Transmission Electron Microscopy-electron Energy Loss Research Articles

Degradation of thermoplastic polymers for fused filament fabrication under (S)TEM electron beam irradiation

The structural characterization of polymers and, in particular, of those used in Additive Manufacturing (AM) technologies, is essential to improve the understanding of their structure-property relationship for promising high-performance applications. For this, (scanning) transmission electron microscopy-electron energy loss spectroscopy, (S)TEM-EELS is an outstanding tool for exploring materials chemical and structural characteristics at high spatial resolution. However, the high beam-sensitivity of soft materials, such as polymers, hinders the possibility of probing in-depth analysis provided by (S)TEM-EELS. In this work, we analyse the electron beam irradiation damage of four polymers commonly used in Fused Filament Fabrication (FFF), namely polylactic acid (PLA), polycaprolactone (PCL), acrylonitrile butadiene styrene (ABS) and acrylonitrile styrene acrylate (ASA). For this, sequential low-loss and core-loss EEL spectra have been recorded, and the related signals have been monitored as a function of the accumulated dose. Our results show that the critical electron doses using the specimen thickness variations are larger for polymers containing aromatic groups (ABS and ASA) than for aliphatic polymers (PLA and PCL). Regarding the different elements, a larger sensitivity to the electron beam of oxygen regarding carbon and nitrogen is also evidenced. Our results have shown that polymer degradation occurs to a larger extent in the initial steps of electron irradiation, for very low accumulated electron doses, meaning that care should be taken in the selection of the microscopy settings to avoid artefacts produced by the electron beam. Degradation pathways for the four polymers studied are discussed.

Effects of irradiation on interfacial strength and microstructure of double-layer mullite and alumina coating on SiC

Silicon carbide (SiC) ceramics hold great potential for use in nuclear-reactor components due to characteristics such as high-temperature strength and low activation. Despite their resistance against corrosion in harsh environments, they suffer elevated corrosion rates under particle irradiation. Thin anti-corrosion coatings, such as mullite bond layers and alumina top layers, are essential to enhance the irradiation stability of SiC. The objective of this study was to evaluate the irradiation stability of a double-layer coating developed to enhance the corrosion resistance of SiC in nuclear applications. The coating, which comprised a mullite bond layer and an alumina top layer, was applied to a SiC substrate using laser chemical vapor deposition. Irradiation experiments were conducted at 300 °C with 5.1-MeV Si ions up to 10 displacements per atom. To assess the interfacial strength, a novel testing method, called the double-notch shear compression testing method, was developed based on ASTM standards and was implemented using a nanoindenter. This approach enabled precise measurements of the mechanical integrity at the interfaces of the coating under irradiation. The results showed an increase in the interfacial strength at the SiC/mullite and alumina/mullite interfaces with irradiation. Microstructural analysis of the fracture surface through scanning electron microscopy–energy-dispersive X-ray spectroscopy revealed that cracks propagated within the mullite layer, indicating the presence of mullite on the fracture surface. Transmission electron microscopy (TEM) images indicated that a 2-µm-thick transition layer existed at the SiC/mullite interface but not at the alumina/mullite interface. The TEM–electron energy loss spectroscopy suggested that the Al-O bonding structures in the transition layer were changed from tetrahedral (AlO4) to octahedral (AlO6) through irradiation, and this structural transition may directly affect the strength.

Correlating the electronic structures of β-Ga2O3 to its crystal tilts induced defects at nanoscale

Crosslinking structural transformation mechanism at nanoscale to the corresponding anisotropically electronic properties is essential to both the atomic-level controlled synthesis and extreme-conditional applications of the ultrawide bandgap semiconductors. Here, we report the first direct observation of bandgap variation at certain microstructural flaws in monoclinic crystal β-Ga2O3 via scanning transmission electron microscopy-electron energy loss spectrum (STEM-EELS) technology with both high energy and spatial resolution. Atomic-scale tilt of the Mosaic blocks relative to [010] zone axis is demonstrated to cause specific point defects accumulation at block boundaries, resulting in the formation of vacancy lines (or clusters) and then the crystal deformation induced flaws. These as-formed tiny Mosaic tilts observed from both in-plane and out-of-plane geometry are correlated to the corresponding electronic structure obtained by density functional calculations, indicating the carrier mobility limits transferred from intrinsic polar optical phonon scattering to the ionized oxygen atoms scattering in the defective monoclinic crystal. These findings provide a new insight on these anisotropic defect formation induced electronic structural variation, paving the way for precise synthesis and development of high-performance ultra-wide bandgap materials.

Magic angle and STEM-EELS mapping of the sp2/sp3 hybridization in heterogeneous carbonaceous materials

Extensive research has been conducted on carbonaceous materials due to their unique combination of physical, chemical, and mechanical properties, which significantly rely on the hybridization of carbon atoms. Scanning transmission electron microscopy-electron energy loss spectroscopy is a powerful technique that enables the identification of carbon allotropes with high spatial resolution, utilizing specific spectral features. However, anisotropic materials like graphite and carbon nanotubes can exhibit variations in these spectral features based on their orientation relative to the electron beam. Optimized experimental conditions, referred to as magic angle conditions, permits overcoming this challenge. By implementing such conditions, we have successfully mapped the hybridization in a heterogeneous system containing three carbon allotropes, i.e. nanodiamonds, multi-walled carbon nanotubes, and lacey carbon. Moreover, a convolutional neural network has been created and trained to accurately identify and map these carbonaceous phases. Thus, the reported innovative approach allows nanoscale mapping of both hybridization and phase distributions for complex heterogeneous carbon systems.

In Situ Monitoring of the Polymerization Kinetics of Organic Pollutants during Persulfate-Based Advanced Oxidation Processes Using Plasmonic Colorimetry.

Using sulfate radicals to initiate polymer production in persulfate-based advanced oxidation processes (AOPs) is an emerging strategy for organics removal. However, our understanding of this process remains limited due to a dearth of efficient methods for in situ and real time monitoring of polymerization kinetics. This study leverages plasmonic colorimetry to monitor the polymerization kinetics of an array of aromatic pollutants in the presence of sulfate radicals. We observed that the formation of polymer shells on the surfaces of gold nanoparticles (AuNPs) led to an increase and red shift in their localized surface plasmon resonance (LSPR) band as a result of an increased refractive index surrounding the AuNP surfaces. This observation aligns with Mie theory simulations and transmission electron microscopy-electron energy loss spectroscopy characterizations. Our study demonstrated that the polymerization kinetics exhibits a significant reliance on the electrophilicity and quantity of benzene rings, the concentration of aromatic pollutants, and the dosage of oxidants. In addition, we found that changes in LSPR band wavelength fit well into a pseudo-first-order kinetic model, providing a comprehensive and quantitative insight into the polymerization kinetics involving diverse organic compounds. This technique holds the potential for optimizing AOP-based water treatment by facilitating the polymerization of aromatic pollutants.

Visualization of the Active Sites of Zinc-Chromium Oxides and the CO/H2 Activation Mechanism in Direct Syngas Conversion.

Despite wide studies demonstrating the versatility of the metal oxide-zeolite (OXZEO) catalyst concept to tackle the selectivity challenge in syngas chemistry, the active sites of metal oxides and the mechanism of CO/H2 activation remain to be elucidated. Herein, we demonstrate experimentally the role of Cr in zinc-chromium oxides and unveil visually, for the first time, the active sites for CO activation employing scanning transmission electron microscopy-electron energy loss spectroscopy using the volumetric density of surface carbon species as a descriptor. The ZnCr2O4 spinel surface with atomic ZnOx overlayer is the most active site for C-O bond dissociation, particularly at the narrow ZnCr2O4(110) facets constrained between the (311) and (111) facets, followed by the Cr-doped wurtzite ZnO surface. In comparison, the surfaces of ZnCr2O4 with aggregated ZnOx overlayers, pure ZnO, and the stoichiometric ZnCr2O4 exhibit a significantly lower activity. In situ synchrotron-based vacuum ultraviolet photoionization mass spectrometric study on different temperature programmed surface reactions with isotopes of C18O, 13CO, and D2 validates direct CO dissociation over ZnCrn oxides in CO, forming CH2 and further to hydrocarbons if H2 is present and CH2CO intermediates in syngas. The activity of CO dissociation and hydrogenation over ZnCrn oxides correlates well with the syngas-to-light-olefins activity of ZnCrn-SAPO-18 composite catalysts as a function of the Cr/Zn ratio.

Charge Distribution across Capped and Uncapped Infinite-Layer Neodymium Nickelate Thin Films.

Charge ordering (CO) phenomena have been widely debated in strongly-correlated electron systems mainly regarding their role in high-temperature superconductivity. Here, the structural and charge distribution in NdNiO2 thin films prepared with and without capping layers, and characterized by the absence and presence of CO are elucidated. The microstructural and spectroscopic analysis is done by scanning transmission electron microscopy-electron energy loss spectroscopy (STEM-EELS) and hard X-ray photoemission spectroscopy (HAXPES). Capped samples show Ni1+ , with an out-of-plane (o-o-p) lattice parameter of around 3.30 Å indicating good stabilization of the infinite-layer structure. Bulk-sensitive HAXPES on Ni-2p shows weak satellite features indicating large charge-transfer energy. The uncapped samples evidence an increase of the o-o-p parameter up to 3.65 Å on the thin film top with a valence toward Ni2+ in this region. Here, 4D-STEM demonstrates (303)-oriented stripes which emerge from partially occupied apical oxygen. Those stripes form quasi-2D coherent domains viewed as rods in the reciprocal space with Δqz ≈ 0.24 reciprocal lattice units (r.l.u.) extension located at Q = ( ) and ( ) r.l.u. The stripes associated with oxygen re-intercalation concomitant with hole doping suggest a possible link to the previously reported CO in infinite-layer nickelate thin films.

Effect of Spatial Heterogeneity on the Unusual Uptake Behavior of Multivariate-Metal-Organic Frameworks.

The uniqueness of multivariate metal-organic frameworks (MTV-MOFs) has been widely explored to discover their unknown opportunities. While mesoscopic apportionments have been studied, macroscopic heterogeneity and its spatial effects remain unexplored in MTV-MOFs. In this study, we investigated the effect of macroscopic heterogeneity on MTV-MOFs on their uptake behaviors by comparing three types of MTV-MOFs having the functional groups in inner, outer, or entire parts of crystals. Their adsorption behavior for carbon dioxide (CO2) and water (H2O) brought out that functional groups located in the outer part of the crystals dominantly influence the sorption behavior of MTV-MOFs. These results are also visualized by observing iodine adsorption in the three types of MTV-MOFs using scanning transmission electron microscopy-electron energy loss spectroscopy. We believe that this finding provides new ways to decipher and design MTV-MOFs for their unusual properties.

Machine Learning-Aided Design of Gold Core-Shell Nanocatalysts toward Enhanced and Selective Photooxygenation.

We demonstrate the use of the machine learning (ML) tools to rapidly and accurately predict the electric field as a guide for designing core-shell Au-silica nanoparticles to enhance 1O2 sensitization and selectivity of organic synthesis. Based on the feature importance analysis, obtained from a deep neural network algorithm, we found a general and linear dependent descriptor (θ ∝ aD0.25t-1, where a, D, and t are the shape constant, size of metal nanoparticles, and distance from the metal surface) for the electric field around the core-shell plasmonic nanoparticle. Directed by the new descriptor, we synthesized gold-silica nanoparticles and validated their plasmonic intensity using scanning transmission electron microscopy-electron energy loss spectroscopy (STEM-EELS) mapping. The nanoparticles with θ = 0.40 demonstrate an ∼3-fold increase in the reaction rate of photooxygenation of anthracene and 4% increase in the selectivity of photooxygenation of dihydroartemisinic acid (DHAA), a long-standing goal in organic synthesis. In addition, the combination of ML and experimental investigations shows the synergetic effect of plasmonic enhancement and fluorescence quenching, leading to enhancement for 1O2 generation. Our results from time-dependent density functional theory (TD-DFT) calculations suggest that the presence of an electric field can favor intersystem crossing (ISC) of methylene blue to enhance 1O2 generation. The strategy reported here provides a data-driven catalyst preparation method that can significantly reduce experimental cost while paving the way for designing photocatalysts for organic drug synthesis.

Atomistic Simulation Informs Interface Engineering of Nanoscale LiCoO2.

Lithium-ion batteries continue to be a critical part of the search for enhanced energy storage solutions. Understanding the stability of interfaces (surfaces and grain boundaries) is one of the most crucial aspects of cathode design to improve the capacity and cyclability of batteries. Interfacial engineering through chemical modification offers the opportunity to create metastable states in the cathodes to inhibit common degradation mechanisms. Here, we demonstrate how atomistic simulations can effectively evaluate dopant interfacial segregation trends and be an effective predictive tool for cathode design despite the intrinsic approximations. We computationally studied two surfaces, {001} and {104}, and grain boundaries, Σ3 and Σ5, of LiCoO2 to investigate the segregation potential and stabilization effect of dopants. Isovalent and aliovalent dopants (Mg2+, Ca2+, Sr2+, Sc3+, Y3+, Gd3+, La3+, Al3+, Ti4+, Sn4+, Zr4+, V5+) were studied by replacing the Co3+ sites in all four of the constructed interfaces. The segregation energies of the dopants increased with the ionic radius of the dopant. They exhibited a linear dependence on the ionic size for divalent, trivalent, and quadrivalent dopants for surfaces and grain boundaries. The magnitude of the segregation potential also depended on the surface chemistry and grain boundary structure, showing higher segregation energies for the Σ5 grain boundary compared with the lower energy Σ3 boundary and higher for the {104} surface compared to the {001}. Lanthanum-doped nanoparticles were synthesized and imaged with scanning transmission electron microscopy-electron energy loss spectroscopy (STEM-EELS) to validate the computational results, revealing the predicted lanthanum enrichment at grain boundaries and both the {001} and the {104} surfaces.

Electron Microscopy Probing Electron-Photon Interactions in SiC Nanowires with Ultrawide Energy and Momentum Match.

Light-matter interactions are commonly probed by optical spectroscopy, which, however, has some fundamental limitations such as diffraction-limited spatial resolution, tiny momentum transfer, and noncontinuous excitation/detection. In this work, through the use of scanning transmission electron microscopy-electron energy loss spectroscopy (STEM-EELS) with ultrawide energy and momentum match and subnanometer spatial resolution, the longitudinal Fabry-Perot (FP) resonating modes and the transverse whispering-gallery modes (WGMs) in individual SiC nanowires are simultaneously excited and detected, which span from near-infrared (∼1.2 μm) to ultraviolet (∼0.2 μm) spectral regime, and the momentum transfer can range up to 108 cm-1. The size effects on the resonant spectra of nanowires are also revealed. This work provides an alternative technique to optical resonating spectroscopy and light-matter interactions in dielectric nanostructures, which is promising for modulating free electrons via photonic structures.

Imaging of isotope diffusion using atomic-scale vibrational spectroscopy.

The spatial resolutions of even the most sensitive isotope analysis techniques based on light or ion probes are limited to a few hundred nanometres. Although vibrational spectroscopy using electron probes has achieved higher spatial resolution1-3, the detection of isotopes at the atomic level4 has been challenging so far. Here we show the unambiguous isotopic imaging of 12C carbon atoms embedded in 13C graphene and the monitoring of their self-diffusion via atomic-level vibrational spectroscopy. We first grow a domain of 12C carbon atoms in a pre-existing crack of 13C graphene, which is then annealed at 600 degrees Celsius for several hours. Using scanning transmission electron microscopy-electron energy loss spectroscopy, we obtain an isotope map that confirms the segregation of 12C atoms that diffused rapidly. The map also indicates that the graphene layer becomes isotopically homogeneous over 100-nanometre regions after 2 hours. Our results demonstrate the high mobility of carbon atoms during growth and annealing via self-diffusion. This imaging technique can provide a fundamental methodology for nanoisotope engineering and monitoring, which will aid in the creation of isotope labels and tracing at the nanoscale.

Fabrication of chemically and structurally abrupt Eu1−xLaxO/SrO/Si interfaces and their analysis by STEM-EELS

Precise control of molecular-beam epitaxy deposition conditions has enabled the fabrication of chemically and structurally abrupt ${\mathrm{Eu}}_{1\ensuremath{-}x}{\mathrm{La}}_{x}\mathrm{O}/\mathrm{SrO}/\mathrm{Si}$ interfaces by interposing as little as half a monolayer of SrO between the La-doped EuO and silicon. The interfaces were analyzed by scanning transmission electron microscopy-electron energy loss spectroscopy; no reaction products were detected. These abrupt interfaces might enable direct ohmic injection of spin-polarized currents from the doped half-metallic semiconductor EuO into silicon.

Phase analysis of hydrogen contained Zr-2.5wt%Nb CANDU pressure tube using plasmon energy loss analysis

Canada Deuterium Uranium (CANDU) reactors use Zr-2.5Nb alloys as the pressure tube material. Ensuring optimal performance of Zr-2.5Nb pressure tubes is vital to the safety of CANDU-type heavy water reactors. Long-term use of the pressure tube induces hydrogen pickup and hydride precipitation, causing delayed hydride cracking. Therefore, to evaluate the mechanical integrity of the pressure tube, it is critical to determine the phase distributions inside the pressure tube during hydrogen charging. This study focuses on visually expressing the hydride phasesusingscanning transmission electron microscopy-electron energy loss spectroscopy (STEM-EELS).By using EELS, two phase characterization possibilities were noted at the phase boundary between α-Zr and δ-hydride: nucleated β-Zr and γ-hydride and a mechanical mixture of α-Zr and δ-hydride. The first characterization possibility was reported previously by Barrowet al., and the second possibility is described comprehensively in this paper.

NiCo2S4 Bi-metal Sulfide Coating on LiNi0.6Co0.2Mn0.2O2 Cathode for High-Performance All-Solid-State Lithium Batteries.

NiCo2S4 nanoparticles (NPs) were dry coated on LiNi0.6Co0.2Mn0.2O2 (NCM622) cathode using a resonant acoustic coating technique to produce all-solid-state lithium batteries. The NiCo2S4 coating improved the electrochemical properties of the NCM622 cathode. In addition, NiCo2S4 eliminated the space-charge layer and the cathode showed an excellent affinity with the interface with a sulfide-based solid electrolyte as an inert material. X-ray diffraction patterns of NCM622 coated with NiCo2S4 showed the same peak separations and lattice parameters as those of bare NCM622. Field-emission scanning electron microscopy and electron dispersive spectroscopy mapping analyses showed that 0.3 wt% NiCo2S4-coated NCM622 had an evenly modified surface with NiCo2S4 NPs. X-ray photoelectron spectroscopy (XPS) revealed that the surface of 0.3 wt% NiCo2S4-coated NCM622 had two different S 2p peaks, a Co–S peak, and Ni and Co peaks, compared to those of bare NCM622. Electrochemical studies with electrochemical impedance spectroscopy and galvanostatic charge–discharge cycle performances showed that NiCo2S4-coated NCM622 retained a higher specific capacity over multiple cycles than bare NCM622. Especially, 0.3 wt% NiCo2S4-coated NCM622 exhibited a capacity retention of 60.6% at a current density of 15 mA/g for 20 cycles, compared to only 37.3% for bare NCM622. Finally, interfacial XPS and transmission electron microscopy-electron energy loss spectroscopy analyses confirmed the stable state of 0.3 wt% NiCo2S4-coated NCM622 with minimal side reactions.

Plasmonic evolution of atomically size-selected Au clusters by electron energy loss spectrum.

The plasmonic response of gold clusters with atom number (N) = 100–70 000 was investigated using scanning transmission electron microscopy-electron energy loss spectroscopy. For decreasing N, the bulk plasmon remains unchanged above N = 887 but then disappears, while the surface plasmon firstly redshifts from 2.4 to 2.3 eV above N = 887 before blueshifting towards 2.6 eV down to N = 300, and finally splitting into three fine features. The surface plasmon's excitation ratio is found to follow N0.669, which is essentially R2. An atomically precise evolution picture of plasmon physics is thus demonstrated according to three regimes: classical plasmon (N = 887–70 000), quantum confinement corrected plasmon (N = 300–887) and molecule related plasmon (N < 300).

Tents, Chairs, Tacos, Kites, and Rods: Shapes and Plasmonic Properties of Singly Twinned Magnesium Nanoparticles.

Nanostructures of some metals can sustain light-driven electron oscillations called localized surface plasmon resonances, or LSPRs, that give rise to absorption, scattering, and local electric field enhancement. Their resonant frequency is dictated by the nanoparticle (NP) shape and size, fueling much research geared toward discovery and control of new structures. LSPR properties also depend on composition; traditional, rare, and expensive noble metals (Ag, Au) are increasingly eclipsed by earth-abundant alternatives, with Mg being an exciting candidate capable of sustaining resonances across the ultraviolet, visible, and near-infrared spectral ranges. Here, we report numerical predictions and experimental verifications of a set of shapes based on Mg NPs displaying various twinning patterns including (101̅1), (101̅2), (101̅3), and (112̅1), that create tent-, chair-, taco-, and kite-shaped NPs, respectively. These are strikingly different from what is obtained for typical plasmonic metals because Mg crystallizes in a hexagonal close packed structure, as opposed to the cubic Al, Cu, Ag, and Au. A numerical survey of the optical response of the various structures, as well as the effect of size and aspect ratio, reveals their rich array of resonances, which are supported by single-particle optical scattering experiments. Further, corresponding numerical and experimental studies of the near-field plasmon distribution via scanning transmission electron microscopy electron-energy loss spectroscopy unravels a mode nature and distribution that are unlike those of either hexagonal plates or cylindrical rods. These NPs, made from earth-abundant Mg, provide interesting ways to control light at the nanoscale across the ultraviolet, visible, and near-infrared spectral ranges.

Manipulation of surface phonon polaritons in SiC nanorods

Surface phonon polaritons (SPhPs) are potentially very attractive for subwavelength control and manipulation of light at the infrared to terahertz wavelengths. Probing their propagation behavior in nanostructures is crucial to guide rational device design. Here, aided by monochromatic scanning transmission electron microscopy-electron energy loss spectroscopy technique, we measure the dispersion relation of SPhPs in individual SiC nanorods and reveal the effects of size and shape. We find that the SPhPs can be modulated by the geometric shape and size of SiC nanorods. The energy of SPhPs shows red-shift with decreasing radius and the surface optical phonon is mainly concentrated on the surface with large radius. Therefore, the fields can be precisely confined in specific positions by varying the size of the nanorod, allowing effective tuning at nanometer scale. The findings of this work are in agreement with dielectric response theory and numerical simulation, and provide novel strategies for manipulating light in polar dielectrics through shape and size control, enabling the design of novel nanoscale phonon-photonic devices.

Spectrally tunable infrared plasmonic F,Sn:In2O3 nanocrystal cubes.

A synthetic challenge in faceted metal oxide nanocrystals (NCs) is realizing tunable localized surface plasmon resonance (LSPR) near-field response in the infrared (IR). Cube-shaped nanoparticles of noble metals exhibit LSPR spectral tunability limited to visible spectral range. Here, we describe the colloidal synthesis of fluorine, tin codoped indium oxide (F,Sn:In2O3) NC cubes with tunable IR range LSPR for around 10 nm particle sizes. Free carrier concentration is tuned through controlled Sn dopant incorporation, where Sn is an aliovalent n-type dopant in the In2O3 lattice. F shapes the NC morphology into cubes by functioning as a surfactant on the {100} crystallographic facets. Cube shaped F,Sn:In2O3 NCs exhibit narrow, shape-dependent multimodal LSPR due to corner, edge, and face centered modes. Monolayer NC arrays are fabricated through a liquid-air interface assembly, further demonstrating tunable LSPR response as NC film nanocavities that can heighten near-field enhancement (NFE). The tunable F,Sn:In2O3 NC near-field is coupled with PbS quantum dots, via the Purcell effect. The detuning frequency between the nanocavity and exciton is varied, resulting in IR near-field dependent enhanced exciton lifetime decay. LSPR near-field tunability is directly visualized through IR range scanning transmission electron microscopy-electron energy loss spectroscopy (STEM-EELS). STEM-EELS mapping of the spatially confined near-field in the F,Sn:In2O3 NC array interparticle gap demonstrates elevated NFE tunability in the arrays.

Dynamic behavior of helium bubbles at high temperature in Si studied by in situ TEM, STEM-EELS, and TDS

We demonstrated, using in situ transmission electron microscopy (TEM), the random motion of helium bubbles formed by irradiation with helium ions in an FZ (011)Si specimen at high temperatures. Analysis of the random motion yielded the bubble diffusivity, which was dependent on the bubble size and the annealing temperature. The helium density in individual bubbles, measured using scanning transmission electron microscopy-electron energy loss spectroscopy, was as high as 70 ± 14 He/nm3. It was dependent on the irradiation temperature, beam flux, and annealing temperature but was independent of the bubble size. The bubbles retained a high density of helium atoms at around 800 K, but a significant reduction of the helium density within individual bubbles took place after annealing above about 970 K. Thermal desorption of helium from the specimen, measured using thermal desorption spectroscopy, occurred at temperatures between 650 K and 1100 K, centering at around 890 K. TEM revealed the corresponding microstructure. On the basis of these results, the thermal desorption profile was discussed in terms of the release of helium at lower temperatures from migrating and coalescing overpressurized bubbles and bubbles migrating to the specimen surface, and at higher temperatures from stationary large bubbles.