A trade-off between efficiency and stability in a class of sky-blue organic light-emitting diodes

The stability of organic light-emitting diodes (OLEDs) is a key requirement for their use in commercial displays. One approach to increase the performance of the blue subpixel is to use thermally activated delayed fluorescence (TADF) emitters, which combine a small singlet-triplet gap, ΔE_ST, and non-trivial spin-orbit coupling to enable the harvesting of both singlet and triplet excitons to produce light; however, device stability is compromised due to long triplet lifetimes that increase the probability of the degradation of materials by biexcitonic and polaron events. In this work, we correlate the efficiency and stability of the device with the device structure, identifying possible origins for the trade-off between device efficiency and stability. By comparing a set of sky-blue emissive devices and fitting drift-diffusion simulations to intermittent optoelectronic measurements during stress-testing, the layers most affected by degradation can be determined. By coupling electrical simulations to an excitonic model, the contributions from biexcitonic and polaronic events to efficiency roll-off can be distinguished. We find that efficient but less stable devices suffer mainly from exciton-exciton annihilation, while stable but less efficient devices have excitons predominantly quenched by polarons. This implies that the device structure is responsible for determining non-radiative exciton pathways, and an efficient and stable OLED structure should aim to minimize exciton accumulation at high brightness.