Lattice-matched GaP layers without extended defects can be grown on Si(001) substrate via a two-step growth procedure, consisting of low-temperature nucleation followed by high-temperature overgrowth. A transient reflectivity experiment on a thin, low-temperature nucleation layer discovered a previously unknown phonon mode at 2 THz upon below-bandgap optical excitation (Metteet al2025Adv. Mater. Interfaces2400573). Here we examine the influence of the two-step growth process on the ultrafast carrier and phonon dynamics of the GaP/Si interface. We find that the discrete electronic state, which governed the interfacial carrier dynamics of the thin nucleation layer, becomes suppressed when a thicker layer is formed by high-temperature overgrowth. The coherent 2 THz oscillation is observed also in the high-temperature overgrown structures, at the constant frequency regardless of the GaP layer thickness. Its resonance behavior closely follows that of the carrier dynamics at the respective growth stage. This supports its assignment to a phonon mode generated at the heterointerface and strongly coupled to the interfacial carriers. The phonon amplitude exhibits a non-monotonic dependence on the GaP layer thickness, and its optical polarization dependence is qualitatively altered by the high-temperature overgrowth, neither of which is accounted for by the carrier-phonon coupling alone. Our results demonstrate that the 2 THz interfacial phonon mode is robust against high-temperature overgrowth, while its amplitude is determined by both coupling to interfacial electronic transitions and atomic-scale structural reorganization at the interface.

2D indium selenide (InSe) is a layered semiconductor with high electron mobility and a tunable band gap ranging from 1.25 eV in the bulk to 2.8 eV in the monolayer limit. However, growing phase-pure InSe remains challenging due to the complex indium-selenium (In-Se) phase diagram. This complexity and the sensitivity of chemical precursors to growth conditions make it difficult to control which In-Se phase forms during synthesis during, e.g., metal-organic chemical vapor deposition (MOCVD). MOCVD is considered the most promising approach for growing InSe, as it enables wafer-scale, uniform, and controllable deposition-key requirements for device integration. We present a systematic investigation of InSe synthesis on c-plane sapphire substrates at low temperatures. By varying Se/In precursor ratio and growth temperature, we create a phase diagram that covers In-rich, equal stoichiometric, and Se-rich InxSey phases. Raman spectroscopy and atomic force microscopy, supported by energy dispersive X-ray spectroscopy and scanning transmission electron microscopy, reveal formation conditions of 2D InSe. The epitaxial alignment is verified by in-plane X-ray diffraction. Samples grown under optimized conditions exhibit a strong optical absorption in the visible range and especially a comparably high electron mobility, underlining the potential of the MOCVD-grown material for future applications.

Polycrystalline materials have numerous applications due to their unique properties, which are often determined by the grain orientation relationships. Hence, quantitative characterization of grain as well as interface orientation is essential to optimize these materials, particularly energy materials. Using scanning transmission electron microscopy (TEM), materials can be analysed in an extremely fine grid of scan points via electron diffraction patterns at each scan point. By matching the diffraction patterns to a simulated database, the crystal orientation of a grain at each scan point can be determined. In this work, we train convolutional neural networks on dynamically simulated diffraction patterns of LiNiO2, an important cathode-active material for lithium-ion batteries, to predict the orientation of grains in terms of three Euler angles for the complete fundamental orientation region. Results demonstrate that these networks outperform the conventional pattern-matching algorithm with increased accuracy and efficiency. The former can be attributed to the fact that these models are trained by data incorporating dynamical effects. This work is an attempt to apply deep learning for the analysis of TEM data to determine the grain orientation and enlighten the great potential of machine learning to accelerate the analysis of electron microscopy data, toward a high-throughput characterization technique.

The properties of polycrystalline materials are strongly influenced by the spatial arrangement and orientations of individual grains within the microstructure, making nanoscale characterization of grain orientation essential. This is also often the case for small grains in the nm regime explored using scanning transmission electron microscopy (STEM). Automated crystal orientation mapping (ACOM) is traditionally performed using spot-like diffraction patterns. In contrast, orientation mapping based on transmission Kikuchi diffraction (TKD) using an aberration-corrected (AC) convergent STEM probe remains relatively underexplored, despite its superior orientation sensitivity and higher spatial resolution. In this work, we present an open-source software-based template-matching approach for orientation mapping using AC-STEM TKD. A master pattern (a simulated angular distribution of Kikuchi band intensities on the unit sphere) is first generated through a dynamical simulation implemented in open-source software. This resulting pattern is subsequently imported into another open-source package for geometric simulations and orientation indexing. We demonstrate the capability of the proposed method by applying it to orientation mapping in BaZr0.4Ce0.4Y0.1Yb0.1O3-δ (BZCYYb4411) fuel-cell material and LiNiO2 (LNO) lithium-ion battery cathode material. The best-matched simulated patterns exhibit strong agreement with experimental data, even under the challenging conditions with limited diffraction space available for matching.

To enhance the range of electric vehicles, research is focused on increasing the nickel content in cathode active materials (CAM), which leads to higher practically achievable specific capacities. As a result, the material LiNiO2 (LNO) has attracted significant interest. In this study, a two-step temperature swing synthesis is employed to produce LNO secondary particles with large primary grains as CAM for solid-state batteries (SSBs). The synthesis involves sintering the material at 800 °C for 1 h, followed by an annealing step at a lower temperature for 6 h. Different batches of annealed LNO, using temperatures of 600 and 700 °C, are compared with unannealed LNO, utilizing various transmission electron microscopy (TEM) techniques. The annealing step contributes to smoother particle surfaces, reduced residual lithium species on particle surfaces, and increased lithium occupancy in the crystal lattice, resulting in higher discharge capacity during the initial cycles. However, higher annealing temperatures also lead to the formation of a thin rock-salt layer on the surface and internal misorientation, likely caused by thermal residual stress during cooling. These effects are more pronounced in the sample annealed at 700 °C compared to 600 °C. Despite the potential drawbacks associated with these factors, LNO annealed at 700 °C achieves the highest discharge capacity, indicating that the benefits of annealing outweigh its disadvantages, at least during the initial cycles.

Two-dimensional materials such as gallium selenide (GaSe) hold promise for optoelectronics due to their tunable bandgaps. Gallium sesquiselenide (Ga2Se3), a related phase with a direct bandgap, is also suitable for integration on silicon due to its matching lattice constant. We demonstrate wafer-scale synthesis of GaxSey by metal-organic chemical vapor deposition, varying growth temperature (450–600 °C) and selenium-to-gallium ratio. To guide phase-pure growth, we construct a phase diagram. Raman spectroscopy confirms phase formation, while mass spectrometry of Di-iso-propyl selenide supports the temperature-dependent phase transition, induced by the actual selenium-to-gallium ratio on the wafer surface. Microscopy reveals distinct morphologies: Ga2Se3 forms epitaxial films with antiphase domains on GaP/Si, while GaSe grows as faceted 〈111〉 -oriented sheets. Optical spectroscopy confirms direct transitions at 1.34 eV (Ga2Se3) and 1.89 eV (GaSe). These results demonstrate controlled MOCVD growth of phase-pure GaxSey and highlight their potential for silicon-based optoelectronic integration.

Accurate grain orientation mapping is essential for understanding and optimising the performance of polycrystalline materials, particularly in energy applications. Lithium nickel oxide (LiNiO2) is a promising cathode material for next-generation lithium-ion batteries, and its electrochemical behaviour is closely linked to microstructural features such as grain size and crystallographic orientation. Traditional orientation mapping methods—manual indexing, template matching (TM), or Hough transform-based techniques—are often slow and noise-sensitive, especially when processing complex or overlapping patterns, creating a bottleneck in large-scale microstructural analysis. This work presents a machine learning-based approach for predicting Euler angles directly from scanning transmission electron microscopy (STEM) diffraction patterns (DPs), enabling automated, high-resolution orientation mapping for nanoscale microstructure analysis. Three deep learning architectures—convolutional neural networks (CNNs), Dense Convolutional Networks (DenseNets), and Shifted Windows (Swin) Transformers—are evaluated using an experimentally acquired dataset labelled via a commercial TM algorithm. While the CNN model serves as a baseline, both DenseNets and Swin Transformers achieve superior performance, with the Swin Transformer yielding the highest evaluation scores and most consistent microstructural predictions. The resulting crystal maps reveal clear grain boundaries and coherent intra-grain orientation distributions, underscoring the potential of advanced machine learning models for high-throughput microstructural characterisation.

Ultrafast carrier and phonon dynamics at the lattice-matched interface of GaP/Si(001) are investigated upon below-bandgap excitation of the GaP layer at different growth stages. Transient reflectivity (TR) signals exhibit an abrupt change upon photoexcitation, revealing ultrafast creation of carriers at the heterointerface and/or in the GaP layer. Temporal evolution and resonance behavior of the interfacial carrier dynamics reveals the dominance of a discrete electronic state for thin low-temperature nucleation layers and its extinction for thicker high-temperature overgrown layers. In addition, a coherent 2-THz oscillation, which was reported previously for the low-temperature nucleation layer, is observed also for the high-temperature overgrown layers. The resonance behavior of the oscillation amplitude is similar to that of the interface carrier dynamics of the respective layers, supporting its assignment as a phonon mode localized at the heterointerface and coupled strongly with the interface carriers. On the other hand, the phonon amplitude exhibits a non-monotonic dependence on the GaP layer thickness, and its optical polarization-dependence is transformed qualitatively by the high-temperature overgrowth, both of which can be explained only qualitatively by the coupling with the interface carrier dynamics. Our observations imply that the 2-THz phonon mode itself is robust against the high-temperature overgrowth, but its amplitude is dominated by the coupling with the interface electronic transition as well as by the atomic reorganization at the interface by the overgrowth.

Momentum-resolved scanning transmission electron microscopy (MRSTEM) is a powerful phase-contrast technique that can map lateral magnetic and electric fields ranging from the micrometer to the subatomic scale. Resolving fields ranging from a few nanometers to a few hundred nanometers, as well as across material interfaces, is particularly important since these fields often determine the functional properties of devices. However, it is also challenging since they are orders of magnitude smaller than atomic electric fields. Thus, subtle changes in diffraction conditions lead to significant changes in the measured MRSTEM signal. One established approach to partially overcome this problem is precession electron diffraction, in which the incident electron beam is continuously precessed while precession-averaged diffraction patterns are acquired. Here, we present an alternative approach in which we sequentially tilt the incident electron beam and record a full diffraction pattern for each tilt and spatial position. This approach requires no hardware modification of the instrument and enables the use of arbitrary beam tilt patterns that can be optimized for specific applications. Furthermore, recording diffraction patterns for every beam tilt allows access to additional information. In this work, we use this information to create virtual large-angle convergent beam electron diffraction patterns to assess MRSTEM data quality and improve field measurements by applying different data analysis methods beyond simple averaging. The presented data acquisition concept can readily be applied to other 4D-STEM applications.

A key issue in the steam electrode development for Proton-conducting Ceramic Electrolysis Cells (PCECs) is the required contact with the electrolyte material as well as thermo-chemical compatibility during cell processing and operation. This study focuses on BaCo0.4Fe0.4Zr0.2O3-delta (BCFZ442) as a potential steam electrode material paired with the state-of-the-art proton conductor BaZr0.7Ce0.2Y0.1O3-delta (BZCY721). The objectives are to identify an optimal material composition amongst several compositional variations and to define its corresponding fabrication conditions for achieving suitable electrodes compatibility, adhesion and performance for further translation into the cell fabrication process. Therefore, we tested 90:10, 70:30 and 50:50 BCFZ442:BZCY721 composites which were heat-treated in the temperature of 800-1200 degrees C. The phase evolution accompanied by thermal and microstructural analyses and in addition to electrical and mechanical characterization yielded interesting results. Extensive analysis indicate compositional, functional and fabrication optimum in the steam electrode development of PCECs.

Nitride-based semiconductors are vital for efficient optoelectronic devices in the ultraviolet to green spectral range. However, producing red-emitting InGaN micro-LEDs is challenging due to lattice mismatch with traditional GaN substrates. This mismatch causes strain relaxation, compositional gradients, and defects in high-indium-content InGaN films. These issues severely limit device efficiency, and the potential of alternative substrates to address these challenges is not fully explored. Here, we show that Al1-xScxN pseudosubstrates with adjustable lattice parameters greatly improve lattice matching of InGaN. Using plasma-assisted molecular beam epitaxy, we grow 120 nm-thick, phase-pure Al1-xScxN layers (0.1 < xSc < 0.2). This enables high-quality deposition of In0.28Ga0.72N layers and a uniform indium distribution compared to growth directly on GaN. AlScN-supported films exhibit no compositional pulling effect common for conventional substrates. This uniformity is confirmed by room-temperature photoluminescence, showing a narrow emission at 538 nm. Our results demonstrate that AlScN pseudosubstrates are promising for future integrated red micro-LED devices.

The type-II band alignment of particular III/V heterostructures is a promising route towards achieving certain wavelengths on specific substrates, which would not be possible with type-I structures. One example is telecommunication lasers on GaAs substrates. This study reports on the progress in combining dilute nitrides and dilute bismides in W-type hetero structures to improve the luminescence intensity, which is a fundamental prerequisite for future incorporation in a laser structure. Increasing the emission wavelength of these structures is challenging and the interface formation is critical, especially as two metastable materials are combined. Here, we investigate the impact of different interface configurations by growing Ga(N,As)/Ga(As,Bi) and Ga(As,Bi)/Ga(N, As) type-II structures. We employ metal-organic vapor phase epitaxy (MOVPE) to grow Ga(N,As) and Ga(As,Bi) layers and investigate the effects of interlayer thicknesses on the structural and optical properties of the hetero structures. The results indicate that - while the introduction of a GaAs interlayer can affect the direct and indirect transitions intensities - it does not significantly improve the interface quality. However, this is strongly influenced by the order in which the materials are grown. The growth of Ga(N,As) on Ga(As,Bi) shows no peculiarities, while the type II transition energy is shifted to lower energies when Ga(As,Bi) is grown on Ga(N,As). High-resolution X-ray diffraction (HR-XRD), photoluminescence (PL) spectroscopy, atomic force microscopy (AFM), and scanning transmission electron microscopy (STEM) were used to characterize the samples.

We investigate the excitonic properties of epitaxially grown WS2 monolayers, bilayers and multilayers on graphene using monochromatic electron energy loss spectroscopy (EELS) in a scanning transmission electron microscope. This material system is particularly attractive for optoelectronic applications, as direct growth from the gas phase offers a scalable route to wafer-sized heterostructures. The combination of nanometer-scale spatial resolution and high spectral quality in EELS allows for a detailed analysis of layer-dependent excitonic features. To complement the experimental results, we perform ab initio simulations based on density functional theory and the Bethe-Salpeter equation. The experimental spectra reveal a systematic redshift of both A and B excitons at the K-valley─centered near 2.0 and 2.4 eV, respectively─as the number of WS2 layers increases. While such redshifts are often attributed to dielectric screening, our ab initio calculations show that the dominant contribution arises from a subtle lattice mismatch between the lower and upper WS2 layers. We trace this mismatch to the heteroepitaxial alignment of the first WS2 layer to the graphene substrate during the growth process. Our results highlight how nanoscale structural distortions in epitaxial 2D materials can strongly influence key excitonic properties, even in the absence of intentional strain or alloying. By combining nanometer-scale electron spectroscopy with advanced theory, we establish a direct link between atomic structure and excitonic response in realistic, nonidealized heterostructures. These findings underscore the importance of microscopic interface effects in the design and scalable fabrication of exciton-based optoelectronic devices.

Silicon offers great promise as a potential anode active material and the optimum alternative to lithium metal in all‐solid‐state lithium‐ion batteries. However, its practical application is limited by severe volume expansion (≈300%) during lithiation, leading to cracking upon delithiation. In this study, the microstructural evolution of microcrystalline silicon electrodes in a solid‐electrolyte‐free environment is investigated using cryogenic scanning transmission electron microscopy (STEM) after electrochemical cycling. A controlled workflow prevents ambient exposure, and cryo‐TEM ensures structural integrity. After the first lithiation, the electrode shows a heterogeneous mix of crystalline Li 15 Si 4 , various amorphous Li x Si phases, and residual crystalline silicon. After the first delithiation, the silicon becomes largely amorphous, showing a heterogeneous texture with pronounced thread‐like features and only traces of crystallinity. By the tenth delithiation, the bulk microstructure is far more uniform, with thread‐like features largely eliminated and persisting only in small regions near grain boundaries. These results indicate that, although silicon begins in a crystalline state, a more homogeneous bulk silicon amorphous microstructure develops only after several cycles. These findings highlight that stabilizing the microstructure and minimizing cracking during cycling requires not only optimization of electrode architecture but also careful selection of the silicon phase.

Cathode active material (CAM) particles and solid electrolyte (SE) – CAM composites for solid‐state batteries (SSBs) are often subjected to elevated temperatures during annealing or co‐sintering. This thermal treatment can affect the material's structure and induce degradation processes, particularly at the SE – CAM interface. To better understand these phenomena and improve material stability and performance, investigations by (scanning) transmission electron microscopy ((S)TEM) under realistic processing conditions, i.e., in an oxygen atmosphere, are desirable. However, preparing electron‐transparent TEM lamellae of SE – CAM composites with intact interfaces is highly challenging. Therefore, an in situ heating methodology is first established using LiNiO 2 (LNO) particles as a model system. In this study, the morphological and structural evolution of thinned LNO particles during heating in an oxygen atmosphere is investigated, employing in situ 4D nanobeam STEM. The in situ observations are complemented with postmortem electron diffraction and spectroscopy measurements. These findings indicate that LNO undergoes structural degradation at temperatures ≈350 °C, transitioning from the layered () structure to a NiO‐type rock‐salt phase (). This onset temperature is significantly higher than that observed in comparable in situ heating experiments conducted in vacuum, highlighting the importance of an oxygen atmosphere for replicating real‐world processing conditions.

Alternating metal-modulated molecular beam epitaxy enables the growth of both self-assembled c-InGaN/GaN quantum wells and fully alloyed c-InGaN layers. In situ reflection high-energy electron diffraction (RHEED) analysis coupled with ex situ structural characterization investigates the growth mechanism and prerequisites for the self-assembled c-InGaN quantum well formation. The data reveal that indium accumulates without incorporating into the underlying c-GaN layer during an indium deposition step. However, the accumulated indium forms c-InGaN during a subsequent GaN growth step consistent with vertical cation segregation. Furthermore, X-ray diffraction, time-of-flight secondary ion mass spectrometry depth profiles, and scanning transmission electron microscopy imaging show homogeneous and well-defined c-InGaN layers. The presented growth mechanism requires high substrate temperatures and gallium fluxes. Still, limit testing suggests that indium contents of up to 37% are feasible. This encourages the implementation of metal-modulated grown c-InGaN in red light-emitting devices. Furthermore, combining RHEED operando diagnostics and a precise understanding of the growth mechanism is vital for progressing toward automated growth of complex heterostructures.

Cubic InGaN alloys are a promising candidate material for next-generation optoelectronic applications as they lack internal fields and promise to cover large parts of the electromagnetic spectrum from the deep UV toward the mid-infrared. This demands high-quality epitaxial growth of cubic InGaN/GaN quantum wells, especially for the red energy range. However, the growth of indium-bearing nitride quantum wells in the metastable cubic phase still poses many challenges. InGaN and GaN are typically grown at different temperatures and with different metal fluxes in molecular beam epitaxy, leading to either long waiting periods for temperature adjustment or growth under suboptimal conditions. Both degrade the crystal quality and optical properties. In this study, we apply a metal-modulated growth approach in molecular beam epitaxy that enables us to grow either self-assembled, phase pure, cubic InGaN/GaN multi quantum wells (MQWs) or homogeneous c-InGaN layers, only by adjusting the shutter duration times for Ga and In. We achieve smooth surfaces and sharp interfaces with a quantum well thickness tunable from 6 to 16 nm and a barrier thickness ranging from 4 to 10 nm. X-ray diffraction confirms >99% phase purity of our cubic layers, while time-of-flight secondary ion mass spectrometry, scanning transmission electron microscopy, and energy-dispersive X-ray spectroscopy provide detailed information on the quantum well composition and strain. Photoluminescence measurements at room temperature demonstrate the emission properties of the samples, with the emission wavelength ranging from 540 to 670 nm. Changing the barrier and QW thickness results in a shift of emission energy of up to 400 meV, which is explained by quantum confinement and strain. The high interface quality and excellent optical properties of the quantum wells without the need for multiple metal sources or long waiting times represent a significant advance in the development of next-generation optoelectronic devices.

In situ transmission electron microscopy (TEM) observations of the metal-organic vapor phase epitaxy (MOVPE) growth promise to enhance the understanding of this complex process. However, a new experimental approach is required, capable of live imaging at the atomic scale and simultaneously reflecting this method's elevated pressures. To this end, a closed gas cell in situ TEM setup is used as a micrometer-scaled MOVPE reactor to grow GaP using tertiary butyl phosphine (TBP) and trimethyl gallium (TMGa). To prove the MOVPE reactor ability of the in situ TEM holder, the thermal decomposition of TBP and TMGa is shown to proceed similarly to conventional reactor setups. Decomposition temperatures align with susceptor temperatures in MOVPE machines. Formed products and their temperature decomposition curves are comparable to previous investigations performed in conventional reactors, even though the setups significantly differ. The obtained results are exploited to grow GaP nanostructures via the MOVPE growth process inside the TEM. To prepare a substrate surface for GaP growth, which is highly challenging, Au-catalyzed vapor-liquid-solid-grown GaP nanowires are grown in the reactor cell. Subsequently, the nanowire's sidewalls serve as MOVPE substrates. These results lay the foundation for crystal growth observation under MOVPE conditions in a TEM.