Solar cells that stack a perovskite layer on top of silicon, or on top of another perovskite, can turn more sunlight into electricity than any single-junction device. The upper layer of these "tandem" cells usually relies on mixed-halide perovskites, in which iodide and bromide ions are blended to tune how much of the solar spectrum the material absorbs. That chemical flexibility comes at a price. The films are highly sensitive to how they are processed, and under light, heat or moisture their internal structure can shift. A well-known example is halide segregation, in which the material separates into iodide-rich and bromide-rich regions, eroding performance and stability.
In a review published in Science Bulletin , researchers from the University of Electronic Science and Technology of China (UESTC) and the Technical University of Munich (TUM), together with colleagues at the Karlsruhe Institute of Technology, the Eastern Institute of Technology in Ningbo, Xiamen University and Northwestern Polytechnical University, argue that overcoming these problems requires watching the films while they change, rather than only inspecting them afterwards. The team presents a systematic methodological guide to synchrotron-based X-ray scattering, a family of techniques that can probe both the surface and the interior of a thin film without damaging it, at length scales ranging from the spacing between atoms to features several micrometers across.
Conventional tools such as electron microscopy or laboratory X-ray diffraction provide valuable snapshots, but they have difficulty capturing buried structures, short-lived intermediate phases and continuous processes. Synchrotron X-rays are bright enough to record scattering patterns in rapid succession, so experiments can be run while a film is being coated, heated, illuminated, exposed to humid air or operated as a working solar cell. Focusing the beam down to the nanoscale also makes it possible to map how the structure varies from point to point across, and through, a film.
The authors organize the field along three complementary axes. The first is the birth of a film. Using in situ grazing-incidence wide-angle X-ray scattering (GIWAXS), researchers can follow a perovskite from precursor solution through nucleation and crystal growth to the final solid layer. Comparing methylammonium-, formamidinium-, cesium-based and mixed-cation systems, the review shows that the halide composition strongly influences whether the desired cubic perovskite phase forms directly, how quickly crystals grow and how they are oriented. Bromide-rich formulations, for example, tend to crystallize faster and more directly into the cubic phase, which the authors attribute to lead-bromide bonds being stronger than the interactions between lead and the common solvent DMSO. High bromide contents can also cause halides to segregate while the film is still forming.
The second axis is ageing. By combining X-ray scattering with standardized stress protocols that control light, temperature, humidity and electrical bias, and by recording device performance at the same time, operando experiments can connect atomic-scale structural changes with the gradual loss of efficiency. In one study discussed in the review, the main diffraction peak of an iodide-bromide perovskite split into two after about an hour of illumination, a signature of separation into iodide-rich and bromide-rich phases, and then merged again when the light was switched off. Other work shows that films with more bromide can degrade faster in humid air, forming a hexagonal phase and hydrated compounds.
The third axis is spatial heterogeneity. Nano-focused X-ray beams can map crystal phase, lattice spacing, orientation and strain across a film surface and through cross-sections, and can be paired with X-ray fluorescence and optical mapping. Bringing the three axes together, the review traces an evidence chain from film formation to device failure: compositional non-uniformities created during crystallization can act as "seeds" that accelerate light-induced halide segregation, and residual lattice strain can retain a spatial "memory" of earlier segregation that guides where it recurs. The authors take care to distinguish direct correlations reported within individual studies from mechanistic pictures assembled across different studies, and identify experiments that combine time-resolved crystallization, nanoscale mapping and operando measurements on the very same film area as a priority for future research.
Beyond the science, the review serves as a practical handbook. It explains how key experimental choices, including sample-to-detector distance, incidence angle, penetration depth and resolution, determine what a measurement can reveal, and describes strategies to limit X-ray radiation damage to these sensitive materials. It also sets out methodological limitations, cautioning, for instance, that apparent peak shifts should not be interpreted as lattice strain without proper geometric correction, and that measurement conditions should be reported in full.
Looking ahead, the authors highlight three directions. The first is building in situ X-ray setups around scalable manufacturing methods such as slot-die coating, blade coating, inkjet printing, vapor deposition and roll-to-roll printing, as well as fast drying steps like gas quenching and intense pulsed light annealing, to close the gap between laboratory research and industrial production. The second is artificial intelligence: a single day of modern in situ experiments can generate millions of scattering images, and machine-learning tools for tasks such as automated diffraction-peak detection could turn this flood of data into rapid feedback for optimizing films. The third is tighter integration of X-ray scattering with complementary techniques, including X-ray fluorescence, X-ray absorption spectroscopy, resonant soft X-ray scattering and electron microscopy.
By providing advanced structural-probing strategies, the authors aim to help the research community overcome the film-quality bottlenecks of perovskite tandem solar cells and accelerate their path to commercial use.
Science Bulletin
Literature review