Disordered Systems and Neural Networks
Theory of Optimal Learning Rate Schedules and Scaling Laws for a Random Feature Model
Blake Bordelon and Francesco Mori
Setting the learning rate (LR) for a deep learning model is a critical part of successful training. Choosing LRs is often done empirically with trial and error. In this work, we explore a solvable model of optimal LR schedules for a powerlaw random feature model trained with
stochastic gradient descent (SGD). We consider the optimal schedule $\eta_T^\star(t)$ where $t$ is the current iterate and $T$ is the training horizon. This schedule is computed both as a numerical optimization problem and also analytically using optimal control theory. Our analysis reveals two regimes which we term the easy phase and hard phase. In the easy phase the optimal schedule is a
polynomial decay $\eta_T^\star(t) \simeq T^{-\xi} (1-t/T)^{\delta}$ where $\xi$ and $\delta$ depend on the properties of the features and task. In the hard phase, the optimal schedule resembles warmup-stable-decay with constant initial LR and annealing performed over a vanishing fraction of training steps. We investigate joint optimization of LR and batch size and find batch ramps can improve the wall-clock time in the easy phase. Beyond SGD, we derive optimal schedules for momentum parameter $\beta(t)$ and show that it improves the loss-scaling exponent in the hard phase. We compare our optimal schedule to various
benchmarks including (1) optimal constant learning rates $\eta_T(t) \sim T^{-\xi}$ (2) optimal power laws $\eta_T(t) \sim T^{-\xi} t^{-\chi}$, finding that our schedule achieves better rates than either of these. Our theory suggests that LR transfer across training horizon depends on the structure of the model and task. For ResNet image classification on CIFAR-5M, the learning curves exhibit hard-phase behavior where optimal base LRs are constant under sufficient annealing. GPT-2 style
transformers trained in
language modeling exhibit easy-phase behavior where optimal LRs shift even under annealing.
Read the full abstract ▾
Materials Science
Scalar machine learning of tensorial quantities – Born effective charges from monopole models
Bernhard Schmiedmayer, Angela Rittsteuer, Tobias Hilpert et al. (+1 more)
Predicting tensorial properties with machine learning models typically requires carefully designed tensorial descriptors. In this work, we introduce an alternative strategy for learning tensorial quantities based on scalar descriptors. We apply this approach to the Born effective charge tensor, showing that scalar (monopole) kernel models can successfully capture its tensorial nature by exploiting the definition of the Born effective charge tensor as the derivative of the polarisation with respect to atomic displacements. We compare this method with tensorial (dipole) kernel models, as established in our previous work, in which the tensorial structure of the Born effective charge is encoded directly in the kernel and obtained via its derivative. Both approaches are then used for charge partitioning, enabling the separation of monopole and dipole contributions. Finally, we demonstrate the effectiveness of the framework by computing finite-temperature infrared spectra for a range of complex materials.
Read the full abstract ▾
All authors Bernhard Schmiedmayer, Angela Rittsteuer, Tobias Hilpert, Georg Kresse
Soft Condensed Matter
Hydrodynamics substantially affects induced structure formation in magnetic fluids
Henning Reinken, Markus Heiber, Takeaki Araki et al. (+1 more)
Magnetorheological fluids consist of micrometer-sized
magnetic particles in a carrier
liquid. Sufficiently strong external
magnetic fields lead to the formation of string-like particle aggregates. We demonstrate that hydrodynamic interactions, that is, mutual couplings via induced flows, play a substantial role during the structuring process. They support the formation of slender chains instead of more compact clusters in the absence of mutual hydrodynamic interactions between the particles. This fundamental insight is substantial from an application perspective, due to the enormous technical importance and potential of structured magnetorheological materials.
Read the full abstract ▾
All authors Henning Reinken, Markus Heiber, Takeaki Araki, Andreas M. Menzel
Quantum Gases
Dimensional crossover of bound complexes in a two-species Bose-Hubbard lattice: Correlations and dynamics
Deepak Gaur, Koushik Mukherjee and Stephanie M. Reimann
We study the
equilibrium and nonequilibrium formation of four-particle complexes in a balanced two-species Bose-Hubbard model with repulsive intraspecies and attractive interspecies interactions. Using exact diagonalization, we characterize the transition from weakly to strongly correlated dimer and tetramer states along the one- to two-dimensional crossover in coupled-chain geometries by combining local correlation signatures with global diagnostics such as the binding energy and interspecies
entanglement entropy. We show that transverse connectivity between chains qualitatively reshapes the phase diagram, substantially enlarging the tetramer region and, in particular, stabilizing weakly bound tetramers when compared to the one-dimensional chains. By tuning the interchain hopping, we identify a transition from a degenerate
manifold of spatially separated dimers to a localized tetramer ground state, driven by the lifting of one-dimensional configurational degeneracies and an associated kinetic-energy gain. By extending our analysis to larger complexes of six and eight particles, we reveal that these binding mechanisms are robust and suggest a general pathway for the formation of higher-order complexes. Finally, we demonstrate interaction and
geometric quench protocols to dynamically prepare these complexes with high fidelity. Our results provide a microscopic framework for engineering and probing few-body bosonic bound states in tunable
lattice geometries.
Read the full abstract ▾
Materials Science · Cross-listed from cs.LG
DMFlow: Disordered Materials Generation by Flow Matching
Liming Wu, Rui Jiao, Qi Li et al. (+4 more)
The design of materials with tailored properties is crucial for technological progress. However, most deep generative models focus exclusively on perfectly ordered
crystals, neglecting the important class of disordered materials. To address this gap, we introduce DMFlow, a generative framework specifically designed for disordered
crystals. Our approach introduces a unified representation for ordered, Substitutionally Disordered (SD), and Positionally Disordered (PD)
crystals, and employs a flow matching model to jointly generate all structural components. A key innovation is a Riemannian flow matching framework with spherical reparameterization, which ensures physically valid disorder weights on the probability simplex. The vector field is learned by a novel Graph Neural Network (GNN) that incorporates physical
symmetries and a specialized message-passing scheme. Finally, a two-stage discretization procedure converts the continuous weights into multi-hot atomic assignments. To support research in this area, we release a
benchmark containing SD, PD, and mixed structures curated from the Crystallography Open Database. Experiments on
Crystal Structure Prediction (CSP) and De Novo Generation (DNG) tasks demonstrate that DMFlow significantly outperforms state-of-the-art baselines adapted from ordered
crystal generation. We hope our work provides a foundation for the AI-driven discovery of disordered materials.
Read the full abstract ▾
All authors Liming Wu, Rui Jiao, Qi Li, Mingze Li, Songyou Li, Shifeng Jin, Wenbing Huang
Soft Condensed Matter
Transport Properties of Active Particles Moving on Adjustable Networks
William G. C. Oropesa, P. de Castro, Hartmut Löwen et al. (+1 more)
Active adaptive matter has attracted considerable interest due to its rich, largely unexplained dynamics and its relevance to a wide range of synthetic and biological materials. An important subclass of such systems consists of active particles that can remodel the network in which they move. Here, we introduce a minimal yet versatile model of active particles moving on an adjustable network. In this model, particles undergo discrete run-and-tumble motion along the links of a triangular
lattice and leave behind a trail of temporarily blocked links. These closed links cannot be traversed by other particles and reopen only after a characteristic healing time. The resulting trail-mediated blocking mechanism is fundamentally distinct from more familiar interactions such as excluded-volume effects. In the high-persistence limit, we find a qualitative contrast between the two mechanisms: while steric blocking leads to reduced diffusivity with increasing persistence, trail-induced blocking causes diffusivity to increase monotonically. We characterize this fundamental difference and the associated, unexpected transport properties, and discuss potential applications of our findings.
Read the full abstract ▾
All authors William G. C. Oropesa, P. de Castro, Hartmut Löwen, Danilo B. Liarte
Disordered Systems and Neural Networks
Effect of Local Topological Changes on Resistance in Spatially-Embedded Disordered Networks
Chenxi Wang, Charles Emmett Maher and Katherine A. Newhall
Disordered materials occur naturally and also provide a broader design space than ordered or crystalline structures. We investigate a two-dimensional disordered network metamaterial constructed from a Delaunay triangulation of an underlying point cloud. Small perturbations in the point cloud induce discrete topological changes. One such change we identify is a Delaunay flip, in which two neighboring Delaunay triangles that form a convex quadrilateral structure with their common edge being one of the two quadrilateral diagonals exchange this diagonal for the other diagonal. These topological changes can cause substantial jumps in the effective resistance measured diagonally across the network, when the change is located near the source or the sink node. The jumps are explained analytically by showing that the change in effective resistance from edge removal or addition depends on the voltage drop across that edge. However, Delaunay flips have less impact on global resistance measurements and in larger networks. These local topological changes are relevant for finite-sized samples and experimentally-measurable properties such as electrical transport. Global characterizations of the network disorder or
topology lack the location-specificity of our observed effects on network transport, and thus may be inadequate for predicting certain experimentally measurable transport properties in disordered network metamaterials, highlighting the importance of localized regions in material design.
Read the full abstract ▾
Mesoscale and Nanoscale Physics · Cross-listed from quant-ph
Dicke Superradiance in Extended 2D Quantum Arrays Coupled to Metasurface Bound States in the Continuum
Daniel Eyles, Emmanuel Lassalle, Adam Stokes et al. (+2 more)
Dicke superradiance is a collective phenomenon where the emission from ensembles of
quantum emitters is coherently enhanced beyond the sum of each emitter's independent emission. Here, we propose a platform that exploits the delocalised nature of a high-Q, non-local mode supported by a dielectric metasurface (a so-called bound-state-in-the-continuum or BIC) to induce superradiant behaviour within an extended two-dimensional array of distant
quantum emitters. We show that these BIC-mediated emitter interactions can span several wavelengths, thus overcoming the traditional subwavelength separation between emitters required in free space. We further show that reaching the idealised Dicke limit is possible in this system, provided that the emitters are coupled to the BIC mode efficiently enough, as quantified through the $\beta$-factor. Moreover, we demonstrate its experimental viability by analysing its robustness to realistic experimental imperfections. This work puts forward optical metasurfaces supporting BICs as a physically viable platform for realising the upper limits of cooperative emission in physically extended
quantum emitter arrays.
Read the full abstract ▾
All authors Daniel Eyles, Emmanuel Lassalle, Adam Stokes, Ahsan Nazir, Ramón Paniagua-Domínguez
Statistical Mechanics
The Most Dispersed Subset of Random Points in $\mathbb{R}^d$
Fabio Deelan Cunden, Noemi Cuppone, Giovanni Gramegna et al. (+1 more)
Consider a population of $N$ individuals, each having $d\geq 1$ different traits, and an additive measure, called dispersion, which rewards large pairwise separations between traits. The goal is to select $M\leq N$ individuals such that their traits are as dispersed as possible. We compute analytically the full statistics (including large deviation tails) of the maximally achievable dispersion among sub-populations of size $M$ when the traits are independent and identically distributed. Two complementary approaches are developed, one based on a mean-field theory for order statistics, and the other on the replica method from the field of disordered systems. In all dimensions $d$, and for rotationally symmetric distributions, the optimal subset for large populations consists of all points lying outside a $d$-dimensional ball whose radius is determined self-consistently. For a single trait ($d=1$), the statistics of the maximal dispersion can be tackled for finite $N,M$ as well. The formulae we obtained are corroborated by numerical simulations on small instances and by heuristic
algorithms that find near-optimal solutions.
Read the full abstract ▾
All authors Fabio Deelan Cunden, Noemi Cuppone, Giovanni Gramegna, Pierpaolo Vivo
Disordered Systems and Neural Networks
Scalable platform enabling reservoir computing with nanoporous oxide memristors for image recognition and time series prediction
Joshua Donald, Ben A. Johnson, Amir Mehrnejat et al. (+5 more)
Typical mammal
brains have some form of random connectivity between
neurons. Reservoir computing, a neural network approach, uses random weights within its processing layer along with built-in recurrent connections and short-term, fading
memory, and is shown to be time and training efficient in processing spatiotemporal signals. Here we prepared a niobium
oxide-based thin film memristor device with intrinsic structural in-homogeneity in the form of random nanopores and performed computational tasks of XOR operations, image recognition, and time series prediction and reconstruction. For the latter task we chose a complex three-dimensional chaotic Lorenz-63 time series. By applying three temporal voltage waveforms individually across the device and training the readout layer with electrical current signals from a three-output physical reservoir, we achieved satisfactory prediction and reconstruction accuracy in comparison to the case of no reservoir. This work highlights the potential for scalable, on-chip devices using all-
oxide reservoir systems, paving the way for energy-efficient neuromorphic electronics dealing with time signals.
Read the full abstract ▾
All authors Joshua Donald, Ben A. Johnson, Amir Mehrnejat, Alex Gabbitas, Arthur G. T. Coveney, Alexander G. Balanov, Sergey Savel'ev, Pavel Borisov
Materials Science
Cross-sectional helium irradiation reveals interface-controlled bubble evolution in Cr/CrAlSiN multilayer coatings on zirconium alloys
Renda Wang, Xue Bai, Xueliang Pei et al. (+9 more)
The irradiation stability of Cr based protective coatings on zirconium
alloys is critical for the development of accident-tolerant fuel claddings. However, conventional surface irradiation often produces shallow, nonuniform damage, obscuring interfacial behavior. In this study, we perform cross-sectional He irradiation to directly examine the interfacial response and He bubble
evolution across Cr monolayer and Cr and CrAlSiN multilayer coatings on Zr substrates. Irradiation was carried out at 500 C and 750 C to doses of 2 and 3 dpa, enabling a direct comparison of temperature-dependent microstructural
evolution. In the Cr monolayer, He implantation produced a homogeneous distribution of nanoscale bubbles throughout the damaged region and large cavities at the Cr and Zr interface, indicating severe Kirkendall-type voiding and interfacial decohesion at elevated temperature. In contrast, the Cr/CrAlSiN multilayer exhibited a periodically modulated bubble distribution, with bubble fragmentation and transformation into nanoscale platelets at CrAlSiN interfaces. A N-enriched Zr(N) interlayer formed spontaneously at the CrAlSiN and Zr interface, effectively suppressing bubble accumulation and interdiffusion. The nanochannel interfaces acted as He sinks and
diffusion barriers, enhancing interfacial bonding and mitigating swelling. This work demonstrates that cross-sectional ion irradiation is a powerful approach for probing interfacial stability in multilayer systems, offering new insights into He-defect interactions and radiation tolerance engineering at buried interfaces. The findings highlight the potential of Cr and CrAlSiN multilayers as advanced coating architectures for high-temperature
nuclear environments.
Read the full abstract ▾
All authors Renda Wang, Xue Bai, Xueliang Pei, Sijie Liu, Chunfan Liu, Ping Yu, Bingsheng Li, Nabil Daghbouj, Tomas Polcar, Fanping Meng, Fangfang Ge, Qing Huang
Quantum Gases
Density Modulations of Zero Sound
Leonardo Pisani
We study the density modulation of an interacting
Fermi gas caused by the uniform motion of an
impurity at zero temperature. For strong enough interaction among
Fermi atoms, the modulation propagates thanks to the excitation of the collective zero sound mode if the
impurity speed is above the zero sound threshold. We are able to assess, via a semi-analytic evaluation, the extent of the zero sound contribution to the density oscillation over and above the incoherent background of particle-hole excitations. Given the strong dependence of the results on the features of the gas interaction potential, we also analyze how they vary depending on its strength, range and shape.
Read the full abstract ▾
Materials Science
Structures of iron and cobalt bimetallic clusters for optimized chemical vapor deposition growth of single-walled carbon nanotubes
Qingmei Hu, Daniel Hedman, Ya Feng et al. (+17 more)
We investigate iron-cobalt (Fe-Co)
alloys as a representative high-performance catalyst system for SWCNT growth in a systematic manner by combining chemical vapor deposition (CVD) experiments with chirality-resolved spectroscopic analysis, as well as molecular dynamics (MD) simulations based on density functional theory-derived machine learning force fields while varying the Fe-Co ratio. Using zeolite-based SWCNTs prepared by alcohol CVD, absorption and photoluminescence spectroscopy, together with two-dimensional excitation-emission fitting was employed to quantify chirality-specific growth efficiency. Two distinct growth regimes were identified. At a relatively low temperature of 600 C, pure Co exhibits the highest catalytic activity, promoting efficient growth of small-diameter (0.7-0.9 nm) SWCNTs. In contrast, at 850 C, the Fe0.75Co0.25
alloy shows a pronounced enhancement in growth efficiency compared with pure Fe, pure Co, and other Fe-Co compositions, also yielding larger diameter tubes (0.9-1.1nm). Similar growth behavior was observed on SiO2 substrates, enabling detailed transmission electron microscopy analysis of catalyst nanoparticles. Electron microscopy and energy-dispersive X-ray spectroscopy reveal that high SWCNT yields correlate with the formation of small, uniform Fe-Co nanoparticles with Co-enriched surfaces, in excellent agreement with MD simulations. Lastly, MD results are summarized in a composition-diameter phase diagram that rationalizes the experimentally observed growth trends. The exceptional performance of the Fe0.75Co0.25 catalyst at high temperature is attributed to the stabilization of small and uniform catalyst clusters, providing mechanistic insight into the synergistic roles of
alloy composition and temperature in SWCNT growth.
Read the full abstract ▾
All authors Qingmei Hu, Daniel Hedman, Ya Feng, Wanyu Dai, Daisuke Asa, Aina Fito-Parera, Yixi Yao, Yongjia Zheng, Kaoru Hisama, Gunjan Auti, Hirofumi Daiguji, Shohei Chiashi, Dmitry Levshov, Wim Wenseleers, Keigo Otsuka, Yan Li, Christophe Bichara, Sofie Cambre, Rong Xiang, Shigeo Maruyama
Materials Science
Automated Extraction of Multicomponent Alloy Data Using Large Language Models for Sustainable Design
Aravindan Kamatchi Sundaram, Mohit Chakraborty, Sai Mani Kumar Devathi et al. (+2 more)
The design of sustainable materials requires access to materials performance and sustainability data from literature corpus in an organized, structured and automated manner. Natural
language processing approaches, particularly large
language models (LLMs), have been explored for materials data extraction from the literature, yet often suffer from limited accuracy or narrow scope. In this work, an LLM-based pipeline is developed to accurately extract
alloy-related information from both textual descriptions and tabular data across the literature on high-
entropy (or multicomponent)
alloys (HEA). Specifically two databases with 37,711 and 148,069 entries respectively are retrieved; one from the literature text, consisting of
alloy composition, processing conditions, characterization methods, and reported properties, and other from the literature tables, consisting of property names, values, and units. The pipeline enhances materials-domain sensitivity through
prompt engineering and
retrieval-augmented generation and achieves F1-scores of 0.83 for textual extraction and 0.88 for tabular extraction, surpassing or matching existing approaches. Application of the pipeline to over 10,000 articles yields the largest publicly available multicomponent
alloy database and reveals compositional and processing-property trends. The database is further employed for sustainability-aware materials selection in three application domains, i.e., lightweighting, soft
magnetic, and corrosion-resistant, identifying multicomponent
alloy candidates with more sustainable production while maintaining or exceeding
benchmark performance. The pipeline developed can be easily generalized to other class of materials, and assist in development of comprehensive, accurate and usable databases for sustainable materials design.
Read the full abstract ▾
All authors Aravindan Kamatchi Sundaram, Mohit Chakraborty, Sai Mani Kumar Devathi, B. Pabitramohan Prusty, Rohit Batra
Statistical Mechanics
Emergent Hawking Radiation and Quantum Sensing in a Quenched Chiral Spin Chain
Nitesh Jaiswal and S. Shankaranarayanan
We investigate the emergence and detection of Hawking radiation (HR) in a 1D
chiral spin chain model, where the gravitational collapse is simulated by a sudden
quantum quench that triggers a horizon-inducing phase transition. While our previous work Jaiswal [2025] established that this model mimics BH formation conditions even when the Hoop
conjecture is seemingly violated, we here focus on the resulting stationary radiation spectrum and its detectability. By mapping the
spin chain dynamics to a Dirac
fermion in a curved (1 + 1)-dimensional spacetime, we analyze the radiation using two complementary approaches: field-theoretic modes and operational
quantum sensors. First, using localized Gaussian wave packets to model realistic detectors, we find that the radiation spectrum exhibits deviations from the ideal Planckian form, analogous to frequency-dependent greybody factors, while retaining robust Poissonian statistics that signal the loss of formation-scale information. Second, we introduce a
qubit coupled to the chain as a stationary Unruh-DeWitt detector. We demonstrate that the
qubit functions as a faithful
quantum sensor of the Hawking temperature only in the weak-coupling regime, where its population dynamics are governed solely by the bath spectral density. In the strong-coupling limit, the probe thermalizes with the global environment, obscuring the horizon-induced thermal signature. These results provide a clear operational protocol for distinguishing genuine analog HR from environmental noise in
quantum simulation platforms.
Read the full abstract ▾
Materials Science
Nonreciprocal topological kink-wave propagation in mechanical metamaterials
Brahim Lemkalli, Qingxiang Ji, Jingyi Zhang et al. (+3 more)
Nonlinear mechanical metamaterials can exhibit emergent transport phenomena that mimic topological protection without relying on linear band
topology. Here, we realize a bifurcation-induced nonreciprocal
lattice that supports robust propagation of elastic kink waves. Each unit is a prestrained, hinged-beam circulator that develops angular momentum bias during snap-through transitions between buckling states, producing an effective breaking of time reversal
symmetry. Coupling such units into a hexagonal array yields a mechanically
chiral network where localized
soliton-like excitations propagate unidirectionally along interfaces and edges, immune to sharp bends. We demonstrate non-dispersive kink transport governed by a SineGordon type field whose effective bias encodes mechanical chirality. This framework bridges bifurcation dynamics and nonreciprocal transport, establishing a nonlinear route toward topological like mechanical functionality without
magnetic or gyroscopic bias.
Read the full abstract ▾
All authors Brahim Lemkalli, Qingxiang Ji, Jingyi Zhang, Richard Craster, Johan Christensen, Muamer Kadic
Quantum Gases
Physics-Informed Neural Networks for the Quantum Droplets in Binary Bose-Einstein Condensates
Dongshuai Liu, Boris A. Malomed and Wen Zhang
Physics-Informed Neural Networks (PINNs), which integrate deep learning with physical prior knowledge, have proven to be a powerful tool for studying the dynamics of high-dimensional nonlinear systems. The present work utilizes PINNs to analyze the existence and
evolution of
quantum droplets (QDs) in a binary Bose-Einstein condensate (BEC), revealing the ability of this technique to accurately predict structural features of the QDs, their multipeak profiles, and dynamical behavior. The stable
evolution of multipole QDs is thus demonstrated. Comparing different network architectures, including the training time, loss values, and $\mathbb{L_{2}}$ error, PINNs accurately predict specific dynamical characteristics of QDs. Furthermore, the PINN robustness is evaluated by the application of PINN to parameter-discovery tasks, considering both clean training data and data contaminated by $1\%$ random noise. The results highlight the efficiency of PINNs in modeling complex
quantum systems and extracting reliable parameters under the noisy conditions.
Read the full abstract ▾
Mesoscale and Nanoscale Physics
Optical signatures of -1/3 fractional quantum anomalous Hall state in twisted MoTe2
Haiyang Pan, Shunshun Yang, Yuzhu Wang et al. (+9 more)
The discovery of fractional charge excitations in new platforms offers crucial insights into strongly correlated
quantum phases. While a range of fractional
quantum anomalous
Hall (FQAH) states have recently been observed in two-dimensional twisted moire systems, the theoretically anticipated filling factor v = -1/3 FQAH state has remained elusive, with debates centering on its nature of charge density wave or a topological Chern
insulator. Here, we report the optical detection of a v = -1/3 FQAH state in twisted MoTe2
bilayers. Using photoluminescence (PL) and reflective
magnetic circular dichroism (RMCD) techniques, we identify
ferromagnetic states at filling factors v = -1, -2/3, and -1/3, all tunable by a vertical electric field. The corresponding Curie temperatures are approximately 11 K, 3.5 K, and 2.4 K, respectively. The -1/3 state emerges over a narrower electric field range and a lower temperature compared to the integer and other fractional states, indicating its fragile nature that may lead to its absence in previous reports. Notably, the PL spectra at v = -1/3 disperse as the out-of-plane
magnetic field increases, consistent with a nontrivial topological origin. Theoretical calculations based on the exact diagonalization method further support the interpretation of this topologically non-trivial state.
Read the full abstract ▾
All authors Haiyang Pan, Shunshun Yang, Yuzhu Wang, Xiangbin Cai, Wei Wang, Yan Zhao, Kenji Watanabe, Takashi Taniguchi, Linlong Zhang, Youwen Liu, Bo Yang, Weibo Gao
Statistical Mechanics
Control protocols for harmonically confined run-and-tumble particles
Marco Baldovin and Alessandro Manacorda
Run-and-tumble particles constitute one of the simplest models of self-propelled active matter, and provide an ideal playground to the understanding of out-of-
equilibrium systems. We consider an idealized setup where one such particle is subject to a harmonic confining potential, and an external
agent can vary in time the tumbling rate and the strength of the trap. We search for time-dependent control protocols steering the system between assigned end states, in a prescribed time interval. To this aim, we propose a description of the dynamics, alternative to the usual ones, in the form of an infinite set of ordinary differential equations. Solutions based on a suitable closure of such hierarchy, which we expect to hold true in the limit of long protocol duration, are discussed and compared with numerical simulations. We also look for the protocol completing the task with the minimal work, on average: the problem can be tackled analytically, again in the regime of slow (but not quasi-static) transformations. The solution provides insightful intuition on the optimal strategies for the control of active matter systems.
Read the full abstract ▾
Statistical Mechanics · Cross-listed from physics.plasm-ph
Nonlinear Dynamical Friction from the Doppler-Shifted Equilibrium Memory Kernel
N. R. Sree Harsha, Zhenyuan Yu, Chuang Ren et al. (+2 more)
We present a statistical mechanics framework for modeling
equilibrium friction coefficients using the Generalized Langevin Equation (GLE). We show that the kernel, obtained via the Fluctuation-Dissipation
Theorem (FDT) from the
stochastic force autocorrelation measured in a thermal
equilibrium state, is sufficient to model the dynamics of the system in a Non-
Equilibrium Steady State (NESS). This approach provides a computationally efficient path to modeling complex
equilibrium friction problems. We apply this framework to the canonical problem of test particle drag in a uniform
plasma. The GLE formalism is shown to naturally capture non-Markovian phenomena through the moments of the kernel, including an effective mass renormalization and oscillatory relaxation. We demonstrate that the standard Chandrasekhar stopping power formula arises naturally as the Markovian limit of this
equilibrium memory kernel. These theoretical predictions are quantitatively validated by direct Particle-in-
Cell simulations, which confirm the predicted oscillatory structure of the
memory kernel. This work thus establishes a practical method for predicting
equilibrium friction properties from first-principles
equilibrium simulations.
Read the full abstract ▾
All authors N. R. Sree Harsha, Zhenyuan Yu, Chuang Ren, Virginia Billings, Michael Huang
Materials Science
Fermi surface geometry and momentum dependent electron-phonon coupling drive the charge density wave in quasi-1D ZrTe$3$
Josu Diego and Matteo Calandra
ZrTe$_3$ is a prototypical quasi-one-dimensional compound undergoing a charge density wave transition via a very sharp Kohn anomaly in
phonon momentum space. While
Fermi surface geometry has long been considered the primary driver of the instability, a full understanding of the
lattice dynamics and electron-
phonon role has remained elusive. Our first principles calculations in the high-
symmetry phase show that the
Fermi surface is correctly reproduced only when the Hubbard interaction on the Te $5p$ orbitals is included, which in turn is essential for the appearance of a soft harmonic
phonon mode at the CDW wavevector. Analyzing the mode and momentum dependence of the electron-
phonon coupling, we find that its variations with
phonon momentum dominate over electronic effects. These results identify unambiguously the CDW origin in ZrTe$_3$ as a cooperative effect of
Fermi surface geometry and momentum-dependent electron-
phonon coupling, with the latter playing the leading role. We further determine the atomic structure in the low-
symmetry CDW phase, revealing a nonchiral modulation. The mechanisms revealed in our work are directly relevant to other quasi-1D systems, including trichalcogenides and compounds hosting Peierls-like chains.
Read the full abstract ▾
Materials Science
Triple Junctions as Dislocation-Like Defects: The Role of Grain Boundary Crystallography Revealed by Experiment and Atomistic Simulation
Tobias Brink, Saba Saood, Peter Schweizer et al. (+2 more)
Grain boundary networks and their
evolution are strongly influenced by triple junctions. The defect nature of these line defects significantly affects the properties of the network, but they have not been fully characterized to date. Here, we use scanning transmission electron microscopy combined with atomistic computer simulations to investigate a triple junction at the atomic scale in an Al thin film with {111} texture. Using sampling methods, we were able to construct the same junction structure as in the experiment within a computer model. We present a technique to calculate the Burgers vector of the triple junction. This allows us to connect the junction's dislocation character to the microscopic degrees of freedom of the joining grain boundaries. The junction line energy in the computer model can then be calculated using an embedded atom method potential. It follows the same laws as a bulk dislocation. Finally, we discovered a range of possible triple junctions for the observed grain boundaries, which vary in the magnitude of their Burgers vector. Interestingly, the experimentally observed junction is not the one with the smallest possible Burgers vector and energy. This suggests that the kinetics of transforming the junction line are likely too slow to be driven by the small energy contribution of the triple junction.
Read the full abstract ▾
All authors Tobias Brink, Saba Saood, Peter Schweizer, Jörg Neugebauer, Gerhard Dehm
Materials Science
Strain tunable anomalous Hall and Nernst conductivities in compensated ferrimagnetic Mn$_3$Al
Guihyun Han, Minkyu Park and S. H. Rhim
The tunability of anomalous
Hall and Nernst conductivities is investigated in the compensated ferrimagnet Mn$_3$Al under isotropic strain ($\eta$) and chemical potential variation using first-principles calculations. At a chemical potential of $\mu = -0.3$ eV, three distinct topological features – Weyl points, nodal lines, and gapped nodal lines – are simultaneously realized along high-
symmetry directions of the Brillouin zone in the framework of
magnetic space group. The anomalous
Hall conductivity (AHC) is found to be predominantly governed by the Berry
curvature in the $k_y k_z$ plane and can be enhanced significantly under tensile strain, reaching $-1200$ $(\Omega~\mathrm{cm})^{-1}$. On the other hand, the anomalous Nernst conductivity (ANC) shows a sign change near the
Fermi level and whose magnitude increases at $\mu = -0.3$ eV with quasi-quadratic strain dependence. Regardless of strain, the underlying bands and
Fermi surface structures remain robust, while the distribution and magnitude of Berry
curvature evolve substantially. These results underscore the potential of Mn$_3$Al, a compensated ferrimagnet, as a platform for Berry
curvature engineering via strain and doping.
Read the full abstract ▾
Materials Science
Microscopic Origin of Polarization-Controlled Magnetization Switching in FePt/BaTiO$_3$
Qurat-ul-ain, Thi H. Ho, Soon Cheol Hong et al. (+2 more)
Electric-field driven magnetization switching in FePt/BaTiO$_3$ (001) is demonstrated through first-principles calculations. The
magnetic easy axis of FePt layer undergoes a transition from in-plane to perpendicular direction upon ferroelectric polarization reversal, a process sensitively controlled by epitaxial strain with threshold strain strain($\eta$) $\eta\approx\%$. At this phenomena, a large interfacial magnetoelectric coupling ($\alpha_I = 3.6 \times 10^{-10}$ G$\cdot$cm$^2$/V) is responsible, stemming from the orbital reconstruction. In particular, the redistribution of Pt-$d$ orbital occupancy alters
spin-orbit coupling, thereby tuning the competition between
magnetic anisotropy ($K_i$) and magnetoelastic energy ($b_1$). Our work clarifies the fundamental physics of strain-engineered magnetoelectricity and suggests a concrete pathway for designing ultra-low-power voltage-controlled
magnetic memory.
Read the full abstract ▾
All authors Qurat-ul-ain, Thi H. Ho, Soon Cheol Hong, Dorj Odkhuu, S. H. Rhim
Materials Science
Potential-Induced Dynamic Coordination of Nonmetal Atoms Directly Bound to Metal Centers in Graphene-Embedded Single-Atom Catalysts and Its Implications
Jiahang Li, Suhang Li, Chong Yan et al. (+3 more)
Electrode-potential-induced dynamic coordination is an essential factor governing the performance of
graphene-embedded single-atom catalysts (SACs). While previous studies have primarily centered on structural dynamics at the metal site, the response of its coordinated nonmetal atoms remains largely unexplored. Here, using Ni SACs with mixed nitrogen/carbon coordination (NiN4-xCx) as representatives, we investigate potential-driven hydrogenation of metal-center-coordinated nonmetallic atoms through constant-potential density functional theory and ab initio molecular dynamics. We find that the C sites directly bound to Ni undergo potential-driven hydrogenation, whereas hydrogenation at N sites is thermodynamically unfavorable. Taking NiNC3 as a representative system, we demonstrate that these hydrogenation processes proceed with accessible kinetic barriers and obey the Brønsted-Evans-Polanyi relation. The resulting dynamic coordination reshapes the Ni 3d and dz2 orbitals, modulates the stability of the active center, and weakens molecular adsorption through combined electronic and steric effects. These findings reveal that electrode potential and solvent not only regulate the metal center but also dynamically reconfigure its coordination environment, offering novel mechanistic insights into potential-induced coordination dynamics and guiding the rational design of coordination-engineered SACs.
Read the full abstract ▾
All authors Jiahang Li, Suhang Li, Chong Yan, Qinzhuang Liu, Jiajun Yu, Dongwei Ma
Materials Science
Electronic States, Spin-Orbit Coupling and Magnetism in Germanium 60{\deg} Dislocations
Veronica Regazzoni, Fabrizio Rovaris, Anna Marzegalli et al. (+2 more)
Defects in
semiconductors have recently attracted renewed interest owing to their potential in novel
quantum applications. Here we investigate the electronic and
magnetic properties induced by 60{\deg} dislocations in Ge. Using large-scale DFT calculations, we determine the band structure for both the shuffle and glide sets in their lowest-energy configurations. The band structure for the shuffle set reveals defect-induced dispersive bands localized within the band gap near the $\Gamma$ point, whereas for the glide set, we observe strong overlap with the conduction band. Defect-induced band splitting evident away from $\Gamma$ reveals Rashba-Dresselhaus
spin-orbit coupling, an effect previously reported only for screw dislocations. Remarkably, we find evidence that specific dislocation arrangements can stabilize
antiferromagnetic ordering with sizable local
magnetic moments and considerable exchange splitting between opposite
spin states. These results uncover rich physics in Ge dislocations through the combination of
spin-orbit coupling and
magnetic ordering, potentially enabling novel defect-based functionalities in Ge devices.
Read the full abstract ▾
All authors Veronica Regazzoni, Fabrizio Rovaris, Anna Marzegalli, Francesco Montalenti, Emilio Scalise
Strongly Correlated Electrons
Emergent Coherence at the Edge of Magnetism: Low-Doped La2-xSrxCuO4+delta Revisited
E.Yu. Beliayev, Y.K. Mishra, I.A. Chichibaba et al. (+3 more)
The La2-xSrxCuO4+delta (LSCO) system provides a unique experimental setting for exploring how magnetism, superconductivity, and disorder jointly shape charge transport in a doped Mott
insulator. Transport measurements in lightly doped and oxygen-enriched LSCO reveal a strongly insulating normal state governed by variable-range hopping, accompanied by pronounced nonlinear current-voltage characteristics and, at low temperatures, current-induced negative differential resistance. With increasing carrier concentration, these features evolve into regimes characterized by granular and percolative superconductivity near the threshold of bulk superconductivity and, eventually, into a homogeneous strange-metal state close to optimal doping. Throughout this
evolution, the transport response shows marked sensitivity to disorder, electronic inhomogeneity, and external control parameters, such as bias current and
magnetic field. Rather than reflecting a sequence of sharply distinct phases, the observed transport regimes form a continuous crossover from a localization-dominated insulating state to granular superconductivity and further to a coherent metallic state. This crossover is driven primarily by the progressive enhancement of electronic screening, inter-region coupling, and superconducting connectivity, rather than by abrupt changes in the underlying microscopic scattering mechanisms. Taken together, the available transport data provide a coherent experimental basis for understanding how disorder and mesoscale electronic inhomogeneity organize charge transport and superconductivity across the LSCO phase diagram, underscoring the central role of percolation and nonequilibrium effects in underdoped cuprates.
Read the full abstract ▾
All authors E.Yu. Beliayev, Y.K. Mishra, I.A. Chichibaba, I.G. Mirzoiev, V.A. Horielyi, A.V. Terekhov
Materials Science
Machine Learning-Driven Crystal System Prediction for Perovskites Using Augmented X-ray Diffraction Data
Ansu Mathew, Ahmer A. B. Baloch, Alamin Yakasai et al. (+3 more)
Prediction of
crystal system from X-ray diffraction (XRD) spectra is a critical task in materials science, particularly for perovskite materials which are known for their diverse applications in photovoltaics, optoelectronics, and catalysis. In this study, we present a machine learning (ML)-driven framework that leverages advanced models, including Time Series Forest (TSF), Random Forest (RF), Extreme Gradient Boosting (XGBoost), Recurrent Neural Network (RNN), Long Short-Term
Memory (LSTM), Gated Recurrent Unit (GRU), and a simple feedforward neural network (NN), to classify
crystal systems, point groups, and space groups from XRD data of perovskite materials. To address class imbalance and enhance model robustness, we integrated feature augmentation strategies such as Synthetic Minority Over-sampling Technique (SMOTE), class weighting, jittering, and spectrum shifting, along with efficient data preprocessing pipelines. The TSF model with SMOTE augmentation achieved strong performance for
crystal system prediction, with a Matthews correlation coefficient (MCC) of 0.9, an F1 score of 0.92, and an accuracy of 97.76%. For point and space group prediction, balanced accuracies above 95% were obtained. The model demonstrated high performance for
symmetry-distinct classes, including cubic
crystal systems, point groups 3m and m-3m, and space groups Pnma and Pnnn. This work highlights the potential of ML for XRD-based structural characterization and accelerated discovery of perovskite materials
Read the full abstract ▾
All authors Ansu Mathew, Ahmer A. B. Baloch, Alamin Yakasai, Hemant Mittal, Vivian Alberts, Jayakumar V. Karunamurthy
Materials Science
ZnCdO:Eu Epitaxially Grown Alloys for Self-Powered Ultrafast Broadband Photodetection
Igor Perlikowski, Eunika Zielony, Aleksandra Wierzbicka et al. (+3 more)
Photodetectors (PDs) are essential in imaging, communication, and sensing technologies. However, their reliance on external power makes them energy-consuming. This creates a strong need for self-powered PDs as a sustainable alternative. ZnO is a promising
semiconductor material due to its pyroelectric properties, stemming from non-centrosymmetric wurtzite
crystal structure, enabling the pyro-phototronic effect that enhances response speed. Properties of ZnO can be tailored via alloying and doping. Thus, this work explores thin layers of ZnCdO:Eu random
alloys grown by molecular beam epitaxy (MBE) on silicon substrates, with varying Cd content. The study shows that doping with Eu notably affects growth kinetics, promoting strong [0001] orientation preference. Moreover, photoluminescence measurements confirm the successful incorporation of Eu3+ ions into the structure. Electrical measurements show that the introduction of Cd eliminates the problem of Schottky barrier formation on the ZnO/Si interface. The n-ZnCdO:Eu/p-Si junctions exhibit rectifying behavior and generate photocurrent across 380-1150 nm wavelength range without external electrical bias. Utilizing the pyro-phototronic effect, these devices achieved ultrafast response times: rise time below 10 us and decay time below 5 us for 405 nm and 650 nm illumination - placing them among the fastest self-powered
oxide-based detectors that do not rely on additional performance-enhancing layers.
Read the full abstract ▾
All authors Igor Perlikowski, Eunika Zielony, Aleksandra Wierzbicka, Anastasiia Lysak, Rafal Jakiela, Ewa Przezdziecka
Disordered Systems and Neural Networks · Cross-listed from cs.LG
Theory of Speciation Transitions in Diffusion Models with General Class Structure
Beatrice Achilli, Marco Benedetti, Giulio Biroli et al. (+1 more)
Diffusion Models generate data by reversing a
stochastic diffusion process, progressively transforming noise into structured samples drawn from a target distribution. Recent theoretical work has shown that this backward dynamics can undergo sharp qualitative transitions, known as speciation transitions, during which trajectories become dynamically committed to data classes. Existing theoretical analyses, however, are limited to settings where classes are identifiable through first moments, such as mixtures of Gaussians with well-separated means. In this work, we develop a general theory of speciation in
diffusion models that applies to arbitrary target distributions admitting well-defined classes. We formalize the notion of class structure through Bayes classification and characterize speciation times in terms of free-
entropy difference between classes. This criterion recovers known results in previously studied Gaussian-mixture models, while extending to situations in which classes are not distinguishable by first moments and may instead differ through higher-order or collective features. Our framework also accommodates multiple classes and predicts the existence of successive speciation times associated with increasingly fine-grained class commitment. We illustrate the theory on two analytically tractable examples: mixtures of one-dimensional Ising models at different temperatures and mixtures of zero-mean Gaussians with distinct covariance structures. In the Ising case, we obtain explicit expressions for speciation times by mapping the problem onto a random-field Ising model and solving it via the replica method. Our results provide a unified and broadly applicable description of speciation transitions in
diffusion-based generative models.
Read the full abstract ▾
All authors Beatrice Achilli, Marco Benedetti, Giulio Biroli, Marc Mézard