Mike Pols

Phonons in chiral crystal structures can be circularly polarized, making them chiral. Chiral phonons carry angular momentum, which is observable in heat currents, and, via coupling to electron spin, in spin currents. Two-dimensional (2D) halide perovskites, versatile direct band gap semiconductors, can easily form chiral structures by incorporating chiral organic cations. As a result, they exhibit phenomena such as chirality-induced spin selectivity (CISS) and the spin Seebeck effect, although the underlying mechanisms remain unclear. Using on-the-fly machine-learning force fields trained against density functional theory calculations, we confirm the presence of chiral phonons, a potential key factor for these effects. Our analysis reveals that low-energy phonons, originating from the inorganic framework, primarily exhibit chirality. Under a temperature gradient, these chiral phonons generate substantial angular momentum, leading to experimentally observable effects. These findings position chiral 2D perovskites as a promising platform for exploring the interplay among phononic, electronic, spintronic, and thermal properties.

With the use of chiral organic cations in two-dimensional metal halide perovskites, chirality can be induced in the metal halide layers, which results in semiconductors with intriguing chiral optical and spin-selective transport properties. The chiral properties strongly depend upon the temperature, despite the basic crystal symmetry not changing fundamentally. We identify a set of descriptors that characterize the chirality of metal halide perovskites, such as MBA\(_{2}\)PbI\(_{4}\), and study their temperature dependence using molecular dynamics simulations with on-the-fly machine-learning force fields obtained from density functional theory calculations. We find that, whereas the arrangement of organic cations remains chiral upon increasing the temperature, the inorganic framework loses this property more rapidly. We ascribe this to the breaking of hydrogen bonds that link the organic with the inorganic substructures, which leads to a loss of chirality transfer.

All-inorganic halide perovskites have received a great deal of attention as attractive alternatives to overcome the stability issues of hybrid halide perovskites that are commonly associated with organic cations. To find a compromise between the optoelectronic properties of CsPbI\(_{3}\) and CsPbBr\(_{3}\), perovskites with CsPb(Br\(_{x}\)I\(_{1-x}\))\(_{3}\) mixed compositions are commonly used. An additional benefit is that without sacrificing the optoelectronic properties for applications such as solar cells or light-emitting diodes, small amounts of Br in CsPbI\(_{3}\) can prevent the inorganic perovskite from degrading to a photo-inactive non-perovskite yellow phase. Despite indications that strain in the perovskite lattice plays a role in the stabilization of the material, a full understanding of such strain is lacking. Here, we develop a reactive force field (ReaxFF) for perovskites starting from our previous work for CsPbI\(_{3}\), and we extend this force field to CsPbBr\(_{3}\) and mixed CsPb(Br\(_{x}\)I\(_{1-x}\))\(_{3}\) compounds. This force field is used in large-scale molecular dynamics simulations to study perovskite phase transitions and the internal ion dynamics associated with the phase transitions. We find that an increase of the Br content lowers the temperature at which the perovskite reaches a cubic structure. Specifically, by substituting Br for I, the smaller ionic radius of Br induces a strain in the lattice that changes the internal dynamics of the octahedra. Importantly, this effect propagates through the perovskite lattice ranging up to distances of 2 nm, explaining why small concentrations of Br in CsPb(Br\(_{x}\)I\(_{1-x}\))\(_{3}\) (\(x\) \( \leq \) 1/4) have a significant impact on the phase stability of mixed halide perovskites.

The migration of defects plays an important role in the stability of halide perovskites. It is challenging to study defect migration with experiments or conventional computer simulations. The former lacks an atomic-scale resolution and the latter suffers from short simulation times or a lack of accuracy. Here, we demonstrate that machine-learned force fields, trained with an on-the-fly active learning scheme against accurate density functional theory calculations, allow us to probe the differences in the dynamical behaviour of halide interstitials and halide vacancies in two closely related compositions CsPbI\(_{3}\) and CsPbBr\(_{3}\). We find that interstitials migrate faster than vacancies, due to the shorter migration paths of interstitials. Both types of defects migrate faster in CsPbI\(_{3}\) than in CsPbBr\(_{3}\). We attribute this to the less compact packing of the ions in CsPbI\(_{3}\), which results in a larger motion of the ions and thus more frequent defect migration jumps.

The commercialization of perovskite solar cells is hindered by the poor long-term stability of the metal halide perovskite (MHP) light-absorbing layer. Solution processing, the common fabrication method for MHPs, produces polycrystalline films with a wide variety of defects, such as point defects, surfaces, and grain boundaries. Although the optoelectronic effects of such defects have been widely studied, the evaluation of their impact on the long-term stability remains challenging. In particular, an understanding of the dynamics of degradation reactions at the atomistic scale is lacking. In this work, using reactive force field (ReaxFF) molecular dynamics simulations, we investigate the effects of defects, in the forms of surfaces, surface defects, and grain boundaries, on the stability of the inorganic halide perovskite CsPbI\(_{3}\). Our simulations establish a stability trend for a variety of surfaces, which correlates well with the occurrence of these surfaces in experiments. We find that a perovskite surface degrades by progressively changing the local geometry of PbI\(_{x}\) octahedra from corner- to edge- to face-sharing. Importantly, we find that Pb dangling bonds and the lack of steric hindrance of I species are two crucial factors that induce degradation reactions. Finally, we show that the stability of these surfaces can be modulated by adjusting their atomistic details, by either creating additional point defects or merging them to form grain boundaries. While in general additional defects, particularly when clustered, have a negative impact on the material stability, some grain boundaries have a stabilizing effect, primarily because of the additional steric hindrance.