Source-dependent epoxide depletion in graphene oxide induced by soft X-rays and low-energy electron irradiation

Technical Summary: Source-Dependent Epoxide Depletion in Graphene Oxide Induced by Soft X-rays and Low-Energy Electron Irradiation

Problem Statement
Graphene oxide (GO) reduction is a critical process for tailoring material properties for electronics, sensing, and energy storage. While radiation-induced reduction offers a reagent-free, spatially controlled alternative to thermal or chemical methods, previous studies have largely relied on global indicators, such as the overall C/O atomic ratio, to assess chemical evolution. This approach overlooks the fact that different oxygen-containing functional groups (hydroxyl, epoxide, carbonyl, carboxyl) may possess distinct radiation sensitivities and reaction pathways. Furthermore, comparing different radiation sources (e.g., soft X-rays vs. low-energy electrons) based solely on exposure time is insufficient due to significant differences in dose rates, penetration depths, and energy-deposition profiles. There is a need to understand the dose-dependent evolution of specific functional groups, particularly epoxides, and to distinguish between real-time kinetics and dose-normalized chemical responses.

Methodology
The study investigated self-supported GO sheets (approx. 10 µm thick) exposed to two distinct radiation sources within an ultra-high-vacuum chamber:

1. Soft X-rays: Non-monochromatic Mg Kα radiation (1253.6 eV) with a maximum exposure of 35 hours.
2. Low-energy electrons: A 500 eV electron beam (0.17 µA) with a maximum exposure of 2 hours.

Chemical evolution was monitored using X-ray photoelectron spectroscopy (XPS) with a Mg Kα source, analyzing detailed C 1s and O 1s core-level spectra. The analysis focused on deconvoluting specific components: graphitic carbon (C–C/C=C), epoxide (C–O–C), hydroxyl, carbonyl, and carboxyl groups.

A critical methodological component was the estimation of the local absorbed dose within the XPS-sensitive near-surface volume (defined as the top 10 nm).

• For X-rays, dose was calculated using photon flux, energy, and soft X-ray attenuation data.
• For electrons, the Kanaya–Okayama (K–O) range relation and a K–O-derived continuous-slowing-down approximation (CSDA) were used to estimate the fraction of incident energy deposited within the first 10 nm.
• Dose rates were calculated as 5.15 MGy h⁻¹ for X-rays and 1150 MGy h⁻¹ for electrons.

Key Results

1. Selective Epoxide Depletion: Both radiation sources induced a chemically selective reduction. The most significant change was the monotonic depletion of epoxide (C–O–C) groups, accompanied by a relative increase in graphitic carbon (C–C/C=C) and hydroxyl-related components.
2. Stability of Other Groups: Carbonyl and carboxyl groups remained nearly unchanged across the investigated dose ranges, indicating they are less sensitive to these specific irradiation conditions than epoxides.
3. Structural Preservation: SEM and XRD analyses confirmed that the irradiation conditions induced chemical modifications without causing severe morphological degradation (cracking or delamination) or destroying the lamellar stacking structure, though a slight decrease in interlayer spacing was observed.
4. Kinetic Modeling: The dose-dependent evolution of the epoxide-to-graphitic carbon ratio followed a phenomenological first-order-like saturating exponential model: $Y_D = Y_\infty(1 - e^{-D/D_c})$.
   • X-rays: Characteristic dose ($D_c$) = $140 \pm 10$ MGy; Saturation yield ($Y_\infty$) = $74 \pm 3\%$.
   • Electrons: Characteristic dose ($D_c$) = $2310 \pm 235$ MGy; Saturation yield ($Y_\infty$) = $72 \pm 5\%$.
   • Note: For electron irradiation, the saturation regime was not experimentally reached within the maximum dose of 2300 MGy (reaching only ~62% of the fitted $Y_\infty$); thus, the electron parameters are model-dependent extrapolations.

Significance and Claims
The paper establishes that while both soft X-rays and low-energy electrons induce qualitatively similar, epoxide-selective reduction in GO, the dose-normalized response is strongly source-dependent.

• Dose Efficiency: Electrons produce apparent faster changes with exposure time due to a significantly higher dose rate. Under the maximum exposure conditions, the nominal local electron dose is approximately 13 times the Mg Kα dose within the adopted K–O-derived CSDA approximation. However, when normalized to the absorbed dose, electrons show significantly lower efficiency in depleting epoxide groups compared to X-rays (indicated by the much higher characteristic dose $D_c$).
• Methodological Insight: The study demonstrates that evaluating radiation-induced GO reduction requires distinguishing between exposure-time kinetics and absorbed-dose-normalized chemical responses. It highlights that epoxide groups serve as the most radiation-sensitive chemical marker for tracking GO reduction, while other oxygen functionalities remain relatively stable.

The authors conclude that the chemical response of GO to ionizing radiation is governed by the specific interaction of the radiation source with the material's functional groups and the local energy-deposition characteristics, rather than just the total exposure time.