TU Wien research published in Science points to a fundamental limitation in the interface between 2D materials and insulating layers in advanced transistors
The chip industry has been searching for years for the next generation of materials that will allow transistors to continue to be miniaturized. One of the great hopes was two-dimensional materials, extremely thin layers one or a few atoms thick, such as graphene and molybdenum disulfide. But new research by Prof. Mehdi Porfat and Prof. Tibor Grasser from TU Wien points to a fundamental problem: It is not enough to examine the active material itself. You also need to examine how it touches, or more precisely, does not touch, the insulating layer above it. (Vienna University of Technology)
In a typical transistor, an electrical gate controls the transition of the semiconductor between a conducting and non-conducting state. To ensure effective control, the gate is separated from the active material by a very thin insulating layer. When the active material is a two-dimensional material and the insulating layer is usually an oxide, a tiny gap is created between them due to weak van der Waals bonds. According to the researchers, this gap is only about 0.14 nanometers, but it is enough to weaken the capacitive coupling between the layers and limit the ability to shrink the device.
A chip is not made of a single material.
The implications for industry are broad. For years, 2D materials have been tested for their intrinsic properties: electron mobility, energy band structure, stability, and the possibility of integration into manufacturing processes. The new study reminds us that a chip is not made of a single material, but of a complete structure. Even a material that looks great on paper may fail when connected to a real insulating layer. According to a preliminary version of the article, the gap could add about 2.7 angstroms to the equivalent oxide thickness, which is to say, it would damage the very parameter that the industry is trying to reduce in future manufacturing generations. (arXiv)
The paper, published in Science on April 16, 2026, under the title “Device-scaling constraints imposed by the van der Waals gap formed in two-dimensional materials,” argues that the gap is not just an engineering nuisance but a physical limitation that needs to be factored into the roadmap for device design. In other words, choosing a 2D material based on its performance alone could lead to large investments in technology that will not be able to meet the miniaturization demands. (PubMed)
The researchers do not present the finding as the end of the road for 2D materials, but rather as a call for more precise design. One direction they suggest is the use of “zipper materials” – structures in which the semiconductor and insulator are joined together in a stronger bond, rather than relying solely on weak van der Waals forces. In such structures, the gap can be reduced or eliminated, thereby improving the electrical control of the transistor.
For the chip industry, the key takeaway is that next-generation materials need to be considered as a complete system: semiconductor, dielectric, interface, heat, currents, and leakage. This is an especially important message in an era where advanced junctions, 3D packaging, and chiplets are making chip design more complex. 2D materials may still play an important role in future electronics, but the new research sets a clear condition: Those who want to use them in truly tiny transistors will need to design the atomic interface as well as the material itself.
Short FAQ
What are 2D materials and why are they of interest to the chip industry?
2D materials are extremely thin layers, one or a few atoms thick. They are considered promising candidates for future transistors because they may enable further miniaturization of electronic components.
What is the problem that TU Wien researchers found?
The researchers found that a tiny gap forms between a 2D material and the insulating layer above it, due to weak van der Waals bonds. Although the gap is only about 0.14 nanometers, it weakens the electrical control of the transistor and may limit miniaturization.
Why is the finding important for the chip industry?
It shows that it is not enough to test the properties of the material itself. The entire structure of the component must be tested: the semiconductor, the insulating layer, and the interface between them. A material that looks promising on paper may fail when integrated into a real chip.
Does the study rule out the use of 2D materials?
No. The researchers are not saying that 2D materials are irrelevant, but that they should be chosen and designed with more care. They suggest looking at structures in which the semiconductor and insulator are more tightly bonded to reduce the atomic gap.
What are “zipper materials”?
These are structures in which the semiconductor material and the insulating layer are more tightly integrated, rather than relying solely on van der Waals forces. Such an approach could improve electrical control of tiny components.
More of the topic in Hayadan:
