Uncovering the structural diversity of two-dimensional zinc oxide

When confined to atomically thin geometries, semiconductor materials may undergo reconstructions that alter their functionalities. It is thus necessary to characterize the atomic structures of semiconductor materials before fitting them into electronic devices, which are increasingly shrinking in size due to performance demands.

Two-dimensional zinc oxide (ZnO) has recently gathered interest in the development of nanoscale electronics. For the past two decades, ZnO has been described by an oversimplified structural model, but new experimental results indicate that the decades-long assumption of ZnO being a flat, graphene-like monolayer is incorrect; rather, covalent and ionic bonding coexist in its intrinsic structure, giving rise to energetically competitive polymorphs.

To understand the structural diversity of ZnO, Jiang et al. conducted a physics-constrained random structure search, which involves applying filters to identify uniquely connected structures. After this, they performed first-principles calculations to reveal ZnO’s ground state structure and low-energy metastable configurations.

“Rather than treating them as simple thin slices of bulk crystals or borrowing planar graphene-like templates, our results show that ultrathin ZnO can adopt a diverse landscape of competing atomic structures shaped by polarity, bonding, coordination, and electrostatics,” said author Yuan Yan. “The structure-search strategy and the resulting structural motif library therefore provide a physics-based roadmap for exploring broader families of polar non-van der Waals semiconductors.”

The authors found the ZnO ground state to be a double-layer honeycomb. Below approximately 50 millielectronvolt per atom, metastable triangular tetrahedral polymorphs emerge, with some exhibiting direct band gaps, strong transport anisotropy, and exceptionally high calculated carrier mobilities that are ideal for electronic and optoelectronic applications.

The authors hope their computational methods help move the field from simply predicting semiconductors towards experimentally identifying, stabilizing, and utilizing their most promising structural forms.

“The atomic models, energetic windows, and predicted properties provided here may help guide targeted experiments using approaches such as substrate templating, interfacial confinement, strain, or controlled growth environments,” said Yan.

Source: “Hidden polymorphic landscape in two-dimensional ZnO,” by Bei Jiang, Dan Wu, Qiuhan Chen, Minglei Sun, Yinchang Ma, Yuan Yan, Chi Cheng, and Jefferson Zhe Liu, Applied Physics Letters (2026). The article can be accessed at https://doi.org/10.1063/5.0345257 .