Accurate Self-Attention Wavefunctions at Large Scale

2026-07-14T17:47:50Z

Self-attention neural networks provide powerful variational wavefunctions that surpass the expressivity of traditional variational ansatze. This expressivity, however, comes with increased computational complexity, raising a pressing question about scalability -- can such wavefunctions retain their accuracy at large system sizes? We apply self-attention wavefunctions to the two-dimensional homogeneous electron gas for up to N=169 particles, obtaining energies systematically lower than state-of-the-art DMC. Direct access to the ground state wavefunction further lets us recover the full collective-mode dispersion of the liquid phase, from the small-q plasmon branch to a roton-like minimum near q=2k_F. Observables at N=91 and N=169 are in near-perfect agreement, indicating convergence to the thermodynamic limit.

Molecular Beam Epitaxy of Al$\mathrm{_{1-x}}$Sc$\mathrm{_{x}}$N Nanowires: Towards Group-III Nitride Piezoelectric Nanogenerators with Enhanced Response

2026-07-14T16:00:32Z

We study the molecular beam epitaxy of self-assembled Al$\mathrm{_{1-x}}$Sc$\mathrm{_{x}}$N nanowires on conductive TiN layers and demonstrate their application in piezoelectric nanogenerators. Wurtzite Al$\mathrm{_{1-x}}$Sc$\mathrm{_{x}}$N nanowires with uniform Sc incorporation are grown across a wide composition range (0

A coupled fully kinetic hydrogen transport and ductile phase-field fracture framework for modeling hydrogen embrittlement

2026-07-14T15:09:48Z

Modeling hydrogen embrittlement (HE) is a long-standing engineering challenge, which has experienced significant developments in recent years. Yet, there is a gap in modeling the effect of the kinetics of hydrogen segregation at dislocations and the resulting interaction between ductile tearing and hydrogen-induced brittle fracture. In this work, a comprehensive chemo-mechanical framework is developed by coupling the fully kinetic hydrogen transport model with the geometric phase-field fracture method. A novel driving force is proposed that utilizes a hyperbolic tangent function of stress triaxiality to ensure that plastic dissipation contributes to fracture only under tensile conditions, phenomenologically representing void-driven ductile damage. The model successfully predicts the hydrogen-dependent shift in damage initiation from the specimen core to the surface. More importantly, hydrogen segregation at dislocations was shown to be crucial for modeling the multiple surface cracking experimentally observed at the necking region. Furthermore, the framework captures the competition between loading rates and diffusion kinetics, resolving the transition from multiple circumferential surface cracking at high strain rates to center-initiated single crack at lower rates. Finally, the model reproduced the experimental J-resistance curves for compact tension specimens, showing the transition from ductile tearing to embrittled crack.

Phase-shifted multicomponent spin-charge nematicity in an altermagnet

2026-07-14T14:37:28Z

Altermagnets host spin-split Fermi surfaces without net magnetization. This intrinsically multicomponent electronic setting raises the possibility that familiar correlated electron phases acquire unconventional spin-charge structure. Here we report the discovery of altermagnetic nematicity in Co0.25NbSe2. Using spectroscopic-imaging scanning tunneling microscopy and spin-polarized scanning tunneling microscopy, we find that the three nominally C3-related directions lose rotational equivalence in the zero-field state, in both charge and spin-sensitive tunneling channels. Strikingly, the dominant spin-sensitive component is shifted by one C3 sector relative to the dominant charge component, revealing a phase-shifted spin-charge nematic response. A phenomenological theory shows that altermagnetic order favors a finite relative phase between the charge and spin-sensitive nematic components -- C3 lattice pinning frustrates this preferred offset and selects the observed phase locking. These results establish altermagnetic nematicity as a new form of multicomponent electronic liquid-crystal order and point to a potentially generic route by which altermagnets can transform conventional correlated phases into symmetry-engineered spin-charge orders.

Symmetry-Informed Deep Learning for Electromagnetic Scattering

2026-07-14T14:22:45Z

Deep learning can accelerate the modeling of electromagnetic devices by replacing costly simulations with neural networks trained to map design parameters to scattering parameters. However, data efficiency remains a central bottleneck, as training data is typically generated through expensive numerical simulations. Here we show that symmetry provides a powerful and largely untapped route to overcoming this limitation in electromagnetic scattering problems. Leveraging the equivariance of Maxwell's equations, we obtain general transformation rules that map symmetries of electromagnetic devices to corresponding transformations of their scattering parameters. This enables both systematic data augmentation and the construction of exactly equivariant neural networks. We implement the framework for both discrete and continuous symmetry groups and demonstrate its effectiveness on photonic-crystal slabs and free-form diffraction gratings. Incorporating symmetry improves data efficiency by an order of magnitude compared to standard architectures, while equivariant models additionally enforce physical constraints exactly. Our approach is general and complementary to existing physics-informed strategies, provides a first-principles framework for constructing physically grounded surrogate models, and establishes symmetry as a unifying inductive bias for data-efficient and physically consistent learning in computational electromagnetics and beyond.

Single-orbital tight-binding model for chiral one-dimensional hybrid organic-inorganic lead halide perovskites

2026-07-14T14:21:05Z

We present a single-orbital tight-binding model for the low-energy electronic states of the chiral one-dimensional hybrid organic-inorganic lead halide perovskite $\mathrm{(}R/S\mathrm{-PEA)PbI}_3$. The model is constructed from a single effective orbital on each of the four symmetry-related sites in the primitive unit cell and incorporates layer, in-plane sublattice, and spin degrees of freedom. Using separate parameter sets for the conduction and valence bands, the effective Hamiltonian reproduces the overall band dispersions obtained from density-functional-theory calculations and quantitatively captures the spin splittings near the band edges. It also captures the leading spin-polarization patterns of the Bloch states, showing that the band-edge spin splitting and spin polarization are encoded in a small number of symmetry-adapted spin-dependent hopping terms. We further analyze the accidental degeneracies of the effective Hamiltonian using screw eigenvalues and antiunitary operators. This analysis separates accidental degeneracies originating from the restricted term content of the effective Hamiltonian from degeneracies enforced by nonsymmorphic screw symmetries and time-reversal symmetry. The present model provides a symmetry-transparent starting point for understanding the band-edge electronic structure of chiral lead halide perovskites and for analyzing optical, spin, and transport responses in this class of materials.

Strain-Tunable Shift Current and Magneto-Optical Kerr Effect in Multiferroic Altermagnet Fe2Mo3O8

2026-07-14T14:14:24Z

Altermagnetism has recently emerged as a compelling frontier in spintronics, seamlessly merging the agile tunability of ferromagnets with the hallmark merits of antiferromagnets. As a prototypical polar multiferroic featuring distinctive altermagnetism, Fe2Mo3O8 hosts an ideal playground for exploring the intricate interplay among ferroelectric polarization, altermagnetic order, and spin-dependent responses. Here, employing first-principles calculations, we systematically investigate the coupling among polarization, spin splitting, shift current, and magneto-optical responses in Fe2Mo3O8. Our findings reveal that switching the ferroelectric polarization not only inverts the sign of the shift current but also comprehensively reshapes the momentum-space spin-splitting texture. Furthermore, the shift current and magneto-optical spectrum exhibits strong tunability under mechanical strain. Remarkably, the application of a-axis uniaxial strain breaks the crystalline symmetry, thereby activating a finite magneto-optical Kerr effect that is otherwise forbidden in the pristine phase.

Photogeneration and signatures of coherent phonons in time-resolved photoemission spectroscopy: First-principles time-dependent adiabatic GW approach

2026-07-14T13:58:32Z

Coherent lattice dynamics can be observed in pump-probe time-resolved and angle-resolved photoemission spectroscopy (TR-ARPES) as a periodic modulation of intensity and energy of photoelectrons over probe time. We present an ab initio time-dependent GW approach including electron-phonon (e-ph) couplings to simulate the photogeneration of coherent phonons and their effects on the TR-ARPES of monolayer MoS2. We demonstrate that state-resolved e-ph coupling strength can be obtained from analyzing coherent phonon modulations on TR-ARPES. Features of both the impulsive stimulated Raman scattering mechanism and the displacive excitation mechanism of coherent phonon generation are identified in our simulations. We clarify their origins and the coincident selection rule of coherent phonon generation and Raman scattering intensity. This method provides supports to analyze coherent phonon dynamics and e-ph couplings in TR-ARPES and enables quantitative engineering of band structure through coherent phonons.

The self-organized vacancy order in Pr$_9$Ge$_{16}$

2026-07-14T13:58:07Z

In this work, we report the discovery of a new crystal structure on the Ge-rich side of the Pr-Ge binary phase diagram. Using a high-temperature flux technique, we grew single crystals of $Pr_9Ge_{16}$, which adopt a previously unreported orthorhombic $Fdd$2 structure type featuring ordered Ge vacancies. We present the anisotropic magnetic properties and identify the crystallographic $b$ axis perpendicular to the crystal plane as the magnetic easy axis. Temperature-dependent resistivity measurements reveal metallic behavior with a distinct anomaly at $T_{\mathrm{C}}$ = 14.3 K. Hall resistivity data indicate that electron-like carriers dominate, with a carrier concentration on the order of $10^{27}~\mathrm{m}^{-3}$. The magnetic order is readily suppressed by a magnetic field of approximately 0.4 T applied along the easy $b$ axis.

Charge-Transfer Induced Reactivity in sp Carbon Atomic Wires: Towards 0-D sp-sp2 Nanostructures

2026-07-14T13:17:04Z

Carbon Atomic Wires (CAWs) are finite linear chains of sp-hybridized carbon atoms. Here the electrochemical reduction of CAWs in the form of polyynes (i.e. with alternated single-triple bonds) is reported. Upon applying a reducing potential to a solution containing polydispersed hydrogen-capped polyynes, the formation of a black precipitate was observed. Electronic absorption spectroscopy confirmed the irreversible reaction of the carbon chains while excluding degradation or side reactions. Subsequent analyses revealed that the precipitate consisted of amorphous carbon nanoparticles with tunable diameters. This control over particle size is attributed to the modulation of growth kinetics through restricted mass transport toward the solid-liquid interface. Raman spectroscopy showed that the resulting material exhibits an amorphous sp-sp2 character, with a retained sp fraction exceeding 60%. Smaller nanoparticles displayed reduced disorder within the sp2 domains and a broader distribution of sp-chain lengths preserved in the amorphous matrix. Additional experiments on size-selected polyynes suggest that this synthesis method allows to better preserve the starting chain length in the final structure. Unlike previously reported amorphous sp-sp2 carbon networks, the nanoparticles produced in this study show remarkable stability under ambient conditions, retaining their sp character for times in excess of six months. These findings pave the way for future applications, particularly as further diameter tuning may enable access to the quantum-dot regime.

Sign-resolved nanoscale readout and control of hidden antiferromagnetic spin order

2026-07-14T13:14:05Z

Antiferromagnetic memories promise ultrafast, stray-field-free information storage. Yet perfect magnetic compensation conceals the information carrier itself: the sign of the Néel vector distinguishing two time-reversed states. Moreover, in future dense memories, the local polarity of Néel domains would need to be read out on the nanoscale. We make this hidden polarity visible in a fully-compensated, high-Néel-temperature, PT-symmetric antiferromagnet by driving interband electric-dipole transitions by mid-infrared near-fields confined at a scanning probe. The excitation generates a Néel-order-dependent quantum-metric photocurrent, a Hall-like signal reversing with Néel order, which we term the optical nonlinear anomalous Hall effect. This optically induced electrical readout maps opposite Néel polarities with sub-100-nm resolution at room temperature and, combined with spin-orbit-torque writing, reveals Néel-texture polarization and reversible domain-wall motion, establishing electrical-write/optoelectronic-read antiferromagnetic functionality.

Antiphase boundary-driven large exchange bias and negative magnetoresistance in NiCo2O4

2026-07-14T13:11:21Z

Nickel-based spinel oxide has recently attracted significant attention due to its remarkable magneto-transport properties, promising various spintronic applications. In this study, we observe an exchange bias effect, where the magnetic hysteresis loop shifts both horizontally and vertically when single-phase nanocrystalline NiCo2O4 is cooled in the presence of an external magnetic field. The magnitude of the exchange bias field is approximately 375 Oe at 5 K, and this effect disappears around 300 K. Furthermore, NiCo2O4 shows semiconducting behaviour in longitudinal resistivity and demonstrates a substantial negative magnetoresistance of ~31.5% at 10 K. The magnetoresistance exhibits a butterfly-shaped behaviour at low magnetic fields and becomes almost linear at high magnetic fields. A detailed analysis of the negative magnetoresistance and exchange bias reveals that both phenomena originate from the formation of antiphase boundaries in the as-synthesized NiCo2O4 sample, as confirmed by the high-resolution transmission electron micrograph.

Emergent exchange bias in ultra-thin La0.67Sr0.33MnO3 films driven by ferro-antiferromagnetic phase coexistence

2026-07-14T11:55:49Z

Ultra-thin La0.67Sr0.33MnO3 (LSMO) films are generally regarded as single-phase ferromagnets, yet their reduced dimensionality enhances the impact of oxygen stoichiometry and local structural distortions. Here we demonstrate that LSMO layers with thicknesses between 3 and 17 nm develop a robust and thickness-independent exchange bias (EB) despite the absence of an engineered ferromagnetic-antiferromagnetic (FM-AFM) interface. Angular magneto-optical Kerr effect measurements reveal a reproducible hysteresis-loop shift that reverses sign upon 180-degree rotation, confirming its intrinsic origin. X-ray photoelectron spectroscopy shows a thickness-independent Mn2+ fraction (about 20 percent), indicating the presence of oxygen-deficient reduced regions embedded within the Mn3+/Mn4+ ferromagnetic matrix. A simple interfacial-exchange model suggests that only a sub-percent fraction of antiferromagnetic (or frustrated spin-ice-like) regions is required to generate the observed EB. These results demonstrate that ultra-thin LSMO can spontaneously host internal FM-AFM interfaces driven by oxygen deficiency, revealing an emergent route to exchange bias in nominally single-phase manganite films.

A critical consideration of X-ray detectors based on Ga2O3: excitation, carrier transport mechanisms and performance standardization

2026-07-14T11:16:30Z

X-ray detection underpins a wide range of applications in medicine, security, industrial inspection, scientific research for non-destructive imaging and material analysis. The rapid development of Ga2O3-based X-ray detectors offers a promising pathway toward next-generation detectors with high sensitivity, low noise, and harsh environment applications, benefiting from its intrinsic material properties such as high density, wide band gap energy, and high thermal-chemical stability. However, the underlying device operating mechanisms, including both carrier excitation and transport processes, have not yet been adequately studied, largely due to the misuse of X-ray sources in previous studies. Besides, benchmarking of device characteristics has been problematic due to experimental or data analysis issues, as well as misunderstandings of the applied equations associated with parameter definitions. In this work, we have designed and performed an instructive research work based on epitaxial beta-Ga2O3:Si and its planar Schottky detectors, measured with energy-tuneable monochromatic X-ray beams on a synchrotron beamline, clarifying the device excitation and carrier transport mechanisms with properly benchmarked device performance. In the end, we propose a set of protocols for correctly measuring and analysing the device performance. The proposed protocols are broadly applicable and can be readily extended to other semiconductor X-ray detectors.

DeepCormack: Fermi surface tomography using model-based data-driven algorithms

2026-07-14T10:10:13Z

The experimental reconstruction of the 3D two-photon momentum density (TPMD) via angular correlation of electron-positron annihilation radiation (ACAR) is a particularly useful method for studying material Fermi surfaces. It does not rely on low temperatures, UHV conditions, or strong magnetic fields, and enables the study of the spin-resolved electronic structure of materials. Yet, it remains a challenging inverse problem. Typically, 10^8 positron annihilation events are measured for 3--6 projections of the TPMD at different angles. The standard reconstruction approach is an ACAR adaptation of Cormack's method (the MCM) that leverages the inherent symmetry in the crystal's structure. However, the poor signal-to-noise ratio means collecting data of sufficient quality for Fermi surface studies can take months per sample. We present DeepCormack, a family of data-driven model-based reconstruction algorithms that augments the MCM by integrating supervised deep-learning models (CNN, MLP, and UNet) at various stages. To overcome the lack of large experimental training sets, we propose a method which leverages singular value decomposition with dynamic mode decomposition to generate realistic synthetic TPMD volumes, requiring only a single reference momentum density computed via density functional theory. On test data, DeepCormack improves reconstruction quality over MCM by about 8.5 dB PSNR at 200M counts and remains stable at reduced counts, enabling significantly faster acquisition times. Generalisation to experimental data depends strongly on how well the training distribution from the reference momentum density matches the sample. We therefore recommend pairing DeepCormack with a DFT calculation of the target material to create sample-specific training data. Our proposed method offers either much higher quality reconstructions, or enables significantly faster ones, on the order of weeks.