Chemists Synthesize Diamondiyne, a Porous 3D Carbon Allotrope Predicted 35 Years Ago

Researchers at the University of Gothenburg, Chalmers University of Technology, and Stockholm University have synthesized diamondiyne, a three-dimensional carbon allotrope that existed only as a theoretical model for 35 years. The breakthrough adds a porous, expanded network to the short list of known 3D carbon structures.[1][4]

The theoretical foundation for this material was laid in the late 1980s, when computational chemists began mapping the nearly endless ways carbon could theoretically bond with itself. The number of predicted carbon structures now stands at 1,635, though the vast majority remain purely hypothetical.[1][4]

The team produced the material as a thin film at the boundary between two liquids, bypassing the extreme conditions usually required to forge 3D carbon. The findings were published in the journal Angewandte Chemie International Edition in September 2026.[2][4]

"It has taken us an incredibly long time, but we finally succeeded," said Karl Börjesson, a professor of physical chemistry at the University of Gothenburg who led the research team. "I am very pleased."[4]

For decades, physical carbon allotropes were largely limited to graphite, diamond, and artificially produced low-dimensional forms like fullerenes, carbon nanotubes, and graphene. Diamondiyne is the first of the newer synthetic group to feature a fully three-dimensional covalent lattice.[4]

Expanding the Diamond Lattice

To create diamondiyne, the researchers had to modify the fundamental bonding architecture that gives conventional diamond its density. In a standard diamond, carbon atoms form a rigid network connected entirely by single bonds, requiring immense geological pressure to assemble.[2][4]

Diamondiyne preserves that overall tetrahedral topology but incorporates carbon atoms with different hybridization states. The structure combines sp³-hybridized carbon atoms, which form single bonds, with sp-hybridized carbon atoms that participate in triple bonds.[2][5]

The triple bonds act as rigid spacers, pushing the tetrahedral nodes further apart. In chemistry, the "-diyne" suffix indicates the presence of two triple bonds, which are inserted between the carbon centers to expand the repeating network.[2]

"The structure creates voids between the tetrahedra," Börjesson explained. "The carbon allotrope can therefore be compared to a porous diamond. But unlike diamonds, diamondiyne does not require high pressure to make the carbon atoms bond to one another."[4]

The Liquid Interface Method

Getting the carbon atoms to adopt this specific geometry proved synthetically challenging. Researchers took the first steps toward making diamondiyne more than 30 years ago, but early attempts struggled with unstable building blocks like tetraethynylmethane.[1][2]

The strong single and triple carbon-carbon bonds also tend to lock into incorrect, amorphous arrangements before a crystal can grow. To solve this, the Swedish team utilized a cascade reaction at the interface of two immiscible liquids, using relatively inexpensive laboratory glassware.[1][3][4]

By confining the reaction to the two-dimensional boundary between the liquids, the researchers constrained the geometrical arrangement of the molecular nodes. This liquid-liquid interface method allowed the carbon atoms to connect based on strict spatial constraints, forming an extended film.[3][5]

The resulting material is not a uniformly crystalline bulk solid. Instead, the synthesis produces amorphous carbon films that contain distinct, highly ordered crystalline regions of diamondiyne embedded within them.[2][5]

Confirming the Crystal Structure

To verify that they had actually built the predicted allotrope, the team relied on high-resolution electron microscopy performed at Stockholm University. The imaging confirmed the presence of the expanded diamond-topology network on the nanoscopic scale.[3][4]

The analysis revealed two asymmetrically interpenetrated 3D networks forming a specific tetragonal space group. The unit cell parameters measured 11.47 angstroms for the lateral dimensions and 16.22 angstroms for the vertical axis, matching the theoretical models.[5]

"We have at least managed to produce repeating patterns of diamondiyne in a volume measuring 10 nanometers on each side, 10 × 10⁻⁹ meters," Börjesson noted. "The images of the patterns correspond well with the theoretical models."[4]

While a 10-nanometer crystal patch is microscopic, it represents a definitive proof of concept. It demonstrates that the robust covalent bonds required for a 3D carbon network can be assembled at room temperature without collapsing into graphite or amorphous soot.[3][4]

Future Applications and Voids

The discovery emerged from a broader effort to develop porous, carbon-based materials that do not rely on metal ions to hold their shape. These covalent organic frameworks offer a lightweight alternative to metal-organic frameworks, which recently earned the 2025 Nobel Prize in Chemistry.[4]

Because diamondiyne forms covalent bonds in all three dimensions, it promises exceptional stability compared to other porous networks. Researchers are now preparing to investigate the material's physical, electrical, and mechanical properties to determine its practical utility.[1][3]

The nanoscale voids created by the triple-bond spacers present the most immediate opportunity for materials science. The open lattice could potentially be used for gas storage, molecular filtration, or as a scaffold for other chemical reactions.[3][4]

"We will now investigate the properties of diamondiyne and how it might be used," Börjesson said. "One exciting aspect of the porous structure is that several diamondiyne structures could potentially be interwoven through each other's voids."[4]

Scaling up the synthesis remains the primary hurdle before diamondiyne can transition from a laboratory curiosity to an industrial material. The current liquid-interface method produces thin films, and researchers must refine the cascade reaction to grow larger, purely crystalline bulk samples.[2][5]