Published: June 07, 2026, 00:47 UTC
For decades, materials scientists have known that you can control the electronic properties of semiconductors by introducing precisely engineered interfaces. Polarization at semiconductor junctions is the backbone of transistors, LEDs, and solar cells.
Metals, by contrast, have been considered stubbornly uniform. Their electronic structure — the dense sea of mobile electrons that defines how they behave in circuits, as catalysts, or as electrodes — was thought to be essentially fixed once you pick the metal.
A team led by the University of Minnesota has just overturned that assumption. In a study published in Nature Communications, researchers showed that by growing a metallic oxide film just a few atomic layers thick and varying its thickness by a mere 2 to 3 nanometers, they could tune the metal’s work function by over 1 electron volt — a swing large enough to fundamentally alter how the material interacts with its surroundings.
The interface effect
The key is not the metal itself, but what happens where it meets another material. The researchers grew thin films of ruthenium dioxide (RuO₂) on top of titanium dioxide (TiO₂) using a technique called hybrid molecular beam epitaxy, which deposits atoms one layer at a time with exquisite precision.
At the interface between the two materials, the atomic lattice of RuO₂ is initially strained — stretched or compressed to match the underlying TiO₂ crystal. As the film grows thicker, this strain relaxes. What the Minnesota team discovered is that this transition from strained to relaxed, which happens at a critical film thickness of roughly 4 nanometers, triggers a dramatic shift in the electronic properties of the metal.
“What is remarkable is the simplicity of the control knob,” said Seung Gyo Jeong, the study’s first author and a postdoctoral researcher at the University of Minnesota. “By changing the film thickness by just a few nanometers, we can modulate the work function by more than 1 eV.”
To put that in perspective: 1 eV is roughly the width of the band gap of silicon, the material that powers the modern electronics industry. Changing a metal’s work function by that amount means you can fundamentally alter how it injects or extracts electrons from adjacent materials — a critical parameter in every electronic device.
Seeing the invisible
The mechanism behind this effect is what the team calls strain-stabilized interfacial polarization. At the RuO₂/TiO₂ interface, the atomic strain causes the metal ions to shift slightly relative to the surrounding oxygen atoms, creating a polarization — a separation of positive and negative charge — inside the metal itself. This kind of polar distortion is well known in insulators and semiconductors, where it produces ferroelectricity. But in metals, where mobile electrons normally screen out any internal electric fields, it was thought to be impossible.
To confirm the effect was real, the team collaborated with researchers at MIT who used a sophisticated imaging technique called multislice electron ptychography to visualize the atomic displacements directly. The images show the ruthenium atoms shifting away from their symmetric positions, creating the polarization that alters the electronic structure.
The team also used Kelvin probe force microscopy to measure the work function changes directly, confirming that the 1 eV shift was reproducible and directly linked to the film thickness.
What this means for technology
Work function is one of the most important parameters in materials science. It determines how electrons move across interfaces — the Schottky barrier height at a metal-semiconductor junction, the catalytic activity of an electrode surface, and the efficiency of charge injection in electronic devices.
“This is a completely new way to think about controlling metals,” said Bharat Jalan, the study’s senior author and a professor at the University of Minnesota. “If we can tune work function by 1 eV simply through thin film engineering, it opens possibilities for designing materials with custom electronic properties.”
Potential applications include:
- Tunable electronics: Electrodes and contacts whose work function can be tailored for specific device architectures, potentially improving the efficiency of transistors and solar cells.
- Catalysis: The surface electronic structure of metals directly affects how they catalyze chemical reactions. A tunable work function means a single material system could be optimized for different reactions.
- Quantum devices: Controlled metallic interfaces are a building block for some quantum computing architectures, where precise control over electronic properties at the nanoscale is essential.
The caveats
This is foundational research, not a deployable technology. Several important limitations remain.
First, the effect has been demonstrated in only one material system — RuO₂ on TiO₂ — and it is not yet clear how generalizable it is to other metals or substrates. Second, the films are grown by molecular beam epitaxy, a technique that requires ultrahigh vacuum and is far too slow and expensive for large-scale manufacturing.
Third, the critical thickness is about 4 nanometers, with the useful tuning window spanning just 2 to 3 nanometers. Fabricating devices with that level of precision at scale is a significant engineering challenge.
The study is a proof of concept: it establishes that the phenomenon exists, that it can be measured directly, and that it produces effects large enough to matter for applications. But practical devices based on this principle are still years away.
What’s next
The Minnesota team is now exploring whether similar strain-stabilized polarization effects exist in other metallic oxide systems. They are also investigating whether the effect can be combined with external electric fields to create dynamically tunable metals — a material whose work function could be switched on demand.
If the phenomenon proves general, it could give materials scientists a new tool: the ability to treat metals not as fixed materials, but as platforms whose electronic properties can be engineered at the atomic scale.
Paper: Seung Gyo Jeong, Bonnie Y. X. Lin, Mengru Jin, et al. “Strain-stabilized interfacial polarization tunes work function over 1 eV in RuO₂/TiO₂ heterostructures.” Nature Communications 17, Article 2516 (2026). DOI: 10.1038/s41467-026-69200-x
Lead institution: University of Minnesota Twin Cities, Department of Chemical Engineering and Materials Science
Funding: U.S. Department of Energy, Air Force Office of Scientific Research