Process-Structure-Property Relationships and Multi-Response Optimization of Z-Direction MEX-Printed Polyphenylene Sulfide

Imagine trying to build a tower out of wet clay, stacking one thin layer after another. If you press the layers together too hard, they might squish and lose their shape; if you press them too lightly, or if the clay dries out before the next layer arrives, the tower will crumble under its own weight. This is the fundamental challenge of a manufacturing technique called material extrusion, where a machine melts a plastic filament and lays it down in precise paths to create a solid object. While this method is excellent for making complex shapes, the resulting parts often have a hidden weakness: the boundaries where one layer meets the next. These boundaries are the most likely places for the object to break, especially if the load is applied from the top down, pulling the layers apart. For everyday plastics, this is a minor inconvenience, but for high-performance materials used in aerospace or automotive industries, a weak layer can mean the difference between a safe component and a catastrophic failure.

One such high-performance material is polyphenylene sulfide, a tough, heat-resistant plastic that can withstand harsh chemicals and extreme temperatures. It is a promising alternative to metal in demanding applications, but printing it successfully is difficult. The material must be heated to very high temperatures to flow, yet if it cools too quickly or if the layers are too thick, the bond between them remains weak. Researchers have long suspected that adjusting the printing settings—how hot the nozzle is, how fast the machine moves, and how thick each layer is—could fix these weak spots. However, a critical question remained unanswered: does making the printed object look denser and more solid on the inside automatically mean it will be stronger?

A team of researchers set out to answer this by printing hundreds of small test pieces using polyphenylene sulfide. They systematically varied three key settings: the temperature of the nozzle, the thickness of the layers, and the speed at which the machine deposited the material. They focused specifically on the "Z-direction," which is the vertical axis where the layers are stacked, because this is the direction where the material is most likely to fail. The team measured how much force the pieces could withstand before breaking, how much they could stretch, and how many tiny air pockets, or voids, were trapped inside them. They also used powerful microscopes to look at the broken surfaces of the pieces to see exactly how the layers had bonded together.

The results revealed a surprising disconnect. The researchers found that the settings that produced the strongest, most durable pieces were not the same as the settings that produced the pieces with the fewest air pockets. The strongest pieces were created when the nozzle was heated to 345 degrees Celsius, the layers were very thin at 0.1 millimeters, and the machine moved slowly at 20 millimeters per second. Under these conditions, the plastic had enough time and heat to flow together, fusing the layers into a cohesive unit. In contrast, the pieces with the lowest amount of internal air pockets were made at a slightly lower temperature of 330 degrees Celsius, with thicker layers of 0.2 millimeters, and a faster speed of 40 millimeters per second.

This finding challenges a common assumption in manufacturing: that a denser object is always a stronger object. The study showed that while a lower number of air pockets is generally good, the shape and location of the remaining defects matter more than the total count. A piece might have very few holes, but if those holes are positioned in a way that creates a weak seam between layers, the part will still break easily. Conversely, a piece with slightly more air pockets might be incredibly strong if the layers are fused so well that the material can transfer stress effectively across the entire structure. The researchers confirmed this by looking at the microscopic surfaces of the broken pieces. The strongest samples showed layers that had melted into one another, with very few gaps between them, while the weaker samples showed clear separations and misaligned layers, even if the total number of holes was similar.

To find the perfect balance, the team used a statistical method to predict the ideal combination of settings that would maximize strength while minimizing air pockets. The model suggested a specific set of conditions: a nozzle temperature of 341.67 degrees Celsius, a layer thickness of 0.1 millimeters, and a printing speed of 20 millimeters per second. Under these optimized conditions, the predicted strength was significantly higher than the average, and the internal structure was dense enough to be reliable. The study also confirmed that the speed of the printer did not significantly affect the amount of air trapped inside the material, a finding that contradicts some previous assumptions about how speed influences density. Instead, the temperature and the thickness of the layers were the dominant factors in determining how many voids formed.

Ultimately, this research provides a clearer roadmap for using high-performance plastics in 3D printing. It demonstrates that simply trying to make a part as dense as possible is not enough; engineers must focus on the quality of the bond between layers. By understanding that the best mechanical performance comes from specific thermal and geometric conditions rather than just minimizing voids, manufacturers can produce safer, more reliable components. The work highlights that the strength of a 3D printed object is not just about what is inside it, but about how the material flows and fuses as it is being built, turning a stack of layers into a single, solid piece.