Over the past decade, organic light-emitting materials have been recognized by both academia and industry as promising components for lightweight, flexible and versatile optoelectronic devices such as OLED displays. However, finding suitably efficient materials is difficult.
To address this challenge, the collaborative research team developed a novel approach that combines machine learning models and quantum-classical molecular design to accelerate the discovery of efficient OLED emitters. The research was published in his May 17 issue of Intelligent Computing, the Science Partner Journal.
The optimal OLED emitters discovered by the authors using this ‘hybrid quantum-classical procedure’ are deuterated derivatives of Alq3, which are highly efficient and synthesizable.
A deuterated OLED emitter is an organic material in which the hydrogen atoms in the emitter molecule have been replaced with deuterium atoms. Although they have the potential to emit light very efficiently, designing such deuterated OLED emitters presents a computational challenge. This challenge arises from the need to optimize the positions of the deuterium atoms within the emitter molecule, and the calculations have to be performed from scratch.
New workflows involving both classical and quantum computers speed up these computations. First, quantum chemical calculations are performed on a classical computer to obtain the ‘quantum efficiency’ of a set of deuterated Alq3 molecules. These data on the luminous efficiencies of various molecules are used to create training and test datasets for building machine learning models that predict the quantum efficiencies of various deuterated Alq molecules.
A machine learning model is then used to construct the system’s energy function, known as the Hamiltonian. Quantum optimization is then performed on a quantum computer using two quantum variational optimization algorithms, variational quantum eigensolver (VQE) and quantum approximate optimization algorithm (QAQA), to achieve optimal quantum efficiency. Assist machine learning to discover molecules with Synthesis constraints are introduced during the quantum optimization process to ensure that the optimized molecule is synthesizable.
To improve the accuracy of predictions on quantum devices, the authors employ a noise-tolerant technique called recursive random variable elimination (RPVE), which states that “quantum devices can be used to find optimal weights with very high accuracy. We succeeded in finding hydrogenated molecules. Furthermore, they point out that combining this new noise-immune technique with two of his selected quantum optimization algorithms could potentially achieve quantum advantages in computing for quantum devices in the near future.
In general, the authors hope that a combined approach of quantum chemistry, machine learning and quantum optimization could create “new opportunities to generate and optimize key molecules for materials informatics.” .
/ Open to the public. This material from the original organization/author may be of the nature of its time and has been edited for clarity, style and length. Mirage.News does not take any organizational positions or positions and all views, positions and conclusions expressed herein are those of the authors only. Read the full article here.
