🔬 materials science

Cumulative X-ray Damage in Bismuth Selenide Examined by Simultaneous TXM and XRD

This study utilizes simultaneous transmission X-ray microscopy and X-ray diffraction at the PAL-XFEL to characterize cumulative X-ray damage in bismuth selenide, revealing a multi-stage degradation process involving rapid vaporization, thermal cycling-induced grain refinement, and a feedback mechanism of accelerated damage driven by grain-boundary formation.

Original authors: Sophie E Parsons, Bernard Kozioziemski, Daewoong Nam, Eric Folsom, Can Yildirim, Sean Breckling, Sungwook Choi, Eric C. Galtier, Arnulfo Gonzalez, Deja Dominguez, Emlyn Frederick, Marylesa M. Howard

Published 2026-08-21

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Sophie E Parsons, Bernard Kozioziemski, Daewoong Nam, Eric Folsom, Can Yildirim, Sean Breckling, Sungwook Choi, Eric C. Galtier, Arnulfo Gonzalez, Deja Dominguez, Emlyn Frederick, Marylesa M. Howard, Sara Jessica Irvine, Kento Katagiri, Sangsoo Kim, Seonghan Kim, Sunam Kim, Stephan Kuschel, R. Stewart McWilliams, Norimasa Ozaki, Alison M. Saunders, Hyunjung Kim, Jon Eggert, Leora Dresselhaus-Marais

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

Some materials are like quiet neighborhoods where electricity flows only on the surface, while the deep interior remains a solid, insulating wall. This strange behavior defines a class of substances called topological insulators, and one of the most promising members is bismuth selenide. Scientists are eager to use this material for next-generation electronics and energy harvesting because its surface conducts electricity with almost no resistance, while its core blocks it completely. However, before these materials can be trusted in real-world devices, researchers must understand how they hold up under extreme stress. One of the most common stressors in high-tech environments is intense radiation, such as the powerful X-rays used to study materials at the atomic level. The question is simple but critical: if you blast a topological insulator with high-energy X-rays, does it simply melt away, or does its unique electronic personality change in a way that ruins its function?

To answer this, a team of researchers took a sample of bismuth selenide to the Pohang Accelerator Laboratory in South Korea, home to one of the world's most powerful X-ray lasers. They did not just fire a single shot and walk away; instead, they subjected the material to a relentless barrage of 27,000 X-ray pulses over the course of fifteen minutes. This was not a passive observation. The scientists set up a unique experiment where they could watch the material change in real time using two different cameras simultaneously. One camera, a transmission X-ray microscope, acted like a high-speed movie camera, capturing images of the sample as the X-rays passed through it. The other camera recorded X-ray diffraction patterns, which act as a fingerprint of the material's internal crystal structure. By watching both the shape of the material and its atomic arrangement at the same time, the team could see exactly how the damage unfolded, pulse by pulse.

The results revealed a two-stage process of destruction that happened much faster and more complexly than anyone had predicted. In the very beginning, within the first 100 pulses, the intense center of the X-ray beam acted like a microscopic drill. It heated the material so rapidly that it vaporized, punching a clean hole straight through the sample. This was the immediate, violent reaction to the energy. But the story did not end there. As the experiment continued for thousands more pulses, a slower, more insidious transformation began to take over the area surrounding the hole. The material that had not been vaporized did not just sit there; it began to break down into smaller and smaller pieces. The original large, perfect crystal structure, which spanned the entire sample, gradually shattered into a chaotic collection of tiny grains, shrinking from the scale of micrometers down to the scale of nanometers.

This shift from a single, unified crystal to a jumbled pile of microscopic grains is significant because it changes how the material behaves. The researchers found that as the grains got smaller, the boundaries between them multiplied. These boundaries act like roadblocks for the electrons that are supposed to flow smoothly along the surface. The team used computer simulations to show that the X-rays penetrated about 13.47 micrometers into the material, heating it to temperatures above 1,600 Kelvin before it cooled down again before the next pulse arrived. This repeated heating and cooling cycle, combined with the creation of millions of new grain boundaries, created a feedback loop. The more the material broke into smaller grains, the more efficiently it absorbed the heat from the X-rays, which in turn caused even more damage in the next pulse.

The study also clarified what this material is not doing. The researchers explicitly ruled out the idea that the material simply turned into a disordered, amorphous soup or that the damage was caused by a sudden electrical explosion. Instead, the evidence showed a clear path of thermal melting and re-solidification. The material melted, flowed, and then froze back into a new, disordered shape. By the end of the experiment, the once-perfect crystal had been transformed into a landscape of tiny, disordered crystals and sputtered streaks of material that looked like frost on a windowpane. The team confirmed this by examining the damaged sample under a scanning electron microscope, which revealed three distinct types of damage: long streaks where material had been blasted away, small prismatic crystals that had regrown, and clusters of chaotic micro-grains.

Ultimately, this work provides a new way to understand how delicate materials survive extreme conditions. The researchers demonstrated that the damage to bismuth selenide is not just about removing material; it is about fundamentally altering its internal architecture. The topological protection that makes the material special is not destroyed by a single blast of energy, but is slowly eroded by the accumulation of tiny defects and grain boundaries over time. This finding suggests that for topological insulators to be useful in future technologies, engineers must design systems that can withstand not just a single hit, but the cumulative effect of repeated thermal cycling. The study establishes a powerful new method for watching these processes happen, combining a microscope and a crystal analyzer to see both the big picture and the tiny details of destruction as they occur.

Technical Summary: Cumulative X-ray Damage in Bismuth Selenide Examined by Simultaneous TXM and XRD

Problem Statement
Bismuth selenide (Bi $_2$ Se $_3$ ) is a topological insulator with significant potential in thermoelectrics, spintronics, and optoelectronics due to its unique electronic structure: an insulating bulk with topologically protected, spin-polarized metallic surface states. While the material's static properties are well-characterized, its response to cumulative radiation damage, particularly under the extreme conditions of X-ray Free Electron Laser (XFEL) irradiation, remains poorly understood. The interaction between the material's dual nature (metallic surface vs. insulating bulk) and high-flux X-rays is complex; unlike pure insulators that may undergo electrostatic rupture or pure metals that undergo thermal ablation, Bi $_2$ Se $_3$ presents a hybrid scenario. Furthermore, existing diagnostic methods face a trade-off: high-resolution local studies often lack bulk context, while bulk measurements lose spatial resolution. There is a critical gap in simultaneously resolving microscale morphological changes and macroscale structural evolution during repeated radiation cycles.

Methodology
The study was conducted at the Pohang Accelerator Laboratory X-Ray Free Electron Laser (PAL-XFEL) using the Nano-crystallography and Coherent Imaging (NCI) beamline. The experimental setup uniquely combined Transmission X-ray Microscopy (TXM) and classical X-ray Diffraction (XRD) to capture complementary data simultaneously over 27,000 successive XFEL pulses.

Experimental Conditions: Samples of Bi $_2$ Se $_3$ with thicknesses of 40 µm and 63 µm were irradiated with 32-fs pulses at 9.7 keV, a repetition rate of 30 Hz, and a fluence of 0.5 mJ per pulse.
Imaging: TXM utilized compound refractive lenses (CRLs) to magnify the transmitted X-ray beam, providing real-time visualization of hole formation and debris evolution with a field of view of approximately 50 µm.
Diffraction: Wide-angle XRD was collected on an area detector to monitor long-range crystal structure, grain size, and strain.
Analysis: The authors employed Williamson-Hall (WH) analysis on the diffraction patterns to decouple crystallite size reduction from microstrain. Additionally, Scanning Electron Microscopy (SEM) was performed on recovered samples to correlate in-situ observations with post-mortem morphology.
Modeling: Finite-element modeling (COMSOL Multiphysics) was used to simulate heat transfer, phase transitions (solid-liquid-vapor), and thermal gradients, assuming a flat-top beam profile and Beer-Lambert attenuation.

Key Results
The study identified distinct, sequential damage mechanisms occurring over different timescales:

Rapid Hole Formation (0–100 pulses): Within the first 100 pulses, the high-intensity core of the X-ray beam caused rapid vaporization, drilling a hole through the sample. Thermal modeling predicted that local temperatures exceed 1600 K, driving vaporization of approximately 16–20 µm; however, experimental transmission data indicated a slightly lower removal rate of ~12 µm for the same flux.
Progressive Grain Refinement (100–27,000 pulses): Following the initial hole drilling, a slower damage pathway dominated the surrounding material. Williamson-Hall analysis revealed a continuous transformation from a single-crystal structure to a nanocrystalline state. Grain sizes decreased from the micron scale to sub-10 nm over thousands of thermal cycles.
Morphological Evolution: SEM analysis of recovered samples identified three distinct morphologies corresponding to different thermal histories: sputter streaks (amorphous material), prismatic crystals, and disordered microcrystals. The "sputter streaks" observed in TXM were linked to material melting (reaching ~710 °C) and ejection.
Feedback Mechanism: The transition to a polycrystalline, nanocrystalline structure increased grain boundary density. The authors posit that these grain boundaries enhance electron-phonon coupling and scatter carriers, effectively lowering the local thermal damage threshold. This creates a feedback loop where subsequent pulses cause disproportionately more damage than previous ones.
Thermal Cycling: Thermal modeling and experimental data confirmed that the sample cools back to near room temperature between pulses (33 ms interval), allowing for re-solidification. The damage is cumulative, driven by repeated melting and recrystallization cycles rather than a single catastrophic event.

Significance and Claims
The paper claims to establish a methodology for studying radiation damage across multiple length scales by simultaneously utilizing TXM and XRD at XFEL repetition rates. This approach resolves a diagnostic gap where previous studies could either quantify bulk dynamics or resolve microscale effects, but not both simultaneously.

Key contributions include:

Mechanism Elucidation: Demonstrating that damage in topological insulators under XFEL irradiation is not solely a surface phenomenon but a bulk process involving the progressive degradation of crystallinity into a nanocrystalline, grain-boundary-dominated microstructure.
Impact on Topological Properties: The results suggest that the topological protection of surface states is compromised not by direct ionization (which might favor Coulomb explosion) but by the accumulation of grain-boundary disorder from repeated thermal cycling. This disorder is expected to degrade coherent, high-mobility topological transport.
Predictive Modeling: The agreement between COMSOL thermal predictions, TXM-derived hole-drilling rates, and WH grain-size evolution validates thermal modeling as a tool for predicting XFEL sample damage. This enables the optimization of dose, spot size, and repetition rate for future in-situ studies.
General Applicability: The authors state that this simultaneous TXM/XRD methodology is applicable to a wide range of materials science challenges, including radiation-hard materials for fusion/fission reactors, pulsed-laser deposition, shock-induced phase transitions, and battery electrode degradation.

The paper concludes that while the local band inversion within individual grains may persist, the global material properties are governed by defect scattering and boundary-limited conduction, fundamentally altering the material's utility under extreme radiation conditions.

Technical Summary: Cumulative X-ray Damage in Bismuth Selenide Examined by Simultaneous TXM and XRD

More like this