Investigation on Light Extraction Behavior of Surface Plasmon-Coupled Deep-Ultraviolet LED in Different Emission Directions

The light extraction behavior of an AlGaN-based deep-ultraviolet LED covered with Al nanoparticles (NPs) is investigated by three-dimensional finite-difference time-domain simulation. For the transmission spectra of s- and p-polarizations in different emission directions, the position of maximum transmittance can be changed from (θ = 0°, λ = 273 nm) to (θ = 0°, λ = 286 nm) by increasing the diameter of Al NPs from 40 nm to 80 nm. In the direction that is greater than the critical angle, the transmittance of s-polarization is very small due to the strong absorption of Al NPs, while the transmittance spectrum of p-polarization can be observed obviously for the 80 nm Al NPs structure. For a ~284 nm AlGaN-based LED with surface plasmon (SP) coupling, although the luminous efficiency is significantly improved due to the improvement of the radiation recombination rate as compared with the conventional LED, the light extraction efficiency (LEE) is lower than 2.61% of the conventional LED without considering the lateral surface extraction and bottom reflection. The LEE is not greater than ~0.98% (~2.12%) for an SP coupling LED with 40 nm (80 nm) Al NPs. The lower LEE can be attributed to the strong absorption of Al NPs.

Inverse design of organic light-emitting diode structure based on deep neural networks

The optical properties of thin-film light emitting diodes (LEDs) are strongly dependent on their structures due to light interference inside the devices. However, the complexity of the design space grows exponentially with the number of design parameters, making it challenging to optimize the optical properties of multilayer LEDs with rigorous electromagnetic simulations. In this work, we demonstrate an artificial neural network that can predict the light extraction efficiency of an organic LED structure in 30 ms, which is ∼103 times faster than the rigorous simulation in a single-treaded execution with root-mean-squared error of 1.86 × 10−3. The effective inference time per structure is brought down to ∼0.6 μs with unaltered error rate with parallelization. We also show that our neural networks can efficiently solve the inverse problem – finding a device design that exhibits the desired light extraction spectrum – within the similar time scale. We investigate the one-to-many mapping issue of the inverse problem and find that the degeneracy can be lifted by incorporating additional emission spectra at different observing angles. Furthermore, the forward neural network is combined with a conventional genetic algorithm to address additional large-scale optimization problems including maximization of light extraction efficiency and minimization of angle dependent color shift. Our approach establishes a platform for tackling computation-heavy optimization tasks with one-time computational cost.