🔬 materials science

A depth resolved investigation of hydrogen uptake in carbon based nanostructures by soft-to-hard photoemission spectroscopy

Using soft-to-hard X-ray photoemission spectroscopy, this study reveals that hydrogen chemisorption on nanoporous graphene and vertically aligned carbon nanotubes is predominantly localized on the surface, forming a thin sp3-rich overlayer with limited penetration into the bulk.

Original authors: Orlando Castellano (Dipartimento di Scienze, Università degli Studi di Roma Tre, Rome, Italy, INFN Sezione di Roma Tre, Rome, Italy), Alice Apponi (INFN Sezione di Roma Tre, Rome, Italy), Luca Cecchin

Published 2026-07-03

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Orlando Castellano (Dipartimento di Scienze, Università degli Studi di Roma Tre, Rome, Italy, INFN Sezione di Roma Tre, Rome, Italy), Alice Apponi (INFN Sezione di Roma Tre, Rome, Italy), Luca Cecchini (INFN Sezione di Roma, Rome, Italy), Daniele Paoloni (Dipartimento di Scienze, Università degli Studi di Roma Tre, Rome, Italy), Simone Ritarossi (Dipartimento di Scienze, Università degli Studi di Roma Tre, Rome, Italy, INFN Sezione di Roma Tre, Rome, Italy), Francesco Pandolfi (INFN Sezione di Roma, Rome, Italy), Ilaria Rago (INFN Sezione di Roma, Rome, Italy), Tien-Lin Lee (Diamond Light Source Ltd, Didcot, United Kingdom), Samuel Jeong (Department of Applied Physics, Institute of Pure and Applied Sciences, University of Tsukuba, Japan), Yoshikazu Ito (Department of Applied Physics, Institute of Pure and Applied Sciences, University of Tsukuba, Japan, Tsukuba Institute for Advanced Research), Carlo Mariani (INFN Sezione di Roma, Rome, Italy, Sapienza Università di Roma, Rome, Italy), Gianluca Cavoto (INFN Sezione di Roma, Rome, Italy, Sapienza Università di Roma, Rome, Italy), Francesco Offi (Dipartimento di Scienze, Università degli Studi di Roma Tre, Rome, Italy, INFN Sezione di Roma Tre, Rome, Italy), Alessandro Ruocco (Dipartimento di Scienze, Università degli Studi di Roma Tre, Rome, Italy, INFN Sezione di Roma Tre, Rome, Italy)

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine carbon materials like graphene and carbon nanotubes as intricate, three-dimensional cities made entirely of carbon atoms. Normally, these atoms are flat and tightly packed, like a sheet of graph paper (scientists call this sp² bonding). The researchers in this paper wanted to see what happens when you introduce hydrogen atoms to these cities. When hydrogen attaches to a carbon atom, it forces that atom to pop up out of the flat sheet, changing its shape to a pyramid-like structure (called sp³ bonding). This change is crucial because it alters how the material conducts electricity and how it might store energy.

The big question the team wanted to answer was: Does the hydrogen just stick to the "skin" of these materials, or does it soak all the way through to the "insides"?

To find out, they used a special kind of X-ray camera called photoemission spectroscopy. Think of this camera like a flashlight with different beam settings:

Soft X-rays are like a dim, close-up flashlight. They only see the very top surface of the material (the "skin").
Hard X-rays are like a powerful, penetrating beam. They can see deep into the material, all the way to the "basement" layers.

By shining these different "flashlights" on their samples, the researchers could take a "depth-resolved" look, essentially peeling back the layers to see where the hydrogen had actually gone.

The Two Cities They Studied

Nanoporous Graphene (NPG): Imagine a sponge made of graphene. It has a continuous, web-like structure with tiny holes everywhere.
Vertically Aligned Carbon Nanotubes (CNTs): Imagine a forest of tiny, hollow straws standing straight up.

What They Found

1. The "Skin" Effect
When they looked at the samples with the "soft" (surface-only) flashlight, they saw a lot of hydrogen. The carbon atoms on the surface had changed their shape (from flat to pyramid) to hold onto the hydrogen. However, when they switched to the "hard" (deep-penetrating) flashlight, the story changed.

In the Graphene Sponge (NPG): The hydrogen was almost entirely stuck to the very outer surface. The "deep" view showed that the inside of the sponge remained mostly untouched. It's as if the hydrogen tried to enter the sponge but got stuck at the entrance, unable to diffuse through the complex, winding internal tunnels. The hydrogenated layer was incredibly thin—less than the thickness of a single sheet of paper (graphene layer).
In the Nanotube Forest (CNTs): The hydrogen did penetrate a bit deeper than in the sponge, but it still didn't go all the way through. It managed to coat the outer walls of the "straws" and maybe squeeze into the tiny gaps between the layers of the straw walls. However, the very inner core of the tubes remained largely pristine. The hydrogenated layer here was about the thickness of one graphene sheet.

2. The "Recoil" Confusion
The researchers also had to deal with a tricky physics quirk. When they used the powerful "hard" X-rays, the energy was so high that it actually gave the carbon atoms a tiny "kick" (recoil), which made the signals look slightly different. They had to use a mathematical correction to make sure they were seeing the hydrogen effect and not just the X-ray kick. Once they fixed this, the picture became clear: the hydrogen really was just staying near the surface.

The Bottom Line

The study concludes that for these specific 3D carbon structures, hydrogen is a "surface dweller."

Even though these materials are designed to have a lot of internal surface area (like a sponge or a forest), the hydrogen atoms generally cannot penetrate deep into the internal volume. They prefer to stick to the outermost accessible layers.

For the Graphene Sponge: The hydrogen stays on the very outside, like a thin coat of paint that doesn't soak into the sponge.
For the Nanotube Forest: The hydrogen coats the outside of the tubes and maybe the immediate gaps between them, but it doesn't reach the deep center of the tubes.

This finding is important because it tells scientists that if they want to use these materials to store hydrogen, they can't just assume the hydrogen will fill the whole 3D volume. Instead, the storage capacity is limited to the "skin" of the material. The paper suggests that while the hydrogen doesn't go deep, the outer surfaces of these materials are still very good at grabbing onto hydrogen, which could be useful for specific types of storage or sensing applications where surface interaction is key.

Technical Summary: Depth-Resolved Investigation of Hydrogen Uptake in Carbon Nanostructures

Problem Statement
Hydrogen chemisorption on graphitic carbon transforms the orbital hybridization of carbon atoms from $sp^2$ to $sp^3$ , fundamentally altering structural and electronic properties. While hydrogenated graphene is a well-studied model system, the behavior of hydrogen uptake in three-dimensional (3D) carbon architectures remains an open question. Specifically, it is unclear whether hydrogen chemisorption is confined to the outermost accessible surfaces or if it penetrates into the internal layers of complex nanostructures. Resolving this spatial distribution is critical for both fundamental understanding of adsorption mechanisms and practical applications, such as hydrogen storage and the development of solid-state tritium targets for neutrino mass measurements (e.g., the PTOLEMY project) or dark matter detectors.

Methodology
The authors investigated two representative 3D carbon nanostructures: nanoporous graphene (NPG) and vertically aligned multi-walled carbon nanotubes (CNTs).

Sample Preparation: NPG was synthesized via chemical vapor deposition (CVD) on a nanoporous nickel template and subsequently hydrogenated in situ using an atomic hydrogen source. Vertically aligned CNTs were grown via CVD and hydrogenated ex situ using a hydrogen plasma.
Spectroscopic Technique: The study employed soft-to-hard X-ray photoemission spectroscopy (XPS) to achieve depth-resolved analysis. Measurements were conducted at photon energies of 0.8, 1.4, 4.1, and 8.1 keV. This range allows for varying the probing depth by exploiting the energy dependence of the inelastic mean free path (IMFP) of photoelectrons.
Data Analysis: The C 1s core-level spectra were analyzed using a global fitting procedure to deconvolve contributions from distinct chemical environments: $sp^2$ (graphitic), $sp^3$ (hydrogenated/defective), carbon oxides, defects, and $\pi$ -plasmon losses.
Recoil Correction: To account for systematic energy shifts and broadening at high photon energies, the authors incorporated a semiclassical treatment of photoemission recoil effects, allowing for a photon-energy-dependent rigid shift of the $sp^2$ binding energy and additional Gaussian broadening.
Modeling: A modified thin-overlayer attenuation model was applied to quantify the depth distribution of hydrogenation, treating the $sp^3$ fraction as a sum of a homogeneous bulk contribution and a surface-confined layer of finite thickness.

Key Results

Spectral Evolution: In both NPG and CNTs, hydrogenation induced a shift in the C 1s lineshape toward higher binding energies and an increase in linewidth, indicative of $sp^3$ $s p^{3}$ formation.
- NPG: At low photon energy (0.8 keV, high surface sensitivity), the hydrogenated NPG (HNPG) spectrum showed a pronounced high-binding-energy tail compared to the pristine sample. As photon energy increased (probing deeper), the spectral differences diminished, and the HNPG and pristine spectra became nearly indistinguishable at 8.1 keV.
- CNTs: Hydrogenated CNTs (HCNTs) showed clear spectral differences from pristine CNTs across all energies, including increased linewidth and a shift toward higher binding energy. However, the magnitude of these differences decreased with increasing photon energy, though a measurable difference persisted even at 8.1 keV.
Quantitative Depth Profiling:
- NPG: The effective hydrogenated overlayer thickness was determined to be $d = 0.13 \pm 0.06$ nm, corresponding to a fraction of a single graphene layer. The homogeneous $sp^3$ contribution remained constant with depth, suggesting hydrogenation is strictly confined to the outermost accessible surface with no significant penetration into the internal porous network.
- CNTs: The effective overlayer thickness was $d = 0.4 \pm 0.1$ nm, approximately one graphene layer. While the $sp^3$ fraction in HCNTs was significantly higher than in pristine CNTs at all depths, the results indicate that hydrogenation is predominantly localized on the outermost walls of the multi-walled tubes, with limited penetration into inner walls or along the tube axis.
Recoil Effects: The study confirmed that the systematic increase in C 1s binding energy with photon energy in pristine samples is dominated by photoemission recoil effects, consistent with free-atom recoil estimates ( $m/M$ ratio).

Significance and Claims
The paper claims to provide a quantitative, depth-resolved clarification of hydrogen distribution in curved and porous graphitic networks. The primary finding is that hydrogen chemisorption in these 3D architectures is predominantly surface-localized.

For NPG, the lack of deep penetration suggests that the bicontinuous monolithic architecture hinders the diffusion of reactive hydrogen species to internal surfaces.
For multi-walled CNTs, the results highlight a distinct difference from single-walled systems; the inner walls are significantly less accessible to hydrogen under plasma exposure, leading to a non-uniform depth profile.

The authors conclude that while these materials offer high surface areas for chemical functionalization, the effective uptake for bulk hydrogen storage is limited by the shallow penetration depth of hydrogen. However, the distinct advantages of each architecture are noted: NPG preserves electrical conductivity while enabling surface functionalization, whereas vertically aligned CNT bundles offer a high density of chemically active sites on their accessible outer surfaces. These findings provide necessary constraints for modeling hydrogen storage capacity and interpreting signals in applications involving hydrogenated carbon targets.

The Two Cities They Studied

What They Found

The Bottom Line

More like this