HOW TO CUT POROSITY IN 3D-PRINTING
A vacuum-assisted extrusion technique aims to boost polymer print performance.
Written by Michael Abrams
The vacuum-assisted extrusion process is tested in large scale additive manufacturing to reduce porosity in the printed parts. Photo: Vipin Kumar/Oak Ridge National Laboratory, U.S. Dept. of Energy
MOST OF TODAY’S CARS AND PLANES don’t include a lot of 3D-printed pieces in structurally important areas. Scale is not the issue. Technology exists to print big—large-format additive manufacturing, or LFAM, machines have been pumping out meter-scale objects for more than a decade now. The primary problem is the porosity found in the finished product.
Now researchers at Oak Ridge National Laboratory (ORNL) have found a solution, one so effective that it may single-handedly allow 3D-printing to be more widely used in the automotive and aerospace industries. And it’s as simple as sucking air.
There are a number of reasons for the pockets that keep a polymer print from being solid. For one thing, pre-printed polymer beads can be produced with some porosity from the get-go. And most polymers used in 3D-printing are hydrophilic and pull moisture from the air. When heated during extrusion the vaporized moisture can create pockets. High temperatures can also degrade the polymer itself, causing off gassing that increases porosity.
But much of the worst porosity is introduced at the macro level: the space between pellets is the culprit. And it can work its way into the extrusion process and, ultimately, into the final part, weakening it drastically.
“So, these pellets are in the shape of either spherical balls or cylindrical tubes and when you put them in a chamber or hopper, they’re not tightly packed—there’s always some air between them,” explained Vipin Kumar, a senior R&D staff scientist at ORNL. “These pellets are gravity fed and when they go inside the barrel they take the surrounding air with them. And that air has a very significant influence on the final porosity.”
There are ways of extracting that air, primarily the use of a vented screw to push the polymer. But those screws, often used with injection molding, are expensive and, what’s worse, large. “We want to do 3D-printing with robots, which means tight space is important,” Kumar said.
The fix that Kumar and his colleagues came up with comes in two stages. First, dry the polymer. This they did at 176 °F for more than eight hours. Stage two is where the big innovation comes in: put a vacuum on the hopper. “Then just turn on the vacuum and extrude the polymer,” Kumar said.
The results of this technique were remarkable. With the small, batch-scale experiments run in the lab, porosity was reduced to under two percent. “Almost like aerospace grade quality,” he explained. “It’s considered to be acceptable in most structural applications.” For larger scale prints there was a 75 percent decrease in porosity.
These scans show printed parts produced with (a) no-vacuum and (b) vacuum-assistance. When the vacuum is used in the material feeding system, porosity dropped by about 75 percent. Photo: Vipin Kumar/Oak Ridge National Laboratory, U.S. Dept. of Energy
“There are fibers as well in the polymer stream and when you apply vacuum, you're putting a different force on them.”
—Vipin Kumar, a senior R&D staff scientist at Oak Ridge National Laboratory
However straightforward the addition of the vacuum, there is some finessing necessary to the process. For instance, it takes a few seconds before the vacuum is at full strength, so extrusion should start a bit after the vacuum is activated. And the vacuum can’t be too strong, or it will overcome the force of gravity and keep the pellets in the hopper. “There has to be a nice balance between how much vacuum you can apply while not disturbing your flow rate,” Kumar said.
It’s possible that the application of the vacuum will have an impact on the strength of 3D-printed objects in other ways, outside of the porosity issue. “There are fibers as well in the polymer stream and when you apply vacuum, you're putting a different force on them,” Kumar continued. “So, either they will align more or they will become more random—that’s an area for future study.”
In addition to looking into the effect of a vacuum on fibers within the polymer, Kumar is hoping to soon use a vacuum on larger scale, continuous systems. “That’s the next step,” he said. “We have an idea, we have a concept, but we need an industry partner to build that one.”
Michael Abrams is a technology writer in Westfield, N.J.
© 2025 The American Society of Mechanical Engineers. All rights reserved.