Effect of Laser-Matter Interaction on Molten Pool Flow and Keyhole Dynamics

Laser-induced keyholing occurs in additive manufacturing and welding processes, but the keyhole dynamics have not been well understood. A multiphase and multiphysics numerical model is used to predict the keyhole shapes recorded in the experimental observations and to predict transient and nonuniform distributions of laser absorption, temperature, and flow velocity in the process. When compared against data from a state-of-the-art dynamic x-ray radiography technique, good agreement is found for the keyhole shapes and fluctuation of the gas-liquid interface, thereby validating the simulation method. A detailed discussion is then given to elucidate the effects of laser absorption on the dynamic behavior of the front and rear keyhole walls. A quantitative comparison of different driving forces on the keyhole is also given to evaluate their significance to the keyhole dynamics.

Figure 1

Schematic of laser-induced keyhole.

Figure 2

Keyhole processing map for SS304 with different laser powers and scanning speeds. The colored boxes highlight the three selected cases of typical keyhole shapes for further investigation. For all three cases, the experiments are performed at the same laser power $P=416\mathrm{W}$ and varying scanning speeds: case 1 at $V=300\mathrm{mm/s}$; case 2 at $V=400\mathrm{mm/s}$; case at 3 $V=800\mathrm{mm/s}$. The inset sketch defines the geometrical parameters used in the course of the study.

Figure 3

Temporal variation of keyhole width in the three selected cases: case 1 at $P=416\mathrm{W}$, $V=300\mathrm{mm/s}$; case 2 at $P=416\mathrm{W}$, $V=400\mathrm{mm/s}$; case 3 at $P=416\mathrm{W}$, $V=800\mathrm{mm/s}$. The solid dotted lines show the instantaneous width, the dashed lines show the average width, and the double-headed arrows show the magnitudes of width fluctuation.

Figure 4

Typical results of 3D simulation of laser-induced keyhole: (a) predicted temperature and velocity on the 2D center plane (velocity in the gas region is hidden for image clarity); (b) distribution of laser absorption on the 3D keyhole wall.

Figure 5

Comparison of keyhole shapes from DXR observation and simulation results for the three selected cases. (a)–(e) DXR images of the keyhole shapes. (f)–(j) Simulated keyhole shapes and distributions of laser absorption. For illustrative purposes, different instances are shown for case 1, which exhibits the most fluctuation out of the three cases. (a)–(c) Case 1 at times $t_0$, $t_0+140\mathrm{ms}$, and $t_0+340\mathrm{ms}$. (f)–(h) Predictions at three selected instances from the simulation.

Figure 6

Distribution of laser absorption intensity: (a)–(c) for all rays, (d)–(f) for fresh rays, and (g)–(h) for reflected rays in the three selected cases.

Figure 7

Temporal variations of (a) total laser absorption and corresponding absorptivity (with respect to the input laser power 416 W), (b) fresh ray absorption, and (c) keyhole wall area with absorption intensity above the critical value $S_{\mathrm{critical}}$, as discusssed in Sec. 3a2. For all three cases, the experiments are performed at the same laser power $P=416\mathrm{W}$ and varying scanning speeds: case 1 at $V=300\mathrm{mm/s}$; case 2 at $V=400\mathrm{mm/s}$; case 3 at $V=800\mathrm{mm/s}$.

Figure 8

Laser absorption on the keyhole wall. (a) The absorption of one fresh laser ray by a linear segment of the front keyhole wall. (b) Laser absorption by keyhole wall. The purple curve shows the local absorptivity as a function of incident angle. The orange curve shows the ratio of the absorbed laser power intensity on the front keyhole wall over the incident laser power intensity from the fresh laser ray. The red (${\overline{\beta }}_1$, ${\beta }_1^U$, ${\beta }_1^D$); blue (${\overline{\beta }}_2$); and green (${\overline{\beta }}_3$, ${\beta }_3^U$, ${\beta }_3^D$) terms mark out the incident angles of the fresh ray on the front keyhole wall for cases 1, 2, and 3, respectively, and will be used in the discussion of Sec. 3b. (c) Nonuniform absorption of fresh rays across a protrusion on the front keyhole wall. Note that the inclination angle and protrusion size are exaggerated in this figure for clarity of illustration.

Figure 9

Three consecutive instants in time showing the generation and downward motion of protrusions on the front keyhole wall in case 1. Protrusion A starts to form on the convex surface in the transition zone between the horizontal top surface and inclined front keyhole wall. Protrusions B and C are moving downward along the front keyhole wall (notice the high velocity of the fluid flow on the front keyhole wall).

Figure 10

Simulation results of the temperature and velocity fields in the three selected cases. (a)–(c) Temperature and velocity fields on the center plane from the simulations. (d)–(f) Horizontal planes looking from the top of the keyhole near the surface of the substrate from the simulations. Note that the velocity field for vapor is not shown. The vortices labeled A, B, and C are discussed in Sec. 3c1.

Figure 11

Sketch of dynamic x-ray radiography experiments. The laser scanning direction is shown by the black arrow.