Prediction method for weak interaction energy of intermolecular non-covalent bond in spiro-containing fluorenyl molecular crystal and training method of prediction model for weak interaction energy of intermolecular non-covalent bond in spiro-containing fluorenyl molecular crystal

Abstract

The invention discloses a training method of a prediction model for weak interaction energy of intermolecular non-covalent bonds in a spiro-containing fluorenyl molecular crystal. According to the method, the dipole moment of molecules, the positive electric energy of the molecules, the intermolecular distance, the relative orientation of the molecules, the number of electrons on the outermost layer of the molecules, the polarizability of the molecules and the number of spiro SP3 carbon contained in the molecules are taken as descriptors; and after a result is compared with a theoretical calculation result, the reliability of a prediction result is proved. Meanwhile, the efficiency of calculating the weak interaction energy of the intermolecular non-covalent bonds in the spiro-containing fluorenyl molecular crystal can be effectively improved, the calculation cost is greatly reduced, and a new way is opened up for researching the weak interaction energy of the intermolecular non-covalent bonds in the spiro-containing fluorenyl molecular crystal.

Description

technical field [0001] The invention relates to the prediction of non-covalent bond weak interaction energy between molecules, and belongs to the fields of computational chemistry, organic chemistry and photoelectric materials. [0002] technical background [0003] The first part: general situation and application value of fluorenyl molecular crystals containing spiro ring. [0004] Fluorene-based molecules containing spiro rings belong to spiro-aromatic compounds and have a unique cross-shaped conformation. The fluorenes and other independent groups arranged orthogonally have their own electronic properties, and their molecular orbitals are separated, and they also have bipolar It can effectively improve the stability of the device, reduce electron coupling and charge recombination, and has become a general-purpose molecular framework in the field of optoelectronic materials. So far, spiroarenes represented by spirobifluorene (SBF) derivatives have been widely used in orga...

AI Technical Summary

Problems solved by technology

[0015] Aiming at the existing difficulties and problems in the prediction of the non-covalent bond weak interaction energy between molecules of fluorenyl molecular crystals containing spiro rings in the prior art, the present invention proposes a method based on deep learning to predict the intermolecular non-covalent bonds in this system. The prediction method of bond weak interaction energy and its prediction model training method, under the premise of ensuring the reliability of the prediction results, improve the intermolecular non-covalent properties of spirofluorene molecular crystal materials with cross-shaped three-dimensional spatial structures in the field of optoelectronic materials. The calculation efficiency of bond-weak interaction energy. This prediction model can solve the problems of existing prediction models, including huge consumption of computing resources and long calculation cycle. By simplifying the calculation process, the prediction time can be shortened and the research and development process of such materials can be accelerated. , to reduce computing costs, so as to realize energy-saving, high-efficiency, and low-cost optoelectronic new material research and development model

Images