occurs resulting in a large apparent dielectric constant ε^/ at low frequencies. The same stoichiometric polarization is also responsible for a pronounced hysteretic behavior during i-V sweeps, whose extent depends on the ratio between the duration of a single sweep cycle and τ^δ. To further corroborate the experimental findings, we also verify these effects by modelling the conduction properties of MAPbI₃ with a relevant equivalent circuit.


[1] T.-Y. Yang et al., Angew. Chem. Int. Ed. 54 [27], 7905-7910 (2015).

Thin-film solar cells based on organometallic halide perovskite absorber layers have become a topic of intense research for high-performance photovoltaics. Since the first report on perovskite-based solar cells by Kojima et al. [1] the efficiency of these cells has increased rapidly up to 20.1% [2] when using the CH₃NH₃PbI₃ perovskite. During the last year significant attention was focused on the so-called photoluminescence (PL) enhancement phenomenon [3, 4]. This effect clearly proves that the actual state of CH₃NH₃PbI₃ in a working device is not exclusively dictated by its deposition parameters. During illumination, defects at the surface and in the bulk of the perovskite crystal can be deactivated, such that PL quantum yield and overall electrical properties (diffusion length, operating voltage etc.) of the material and device are enhanced.
The precise mechanism behind such PL enhancement is still unclear, however. Based on experiments in controlled atmospheres, oxygen was identified to play a crucial role in this process [4]. But the same effect was also obtained on finished solar cells [3], i.e. without direct contact between the absorber and ambient atmosphere. We have observed a similar discrepancy for PL measurements, where we tuned the excitation volume by changing the excitation laser wavelength, and thus absorption depth in the layer. This intriguing result motivated us to probe the PL enhancement effect by complementary characterization methods. Since the radiative recombination scales with the electrical quality of the perovskite layer, the PL data were compared to the electronic properties of perovskite films, simultaneously acquired by Fourier-transform photocurrent spectroscopy (FTPS). The latter technique was recently used to demonstrate the high material quality of perovskite films, as evidenced by their sharp absorption edge [5]. This time, we will discuss in detail the change of the absorption edge with illumination (defect detection) and electrical current. Surprisingly, even if the local electrical quality improves, the photo-current detected during FTPS measurement decreases with illumination time. We introduce a microscopic model based on grain boundaries to explain the measured dependences.
[1] A. Kojima, K. Teshima, Y. Shirai and T. Miyasaka: J. Am. Chem. Soc. 131 (2009) 6050.
[2] NREL chart – Best Research-Cells Efficiencies.
[3] S. D. Stranks, V. M. Burlakov, T. Leijtens, J. M. Ball, A. Goriely, and H. J. Snaith: PHYS. REV. APPLIED 2 (2014) 034007.
[4] Y. Tian, M. Peter, E. Unger, M. Abdellah et al.: Phys. Chem. Chem. Phys. 17 (2015) 24978.
[5] S. De Wolf, J. Holovsky, S.-J. Moon, P. Löper, B. Niesen, M. Ledinsky, et al.: J. Phys. Chem. Lett. 5 (2014) 1035.

Hybrid perovskite based solar cells have already exceeded 20.1% certified power conversion efficiencies. However, a clear understanding of the relationship between the optoelectronic properties of these materials and their morphology is lacking due to the large degree of structural variability within samples. Hybrid perovskites are usually deposited as polycrystalline thin-films with variable mesoscale morphology. Depending on the growth conditions, the obtained grain size typically ranges from tens to thousands of nm. To overcome this issue, perovskite single crystals are the ideal sample for understanding the inherent properties of the material, providing insight into their full potential once embodied in optoelectronic devices. Several recent studies have reported conflicting optical properties for MAPbI₃ (MA = CH₃NH₃⁺) single crystals.

Here we provide a unifying picture, by highlighting the variation of the structural and optical properties across the crystal surface at a micrometer scale and by combining optical and vibrational spectroscopy combined with confocal microscopy and ab-initio molecular dynamics simulations

We identify a structural inhomogeneity on a micrometer scale across the surface of a CH3NH3PbI3 perovskite single crystal. At the crystal edge a local distortion of the crystal lattice is responsible for a widening of the optical bandgap and a faster photo-carrier recombination. These effects are inherently present at the edge of the crystal, and are further enhanced upon water intercalation, as a preliminary step in the hydration of the perovskite material.

The rapid emergence of efficient solar cells based on metal-organic halide perovskites has stimulated research aimed at understanding the physical properties of these novel active layer materials. For example, the carrier diffusion length in methylammonium lead iodide (MAPbI) exceeds 1 μm, but this unusually high value for a solution-processed material is poorly understood. We recently developed first-principles calculations of the carrier lifetimes and diffusion lengths in semiconductors [1,2], which require accurate knowledge of the bandstructure.
In this talk, we show that in MAPbI the structure strongly affects the bandstructure and band edges, and that density functional theory (DFT) is unable to predict the room temperature tetragonal structure due to the polymorphism of MAPbI. The Rashba splitting induced by the spin-orbit interaction, and the DFT band gap and effective masses, all depend strongly on the chosen structure, a point that previous work failed to address. Working with multiple stochastic realizations of large unit cells with random orientations of the methylammonium ions, we compute average effective masses and show that the band gap depends linearly on the effective mass. The average Rashba coefficient we find is an order of magnitude smaller than previously reported, and the band edges are almost parabolic. Our structures possess the I4/mcm symmetry found experimentally, and are free of the spurious Pb off-centering assumed in previous work. We identify the correct starting point for accurate bandstructure calculations using the GW method, and thus the starting point to compute the carrier lifetime and diffusion length.
[1] M. Bernardi, D. Vigil-Fowler, J. Lischner, J. B. Neaton, and S. G. Louie, Phys. Rev. Lett. 112, 257402 (2014).
[2] M. Bernardi, D. Vigil-Fowler, C.S. Ong, J. B. Neaton, and S. G. Louie, PNAS 112, 5291 (2015)

Polycrystalline CH₃NH₃PbI₃ perovskite layers are employed in highly efficient solar cells to absorb light, to generate free charge carriers and to transport these charge carriers towards the respective electrodes. The influence of the perovskite crystal orientation on the charge carrier transport is yet to be fully understood. Currently, most publications focus on morphological features, surface roughness and substrate coverage of perovskite layers. Understanding the electronic properties on a sub-micrometer scale will foster design rationales for optimized charge carrier transport in multi-crystalline organometal halide perovskites.
In this work, we identify surface features and crystal grains by Atomic Force Microscopy (AFM) and correlate their orientation with local optoelectronic properties. The work function as determined by Kelvin Probe Force Microscopy (KPFM), can vary several hundred mV between neighbouring grains which may originate from different crystal faces stemming from the growth process during layer deposition from solution and post treatment. This variation in the work function influences local charge carrier transport at the interface to the adjacent buffer layer. Any mismatch of energy levels leads to charge carrier accumulation in the device and locally increased recombination losses. To correlate optoelectronic properties and crystal structures, we determine the crystal orientation of single grains employing Electron Backscattering Diffraction (EBSD), thereby calculating the crystal face of the surface. By combining these methods with Conductive Atomic Force Microscopy (c-AFM) we spatially correlate charge carrier transport and crystal orientation.
These microscopic observations allow for further improvement of perovskite layer deposition and to specifically optimize crystal growth in order to enhance charge carrier transport and reduce recombination losses within solar cells.

3D organic-inorganic halide perovskites (OIHPs) are compounds characterized by the formula AMX₃ (A=organic cation; M=Ge, Sn, Pb; X=halide) whose hybrid nature is conferred by the presence of organic cations that fit the semiconductor network cavities according to well established tolerance size parameters [1]. The renewed interest for 3D OIHPs (2D OIHPs were already investigated for optoelectronics applications [2]) stems from their superior features as light harvester in photovoltaic (PV) devices due to a wide range of unique properties [3, 4]. From the very first assembled perovskite based cells [5] characterized by photoconversion efficiencies (PCEs) of ~3.5%, impressive breakthroughs have been achieved, leading to devices with PCE > 20% [6]. Methylammonium lead iodide (MAPbI₃, MA=⁺CH₃NH₃) represents the most widely employed OIHP due to its high compatibility with solution-based processing, a high absorption coefficient, and a bandgap close to the optimal one for single junction solar cells even if OIHPs characterized by the presence of other cations may find applicability in solar cells [7].
While nowadays OIHP bulk properties are massively investigated as testified by the huge number of papers recently appeared in literature, there is total lack of knowledge of the chemico-physical properties of OIHP clusters and quantum dots. This is surprising in view of their possible application not only in PV but also in lasing [8] and as quantum emitters [9]. In the present contribution by means of a first principle campaign of calculations, we offer the first description of the structural and electronic properties of variable size OIHP clusters focusing on analogies and differences with bulk ones.

[1] (a) V. M. Goldschmidt, 1926 Naturwissenschaften, 14 477; (b) G. Kieslich et al., 2014 Chem. Sci. 5 4712.
[2] (a) D. B.Mitzi et al., 1995 Science, 267 1473; (b) D. B. Mitzi, 2000 Inorg. Chem., 39 6107.
[3] (a) J.-H. Heo et al., 2013 Nature Photonics 7 486 ; (b) L. Etgar et al., 2012 J. Am. Chem. Soc. 134 17396; (c) W. A. Laban et al., 2013 Energy Environ. Sci. 6 3249; (d) M. M. Lee et al., 2012 Science 338 643; (e) G. Giorgi et al., 2013 J. Phys. Chem. Lett. 4 4213.
[4] (a) G. Xing et al., 2013 Science 342 344; (b) S. D. Stranks et al., 2013 Science 342 341.
[5] A. Kojima et al., 2009 J. Am. Chem. Soc. 131 6050.
[6] http://www.nrel.gov/ncpv/images/efficiency_chart.jpg
[7] (a) A. Amat et al., 2014 Nano Lett. 14 3608; (b) G. Giorgi et al., 2015 J. Phys. Chem. C 119 4694 (Editor’s choice).
[8] A. Fu et al., 2015 Nature Materials, 14 557.
[9] Y.-S. Park et al., ACS Nano, just accepted.

We use symmetry analysis, density functional calculations and k.p modeling to scrutinize Rashba and Dresselhaus effects in hybrid organic-inorganic halide perovskites [1]. This class of perovskite currently acts as the superstar of photovoltaic technology. Due to a large spin-orbit coupling [2], it has also demonstrated great potential for spin-based applications [3]. With a detailed study of the electronic structures of a variety of system as well as the symmetries [4] we detail the origin of the spin splitting in two- and three-dimensional hybrid perovskites. Finally, we show that layered structures of CH₃NH₃PbX₃ (X=I, Br) leads to a splitting that can be controlled by an applied electric field. From these observations we propose a setup for a spin-FET, opening the door to a perovskite-based spintronics.

[1] M. Kepenekian, R. Robles, C. Katan, D. Sapori, L. Pedesseau, J. Even, ACS Nano, DOI: 10.1021/acsnano.5b04409.
[2] J. Even et al., J. Phys. Chem. Lett. 4, 2999 (2013).
[3] J. Even et al., Phys. Status Solidi RRL 8, 31 (2014). M. Kim et al., Proc. Natl. Acad. Sci. U.S.A. 111, 6900 (2014). F. Brivio et al., Phys. Rev. B 89, 155204 (2014). A. Stroppa et al., Nature Commun. 5, 5900 (2014).
[4] C. Katan et al., J. Mater. Chem. A 3, 9232 (2015). J. Even et al., J. Phys. Chem. C 119, 10161 (2015).

Organometal trihalide perovskites (OTP) attract interest because bycoupling exceptional optoelectronic properties with low-cost, solution-processability they yieldi high efficiency phototovoltaic devices. Although OTP properties are remarkable, it generally believed that they could be improved further. In this talk, I will characterize the light absorption properties OTP films and devices from the visible to the mid-IR using the photothermal induced resonance (PTIR) technique. PTIR uses an atomic force microscope tip in contact mode to transduce the sample thermal expansion, induced by light absorption.
Very recently, it was observed that the photovoltaic effect of OTP film sandwiched between two gold electrodes could be switched on and off repeatedly by applying a small voltage.[1] It was suggested that ion migration may be responsible for this novel effect. The topography and PTIR (due to methyl ammonium ion absorption) images before poling were homogeneous, reflecting the uniform distribution of the methyl ammonium ions (MA+) in the film. After electrical poling the topography images did not show significant changes. In contrast, the corresponding PTIR maps, show stronger MA+ intensity in proximity of the cathode as a function of the poling time. Since the PTIR signal is proportional to the sample absorbed energy, the PTIR maps indicate the redistribution in the concentration of the MA ions due to the electromigration of the MA ions [2].
We recently extended the PTIR spectra range to the visible, where light absorption can be used to study electronic transitions and the local bandgap. The incorporation of Cl- in mixed halide perovskite with general formula MAPbI_3-xCl_x is debated. Cl- incorporation in mixed halide perovskites alters the bandgap which can be used as a proxy for determining the local composition with PTIR. The as prepared sample, consist of a mixture of Cl-rich and Cl-poor phases with ≈ 0.06