Epitaxial Hexagonal Boron Nitride for Hydrogen Generation by Radiolysis of Interfacial Water

Abstract

Hydrogen is an important building block in global strategies toward a future green energy system. To make this transition possible, intense scientific efforts are needed, also in the field of materials science. Two-dimensional crystals, such as hexagonal boron nitride (hBN), are very promising in this regard, as it has been demonstrated that micrometer-sized flakes are excellent barriers to molecular hydrogen. However, it remains an open question whether large-area layers fabricated by industrially relevant methods preserve such promising properties. In this work, we show that electron-beam-induced splitting of water creates hBN bubbles that effectively store molecular hydrogen for weeks and under extreme mechanical deformation. We demonstrate that epitaxial hBN allows direct visualization and monitoring of the process of hydrogen generation by radiolysis of interfacial water. Our findings show that hBN is not only a potential candidate for hydrogen storage but also holds promise for the development of unconventional hydrogen production schemes.

Keywords: Raman spectroscopy; bubbles; deuterium; hydrogen barrier; hydrogen production; hydrogen storage.

Figure 1

Bubble formation mechanism. (a) Schematic illustration of the bubble formation. hBN is grown by MOVPE at temperatures above 1000 °C. After the growth the sample is cooled to room temperature, which leads to the formation of hBN wrinkles. The sample is removed from the reactor and exposed to ambient conditions. Electron beam exposure in an SEM leads to bubble formation. (b) AFM image of a typical wrinkle pattern. (c) Evolution of the SEM image as a function of exposure time. The acceleration voltage was 5 kV and the current was 1.4 nA. The whole image sequence took about 20 s. The white scale bars correspond to a length of 5 μm. (d) Optical microscope image of bubbles exposed in the shape of an ”hBN” label. The reddish dark shadow marks areas that were exposed by the electron beam. The white scale bar corresponds to a length of 20 μm. The red square indicates the region measured by AFM in (e). The AFM image shows that the wrinkle pattern vanishes on the bubbles due to strain relaxation, while it remains clearly visible elsewhere.

Figure 2

Evidence of molecular hydrogen in Raman spectroscopy. (a) Raman line scan across a hydrogen-filled bubble. It can be clearly seen that two lines are present only on the bubble. These lines correspond to the lines of molecular hydrogen S(0) at 354 cm^–1 and S(1) at 587 cm^–1. The Raman spectra are shifted vertically for clarity. (b) Optical microscope image of the studied bubble showing the laser spot (532 nm). The dashed line indicates the direction of the line scan. (c) Additional rovibrational lines measured on the bubble not shown in the line scan. It was possible to identify the first four lines S(0)–S(3) of the (0–0) transitions and line Q(1) of the (1–0) transitions.

Figure 3

Origin of the hydrogen. (a) The sample was mounted upside down in a container filled with heavy water (D₂O) for 23 days. Afterward the sample was directly mounted in the SEM and exposed by e-beam irradiation to form bubbles as shown in (a). (b) Raman spectra of two typical points on (next to) the bubble are shown as red (green) curves. The spectrum next to the bubble shows only Raman bands related to sapphire (the Raman band of hBN is at a higher energy; see the Supporting Information). For the measurement on the bubble not only a signal related to molecular hydrogen but also a signal related to hydrogen deuteride (HD) can be observed. (c) After subtracting the sapphire signal, three peaks (S(0), S(1), and S(2)) related to HD can be clearly observed.

Figure 4

Evolution of the Raman signal of molecular hydrogen. (a) The graph presents two spectra taken at the same point of the bubble directly after electron beam exposure and after 4 weeks. The Raman lines did not decrease in intensity. This means that the hydrogen is still confined in the bubble after 1 month under ambient conditions. The Raman spectrum of a bare sapphire substrate is shown for comparison. (b) Optical microscope images of a large bubble for different ambient pressures. The bubble was inflated and deflated by automatically cycling the pressure between 100 and 400 mbar (see Video 3 in the Supporting Information for a video of a typical cycle). (c) After certain numbers of pressure cycles Raman measurements were performed. Molecular hydrogen was still present after 551 cycles.