Microanalyzing 3-D Printed Metals

The scientific and popular presses alike regularly carry stories about the seemingly limitless range of products that three-dimensional printing can make, from apparel to car parts to food. This “anything is possible” aura has enabled 3-D printing to grow by leaps and bounds in just the past few years.

Still, the growth hasn’t been uniform when it comes to the types of materials used with the layer-by-layer fabrication method. Although manufacturers in the aerospace, the biomedical, and other industries have demonstrated that 3-D printing, or “additive manufacturing,” can make advanced metal parts, the overwhelming majority of 3-D printed products remain polymer based.

“The reason we don’t have more high-end metal parts today is because of technical challenges in additive manufacturing,” according to John A. Slotwinski, an additive manufacturing engineer at Johns Hopkins University’s Applied Physics Laboratory. To address those challenges, researchers have begun analyzing in detail the physical and chemical properties of metal parts made via additive manufacturing. Those studies are showing that in some cases, metal products exhibit poor mechanical performance as a result of microscopic defects and other small-scale features. A number of scientists who focus on those kinds of analyses gathered in Portland, Ore., last month to present their latest findings at a microscopy and microanalysis conference organized by the Microscopy Society of America.

The procedure for printing metal products starts the same way as with plastic parts, explained one of the meeting attendees, Harvey A. West II, a research associate professor at the Center for Additive Manufacturing & Logistics at North Carolina State University (NCSU), Raleigh. Both procedures rely on a computer-aided design (CAD) program to mathematically depict a 3-D object as a stack of 2-D cross sections. The cross sections serve as blueprints for each layer, instructing an instrument to build the object, layer-by-layer, from the bottom up.

In one standard method for making metal parts, an automated arm coats a work surface with a thin layer of powdered metal. Then a sharply focused, intense electron or laser beam traces the 2-D pattern in the powder, melting and fusing the powder in its path. The instrument lowers the work surface slightly, reapplies powder, and traces the next 2-D pattern, adding another metal layer on top of the one below.

Additive manufacturing aficionados such as West see opportunities for using this technology to engineer high-value metal parts on demand. For example, if an army transported this type of 3-D printer to the battlefield or a navy stationed them on aircraft carriers, the instruments could make critical replacement parts for military machines as needed. That flexibility would avoid the logistical hassle and high cost of manufacturing, transporting, and warehousing large inventories of spare parts.

But as William Harris, another meeting attendee, noted, using additive manufacturing in that way requires a high level of confidence that the technology can reliably make parts with the required properties. “You can program a 3-D printer to make almost anything,” said Harris, who is an X-ray microscopy specialist with Pleasanton, Calif.-based analytical instrument maker Carl Zeiss. “But how well will the products perform the intended function? How long will they last? And if they fail, how will they fail?”

To answer those kinds of questions, West, NCSU’s Timothy J. Horn, and colleagues used a range of analytical methods, including mechanical strength tests as well as optical and electron microscopy, to evaluate 3-D printed parts made of Ti6Al4V. The titanium alloy contains 6 wt % aluminum and 4 wt % vanadium, and is widely used in aerospace, marine, and biomedical applications.

The team found that some specimens withstood hundreds of thousands of cycles of fatigue testing before breaking or cracking. Others failed quickly. Analysis of the weak specimens revealed various types of microscopic defects. For example, some micrometer-sized cracks contained Ti6Al4V particles that did not melt and fuse during printing. Other cracks were lined with contaminant particles, including metals from previous printing experiments.

Porosity is another type of defect that can cripple printed metal parts. Sometimes, West explains, the fabrication process forms microscopic gas pockets or void spaces inside the product. These regions have Swiss-cheese-like structures and are less dense and weaker than their surroundings.

One source of the unwanted pores is gas trapped inside the metal powders that feed the 3-D printers. The gas may come from a process for manufacturing fine powders that involves injecting gas into a stream of molten metal to atomize the metal. When the powder is zapped in the 3-D printer, the particles heat up, melt, and then cool so quickly—the temperature change may exceed 10,000 °C per second, according to Slotwinski—that trapped gases don’t have time to escape.

A lack of uniformity in the size and shape of the metal particles also contributes to porosity. The mismatched geometry prevents particles from packing densely and leaves microscopic voids and air pockets in the rapidly fusing metal. In a recent study, Slotwinski, together with Edward J. Garboczi and coworkers at the National Institute of Standards & Technology, characterized such problems in stainless steel and cobalt chrome powders (J. Res. Natl. Inst. Stand. Technol. 2014, DOI: 10.6028/jres.119.018).

Meanwhile, in another study, Nesma T. Aboulkhair and coworkers at the University of Nottingham, in England, tried to find ways to minimize porosity in 3-D printed parts made from a strong, stiff aluminum alloy known as AlSi10Mg. The key finding was that the parameters that define the printing process—such as the beam power, beam diameter, scan speed, particle size, and powder layer thickness—all affect the product’s microstructure. Yet by carefully optimizing those parameters, it is possible to print aluminum alloy parts that are 99.8% dense (Add. Manufact. 2014, DOI: 10.1016/j.addma.2014.08.001).

But porosity in printed metal parts isn’t always a bad thing. Additive manufacturing specialists and veterinary researchers at NCSU sought to take advantage of porosity in bone-implanted prostheses. The team successfully implanted flexible titanium alloy mesh structures into several cats and dogs. The researchers think that the parts’ porosity could promote tissue growth at the implant site and protect against implant failure typically caused by the large difference in stiffness between bone and metallic implants (Add. Manufact. 2014, DOI: 10.1016/j.addma.2014.05.001).

Additive manufacturing has shown itself capable of making highly complex metal parts. But will the technology be used to make high-value parts for demanding engineering applications? “Absolutely,” Slotwinski said. “It’s just a matter of timing.” Researchers are working to optimize the fabrication process and control the product microstructure, he adds, “When we overcome these technical challenges, the technology will become more pervasive.”