The Influence of Quantum Dot Size Effect on Optical Properties is Analyzed by Deep Learning Algorithm

As an important member of nano-materials, quantum dots' unique quantum size effect significantly regulates electronic structure and optical properties, such as broadening absorption spectrum and tunability of emission wavelength. However, traditional research methods mostly rely on trial and error, lacking efficient and accurate theoretical prediction means. In this study, quantum dots with different sizes were prepared by wet chemical synthesis, and their absorption and emission spectra were collected. Subsequently, a hybrid deep learning model based on Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM) was constructed to predict the optical properties of quantum dots using size parameters and spectral data as inputs. The experimental results show that with the increase of quantum dot size, both absorption peak and emission peak shift red, which verifies the existence of quantum size effect. The deep learning model performed well on the test set, with an accuracy rate of 92.3%, the recall rates of absorption spectrum and emission spectrum were 89.7% and 91.2%, and the F1 score was 90.9% and 91.4%, respectively. This study not only reveals the complex relationship between quantum dot size and optical properties, but also provides scientific basis for rational design and performance optimization of quantum dot materials, and shows the application potential of deep learning in the field of nano-materials.