A visual CFD-assisted approach for precise measurement of trace alkali metals in sealed cells

Accurately quantifying microgram-level alkali metal amounts inside cells remains a significant challenge. To address this, this study proposes an innovative method integrating computational fluid dynamics (CFD) simulation and visual image processing. First, CFD simulation analyzes the formation and dynamic characteristics of alkali droplets under thermal gradients, laying a foundation for theoretical modeling and image acquisition optimization. Second, a spherical cap geometric model is constructed based on droplet morphology; the contact angle is determined by combining the Marangoni effect and contour features, establishing a quantitative relationship between model parameters and alkali metal mass. Finally, in image processing, techniques including illumination correction, edge enhancement, and adaptive threshold segmentation are adopted to resolve issues such as uneven lighting and blurred edges, enabling accurate extraction of parameters (e.g., droplet radius, contact angle). Experimental results show this method is highly effective for measuring minute alkali metal amounts in cells, with relative standard deviation (RSD) less than 10% for microgram-level measurements, which has been successfully cross-validated against traditional calorimetry. It provides a practical solution for detection of minute alkali metals in cells and holds promising application potential in optoelectronic device manufacturing and quantum sensing.