Island Interface Porosity in PBF-LB 316L: Effects of Island Overlap and Interlayer Shift in Chessboard Scanning

In the world of modern manufacturing, there is a technique that builds solid metal objects layer by layer, directly from a digital design, without the need for molds or heavy cutting tools. This process, known as powder bed fusion, works by spreading a thin layer of fine metal powder across a platform and then using a powerful laser to melt specific spots, fusing the powder into a solid shape. As the laser moves, it traces out the geometry of the part, and once a layer is complete, the platform drops slightly, a new layer of powder is spread, and the process repeats until the object is finished. While this method allows engineers to create incredibly complex shapes that would be impossible to make any other way, the resulting parts are not always perfect. The rapid heating and cooling inherent in the process can leave behind tiny, invisible holes called pores. These voids are a serious problem because they weaken the metal, making it more likely to crack or fail under stress, which is a critical concern for industries like aerospace and medicine where safety is paramount.

To manage the intense heat generated during printing, manufacturers often use a strategy called chessboard scanning. Instead of letting the laser draw one long, continuous line across the entire part, the software breaks the surface into a grid of small squares, or islands. The laser fills one square, then jumps to another, alternating directions to prevent heat from building up in one spot. However, this approach creates new boundaries where these squares meet. If the laser does not melt enough of the powder at these edges, the squares fail to fuse properly, leaving a line of weak spots and holes right where the islands touch. For years, engineers have tried to fix this by shifting the grid pattern slightly between each new layer, hoping that the new layer would melt over the gaps left by the previous one. But a recent study suggests that this shifting alone is not the whole answer, and that a different adjustment—how much the laser traces overlap the edges of the squares within the same layer—plays a far more critical role.

Researchers at the Federal University of Technology in Paraná and Alkimat Tecnologia set out to investigate exactly how these two factors interact when printing stainless steel. They focused on a specific type of steel, AISI 316L, which is widely used for its strength and resistance to rust. Using a machine equipped with a 200-watt laser, they printed small cubes of metal, varying two specific settings to see how they affected the number of holes inside the material. The first setting was the "island shift," where they moved the entire grid pattern by one millimeter between layers. The second setting was the "island overlap," which controlled how much the laser path extended into the neighboring square while still printing the current one. In simpler terms, they tested whether just moving the pattern up and down was enough to seal the gaps, or if they needed to deliberately make the laser trace over the edges of the squares more aggressively to ensure a solid bond.

The team printed dozens of samples, carefully measuring the internal structure of each one by slicing them open and examining the cross-sections under a microscope. They found that simply shifting the grid between layers, a common practice in the industry, had very little effect on fixing the holes if the overlap between the squares was too small. When they used a standard, minimal overlap, the shifting did not stop the lack-of-fusion defects from forming at the island boundaries. The holes remained concentrated in a line right where the squares met, regardless of how the layers were offset. This finding challenged the assumption that shifting the pattern alone was a sufficient cure for these specific defects.

However, when the researchers increased the amount of overlap, the results changed dramatically. By adjusting the laser to trace further into the neighboring squares, they forced the metal to melt and fuse more thoroughly at the boundaries. As they increased this overlap, the concentration of holes at the island interfaces disappeared. The most significant improvement occurred when they set the overlap to 135 percent, a setting where the laser path extended well beyond the edge of its designated square. At this level, the holes were no longer clustered in a weak line; instead, they were spread out evenly throughout the material, and the total number of voids dropped significantly. When they combined this high overlap with the one-millimeter layer shift, the results were even better, reducing the average porosity to just 0.10 percent.

The study also looked at the time it took to print these parts. Increasing the overlap meant the laser had to travel slightly further, which added a tiny amount of time to the process, but the increase was negligible. The researchers concluded that the most effective way to produce strong, dense metal parts using this chessboard strategy is not to rely on shifting the layers, but to ensure that the laser overlaps the edges of the islands enough to guarantee a complete melt. While the shifting helps to distribute any remaining defects, it cannot fix the problem if the initial fusion between the squares is incomplete. This insight offers a clear, practical path for engineers to improve the quality of metal 3D printed parts, ensuring they are free from the hidden weaknesses that can lead to failure.