Remote monitoring of powder dehumidification by speckle pattern analysis

Non-invasive monitoring of heterogeneous powders is a required task in several food science and industrial applications. We introduce laser speckle analysis (LSA) as a remote, non-contact, non-invasive, cost-effective, and speedy approach for tracking powder dehumidification. LSA is based on the statistical analysis of dynamic speckle patterns that are generated from scattering samples upon laser light illumination. In this work, LSA is applied to monitor dehumidification dynamics in cornstarch powder, chosen as a representative hygroscopic food system. Cornstarch is selected due to its industrial relevance as one of the most widely used hygroscopic food powders, where moisture content strongly governs flowability, caking, and shelf life. In addition, its highly diffusive structure provides a suitable testbed for LSA experiments. As drying progresses, the speckle-derived metrics exhibit specific trends that indicate a gradual reduction of internal dynamics. The optical observations are validated by gravimetric measurements and 3D profilometry. The results present a two-stage drying profile, characterized by an initial phase of rapid mass loss followed by a slower approach toward equilibrium. Collectively, the LSA results, in agreement with the validating experiments, reveal the formation and stabilization of surface micro-fractures over time. The technique effectively captures both dynamic and structural transitions, offering a practical solution for process monitoring and quality control in industrial settings, and has the potential to serve as a bench-top analysis device.