Titanium is not supposed to float. It is about four and a half times denser than water, and every metal lattice ever printed, no matter how airy, eventually fills with water and sinks. This week a team at RMIT University in Melbourne published a fix, and the fix is almost embarrassingly simple: print the lattice with hollow struts, then fill only the struts with foam. The result floats, keeps floating after it cracks, and outperforms the stainless steel and plastic used in buoys and jetties. Here is what they did, and what it teaches anyone designing 3D printed parts, whether you print metal or not.
What RMIT actually built
According to RMIT's September 3 announcement, the Centre for Additive Manufacturing printed a titanium lattice made of hollow, interconnected struts and injected polyurethane foam into those struts. Water flows freely through the open cells of the lattice, but the sealed, foam-filled struts trap gas and keep the whole structure buoyant. The study, published in Advanced Materials (DOI 10.1002/adma.74641), is described as the first reported demonstration of a floating metal-hybrid lattice metamaterial.
The numbers are worth repeating. Compared at the same overall density, the lattice tested 70 percent stronger than the stainless steel or high-density polyethylene commonly used in marine hardware. After two weeks in natural seawater from Port Phillip Bay, it lost 0.15 percent of its mass and less than 1 percent of its strength. Samples floated in freshwater for more than two months. The team also printed a working buoy that stayed stable in a seawater tank tilted to 45 degrees with no sealed casing, no coating, and no added flotation, as VoxelMatters reported.
The part that should make engineers sit up is the damage tolerance. Lead researcher Dr Jordan Noronha said the lattice stayed afloat after cracking, after failure at key connection points, and after the fracture of an entire lattice layer. It only sank once it was severely crushed and compacted. A conventional hollow marine structure floods the moment it cracks. This one has hundreds of tiny sealed cells, so no single crack can flood it.
Skeletal density: the design rule you can steal
The team introduced a metric they call skeletal density. Normal density math counts the empty space inside a lattice as part of the volume, which is misleading when water can flow into that space. Skeletal density counts only the material that actually excludes water: the titanium walls and the sealed, foam-filled channels. The rule is then simple. If the skeletal density is lower than the density of the surrounding liquid, the structure floats, even if water pours through every external opening.
That framing applies far beyond boats. Project leader Distinguished Professor Ma Qian noted that swapping the material inside the titanium framework could tune a similar structure for energy absorption, thermal management, or vibration control. The hollow strut is a container, and what you put in it is a design variable.
I need a part that floats but has to be strong. Can you 3D print something like this?
Yes, we can 3D print strong buoyant parts today, though not in titanium: we design closed-cell or foam-filled polymer geometries on our desktop machines, and for metal lattice work like RMIT's we help you prototype the geometry in polymer before you commit to a metal build. RMIT's trick is a geometry trick more than a material trick. Hollow struts with a sealed core print fine in PETG, ABS, or PCTG with the channel modeled in CAD, and we check fit and buoyancy on the bench before anything goes to metal. A lattice that behaves badly in plastic will not magically behave well in titanium. Start with our Build and Ship page to send us the requirements.
My lattice prints look great but they crack under load. What am I doing wrong?
Most lattice cracking on desktop printers comes from strut junctions that are too thin for the nozzle to fully fuse, so widen the nodes, orient the part so the load runs along the extruded roads rather than across layer lines, and add perimeters instead of infill. RMIT's structure survived junction failures because it had redundancy: many struts, many sealed cells, no single point that could take the whole part down. You can borrow that. Design lattices with more, thinner load paths rather than a few heavy ones, and keep every strut at least three nozzle widths across so the slicer has room to lay real perimeters. If you are pushing into high-temperature or chemically harsh environments, our PEEK and ULTEM printing service handles the materials that ordinary desktop filaments cannot.
Why a shop in Rockville cares about a buoy in Melbourne
We do not print titanium, and a foam-filled buoy is not a typical Tuesday order. What we do all day is design and print small, strong, one-off parts that have to work the first time, and this study is a clean lesson in how geometry beats brute material. It also points at a workflow we use constantly: model the internal channel, print the hollow version, fill or seal it as a second step. Sealed float chambers in pond hardware, silicone-filled sensor mounts, and hollow-core brackets with injected epoxy all follow the same pattern. If the manufacturer of a part like that has stopped making it, our 3D scanning and reverse engineering service can capture it, rebuild the CAD, and print a replacement in a material suited to where it lives.
Brands building marine sensors, drone floats, or lattice-based hardware: we review products in this space. If you want us to put your part through real-world testing and write about it, email us at info@dcadditivepros.com.
Frequently asked questions
Can 3D printed metal actually float on water?
Yes, if the geometry is right. RMIT's 2026 study printed a titanium lattice with hollow struts filled with polyurethane foam, and it floated in freshwater for more than two months and stayed buoyant even after cracking. The metal itself still sinks; the sealed foam-filled struts are what keep the structure up.
What is skeletal density in 3D printed lattice design?
Skeletal density counts only the parts of a structure that exclude water, such as solid walls and sealed channels, and ignores open cells that water can flow into. If the skeletal density is lower than the surrounding liquid, the part floats even with water flowing through it.
Can I 3D print a buoyant part in plastic instead of titanium?
Yes. The same hollow-strut, sealed-core approach works in PETG, ABS, or PCTG on a desktop printer, and it is a smart way to prototype a lattice before paying for a metal build. Model the internal channels in CAD and seal or fill them as a second step.
How strong was RMIT's floating titanium lattice compared to steel?
At the same overall density, the lattice tested 70 percent stronger than the stainless steel or high-density polyethylene used in marine hardware, and after two weeks in seawater it lost only 0.15 percent of its mass and under 1 percent of its strength.