Photoluminescence in halide perovskite films with shallow traps

One of the main reasons why lead halide perovskites are ideal candidates for photovoltaic and optoelectronic applications is that most of the defects are shallow. However, the possibility of lots of shallow defects was often seen as a minor issue by the spectroscopy and device physics communities, who usually assumed just deep defects to explain their data. The idea was that, in most halide perovskites, a small number of deep defects may be the main cause of the overall recombination rate, even though there are more shallow defects and possibly mobile ions present. Recombination via defects was generally described with linear dependencies in carrier density. However, looking again at the full equations for recombination via defects, the SRH model immediately shows that recombination via defects can scale linearly or quadratic (or anything in between) with carrier density. The behaviour of a given trap depends mainly on its energetic position in the band gap (shallow vs. deep), its density (does it contribute to the charge neutrality equation?), and its charge state (does it dope the sample in the dark, under illumination, or basically never?).

While the theory behind semiconductor statistics is more than 70 years old, it is still used today to study recombination in certain types of semiconductors. For quite intrinsic semiconductors that have lots of shallow traps, exhibit photodoping and power law decays the re-evaluation of the theory helps to understand how to analyse data that is sensitive to the recombination dynamics of the sample. We investigated photoluminescence-based techniques in steady state (PL quantum yield as a function of light intensity) and in the time domain (transient PL with differential lifetimes). We start with simple situations and work our way up to more complex ones and we discuss how the SRH model, when used with the charge neutrality condition, can lead to different results for the PL quantum yield and the PL decay. We provide analytic approximations (dashed) as well as numeric simulations (solid lines).

The most important finding is that the presence of shallow traps can be deduced from the dependence of the respective data on the carrier density or the Fermi-level splitting present for each recorded data point. Especially in transient photoluminescence measurements it is common to see power-law decays with increasing differential lifetimes at towards low Fermi level splitting coincide with PL quantum yields that are much lower than 1 (recombination coefficient above k_rad). This can easily be explained by recombination via shallow traps.

In an intrinsic semiconductor, the shallower a trap is, the faster the recombination rate will increase with the number of carriers, and the more it will behave like radiative recombination in any transient experiment. Here, we discuss how the depth of the trap influence steady-state and transient photoluminescence

Further informations you can find here:
J. Hüpkes, U. Rau, and T. Kirchartz, “ Impact of Trap Depth on the Steady-State and Transient Photoluminescence in Halide Perovskite Films.” Adv. Energy Mater. (2025): e03157. https://doi.org/10.1002/aenm.202503157