Linking Consistency in Experimental Results to Surface Topography: A Novel Hypothesis on Laser-Ignited Combustion in Random Particle Packings

Laser-ignited particle combustion is critical to energy, aerospace, and defense applications, yet progress in deciphering its physicochemical mechanisms is hindered by poor reproducibility of combustion data from randomly packed samples. Classical theories attribute this inconsistency to variations in bulk packing density. Here, we propose a novel hypothesis that the consistency of the uppermost surface layer topography is the dominant determinant of combustion reproducibility. A two-/three- dimensional discrete element method (DEM) reveals that increasing packing layers markedly reduces surface topography conformity, while gravitational settling maintains bulk packing density near its theoretical maximum. We introduce a constrained droplet method combining multistage sieving, equal-circle packing in a circle theory, and precision droplet deposition. This approach simultaneously yields particles with a narrow size distribution and enables programmable tuning of packing layer counts, producing multilayer particulate packings that accurately reproduce the structural specifications of DEM model. In situ laser ignition diagnostics show that key combustion metrics—including AlO emission spectral profiles, temporal evolution of spectral full width at half maximum (FWHM), ignition delay, and combustion duration—exhibit a rapid decay in consistency with increasing layer count, closely matching the simulated decline in surface topography conformity. These results are attributable to the nanoscale optical penetration depth of the laser, which anchors the nascent reaction front geometry to the pre-combustion surface topography. In contrast, the near- theoretical- maximum bulk packing exerts a less important influence on this interfacial geometry during subsequent propagation. This work also provides a standardized sample preparation protocol, and enables accurate quantification of key combustion parameters that are inaccessible via conventional methods.