Researchers at the University of Groningen and TU Wien have identified the physical mechanism behind the unusually long "hot carrier" cooling times previously measured in tin-based perovskites, a property considered essential for a next-generation solar cell architecture aimed at exceeding the standard efficiency ceiling for single-junction devices.
Band filling and the hot phonon bottleneck combine to stretch carrier cooling times in tin perovskites into the nanosecond range. Credit: Faber et al., ACS Energy Letters
A hot-carrier solar cell (HCSC) works by extracting photo-excited charge carriers before they shed their surplus energy as heat, potentially pushing efficiency past the Shockley-Queisser limit of about 33% that caps conventional single-junction cells. The approach depends on carriers staying "hot" for nanoseconds rather than the picosecond timescales typical of most semiconductors. Several recent experimental studies on tin-based metal halide perovskites reported cooling times reaching several nanoseconds, but the physics driving this behavior had remained unclear.
Using ensemble Monte Carlo simulations benchmarked against previously published time-resolved photoluminescence measurements of CsSnI3, the team showed that the ultra-long cooling times arise only when two effects act together: the dynamic Burstein-Moss effect, in which low-energy states fill up with excited carriers and block further relaxation, and the hot phonon bottleneck, in which carriers reabsorb heat-carrying phonons faster than the phonons can decay away. Neither effect alone was sufficient to reproduce the measured cooling times.
In the simulations, cooling stretched from about 4 picoseconds with both effects switched off to about 470 picoseconds with both active at a carrier density of 3×10¹⁸ cm⁻³, in line with an experimentally measured value of about 636 picoseconds. The team also found that carriers affected by this mechanism remain "hot" in terms of their Fermi-Dirac temperature, not just their average kinetic energy, staying above 600 K even 200 picoseconds after excitation at the highest carrier densities tested - a distinction that matters because it is this thermodynamic temperature, rather than raw carrier energy, that determines the voltage boost an HCSC could theoretically deliver.
Based on these findings, the team proposed three material design criteria for future hot-carrier absorbers: a low effective mass to maximize the Burstein-Moss shift, which points toward tin-based compositions over their lead-based counterparts; a soft phononic lattice with low-frequency optical modes to lower the threshold for the phonon bottleneck to form; and high material purity, since defects tend to open additional pathways that accelerate rather than extend carrier cooling.
"The measurements were clear, but we didn't understand the physics behind this," said Jan Anton Koster, professor of Physics of Novel Semiconductors and Devices at the University of Groningen and the study's corresponding author, noting the team "even started to doubt the measurements ourselves." Adding the hot phonon bottleneck to the simulations "slowed the loss of energy, but not enough to explain our measurements," Koster said. Only once band filling was included as well did the simulated energy loss stretch into the nanosecond range, matching the experiments carried out by co-author Maria Antonietta Loi, professor of Photophysics and Optoelectronics at the university.
