Researchers at RWTH Aachen University have spent more than a decade in a disagreement over how atoms bond in chalcogenide phase change materials, dividing the scientific community into opposing hypervalent and metavalent camps.
Atomic Bonding Theories Divide RWTH Aachen Researchers
For more than ten years, Richard Dronskowski and Matthias Wuttig, both based at RWTH Aachen University in Germany, have championed contrasting theories regarding the atomic structure of chalcogenide phase change materials. Although the two scientists once collaborated, years of heated academic exchanges mean they are no longer on speaking terms, separated by a profound theoretical divide despite offices located just a few kilometers apart. Researchers in the UK, Spain, Belgium, and South Korea are similarly invested in resolving the atomic puzzle to explain the unusual physical properties of these compounds.
The debate centers on why these materials adopt an unusual crystal structure despite featuring germanium, tellurium, and antimony atoms with very similar electronegativity values. Dronskowski contends that the atomic arrangement is easily solved by the fact that you eventually realise there are too many electrons
, leading atoms to form linear three-centre four-electron bonds.
PhD Positions: Adaptive Soft Matter (Project A1) at RWTH Aachen University, GermanyAre you passionate about interdisciplinary research at the intersection of soft matter physics, chemistry, and stem cell biology? The EMERGE graduate school at RWTH Aachhttps://t.co/9CpFLyFutf
I don’t have an idea why this is hard to swallow for some people. It has been observed decades before and it’s an exciting discovery that you have the same bonding [as in other electron rich materials] also in phase change materials.
Richard Dronskowski, RWTH Aachen University
To support his analysis, Dronskowski points to atom probe tomography experiments where crystals are bombarded with gaseous cations, which yields clusters of atoms hitting the detector. He argues that analyzing the electron wave function in compounds like GeTe reveals that germanium and tellurium donate all their valence electrons, requiring the system to distribute excess electrons across multiple atoms in hypervalent bonds by looking at the wave function of the electrons.
Disputing Metavalency and the Limits of Solid-State Bonding
Wuttig rejects the hypervalent model, arguing that electron-rich hypervalent bonding cannot exist in solid structures. He maintains that creating three-dimensional three-centre four-electron bonds would demand 12 electrons per unit cell, exceeding the maximum possible number by two.
Other international specialists have weighed in on the dispute. Stephen Elliott at the University of Cambridge notes that when color coding is removed from Wuttig’s plots, the proposed metavalent region blends seamlessly with ordinary covalently bonded materials. Elliott supports the hypervalent interpretation, pointing out that although hypervalent bonding is rarely invoked in extended solid-state materials, it remains well known across chemistry.
Wuttig initially described his concept as resonant bonding before changing the terminology to prevent confusion with molecules like benzene, a shift that critics argue masks an old idea. [Wuttig] is still claiming that this metavalent bonding is a previously unknown form of bonding,
Elliott said, describing that proposition as quite a big statement, frankly.
Neuromorphic Computing Applications Expand Material Demand
While theorists debate the fundamentals, practical applications for phase change materials continue to accelerate. By tuning the crystallinity of the material, engineers can generate a continuous spectrum of states between fully amorphous and fully crystalline configurations.
This continuous scaling permits data representation beyond binary 0s and 1s. Synthetic neurons constructed from these materials can execute multiple tasks simultaneously, significantly speeding up complex calculations. Wuttig acknowledges the broader technological implications, noting that it’s obvious these materials defy our conventional wisdom of what’s metal, what’s covalent [and] what’s ionic [bonding].
Commercial Markets and Industrial Expansion in the United States
Beyond experimental computing, commercial markets for phase change materials are expanding rapidly. The U.S. market was valued at USD 0.72 billion in 2024 and is projected to reach USD 1.54 billion by 2032, achieving a 10.0% compound annual growth rate that outpaces the broader American specialty chemicals sector. Growth is driven primarily by building envelope retrofits and cold chain logistics rather than the textile and battery manufacturing applications dominant in Asia.
Domestically, consumption relies on paraffin-based formulations derived from refinery byproducts in Texas and Louisiana. Hyperscale data center operators in Loudoun County, Virginia, are actively piloting paraffin thermal buffers to shift chilled-water loads away from peak utility rates, establishing data center cooling as the fastest-growing domestic application. Meanwhile, commercial building retrofit panels represent an addressable spend of millions annually across U.S. commercial renovations.
Venture Funding Fuels Global Fiber Production Lines
International startup activity is also scaling up to meet functional material demands. Phase Change Energy Storage (Beijing) Technology has secured US$13.8M in a Series A round, announced in October, following an earlier report noting a Series A raise. Backers for the round included Hangzhou City Investment, Linping State Investment, and Hangzhou Financial Investment, alongside prior backer Ventech China.
Founded in 2020 by Song Pengcheng—who holds a materials doctorate from Tsinghua University and previously worked at the Nuclear Power Institute of China—the company specializes in high-temperature melt spinning. The new capital will finance the construction of a 10,000-tonne production line for AC-TEX phase-change temperature-regulating fiber, supporting brand expansion and licensing initiatives across its phase-change, radiative cooling, and aerogel product lines.