Argonne National Laboratory

Intense X-rays expose tiny flaws in 3-D printed titanium that can lead to breakage over time

By Katie Elyce JonesMarch 4, 2016

Titanium is strong but light — a desirable property among metals. In the twentieth century, titanium was used in military aircraft and equipment and commercial jets. Today, we find this tough and flexible metal all around us — in sports gear, tools, surgical and dental implants, prosthetics, eyeglasses and jewelry.

One of the manufacturing revolutions making this explosion of titanium parts possible is additive manufacturing, or 3-D printing. Printing titanium alloys, or other metal alloys for that matter, reduces waste and cost and enables a much wider range of designs. However, the powder-based printing methods used for titanium alloys also increase porosity — the quantity and size of pores—in the final product. Porosity can decrease the material's resistance to fatigue, or cyclic strain, leading to breakage.

To understand the cause of porosity in 3-D printed titanium alloys, researchers from Carnegie Mellon University (CMU) came to the U.S. Department of Energy's (DOE's) Argonne National Laboratory for the intense synchrotron X-rays and a rapid imaging tool known as microtomography at the Advanced Photon Source (APS), a DOE Office of Science User Facility located at Argonne.

"Like any other metal, titanium has a certain amount of fatigue resistance until it cracks or breaks," said Anthony Rollett, professor of materials science and engineering at CMU and co-author of the study published in The Journal of The Minerals, Metals & Materials Society. "The more porosity in the printed metal, the more its resistance to fatigue is decreased."

By inspecting the most common titanium alloy, Ti-6Al-4V, at the micron-scale — for comparison, that is the size of most biological cells or smaller — researchers quantified the number, volume and distribution of pores in samples of Ti-6Al-4V printed using a range of parameters. Additively manufactured Ti-6Al-4V includes six percent aluminum and four percent vanadium and is popular in the aerospace and biomedical industries where speed of manufacture and unique designs are important.

For 3-D printing, metals are usually atomized into powders first. Ti-6Al-4V powders are printed by using either selective laser melting or electron-beam melting (EBM), which is the focus of this study and uses the high power and penetration of electrons to melt the powders layer by layer, heating and compressing them into the desired structure.

As the powder heats, gases trapped in the material can create pores like bubbles that are pinpoints of structural weakness. These pores can be anywhere from a few microns to a few hundred microns in size and are not distributed uniformly throughout the material, so researchers needed a close look at a big sample to effectively characterize porosity.

That is why they needed the APS. "We can observe hundreds or even thousands of pores at a high resolution of about two microns," Rollett said. This study is one of the first to use such high resolutions for studying 3-D printed Ti-6Al-4V.

The CMU team conducted their work at the APS X-ray Science Division (XSD) 2-BM beamline, an X-ray microtomography imaging station that can reconstruct a view of a sample in three spatial dimensions and one temporal dimension to capture slow dynamics. The CMU team, however, focused on static views of Ti-6Al-4V.

The objective was to look at several Ti-6Al-4V samples printed at different specifications, including changes in electron beam power level, speed and spacing. Researchers theorized there would be a "sweet spot" at which they could set printing parameters to significantly reduce or eliminate porosity by controlling the size of the melt pool—the area melted by the beam.

"Relative to printing speed and spacing, if you decrease the power level and the melt pool becomes too small, you may leave behind unmelted powder, which is a source of porosity," Rollett said. "However, if you increase the power level too much, you risk creating deep holes, called keyholes, with the electron beam that also leave behind voids."

With a commercial EBM machine, the team printed five cubes of Ti-6Al-4V with melt pools ranging from four times to one-fourth the area of the relative melt pool. The larger the melt pool, the slower the speed function. Then they extracted 1-by-15-millimeter samples for imaging. They also imaged a sample of preprinted powder.

"The APS microtomography system is capable of scanning objects in minutes, compared to laboratory systems that can take several hours," said Xianghui Xiao, physicist in the Imaging Group of Argonne's XSD.

Microtomography reconstructs 3-D views by combining 2-D images taken at many angles, similar to how a computed tomography (CT) scan works in a hospital, Xiao said. The bright, high-energy X-rays generated at the APS highly interact with a material, providing information on both the density and chemistry of a sample, and the open-source APS tomography software, TomoPy, rapidly analyzes and reconstructs the 3-D structure.

For each Ti-6Al-4V sample, 1,500 images in 2-D were scanned in only two minutes. Had the team used a laboratory technique like electron microscopy, Rollett said, they would have had to study a smaller volume at a resolution roughly 10 times lower with much longer counting times.

When they quantified pore shape, volume and distribution for all of the 3-D printed samples, what they discovered was a little bitter: there was no sweet spot for printing flawless Ti-6Al-4V.

"Porosity was present in every piece," Rollett said. "To us it was a surprise that it was always there."

As expected, printing parameters did significantly impact porosity — they just did not eliminate it. Printing larger melt pool areas at lower speed resulted in fewer, smaller pores overall, but those samples also showed clustering of pores at the surface, which is still detrimental to fatigue resistance.

With these results, the researchers now think they should turn their attention to the titanium alloy powder rather than printing parameters. Rollett said their results suggest that porosity initiates in powder processing. The images of the pre-printed powder sample revealed smaller pores that may have expanded during printing.

"Our next step might be to ask if there is a better way to process the powder before printing," Rollett said.

Funding for this research was provided by America Makes and the National Science Foundation.

Argonne National Laboratory seeks solutions to pressing national problems in science and technology. The nation’s first national laboratory, Argonne conducts leading-edge basic and applied scientific research in virtually every scientific discipline. Argonne researchers work closely with researchers from hundreds of companies, universities, and federal, state and municipal agencies to help them solve their specific problems, advance America’s scientific leadership and prepare the nation for a better future. With employees from more than 60 nations, Argonne is managed by UChicago Argonne, LLC for the U.S. Department of Energy’s Office of Science.

The U.S. Department of Energy’s Office of Science is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time. For more information, visit the Office of Science website.