A quiet shift in silicon manufacturing could be taking shape inside a new study on porous reaction-bonded silicon nitride, a ceramic material valued for its strength, thermal stability and resistance to chemical attack. Researchers T.G. Aguirre, E. Cakmak, M.D. Richardson and colleagues are investigating how the size of additively manufactured silicon components affects their conversion into silicon nitride through a pressureless nitriding process. Published in npj Advanced Manufacturing, the work connects binder jet 3D printing with a manufacturing route that could make complex ceramic structures easier to produce.

The central idea is deceptively simple: print a component from silicon powder, then expose it to nitrogen so that the silicon reacts and forms silicon nitride. In practice, however, the transformation is governed by a network of interacting physical and chemical processes. Nitrogen must penetrate the part, reach reactive silicon surfaces and diffuse through newly formed ceramic layers. Heat must also move through the structure, while gases generated or trapped during processing must escape. As component dimensions change, these processes can behave very differently.

Binder jet additive manufacturing offers a distinctive route for creating such parts. Instead of melting powder with a laser or electron beam, the process selectively deposits a liquid binder to join powder particles layer by layer. The resulting “green” component is then typically dried and thermally treated to remove the binder, leaving a fragile, porous body. For silicon-based structures, that interconnected porosity can become a processing advantage, allowing nitrogen to travel through the material during the subsequent reaction.

The researchers’ focus on size effects addresses a major challenge in scaling ceramic additive manufacturing from laboratory specimens to useful components. A thin printed sample may nitridize relatively quickly because nitrogen has only a short distance to travel. A thicker or larger part presents a far more demanding environment. Its interior can receive nitrogen more slowly, and the reaction front may progress unevenly from the outer surface inward. Differences in heating rate, temperature distribution and gas transport can produce variations in conversion, density and final microstructure.

The term “pressureless nitriding” is crucial to the study’s technological appeal. Many advanced ceramic manufacturing methods depend on external pressure, specialized furnaces or high-energy consolidation steps to achieve the desired structure. A pressureless route instead relies on controlled heating and the chemical affinity between silicon and nitrogen. If the reaction can be managed uniformly, manufacturers could potentially create intricate silicon nitride components without applying mechanical pressure during conversion, simplifying equipment requirements and expanding the range of printable geometries.

Silicon nitride is an especially attractive target because it combines several properties that are difficult to obtain in a single engineering material. It can withstand high temperatures, resist wear and corrosion, and maintain useful mechanical performance under demanding conditions. Those characteristics have made silicon nitride relevant to applications ranging from bearings and cutting tools to heat-resistant components and advanced energy systems. A porous version offers additional possibilities, including lightweight structures, filtration, thermal management and engineered surfaces where controlled pathways through the material are valuable.

The study’s attention to porosity is equally important. In reaction-bonded ceramics, pores are not merely defects; they can serve as channels for reactant transport. At the same time, excessive or poorly distributed porosity can weaken the final component and interfere with complete conversion. The scientific challenge is therefore to balance openness and strength. The printed silicon body must remain permeable enough for nitrogen to reach its interior, while the reaction must generate a coherent silicon nitride framework rather than leaving unreacted silicon or creating damaging gradients.

By examining components of different sizes, the researchers are addressing a question with direct implications for industrial design: can processing rules developed for small printed samples be applied to larger parts? The answer depends on how reaction kinetics, diffusion and heat transfer scale with geometry. A process that appears reliable in a small coupon may produce a different result in a thick wall, enclosed channel or complex lattice. Understanding those differences can help engineers determine allowable dimensions, optimize furnace schedules and design structures around the chemistry of nitridation rather than treating manufacturing as a purely geometric exercise.

The broader significance of the work lies in the convergence of three fields: powder-based additive manufacturing, reaction-bonded ceramics and transport-driven materials processing. Instead of printing a finished ceramic directly, the approach uses additive manufacturing to create a carefully engineered precursor and then transforms it chemically. That strategy could unlock shapes that are difficult to machine from dense silicon nitride, while reducing the need for conventional molds and tooling. As manufacturers seek lighter, more complex and more resource-efficient components, controlling size-dependent reactions may become as important as improving printer resolution.

The research arrives at a moment when additive manufacturing is moving beyond demonstrations of intricate shapes toward repeatable production of functional materials. Its emphasis on pressureless nitriding of binder jet additively manufactured silicon highlights a practical bottleneck often hidden behind spectacular 3D-printing images: the printed object is only the beginning. The final performance depends on what happens during post-processing, when gases, heat and chemical reactions determine the internal structure. By clarifying how component size influences that transformation, Aguirre, Cakmak, Richardson and their collaborators are helping map a route from powder and binder to porous silicon nitride parts designed for the high-temperature technologies of the future.

Subject of Research: Porous reaction-bonded silicon nitride produced through pressureless nitriding of binder jet additively manufactured silicon, with a focus on size effects.

Article Title: Porous reaction-bonded silicon nitride: Size effects for pressureless nitriding of binder jet additively manufactured silicon

Article References: Aguirre, T.G., Cakmak, E., Richardson, M.D. et al. “Porous reaction-bonded silicon nitride: Size effects for pressureless nitriding of binder jet additively manufactured silicon.” npj Advanced Manufacturing (2026). https://doi.org/10.1038/s44334-026-00110-y

Image Credits: AI Generated

DOI: 10.1038/s44334-026-00110-y

Keywords: Silicon nitride, reaction-bonded ceramics, pressureless nitriding, binder jet additive manufacturing, silicon, porous materials, size effects.

Researchers at the U.S. Department of Energy’s (DOE) Oak Ridge National Laboratory (ORNL) have created a new tool

Researchers have demonstrated an unusual new form of 3D printing capable of building entirely new structures directly onto—and even around—existing objects in just seconds. In a series of experiments, the technique printed microscopic fluid channels around embedded glass spheres, fabricated precision gears onto polished steel rods, and even built a miniature lens directly onto the surface of a working LED.

The breakthrough, described in a paper published in Nature Communications, relies on an advanced form of tomographic volumetric additive manufacturing (TVAM) that allows researchers to “overprint” new structures onto components made from a wide range of materials, including metal, glass, and preassembled microfluidic components. Instead of constructing objects layer by layer like conventional 3D printers, the system projects precisely calculated light patterns into a rotating vat of photosensitive resin, causing an entire three-dimensional object to solidify almost simultaneously within seconds.

The advancement could represent a major step toward making volumetric 3D printing practical in real-world manufacturing environments, where new components often must be integrated with parts that already exist.

It additionally highlights the growing sophistication of computational techniques that can accurately model how light behaves as it passes through materials that absorb, scatter, refract, or reflect light.

“Tomographic Volumetric Additive Manufacturing (TVAM) is a light-based 3D printing technique capable of producing centimeter-scale objects within seconds,” researchers write. “This work demonstrates a broad range of overprinting scenarios, where new structures are directly printed onto or around pre-existing components made from different materials.”

Unlike conventional stereolithography or fused-filament printers that gradually build an object from one direction, tomographic volumetric printing projects hundreds of two-dimensional light patterns into a rotating container filled with liquid resin.

As the resin accumulates enough light energy, the desired three-dimensional structure appears almost simultaneously throughout the volume rather than emerging layer by layer. The approach can produce centimeter-scale objects in mere seconds.

However, until now, one of the technology’s greatest challenges has been accurately calculating how light travels through complicated environments.

Existing objects inside the resin can block, scatter, or bend incoming light, degrading print quality. Conventional models often assume these objects are either completely transparent or completely opaque. This is an oversimplification that breaks down in realistic manufacturing scenarios.

To overcome those limitations, the team developed an upgraded computational framework called Dr.TVAM. Built upon physically based ray tracing, the software models how light propagates through materials with different optical properties while simultaneously optimizing the projected light patterns needed to create the desired object.

Because the framework accounts for reflection, refraction, scattering, absorption, and arbitrary container shapes, it can successfully print structures around components that previously would have interfered with the process.

One of the study’s most compelling demonstrations involved manufacturing microfluidic channels directly inside sealed, preassembled square chambers. Microfluidic devices contain microscopic channels that transport tiny volumes of liquids and are widely used in biomedical research, diagnostics, and organ-on-a-chip systems.

Instead of fabricating the channels separately and assembling the device afterward—a process that can introduce contamination or leaks—the researchers successfully printed perfusable channels directly inside an already assembled chamber while accounting for inlet and outlet structures that would otherwise obstruct the light.

They demonstrated straight channels, branching networks, and spiral geometries capable of carrying fluid after fabrication. Supplementary demonstrations showed colored dye successfully flowing through the finished structures.

The researchers then pushed the concept even further by creating what they describe as context-aware printing. In this experiment, they embedded two tiny glass spheres inside a hydrogel-filled chamber, rapidly determined each sphere’s three-dimensional position using two camera images, and generated customized print patterns in less than a minute.

Those optimized projections produced hollow channels connecting each sphere to inlet and outlet ports while simultaneously printing protective cavities around the spheres themselves. The complete workflow—from detecting the embedded objects to generating printing instructions and fabricating the final structure—required less than three minutes.

That ability could eventually enable highly customized biomedical manufacturing, where printers automatically adapt to the precise location of living tissues, organoids, or other biological materials rather than requiring perfect positioning beforehand.

Another experiment highlighted why precisely modeling light matters.

Previous demonstrations of volumetric overprinting treated metal components as though they completely absorbed incoming light. In reality, polished metal reflects and scatters significant amounts of light, creating distortions that reduce print quality.

Researchers tested both assumptions by printing a gear directly onto a polished steel rod. When the software ignored scattered light, the gear’s fine holes became over-polymerized and poorly defined.

After including realistic scattering behavior into the optical model, however, the resulting gear preserved those elaborate features far more accurately, increasing simulated print fidelity substantially. The improvement was confirmed experimentally.

Researchers even went as far as demonstrating that the process could print directly onto an electronic component.

Using an LED-based volumetric printing system, the researchers fabricated both a miniature lens and a small cross-shaped symbol directly onto the surface of a red LED. When illuminated, the printed lens projected the symbol onto a screen, successfully transforming the LED into a tiny optical projector without requiring additional assembly. The experiment showcased how optical elements themselves could potentially be integrated directly onto existing electronic hardware during manufacturing.

Beyond the individual demonstrations, the work additionally underscores an important trend in additive manufacturing: increasingly sophisticated software is becoming just as important as advances in hardware.

Instead of relying on simplified mathematical approximations, the new framework performs physically based optical simulations capable of modeling arbitrary geometries along with material properties.

The software has been released as open source, along with configuration files and experimental data, allowing other researchers to reproduce and expand upon the 3D printing work.

Although the current demonstrations remain laboratory-scale, the possible implications could extend well beyond experimental 3D printing. Biomedical researchers could eventually fabricate sterile microfluidic systems without multi-step assembly. Optical engineers may integrate lenses directly onto light sources. Manufacturers could add custom features to finished components without redesigning entire production processes.

Researchers also note that future improvements could further increase realism through incorporating additional physical effects, including changes in refractive index during polymerization and the diffusion of chemical inhibitors during printing. Those refinements could expand the technology’s accuracy and enable even more complex manufacturing scenarios.

“In summary, Dr.TVAM, our computational framework, has shown to be capable of modeling different optical situations for TVAM,” researchers write. “We believe this significantly expands the application space for TVAM, enabling the creation of complex, functionalized, and multi-component devices from applications in mechanics, optics, or biofabrication.”

Study: Wechsler F, et al. “Overprinting with tomographic volumetric additive manufacturing.” Nature Communications (2026).

Tim McMillan is a retired law enforcement executive, investigative reporter and co-founder of The Debrief. His writing typically focuses on defense, national security, the Intelligence Community and topics related to psychology. You can follow Tim on Twitter: @LtTimMcMillan.  Tim can be reached by email: tim@thedebrief.org or through encrypted email: LtTimMcMillan@protonmail.com