A low-cost desktop 3D printer can extrude a polymer filament packed with bronze powder—but the final part only becomes metal after debinding and sintering. So which printing settings determine whether that bronze is useful or full of defects?

Typical additive manufacturing processes for producing metal and ceramic parts are highly energy-consuming and expensive to install and maintain. Material extrusion additive manufacturing technologies are conventionally used to produce polymeric parts but only marginally to process metallic materials.

A feasible alternative is to process polymeric filaments loaded with metal particles, followed by debinding and sintering to join the metal particles and obtain the final parts. This study investigates the printability of commercial CuSn12 filament by evaluating process parameters’ influence on density, shrinkage, porosity, and mechanical properties using a low-cost desktop three-dimensional printer.

The common metal processes discussed here are powder bed fusion and direct energy deposition. These technologies are generally high-energy consuming, expensive to install and maintain, and require a cooling system and an inert gas environment to prevent oxidation.

They also cause significant powder waste during manufacturing, and that waste is a handling hazard. Material extrusion additive manufacturing is less commonly used to process metallic materials, although it is conventionally used to produce polymeric parts.

Among additive manufacturing techniques, material extrusion additive manufacturing promises to be the most cost-effective option because of smaller investment costs, faster building rates, and greater production flexibility. Material extrusion additive manufacturing minimizes material waste because all of the powder contained within the filament is used during printing.

Its approaches are filament-based, plunger-based, and screw-based; filament-based techniques are also known as fused filament fabrication. This study carries out a preliminary investigation of Bronze Filamet filament manufactured by The Virtual Foundry.

The work provides experimental insights into CuSn12 filament printability and the effects of printing parameters on the mechanical properties of additively manufactured parts. There is scarce scientific evidence about the mechanical properties of highly filled CuSn12 filaments produced by material extrusion additive manufacturing and sintered in an air environment.

Bronze parts have engineering applications and are often used in liquid environments because of their corrosion resistance, including marine and railway components such as ship propellers, valves, and pumps. The study investigates how significant printing parameters influence porosity in additively manufactured metal parts using a low-cost desktop fused filament fabrication machine.

Very few studies have investigated how process parameters influence the mechanical properties of samples obtained using bronze Filamet. Earlier studies considered layer thickness, nozzle temperature, printing speed, printing and sintering directions, shrinkage, tensile properties, porosity, microstructure, phase analysis, flexural strength, and elastic modulus.

This study develops a systematic and multiscale approach that considers nozzle temperature, flow rate, infill pattern, and ironing. The filament contained spherical metal particles with dimensions varying from ten to fifty micrometers, dispersed within the polymeric binder.

The filament contained eighty-seven to ninety percent metal powder by weight and had a density of four point twenty-nine to four point five grams per cubic centimeter. Energy-dispersive spectroscopy confirmed that the bronze powder was composed of eighty-eight to ninety-one percent copper and nine to twelve percent tin by weight.

Nine sets consisting of five dog-bone specimens were three-dimensionally printed using different combinations of parameters, and each set was assigned a letter from A to I. Set A followed the supplier’s guidelines and served as the base set for comparison.

Its flow rate of two point sixteen cubic millimeters per second produced specimens with excessive voids, so preliminary tests increased the flow rate to two point ninety-two cubic millimeters per second, or one hundred thirty-five percent of the initial value. That value produced better specimen infill with no visible macroscopic inter-raster and inter-layer voids, and all sets from B to I shared it.

The studied parameters were nozzle temperature, the pattern of intermediate layers, and ironing, each varied on two levels. Figure three schematically shows the cross-section of the adopted FFF direct-drive extruder. Filament feeds from the spool through the extruder motor and roller gears, then passes through a PTFE tube, coupler, heat sink, four-millimeter heat break, and heated block before exiting the nozzle.

This layout matters because the material has a high metal-particle content and is described as brittle, making direct feeding to the printhead important for the printing process. Sintering used a Metal three-dimensional printing Starter kiln made by Sapphire three-dimensional.

The kiln had a chamber measuring one hundred fifty-two by one hundred fifty-two by one hundred fifty-nine millimeters and a maximum temperature of one thousand three hundred seventy degrees Celsius. Debinding and sintering were carried out sequentially as two post-processing thermal treatments.

The first heating phase debound the PLA by turning it into a gas that was vented from the kiln. A second cycle used a greater maximum temperature to sinter the metal particles and create the brown part. The Shapiro-Wilk normality test, algorithm AS R94, was used to verify whether the collected data followed a normal distribution.

Where applicable, one-way analysis of variance was combined with Tukey’s Honestly Significant Difference test to evaluate the significance of data differences. Printability issues were assessed among all the sets listed in Table 2.

Sets H and I were removed from the statistical analysis because their green samples presented significant printability issues, including over-extrusion and flashes. Set A showed excessive inter-raster voids, but it remained in the analysis because it was used as the reference set.

The figure shows distinct physical signatures across the three specimens: set A has visible longitudinal gaps between neighboring rasters, while sets H and I have material spreading beyond the intended outline. In the lower specimen, the excess is especially irregular along the upper edge and at the widened left end, compared with the more uniformly filled profile in the middle specimen.

The Pareto chart reports the absolute standardized effects of the parameters on ultimate tensile strength in descending order. Bars greater than the red reference line indicate a significant effect on ultimate tensile strength with a confidence level of ninety-five percent.

The significant effects are the pattern-ironing interaction and the main effects of ironing and pattern. The chart’s upper rows show the interaction bar extending past five standardized-effect units, while the next two bars end around three and a half.

By contrast, the temperature-related bars remain close to zero and well left of the red threshold. For green-sample density, set A showed a statistically significant difference compared with the other sets, with an F-value of forty-six point fifty-four and a p-value below zero point zero zero one.

The density values of sets B to E were consistently higher than those of set A, the reference set. Considering both the density of the sintered parts and their shrinkage, sets B and C could represent the best trade-off. In general, the results were not characterized by a large spread, particularly for density.

The boxplot reports the porosity of the sintered samples. Set A had the lowest porosity, approximately sixteen percent, meaning that sixteen percent of the analyzed-section area was composed of voids. The highest porosities were associated with sets B and E, whose values were about double that of set A; sets C and D had intermediate porosities.

The low porosity of set A must be considered together with the considerable shrinkage described for that set. Figure eleven compares sets A and B before sintering, after sintering, and after MATLAB processing of resin-embedded sections. The red regions in the final panels identify porosity, making the internal pore distribution visible rather than relying only on the specimen surface.

The authors note that set A shows more porosity in the green part, but less after sintering than set B, linking the image to changes during debinding and sintering. The vertical positions of the boxes make the separation especially clear: A is isolated near the bottom, while B and E sit near the top of the scale.

C and D occupy the middle, with D’s box slightly below C’s; the whiskers and internal markers show relatively compact distributions within each set. Ultimate tensile strengths for all parameter sets were significantly higher than the PLA tensile strength of about fifty megapascals.

The tested metal specimens had, in the worst case, an average tensile strength of about eighty megapascals. Sets C and E had higher average failure strengths than set A. The table reports mean and standard deviation values for elastic modulus, tensile strength, and elongation at break for all tested sets.

Sets C and E behaved similarly and produced the best results in terms of strength and strain, while set A had an intermediate level of strength and strain. Sets B and D had the lowest values for elastic modulus, ultimate tensile strength, and elongation at break.

Statistical analysis of elongation at break found the most significant difference between sets B and C, with an F-value of three point thirty-two and a p-value below zero point zero five. Material extrusion additive manufacturing represents a valid alternative for low-cost metal production, while retaining the advantages of other additive manufacturing technologies and providing flexibility with different metals.

The study tested nozzle temperature, the pattern of intermediate layers, and ironing, comparing them with a reference set that used the filament manufacturer’s suggested printing parameters. The best results came from tuning the printing parameters rather than simply following the supplier’s baseline.

In particular, flow rate, pattern, and ironing strongly shaped density, porosity, shrinkage, and mechanical performance.