Multi-physics modeling of contour scanning mechanisms affecting surface and subsurface features in laser powder bed fused 304L steel

Abstract

Surface and subsurface defects in laser powder bed fusion affect fatigue performance, limiting its application in safety-critical components. Applying contour scanning potentially improves side surface quality and near-surface features, but the involved multi-scale physical mechanisms for defect formation and suppression remain unclear. In this study, we develop a comprehensive and high-fidelity numerical simulation model that simultaneously captures melt pool dynamics, surface roughness formation, and subsurface defect evolution during contour scanning. Model predictions show good agreement with experimental results of melt pool geometry, surface roughness, and near-surface defects. The results demonstrate that controlling energy density near the keyhole regime improves melt pool stability and track uniformity, thereby reducing surface roughness and suppressing the formation of lack-of-fusion and keyhole-induced pores. A mechanism is identified where gas bubbles undergo expansion, shrinkage, and convection-driven motion, influenced by vapor condensation, Marangoni flow, and buoyancy. These dynamics determine whether bubbles coalesce, migrate, or become entrapped within the melt pool. Optimized contour parameters reduce surface roughness by more than 50 % (Sa from ∼10 to ∼4 μm; Sq from ∼10 to ∼5 μm), and extend fatigue life by over threefold (from ∼7000 to ∼25,000 cycles) relative to suboptimal conditions. Moreover, the remelting effect of multi-layer scanning promotes pore closure at intermediate depths but increases porosity near the top layers. A gradient-decreasing laser energy strategy is proposed to balance these effects. The identified contour mechanisms provide a scientific basis for developing in-situ laser control strategies aimed at enhancing surface quality and fatigue resistance in additively manufactured metallic components.