In the context of an imperious energy transition that requires progress towards decarbonization of energy production, renewable energy sources are being developed, among which solar technologies. Each year, the sun provides the Earth with an amount of energy equivalent to more than 8,000 times the world's energy consumption. However, this energy is not very concentrated, hence the technological challenges involved in its efficient recovery. In 2018, solar (photovoltaic and thermal) accounted for 3 % of the world's electricity production. Over the last ten years, its growth rate has been close to 50 %, mainly due to the massive development of photovoltaics. The most commonly used semiconductor in photovoltaic technologies is silicon. But over the last ten years or so, important work has been conducted on other very promising materials such as halogenated hybrid perovskites. These, however, present challenges of stability and reproducibility.
MAPbI
3 (MA = CH
3NH
3) is the emblematic perovskite material for photovoltaic applications. But the low value of its formation energy leads to a strong dependence on the synthesis conditions, making the reproducibility of the chemical and structural properties of the synthesized thin films a challenge for the experimenters. However, mastering these properties is a prerequisite for optimizing the performance of photovoltaic cells.
Within the framework of a
collaboration, teams from the IRIG and the National Institute of Solar Energy (INES) have studied the mechanisms of formation of thin layers of MAPbI
33. They highlighted the effects of the intrinsic instability of iodinated perovskite, which decomposes under the effect of strain. In this study, researchers elucidate, by
in situ X-ray diffraction (XRD), the crystallization process of MAPbI
3 thin films in the presence of chlorine. Chlorine is known to have a beneficial effect on crystal quality, but the mechanisms are not yet clearly identified yet.
These teams demonstrate (
Figure 1) that methyl ammonium iodide (MAI) and lead chloride PbCl
2 spontaneously form a MAPbCl
3 perovskite layer, which is gradually transformed into a MAPbI
3 layer by halogen substitution during thermal annealing at 100 °C. The MAPbI
3 layer thus formed is highly strained due to the difference in lattice parameter between the two perovskites, so that above a critical threshold MAPbI
3 decomposes to form, among other compounds, lead iodide Pbl
2.
Photovoltaic cells were made from thin layers of MAPbI
3 whose synthesis was stopped at different stages of the ion substitution process. The influence of the chemical composition (presence or not of PbI
2) and the crystalline properties (in particular the strain) of the active layer on the performance of the devices could thus be studied.
Figure 1: Evolution of diffractograms obtained during annealing at 100 °C. The colors represent the successive stages during the annealing process: starting from a MAPbCl
3 layer, the first stage (dark brown) is characterized by the coexistence of the two perovskites MAPbCl
3 and MAPbI
3; it is followed by a stage in which the PbI
2 degradation phase is formed (light brown); the chlorinated perovskite then finally disappears in a third stage (green).
In addition, as frequently observed with compounds with a perovskite structure, MAPbI
3 is a
ferroelectric material: it undergoes a
ferroelastic cubic/tetragonal phase transition at around 57 °C (
Figure 2). Researchers have highlighted an important consequence of this ferroelasticity which induces a variability in the crystalline properties: the strained state of MAPbI
3 reached during its formation at 100 °C, in the cubic phase, determines the crystalline orientation of the layer at room temperature, in the tetragonal phase.
Figure 2: Around 57 °C the MAPbI
3 material undergoes a structural phase transition from high-temperature cubic symmetry to tetragonal symmetry.
This study confirms the intrinsic nature of the instability and structural variability of MAPbI
3 thin films. The introduction of different types of cations instead of MA alone is a promising way to improve stability. The study of such complex materials will benefit from the approach developed in this study, correlating crystallization mechanisms, structural properties and device efficiency.
The
perovskite crystal structure is common to many oxides of the general formula ABO3. This type of material has long generated a great deal of interest, in particular because of its frequent ferroelectric properties. For photovoltaic applications, these are not oxides but materials in which A and B are organic and inorganic ions respectively, and the oxygen is replaced by halogens, as in the emblematic compound MAPbI3 (MA = CH3NH3).
Collaboration: SGX and STEP teams, respectively from the MEM and SyMMES laboratories of IRIG, and the SMPV/LMPO team from the Institut National de l'Énergie Solaire.
A
ferroelectric material has several energy-equivalent stable states. The transition from one stable state to another can be achieved by the temporary application of an external magnetic, electrical or mechanical force respectively in the case of ferromagnetic, ferroelectric and ferroelastic materials.
MAPbI3 and ferroelasticity. During the structural phase transition that MAPbI3 undergoes at 57 °C, the transformation of the highly symmetrical cubic lattice into a less symmetrical tetragonal lattice induces, for a specific orientation of the cubic lattice, several possible equivalent orientations of the tetragonal lattice, hence the existence of crystalline defects called twins. A ferroelastic material is characterized, among other things, by the fact that in the phase of low symmetry, the application of a constraint can transform the orientation of a crystalline domain into an equivalent different orientation.