Machine learning-driven multi-level composite SERS platform for trace detection of chlorogenic acid as pharmacodynamic substance in honeysuckle

We reported a multi-level composite SERS platform which could provide a new method for accurate and quantitative trace detection for pharmacodynamic substance in traditional Chinese medicine (TCM). Using natural dragonfly wings as the template, Ag NPs were deposited on V-Ti nano-columns through controllable magnetron sputtering technology. Ag30@V-Ti20@D.W possessed a uniform nanostructure and exhibited High-performance SERS signal sensing capability. The detection limit (LOD) of R6G was 1 × 10−10 M. Through machine learning, we accurately classified and recognized R6G at different concentrations. The multi-level composite structure provided a wide range of “hot spots” areas and ensured good reproducibility with a relative standard deviation (RSD) value of 7.13%. When quantitatively detected chlorogenic acid, an active ingredient in honeysuckle, LOD value of 1 × 10−6 g/l was obtained. In the future, the combination of SERS spectroscopy technology and machine learning will provide theoretical and experimental support for the development of new strategies for detecting Other pharmacodynamic substances in TCM.