Titration Experiment Teaching of Analytical Chemistry Based on AI and Deep Learning
YAN Xiao-Yi1,2**, YIN Ze-Long1, YANG Ran-Zi-Jun1, KANG Yu-Xiong1, LI Jun-Chen1, WANG Ting-Zhe1, MA Qiang1,2**, SONG Zhi-Guang1,2
1. College of Chemistry, Jilin University, Changchun 130012, China; 2. National Experimental Teaching Demonstrating Center of Chemistry, Jilin University, Changchun 130012, China
Abstract To address the issues of traditional titration experiments,such as time-consuming procedures,subjective endpoint determination,and a teaching focus limited to endpoint observation rather than in-depth investigation of the entire reaction process,this experiment deeply integrated computer vision technology with classical titration analysis,proposing a deep learning-based method for analyzing titration reaction kinetics.By simulating manual titration operations with a magnetic stirrer and combining smartphone video capture with YOLO-series convolutional neural network models,precise identification and measurement of color change-decolorization phenomena and pH variations during the titration process was achieved.Based on a pre-constructed database,this study utilized machine learning methods to determine the time difference between indicator coloration and decoloration in the titration system.A three-dimensional relationship among titrant volume,colordecolorationtime,and pH value was established,and the titration reaction kinetic was further investigated.By leveraging artificial intelligence and other advanced technologies,this study overcomes the limitations of conventional experimental operations.The interdisciplinary methodology expands the pedagogical dimensions and provides a novel paradigm for the digital design of analytical chemistry experiments.
YAN Xiao-Yi, YIN Ze-Long, YANG Ran-Zi-Jun, KANG Yu-Xiong, LI Jun-Chen, WANG Ting-Zhe, MA Qiang, SONG Zhi-Guang. Titration Experiment Teaching of Analytical Chemistry Based on AI and Deep Learning[J]. Chinese Journal of Chemical Education, 2026, 47(8): 57-64.
