Probing Phase Diagrams of Ordered Two-Dimensional Ice

Abstract

Water, a ubiquitous and fundamental substance, plays a critical role across a wide range of disciplines from physics and chemistry to biology and engineering. Despite theoretical predictions of several phases of two-dimensional (2D) ice confined between idealized hydrophobic walls, experimental validation has been limited to the square phase, whose structural origin remains controversial. Here, we propose a realistic nanoconfinement setup using wide-bandgap hexagonal boron nitride (h-BN) as the capping layer and Cu(111) as the substrate. This protocol enables scanning tunneling microscope (STM) to resolve the atomic-scale arrangement of water molecules beneath the h-BN layer, overcoming the limitations of conventional techniques. Simulated STM images unambiguously identify all ordered flat 2D ice phases, as well as coexisting phases, and effectively distinguish them from potential contaminants. These findings establish a robust framework for experiment to systematically probe the phase structures of 2D ice, opening an avenue for studying nanoconfined water under ambient conditions.

Figures (7)

• Figure 1: Configurations for 2D ice phases. a, A water monolayer confined between two graphene monolayers, with the bottom sheet supported on a Cu grid. With this setup, experimental measurements via transmission electron microscopy (TEM) speculated a square arrangement of the water molecules. b, Hydrophobic walls were used in theoretical simulations to explore phase diagrams of nanoconfined 2D ice. This idealized configuration approximates the van der Waals interactions between nanoslits and water molecules but does not correspond to a realistic experimental setup. c, The setup proposed in this work, where a water monolayer is confined between a realistic Cu(111) substrate and a scanning tunneling microscope (STM)-transparent h-BN capping layer. The publication years of the respective studies are labeled in parentheses, and the proposed or observed phases are labeled in the braces.
• Figure 1: Detailed Projected DOS for the NaCl monolayer capped by h-BN. The cyan and blue lines represent the projected DOS contributed from the total NaCl layer and that from the Cl atoms, respectively. The projected DOS of the h-BN (gray region) capping layer is also depicted for comparison.
• Figure 2: Identification of the square 2D ice. a, Top view of the optimized geometry of the square 2D ice confined between a h-BN layer and a Cu(111) surface, with a confinement width of 5.7 Å. The lattice constants (in Å) of the heterogeneous layers and the supercell of Cu(111) are labeled. b, Projected density of state (DOS) for the square 2D ice system with a single h-BN layer (top) and a single graphene layer (1LG, bottom) as the capping layer. c, Simulated STM images for confined square 2D ice with h-BN (left) and graphene (right) as the capping layer. d, Top view after replacing the water layer in panel (a) with a NaCl layer. e, Projected DOS for the system of NaCl capped by h-BN. f, Simulated STM images of confined NaCl with a single h-BN layer (left) and a single graphene (right) as the capping layer. All STM simulations were performed at a sample bias of --0.5 V.
• Figure 2: Identification of an alternative square 2D ice. a, Top view of the optimized geometry of the square 2D ice that has the primitive water unit cell reported in Ref. Chen2016, confined between a single h-BN layer and Cu(111) surface with a confinement width of 5.5 Å. The lattice constants (in Å) of the heterogeneous layers and the supercell of Cu(111) are labeled. b, Projected density of state (DOS) for the alternative square 2D ice systems with a single h-BN layer (top) and a single graphene layer (1LG, bottom) as the capping layer. c, Simulated STM images for confined square 2D ice with h-BN (left) and graphene (right) as the capping layer. d, Top view after replacing water with NaCl in panel (a). e, Projected DOS for the NaCl systems capped by h-BN (top) and graphene layer (1LG, bottom) capping NaCl. f, Simulated STM images of confined NaCl with a single h-BN layer (left) and a single graphene layer (right) as the capping layer. All STM simulations were performed at a sample bias of --0.5 V.
• Figure 3: Probing other 2D ice structures. Top view of the optimized geometries (left) and the corresponding simulated STM images (right) for the zigzag (a), square/zigzag coexisting (b), pentagonal with rectangle supercell (c), and pentagonal with square supercell (d) 2D ice confined between the h-BN and Cu(111) surface. The lattice constants (in Å) of the heterogeneous layers and the supercell of Cu(111) are labeled. All STM simulations were performed at a bias voltage of --0.5 V.