Drivers of chemical diffusion of hydrogen in the thin transition metallic glass V 80 Zr 20

Lennart Spode, Ola Hartmann, Gunnar K. Pálsson · 2025

Read Drivers of chemical diffusion of hydrogen in the thin transition metallic glass V 80 Zr 20 by Lennart Spode, Ola Hartmann, Gunnar K. Pálsson (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-13103.txt.

Introduction

Lennart Spode, Ola Hartmann, and Gunnar Pálsson show how to measure hydrogen diffusion inside a thin metallic glass film using light, in a 2025 paper. The glass, a vanadium-zirconium mix, takes up hydrogen, and tracking how fast it spreads has been hard without a phase change to watch. Their method reads the film's light transmission to extract hydrogen profiles in space and time at once.

From those profiles they derive how the energy barrier for diffusion shifts with hydrogen content. The shift matches hydrogen in disordered bulk vanadium, not in neat crystal films of like thickness. Diffusion speeds up as hydrogen rises in the glass, the reverse of the crystal case.

Catalog record

CREATORLennart Spode, Ola Hartmann, Gunnar K. Pálsson
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE29 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-13103

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Drivers of chemical diffusion of hydrogen in the thin transition metallic glass V 80 Zr 20 Lennart Spode, Ola Hartmann, Gunnar K. Pálsson Abstract We demonstrate the feasibility of using optical transmission to determine concentration-dependent hydrogen diffusion coefficients and activation energies of thin metallic glass films over a wide range of temperatures and concentrations. The hydrogen concentration’s temporal and spatial profiles are simultaneously extracted without requiring a metal-insulator transition or hydride formation. The concentration-dependent activation energy is extracted and is found to exhibit similar concentration dependence as hydrogen in the disordered bulk α \alpha -phase of vanadium. The activation energy is in stark contrast, however, to the activation energy found in epitaxial vanadium films of similar thickness. The chemical diffusion increases with hydrogen concentration in the glass, whereas it decreases in the crystalline case owing to the differences in the thermodynamic factors. This technique provides detailed insight into the concentration dependence of hydrogen diffusion and constrains theories treating correlated motion of hydrogen and short-range hydrogen-hydrogen interaction. The method can be applied to other materials that show an optical response to hydrogen uptake. I Introduction Hydrogen interactions with metals are intriguing from both a fundamental [ 1 , 2 , 3 ] and an applied point of view such as for hydrogen storage and material durability [ 4 ] . Specifically, hydrogen diffusion in disordered metals exhibits experimental anomalies such as non-Arrhenian diffusion [ 5 , 6 ] . Kirchheim, Richards and others have developed theoretical models for describing the motion of hydrogen in metallic glasses [ 7 , 8 , 9 ] , however, many aspects of the mechanisms that drive it are still unknown, like the influence of the coupling to a thermal bath [ 10 ] , the dynamics of self-trapping, the appropriateness of the description of hydrogen as a small polaron quasi particle [ 11 , 12 ] , or what drives the non-Arrhenius-like temperature dependence of the diffusion coefficient D . Furthermore, the potential role of topological disorder, mobility correlation, short-range hydrogen-hydrogen interaction and its effects on blocking and site-availability, and the fractal character of the network of diffusive paths remain poorly understood. This is partly due to the difficulty of accurately modeling the influence of the amorphous host structure and partly due to a lack of experimental data required to test the different models. To validate candidate models, it is highly desirable to have consistent data over a wide range of both concentrations and temperatures for both the diffusion coefficient and the activation energy. The diffusion of the absorbed hydrogen has generally been described with over-barrier hopping between local interstitial sites [ 13 , 12 ] . However, the motion of hydrogen in metals exhibits a plethora of other interesting phenomena, ranging from quantum tunneling and Kondo physics to critical slowing down [ 13 , 14 ] . A strain field emerges during the dynamic self-trapping process, which together with the charge, can be viewed as a small polaron [ 7 , 9 ] , a quasi-particle with peculiar properties. Ab-initio molecular dynamics calculations have even shown that hydrogen can decouple from its strain field, bringing about a new range of diffusive behavior [ 12 ] . Thin films offer an interesting playground to investigate metallic glasses due to the high achievable quenching rates during their growth, making it possible to grow films with a high degree of uniformity while avoiding inclusions, voids and other defects [ 15 ] . They allow well-controlled composition, surface termination [ 16 , 17 ] , and material interfaces [ 18 ] . Furthermore, metallic glasses can be realized with compositions not accessible with bulk methods [ 19 ] , offering access to unexplored territory. These meta-stable phases can occur in regions forbidden in the corresponding crystalline phase diagram [ 19 ] . However, measuring the diffusion coefficient in thin films is currently not possible using standard techniques such as the Gorsky effect, nuclear magnetic resonance, or quasi-elastic neutron scattering due to the low sample volume. In 2012, a method was demonstrated that allows to determine hydrogen diffusion in crystalline vanadium films optically, by relying on the linear dependence between optical transmission and hydrogen concentration [ 20 ] . A great advantage is that both the temporal and spatial profiles are obtained simultaneously, thereby providing more accurate diffusion coefficients. This enabled a number of studies using the technique to look at diffusion in artificially made structures such as superlattices [ 21 , 20 , 22 , 23 ] . This method furthermore, circumvents the need for using indicator layers [ 24 ] , which provide indirect measurements of hydrogen concentration. Another advantage of the technique is that the concentration dependence over a wide range can be determined from a single measurement, making it unnecessary to use many different samples with different hydrogen concentrations. The band structure of a metallic glass is not a well-defined concept, and inter-band transitions that drive optical absorption are thus present in a wider energy range [ 23 ] . Therefore, it needs to be shown whether the method can be applied to disordered materials. In the following, we demonstrate the feasibility of using this contactless, non-destructive, and indicator-free method to determine diffusion coefficients as a function of concentration to study the diffusion of hydrogen in metallic glasses. II Methods II.1 Sample design and growth The samples were produced using DC and AC magnetron sputtering. The design is shown in Fig. 1 . Each sample consisted of 80\, nm of V 0.8 Zr 0.2 , grown onto a 10\, mm \times 10\, mm \times 0.5\, mm double-side polished a-SiO 2 substrate. The film was covered with 6\, nm of palladium and the center 8\, mm \times 10\, mm were subsequently covered using a mask to allow a 50\, nm thick Al 2 O 3 layer to be grown. Vanadium, zirconium and Al 2 O 3 were grown in a chamber having base pressure of 1.5\times 10^{-10}\, mbar and argon pressure of 3.33\times 10^{-3}\, mbar, whereas the palladium layer was grown in an adjacent chamber with a base pressure of 5\times 10^{-8}\, mbar and argon pressure of 6.66\times 10^{-3}\, mbar. The power-dependent growth rate was calibrated for each target to determine the required sputtering times and power ratios for the combinatorial sputtering process. The sample thicknesses were determined using x-ray reflectometry and extracted by fitting a slab model to the reflectivity data using Gen-X [ 25 , 26 ] . The substrates were baked for 50 minutes at 350\,^{\circ} C, to reduce possible deposits of water and organic contaminants. During growth, the temperature was decreased to 27\,^{\circ} C to ensure maximally amorphous sample growth [ 27 , 28 ] . Sample composition and thickness for the V-Zr system have been verified previously using X-ray reflectometry, X-ray diffraction and Rutherford backscattering [ 23 , 19 ] , verifying the growth procedure to gain high-quality, X-ray amorphous samples. The palladium layer protects the underlying V 0.8 Zr 0.2 from oxidation and catalyzes the hydrogen dissociation without taking up hydrogen in the pressure and temperature range applied [ 29 ] . Al 2 O 3 is inert towards reactions with hydrogen [ 16 ] and, therefore, acts as a

Read the complete text