Grain-based morphological simulation via fractal theory with experimental verification and corresponding optical properties in laser melting deposition additive manufacturing: A demystified approach

Abstract

Laser melting deposition (LMD) stands for a challenging and promising 3D printing technology to fabricate near-net-shape parts. In this study, a 3D modified fractal model (3DMFM) has been developed by considering the surface texture orientation in the technical manufactured surfaces. Fractal parameters, including fractal dimensions and periodic lengths in the 3DMFM, have been solved for the grain-based morphologies developed by LMD. In combination with statistical parameters, these fractal parameters were applied in 3DMFM for analytical simulations. Single-track depositions of AISI 304 stainless steel powder on AISI 304 stainless steel substrate were subjected to scanning electron microscopy investigations for experimental validation of the proposed model. A customized investigational setup, having a He-Ne laser system (632.8 nm and 7.75 mW), was designed to identify the optical properties of each deposited layer. Three types of grains were identified: round quasi-continuous (RQC), elongated lath-shaped (ELS), and a combination of the two. The RQC grains were transformed into ELS grains with an increment in laser scanning speed. The RQC and ELS combination was converted into RQC grains with an increase in the powder feeding rate. RQC changed into ELS grains whenever the laser power was elevated. An increase in skewness and a decrease in kurtosis were observed whenever laser scanning speed or powder feeding rate was increased. A laser power amplification also determined a rise of skewness and a reduction in the kurtosis values. A rise in either laser scanning speed or powder feeding rate lowered the surface roughness, while an inverse correlation was observed between laser power and surface roughness. An increase in laser scanning speed or laser power increased the fractal dimensions (Dx and Dy). However, the powder feeding rate was inversely correlated with Dx and Dy. In the case of periodic lengths (Lx and Ly), the scanning speed, powder feeding rate, and laser power exhibited an inverse behavior with Dx and Dy. Larger magnitudes of Dx or Dy were resulted due to an increased optical reflection or a decreased optical absorption, while the opposite results were attained in the case of periodic lengths inclination (Lx and Ly). The developed fractal model can predict and select the best LMD process parameters in terms of quality control and time/cost-effectiveness.