Geophysics
Post-FWI Injection of Learned Priors Using a Flow Matching Model
Hao Zhang and Tariq Alkhalifah
Full Waveform Inversion (FWI) is a powerful tool for subsurface velocity reconstruction but remains highly ill-posed, sensitive to acquisition limitations, often requiring some form of regularization to reduce artifacts and enhance resolution. While recent developments have shown that generative models can inject learned priors directly into the FWI optimization process, such approaches typically require additional, computationally expensive inversion iterations. In this study, we propose a post-FWI refinement strategy based on a Flow Matching (FM) generative model, which leverages learned geological priors without re-running FWI. The method guides the deterministic generative process using the FWI result, as well as well logs, if available. Synthetic and field data experiments demonstrate that we can inject well information and our geological expectations (prior) into the provided FWI result, and thus, we can effectively enhance its resolution and geological quality. In fact, the well prior even managed to alter the model depth to fit the well information, which is a form of correcting for depth misties.
Read the full abstract ▾
Applied Physics · Cross-listed from cond-mat.mtrl-sci
Elemental Germanium Phase-Change Memory
Till Zellweger, Marko Mladenović, Kevin Portner et al. (+7 more)
Phase-change
memory (PCM) is a mature technology for fast, scalable, non-volatile data storage, with applications spanning embedded
memory, as well as in-
memory and neuromorphic computing. PCM predominantly relies on chalcogenide
alloys, with $\mathrm{Ge_2Sb_2Te_5}$ (GST) as the industry standard. Yet in these
alloys, the individual Ge, Sb, and Te atoms redistribute upon cycling, causing
stochastic operation and ultimately device failure. To address this issue, elemental antimony was proposed as a PCM material, but it exhibits a metastable amorphous state that prevents reliable data retention. Moreover, tellurium and antimony can contaminate complementary metal-
oxide-
semiconductor (CMOS) production lines or act as unintended dopants, restricting manufacturing of PCM to dedicated fabs. Here we introduce elemental germanium (Ge) as a CMOS-native phase-change material that overcomes these fundamental limitations. In a vertical PCM
cell architecture, Ge enables sub-nanosecond crystallization (240 ps, 40 times faster than GST), non-volatile data storage with excellent thermal stability ($>$110 {\deg}C for 10 years vs. $\sim$87 {\deg}C for GST), and a resistance drift coefficient approximately 60% lower than in GST. These results establish pure Ge, a standard
semiconductor, as an alternative to chalcogenide phase-change materials, achieving superior performance in key metrics and enabling phase-change
memory to be fabricated in standard
semiconductor facilities.
Read the full abstract ▾
All authors Till Zellweger, Marko Mladenović, Kevin Portner, Christoph Weilenmann, Michael Stiefel, Hanglin He, Klemens Bauer, Luiz Felipe Aguinsky, Mathieu Luisier, Alexandros Emboras
Optics
Nonreciprocal subdiffraction imaging with staggered gyromagnetic photonic crystals
Rui Ding, Tianshu Zhang, Jiarui Yu et al. (+4 more)
Subdiffraction imaging serves as a novel technique to enhance the optical imaging resolution, where a backscattering-free approach remains so far unavailable. Here, we theoretically predict and experimentally demonstrate that the gyromagnetic photonic
crystals applied with staggered
magnetic fields support nonreciprocal light propagation beyond the diffraction limit. Broadband subdiffraction imaging was experimentally observed to span the frequency window traversing a pair of spectrally separated Dirac points, where near-flat equi-frequency contours (EFCs) emerged while experiencing an enforced shape transition. Our findings establish a practical paradigm for constructing backscattering-immune super-resolution imaging systems based on topological photonic
crystal platforms.
Read the full abstract ▾
All authors Rui Ding, Tianshu Zhang, Jiarui Yu, Namitha Nandakumar, Quanlong Yang, Mudi Wang, D.Y. Wang
Physics and Society
Social learning drives underprioritization of collective challenges
Russ Yoon and Vicky Chuqiao Yang
Societies often struggle to prioritize important challenges in a timely manner, with substantial costs from delayed action on issues like
climate change and pandemic mitigation. A persistent puzzle is that broad concern on issues often fails to translate into collective priority. We argue that a key driver lies in how concern is formed across competing issue domains. Some issues depend heavily on social learning, where individuals infer importance from others, often because direct experience is limited. Others depend more on individual learning from firsthand experience. We develop a dynamic model in which two subgroups form issue-specific concerns through individual and social learning, and these concerns are aggregated into collective priority. The model yields three insights. First, with two issues of equal objective severity, the one that depends more on social learning tends to be underprioritized when both issues are severe. Second, gradual increases in severity delay reprioritization of the issue, with the delay growing as reliance on social learning increases. Third, this bias can be reduced by reducing social learning or by increasing intergroup learning beyond a critical threshold. These results offer a general mechanism for why severe problems can remain neglected in collective action despite widespread concern, and why intergroup interaction or experiential simulations may help align collective priorities with objective risks.
Read the full abstract ▾
Biological Physics
Plasma-activated water and phenolic compounds: A potent combined strategy against yeast resilience
Bernard Gitura Kimani, Ramin Mehrabifard, Oleksandr Galmiz et al. (+1 more)
Effective fungal control is essential to prevent spoilage, contamination, and infections in food, healthcare, and industrial environments. This study explores the antifungal activity of
plasma-activated water (PAW), generated by transient spark (TS) electrical discharge, combined with natural phenolic bioactive compounds against the planktonic growth, biofilm formation, and surface adhesion of the yeasts Saccharomyces cerevisiae and Wickerhamomyces anomalus. The phenolic compounds were tested at concentrations of 1 or 2 mg/mL for planktonic growth after determining their minimum inhibitory concentrations. Biofilms were grown on sterile microscope slides for 48 h, stained with acridine orange or calcofluor, and examined using fluorescence microscopy. PAW was used immediately after the
plasma treatment of tap water to prepare PAW-phenolic solutions. The combinations of PAW with cinnamic acid and vanillin demonstrated the strongest antiyeast activity against planktonic growth, biofilm formation, and surface adhesion. Biofilms exposed to the PAW-phenolic combinations were notably fragmented, and the residual
cells exhibited structural damage. Complete inhibition of biofilm formation was observed when 500 microg/mL cinnamic acid was combined with PAW. These findings demonstrate the potential of combining PAW with selected phenolic compounds as an effective and sustainable strategy for controlling fungal growth.
Read the full abstract ▾
All authors Bernard Gitura Kimani, Ramin Mehrabifard, Oleksandr Galmiz, Zdenko Machala
Applied Physics
Variable Emissivity Modeling for Sustainable Lunar Surface Habitats
S. Keller, S. Stewart, S. Taylor et al. (+1 more)
Lunar habitats will be one of the first platforms to enable long-term human presence beyond Low Earth Orbit. These structures act as a stepping stone for exploring our
solar system while simultaneously enabling lunar resource utilization, low-energy cryopreservation, and various other applications. These habitats must be designed to withstand the extreme thermal variation of the lunar surface caused by the changing orientation with respect to the Sun and Earth. White paints and multi-layered insulation are conventionally used to minimize
solar heating, yet this approach is static and results in a structure that requires internal heating to survive lunar night. An adaptive approach to control absorbed and emitted radiation allows for highly efficient daytime cooling and improved nighttime heat retention. Louvers and shutters have been employed to switch between high- and low-emissivity states; however, this approach relies on ensuring moving parts are resilient to
dust contamination. Alternatively, variable emissivity materials are a solid-state solution with no moving parts. The emissivity of these materials can be switched passively based on surface temperature, or actively as a result of applied voltage. Despite their potential to reduce power consumption and increase thermal stability, variable emissivity materials have yet to be explored on the lunar surface. We first present a finite element modeling approach to predict the thermal performance of simplified habitats in realistic lunar environments. We then demonstrate the benefit of variable emissivity materials for thermal stability and lunar night survival by comparing them to traditional constant emissivity coatings. By using variable emissivity materials, we envision near-constant temperature lunar habitats with significantly reduced internal heating requirements.
Read the full abstract ▾
All authors S. Keller, S. Stewart, S. Taylor, O. Ilic
Computational Physics
Tunable mesoscopic numerical model for bacterial biofilms
J. Martin-Roca, B. Wu-Zhang, J. Oller et al. (+2 more)
We present a tunable mesoscale model to provide a basis for future rheological calculations of of bacterial biofilms, explicitly incorporating reversible crosslinking within the extracellular polymeric substance (EPS)
matrix. Using a Dissipative Particle Dynamics framework combined with a Gillespie-inspired
algorithm, bonds between
polymers and bacteria dynamically form and break, capturing the intrinsically evolving nature of the network. We show that biofilm structure is governed by a competition between
polymer-
polymer and
polymer-bacteria crosslinks, controlled by binding energy, linker availability, and bond stiffness and provide a minimal model that helps to understand the competition between both species.
Read the full abstract ▾
All authors J. Martin-Roca, B. Wu-Zhang, J. Oller, J. Ramirez, C. Valeriani
Fluid Dynamics · Cross-listed from math.NA
No Free Lunch in Flow Surrogates under Time-Varying Boundary Conditions: A Two-Regime Study
Georg Winkler and Martin Stoll
We test whether an architecture that succeeds on a simple flow regime also succeeds on a richer one, with each trained separately on each regime. We explore two transient flows under time-varying boundary conditions: the three-dimensional slurry film in chemical-mechanical planarisation (CMP), central to
semiconductor manufacturing, and the two-dimensional Kármán
vortex street (KVS). Eight surrogate models on one shared pipeline differ in whether they learn the full field or a latent representation, and in whether they predict in one shot or step by step. No single architecture wins both regimes. On the film, a one-shot full-field model reconstructs the cumulative wall shear stress to 2.7% relative error. On the wake, a latent autoregressive DeepONet retains 90% of the shedding power that direct and one-shot models damp to almost zero. The treatment of time decides the outcome. The self-sustained wake calls for autoregressive feedback and the boundary-driven film for a direct map. Pointwise RMSE hides the damped oscillation on the wake, compresses the sixfold lead on the film's process target, and picks the damped model under wake extrapolation. The evaluation scores five physical questions. Trained surrogates answer queries 10^3 to 10^4 times faster than the finite-element solver and pay off from the first query beyond the training set on the film and from the third on the wake. Neither the winning architecture nor its validation holds across regimes. The choice of surrogate should follow the dynamical character of the target flow, and its validation should resolve the failure modes.
Read the full abstract ▾
Optics
Carrier-Envelope Phase Control of Orbital Angular Momentum in Solid-State High-Harmonic Generation
Camilo Granados, Rajaram Shrestha, Bikash Kumar Das et al. (+5 more)
High-order harmonic generation (HHG) driven by optical
vortices is a powerful route to produce structured light in different spectral regions. The nonlinear process transfers orbital angular momentum (OAM) from the driving field to the emitted harmonics according to the scaling law $l_q = q\times l$, a consequence of the rotational invariance and angular momentum conservation. Here, we show that, in the regime of few-cycle pulses, the topological charge (TC) of the harmonic radiation detected within a finite spectral window is no longer fixed by this scaling law alone, but is governed by the interplay between broken
crystal inversion
symmetry and carrier-envelope phase (CEP)-sensitive sub-cycle electron dynamics. By driving HHG in a ZnO
crystal with few-cycle ($\approx 1.5$ cycles)
vortex beams centered at 3.2~$\mu$m, we observed that the measured TC becomes strongly CEP-dependent, switching between adjacent integer values, but only when the inversion
symmetry is broken and the harmonic emission is CEP-sensitive. The TC switching vanishes when either condition is removed. Numerical analysis reveals that the TC switching originates from a CEP-controlled redistribution of spectral weight among spectrally overlapping harmonic orders, which changes the dominant OAM channel within the detection window. These results identify the CEP as a degree of freedom for tailoring the topological structure of high-harmonic radiation, pointing toward waveform-controlled structured attosecond light sources.
Read the full abstract ▾
All authors Camilo Granados, Rajaram Shrestha, Bikash Kumar Das, Debobrata Rajak, Eric Cormier, Bálint Kiss, Carmelo Rosales-Guzman, Wenlong Gao
Chemical Physics · Cross-listed from cs.LG
Variational-Ising-Attention:Tailored Attention Matters for Science
Rui Wang
Attention enables context modeling via query-key scoring with softmax normalization. Driven by industrial long-context demands, mainstream research has converged toward sparsity and efficiency, yet softmax's independence assumption persists. For scientific tasks unburdened by long-
token constraints, however, richer structured coupling may often be essential, making tailored
attention both viable and more appropriate. To this end, we propose Variational-Ising-
Attention (VIA), which augments softmax normalization with an interacting Ising model;
attention patterns emerge from learnable pairwise couplings via variational mean-field inference, extending
attention from a ranking over isolated items to a collective state over interacting entities. We instantiate VIA on retrosynthesis reaction center prediction and, as a controlled internal ablation, on
protein residue contact prediction, two structured prediction tasks governed by cooperative constraints: cooperative bond-breaking for retrosynthesis and inter-residue interactions for
protein contact prediction. Comprehensive experiments across model variants, coupled with mechanistic analyses, demonstrate that VIA substantially outperforms standard softmax
attention. More broadly, our findings suggest that for scientific problems, the optimal solution is not general-purpose efficiency, but appropriately tailored
attention aligned with intrinsic domain structure. This work provides a theoretically grounded and empirically validated instantiation of this paradigm.
Read the full abstract ▾
Fluid Dynamics · Cross-listed from math.NA
Central-Hermite Sensing and Collision for Frame-Robust Order-Resolved Relaxation on D3Q125
Bjørn Wu
Raw-Hermite sensing and collision on a fixed discrete-velocity set can convert a uniform translation into artificial coupling between nominally distinct nonequilibrium orders. We develop a central-Hermite formulation for a D3Q125 kinetic model with order-resolved log-Gaussian relaxation and compare three variants: raw sensing/raw collision (A), central sensing/raw collision (B), and central sensing/central collision (C). In homogeneous translated second-order perturbations, model C preserves third- and fourth-order modal purity to machine precision, whereas A and B develop boost-dependent cross-order content. Across a grid-CFL-boost
matrix, model C reduces the post-transport collision frame discrepancy relative to A by 65.342-98.102% (median 81.131%) in the total relative L-infinity measure. Long-time calculations remain positive and conservative to numerical precision, although the accumulated benefit is configuration dependent because transport continually re-injects frame error. A transport study further reveals a clear trade-off: central-Hermite interface reconstruction strongly suppresses the third-order discrepancy but amplifies the fourth-order discrepancy. The fully central-Hermite collision therefore substantially reduces collision-induced cross-order frame discrepancy, while residual dependence remains due to discrete transport and finite velocity-space representation. This moment-space improvement does not by itself establish a comparable reduction in macroscopic Galilean transport error.
Read the full abstract ▾
Atmospheric and Oceanic Physics
Maximum updraft velocity beyond CAPE: the role of boundary layer dynamics and pressure perturbations
Bety Pechacova, Alejandro Casallas, Tom Beucler et al. (+2 more)
Deep convective updraft velocities play a key role in the Earth's
climate system, influencing precipitation extremes, lightning, and the planetary energy budget. While Convective Available Potential Energy (CAPE) is widely used to explain maximum updraft velocity ($w_{\max}$), CAPE is an imperfect predictor as updrafts are also influenced by entrainment, boundary layer dynamics, pressure perturbations, and condensate loading. However, the relative importance of these processes and how they interact to set $w_{\max}$ in individual clouds remains unclear. Here, we use equation learning to identify compact, physically interpretable relationships linking environmental and in-cloud conditions to $w_{\max}$ in individual tracked clouds across idealized radiative-convective
equilibrium regimes spanning a range of sea surface temperatures and radiative cooling rates. For pre-storm prediction, CAPE and local mean boundary layer vertical velocity ($\overline{w_{\mathrm{bl}}}$) together explain nearly half the variance in $w_{\max}$ across regimes ($R^2=0.47$). While CAPE captures regime-mean differences, it has little predictive value within a single simulation. $\overline{w_{\mathrm{bl}}}$ is essential for capturing cloud-to-cloud variability, including the suppression of $w_{\max}$ even at high CAPE values. At the time of peak intensity, a simple approximate Bernoulli-like invariant combining maximum pressure perturbation and maximum cloud condensate explains 89% of the variance ($R^2=0.89$). The tight link between $w_{\max}$ and pressure perturbation supports the sticky thermals hypothesis and highlights the importance of dynamic pressure effects, often neglected in updraft theories. These results highlight $\overline{w_{\mathrm{bl}}}$ as an important regulator of convective intensity alongside CAPE, and demonstrate that dynamic pressure plays an important role within individual updrafts.
Read the full abstract ▾
All authors Bety Pechacova, Alejandro Casallas, Tom Beucler, Lokahith Agasthya, Caroline Muller
Physics Education
The Effect of High-Frequency, Automatically-marked Formative Assessments on Student Outcomes in A-Level Sciences
Matey Yordanov, Mikhail Bychkov and Andrei Kuchma
Traditional human marking in upper-secondary STEM education creates a structural bottleneck that restricts the frequency of formative mock examinations. This quasi-experimental, mixed-methods longitudinal study (N = 142) investigates the efficacy of deploying a fully automated, handwritten assessment marking platform to remove this bottleneck. Students preparing for STEM A-levels (Mathematics, Further Mathematics, Biology, Chemistry, Physics) were divided into control (N = 70, human marking) and intervention (N = 72, automated marking) groups over a 16-week term. Results indicate that the intervention group achieved a statistically significant improvement in final mock A-level marks (p < .01), scoring on average 16.8% higher than the control group. Furthermore, assessment turnaround time dropped from 11.2 days to < 0.1 days, enabling a fourfold increase in practice volume. The automated system's capacity for granular marginalia, specifically its Error Carried Forward (ECF) logic, accelerated the correction of procedural misconceptions and significantly boosted student self-efficacy. Practically, these findings validate that automated marking systems serve as high-speed formative loop accelerators, providing international schools with a highly scalable, low-cost intervention to elevate institutional exam outcomes without necessitating increases in staff headcount.
Read the full abstract ▾
Atmospheric and Oceanic Physics
Global-mean surface air temperature change overestimates global warming rate
Jeremy Cheuk-Hin Leung, Qiuying Gan and Banglin Zhang
Current
climate policies are targeted at slowing down the global warming rate, or the global mean surface air temperature (SAT) change ({\Delta}Tmean), which is measured by the arithmetic mean approach under the assumption that SAT changes are symmetrically distributed. However, in reality, the SAT change is asymmetric in nature and its influence on the {\Delta}Tmean interpretation seldom received
attention in previous research about
climate change. This study theorizes, based on the image histogram approach, that while {\Delta}Tmean measures the Earth's overall SAT change, it yields a value larger than the global-scale SAT change ({\Delta}Tgs) because of the asymmetrical distribution of SAT change. Results show that {\Delta}Tmean is greater than that based on {\Delta}Tgs by 0.19-0.23{\deg}C from 2000-2019, relative to the global SAT in 1979. In future
climate projections, where more significant inhomogeneous warming is expected, the disagreement between {\Delta}Tmean and {\Delta}Tgs reach 0.27-0.54{\deg}C by the end of the 21st century (2080-2099) under different emission scenarios. The large difference between {\Delta}Tmean and {\Delta}Tgs implies that there is a net positive regional-scale warming effect over the globe which is mainly contributed by Arctic Amplification. This paper constitutes a warning that extreme regional warming effects could have large impacts on the interpretation of {\Delta}Tmean, which is often considered in
climate assessments and policies making, and illustrates the limitations and cautions inherent in using {\Delta}Tmean as the only indicator of the global warming rate.
Read the full abstract ▾
Optics · Cross-listed from quant-ph
One Aperture, Two Colors: Depth-Allocated Bilayer Silicon Nitride Gratings for Trapped-Ion Addressing
Gyanendra Yadav
Scaling multicolor
quantum control on chip presents a
geometric conflict: each added wavelength traditionally requires another routing corridor, emitter, and electrode opening, consuming trap-surface area and exposing charge-susceptible dielectric near the
quantum target. This work introduces an all-planar, vertically stacked silicon-nitride architecture that co-registers independently fed optical transitions within one 24 x 24 $\mu$m footprint. For $^{40}\text{Ca}^+$, the device focuses the 729.4 nm
qubit and 854.2 nm repump wavelengths onto a shared site 70 $\mu$m above the upper film. A hierarchical design combines a differentiable
cell search for a three-level scatterer, analytic confocal phase synthesis, and deterministic depth allocation across two etches per layer, preserving an explicit 125 nm propagation-axis line-and-space floor without grayscale patterning. Retargeting the upper-layer phase co-registers the focal maxima to within 0.1 $\mu$m in mesh-converged Tidy3D and independently constructed Lumerical FDTD models, retaining >99.9% of each color's peak intensity at the common coordinate with focusing efficiencies of 0.410 and 0.168. At 5 $\mu$m ion spacing, the
qubit channel reaches -24.8 dB point crosstalk, remains below -18.0 dB across a 0.5 $\mu$m-radius positioning disk, and yields -23.0 dB symmetric-traceless field-gradient suppression. In the analytic House conductor geometry, the shared footprint provides 15.3 $\mu$m clearance to the nearest
radio-frequency rail with zero RF-null displacement. By uniting separate optical inputs and transition-specific wavefronts within one ion-facing aperture, this hierarchy establishes a transferable template for addressing other atomic wavelength pairs and dense multicolor
quantum interfaces.
Read the full abstract ▾
Plasma Physics
Implementation and first application of EMC3-EIRENE on DTT for assessing the heat load on the ICRH antenna
H.S. Wu, Y. Feng, F. Subba et al. (+5 more)
This paper reports on the implementation process of the three-dimensional (3D) edge
plasma transport code EMC3-EIRENE on the Divertor Tokamak Test (DTT) facility and the results of its first application, with a focus on assessing the heat load on the Ion Cyclotron Resonance Heating (ICRH)
antenna surfaces. Using axisymmetric geometries and the SOLPS-ITER simulation results as a reference, we first evaluate the performance of the EMC3-EIRENE in describing axisymmetric
plasmas to ensure correct code modelling setup for later complex 3D applications. We then incorporate the ICRH
antenna structure in our 3D simulations, assuming different toroidal
symmetry for the
antenna to determine whether the heat load assessment can be carried out with reduced computational effort, and to what extent a 3D assessment deviates from a 2D approximation. The predictions of the 3D
antenna heat load distribution are obtained and the peak values on the top plate and the two side plates reach 0.9, 2.1 and 3.8 MW/m2, respectively. Finally, 3D gas puffing is included to evaluate its impact on the
antenna heat load. The corresponding 3D edge
plasma behavior is presented under the coupled effects of gas puffing and the
antenna geometrical structure.
Read the full abstract ▾
All authors H.S. Wu, Y. Feng, F. Subba, S. Ceccuzzi, P. Innocente, M.M. Robaldo, A.A. Tuccillo, R. Zanino
Optics
Scaling Limits of Multichannel Spectral Routers for Snapshot Imaging
Junseo Han, Seunghyun Lee, Donghyun Kim et al. (+1 more)
Inverse-designed spectral routers can enable compact snapshot spectral imaging by directing different wavelengths to designated detector sub-pixels without mechanical scanning or absorptive filters. Here, we examine how routing performance varies as the number of spectral channels increases while the visible bandwidth remains fixed. A delay-bandwidth analysis relates channel count, worst-channel efficiency, and device thickness, explaining the growing optical capacity required to generate many spatially distinct spectral outputs. We then use adjoint-based
topology optimization and three-dimensional finite-difference time-domain simulations to design TiO2/SiO2 routers with 9, 16, 25, and 36 channels under matched material, bandwidth, and thickness conditions. The average routing efficiency increases with sub-pixel size and approaches a plateau, while the maximum near-plateau efficiency decreases monotonically from 97.0% for 9 channels to 82.3% for 36 channels. At a fixed channel count, compressing the wavelength spacing to 8 nm or rearranging the wavelength-to-sub-pixel assignment changes the efficiency by less than one percentage point. These results indicate that the observed efficiency penalty is governed mainly by the number of wavelength-dependent routing constraints rather than by wavelength spacing or local wavelength arrangement. The findings quantify the trade-off between spectral sampling and optical throughput in compact snapshot spectral imagers.
Read the full abstract ▾
All authors Junseo Han, Seunghyun Lee, Donghyun Kim, Haejun Chung
Computational Physics · Cross-listed from cs.LG
Physics-Informed Neural Networks for Discovering Periodic Orbits in the Gravitational Three-Body Problem
Nikolaos Kollias and Nikolaos Matzakos
Locating periodic solutions of chaotic dynamical systems normally requires an initial guess close enough to the target orbit for numerical continuation or gradient-based search to converge. We show that Physics-Informed Neural Networks (PINNs) trained on sparse, noisy observations without initial conditions recover periodic orbits of the gravitational three-body problem, including orbit families absent from the training data. The method rests on a second-order ODE formulation, fixed-frequency
Fourier features, percentile-based adaptive refinement, and a trainable scaling parameter, each validated on forward problems. Across two 100-seed ensembles, $23$–$25\%$ of runs converge to families not present in the training data. We then ask what determines which family emerges. Two $\chi^2$ tests give a consistent answer: changing the training data source significantly shifts the distribution of recovered families ($p < 0.001$, Cramér's $V = 0.339$), whereas switching between the two initialization distributions tested does not ($p = 0.620$, $V = 0.094$). The random seed selects which family a given run recovers; the distribution the weights are drawn from does not shift the aggregate frequencies, but the training data does. The evidence is empirical: we do not characterize the loss landscape analytically, and PINNs remain slower than conventional integrators on well-posed initial-value problems. What the experiments establish is that the recovered orbits are verifiable rather than merely plausible: the identified ones refine to genuine periodic solutions, a network trained on Lagrange data recovers the figure-eight choreography (Li–Liao class I.A.1, matched to seven significant digits in $T^*$), and one trained on figure-eight data recovers a Broucke–Hadjidemetriou–Hénon orbit closing to $\delta_T < 10^{-9}$.
Read the full abstract ▾
Atomic Physics · Cross-listed from quant-ph
Photon pair antibunching and second-order correlations between pair events
Chien-Chang Chen, Ite A. Yu and Hsi-Sheng Goan
We introduce the pair second-order correlation function $g_{\textrm{pairs}}^{\left(2\right)}=\left\langle \left(P^{\dagger}\right)^{2}P^{2}\right\rangle /\left\langle P^{\dagger}P\right\rangle ^{2}$, defined through the pair operator $P^{\dagger}=a^{\dagger}b^{\dagger}$, to characterize second-order correlations and pair bunching and antibunching in
photon-pair creation processes. This quantity directly probes correlations between pair-generation events within a single two-mode
quantum state, providing access to the intrinsic pair-generation process beyond conventional single-mode or heralded second-order correlations, which do not directly capture correlations between pair events. Values of $g_{\textrm{pairs}}^{\left(2\right)}$ greater than, equal to, or less than unity correspond respectively to pair bunching, Poissonian pair statistics, and pair antibunching. Using the Cauchy–Schwarz inequality, we further show that all classical two-mode fields described by a positive Glauber–Sudarshan \ensuremath{P}-function satisfy $g_{\textrm{pairs}}^{\left(2\right)}\geq1$, so that pair antibunching is classically forbidden and constitutes an unambiguous signature of nonclassicality. We evaluate $g_{\textrm{pairs}}^{\left(2\right)}$ for several representative
quantum states and show, in particular, that even arbitrarily weak two-mode squeezed
vacuum states exhibit pair bunching. Comparison with heralded second-order correlations highlights the complementary information provided by these observables. The proposed correlation function is experimentally accessible via standard coincidence measurements, requires no phase reference or state reconstruction, and remains invariant under uniform loss.
Read the full abstract ▾
Optics
When Every Simulation Counts: Value-Based Reinforcement Learning for Accelerated Photonics Inverse Design
Longying Wen, Feiyang Wu, Jinglin Yu et al. (+3 more)
Photonic-
crystal surface-emitting
lasers (PCSELs) can combine high-power operation with narrow-divergence surface emission, but optimizing coupled parameters requires costly full-wave simulations. Deep Q-network (DQN) optimization can reuse simulated transitions to guide edits, yet which value-learning mechanisms remain reliable under tight simulation budgets is unknown. We address this gap by comparing baseline DQN and six value-based variants for a seven-variable PCSEL design under a shared objective, simulator, 83-call budget, and four matched initializations. Beyond endpoints, we analyze sample efficiency, policy behavior, and physical response to separate learning gains from favorable starts or exploratory jumps. Dueling DQN is the only variant to improve all four seeds. Relative to the first evaluated designs, its selected structures increase the mean quality factor () from to (), reduce wavelength error by 64%, and increase upward power by 47%; compared with baseline DQN, they achieve a higher mean under the same budget. Other variants yield no consistent improvement; Double DQN reproduces baseline trajectories, while Rainbow-lite shows high upside but strong seed dependence. These results identify Dueling DQN as the most reliable configuration tested for simulation-budget-limited PCSEL inverse design and provide a reproducible framework for attributing algorithmic gains in scientific optimization. The source code is publicly available at https://github.com/Longying-Wen/PCSEL-RL.
Read the full abstract ▾
All authors Longying Wen, Feiyang Wu, Jinglin Yu, Chongxian Yuan, Renjie Li, Zhaoyu Zhang
Computational Physics · Cross-listed from cond-mat.mtrl-sci
Molecular dynamics with a first-principles-validated universal machine-learning potential reveals dynamic elementary processes of growth-related adspecies on GaN(0001)
Yoshito Takaesu, Akira Kusaba, Junko Ishii et al. (+2 more)
Atomic-scale understanding of the surface elementary processes in metalorganic vapor phase epitaxy (MOVPE) of GaN has so far relied on static density-functional-theory (DFT) energetics and on first-principles molecular dynamics (FPMD) limited to a few tens of picoseconds. Here we combine FPMD with a universal machine-learning interatomic potential (MLIP), UMA, to follow the dynamics of growth-related adspecies on GaN(0001) over time scales inaccessible to purely first-principles approaches. FPMD simulations of a GaNH admolecule coexisting with H adatoms reveal a hitherto unrecognized
diffusion mode, in which the N atom lifts the Ga atom of the GaNH unit off the surface layer during migration, and show that the lifted Ga abstracts an H adatom from the surface, events invisible to static DFT. Single-point UMA calculations on FPMD snapshots reproduce the first-principles relative energies along the trajectory (RMSE of about 8.5 meV/atom) without any retraining. Long-time MLIP-based MD (150 ps) then reveals dynamics never observed within the FPMD window: site-to-site H-adatom hopping, which gates the migration paths of the growth unit, and reversible dissociation of the GaNH unit into independently migrating Ga and NH adspecies. This work constitutes, to our knowledge, the first application of an MLIP to the molecular dynamics of GaN MOVPE.
Read the full abstract ▾
All authors Yoshito Takaesu, Akira Kusaba, Junko Ishii, Shigenori Matsushima, Yoshihiro Kangawa
Physics and Society
From Pen to Palette: Mathematical Analysis of van Gogh's Psychological Trajectory
Anna Singley, Eli Goldwyn, Mark Pitzer et al. (+1 more)
Vincent van Gogh's prolific artistic output and personal correspondence offer a unique window into the psychological trajectory of a 19th-century artist. This study applies computational methods to analyze emotional patterns across van Gogh's final decade through sentiment analysis of over 700 letter segments and fractal dimension analysis of 569 paintings. Using critical slowing down (CSD) theory from dynamical systems, we examine whether van Gogh's psychological system exhibited mathematical signatures of approaching tipping points. Our analysis reveals a statistically significant correlation between visual complexity and death sentiment ($r = 0.193, p < 0.001, n = 524$), alongside evidence for CSD patterns in multiple psychological dimensions before major crises. The 1888 ear incident shows coordinated variance spikes across acquired capability, desire for death, and visual complexity preceding the event, followed by stabilization before the crisis. These findings demonstrate that mathematical approaches can detect psychological transition signatures in historical figures while revealing the complex multi-modal nature of psychological systems.
Read the full abstract ▾
All authors Anna Singley, Eli Goldwyn, Mark Pitzer, Jessica Logue
Medical Physics
Temperature dependence of Bloch equations
David H. Gultekin
We review the temperature dependence of Bloch equations. The temperature dependence of
nuclear magnetization in thermodynamic
equilibrium and relaxation are reviewed, and relations are developed to account for the temperature effects in a
nuclear magnetic resonance experiment. It is found that the temperature dependence of the Bloch equations precisely accounts for the temperature effects observed in a
nuclear magnetic resonance experiment.
Read the full abstract ▾
Applied Physics · Cross-listed from astro-ph.IM
Technologies and novel components for broadband splitting and coupling in pairwise and nulling interferometry
Harry-Dean Kenchington Goldsmith, Nemanja Jovanovic, Anusha Pai Asnodkar et al. (+9 more)
Passive and active photonic components are central to the continued development of astronomical photonic integrated
circuits (PICs), although achieving broadband achromatic performance remains a significant challenge. This work presents broadband evanescent tri-couplers, tapered directional couplers, and a chromatically controlled achromatic intensity modulator for astronomical interferometry in the J- and H-bands. Silicon nitride and silicon
oxide provide complementary low-loss platforms, while customised tapered components enable broadband operation across the 0.95-1.8 microns range. For the silicon nitride platform developed by STMicroelectronics, tapered tri-couplers and directional couplers were designed as replacements for conventional components in a pairwise beam combiner for the
PLANETS project. Optimised tapered tri-couplers achieve less than 1% excess loss across the J-band, while tapered directional couplers provide broadband 40:60 splitting suitable for beam combination. For the silicon
oxide platform from Enablence, a two-dimensional tapered tri-coupler was investigated for nulling interferometry. The device provides broadband starlight suppression while limiting
exoplanet throughput loss to less than 2.2% across the H-band and simultaneously retaining broadband phase-sensing capability for fringe tracking. The chromatically controlled achromatic intensity modulator is introduced as a combination of a tapered directional coupler with a thermo-optic phase shifter, the device provides programmable broadband intensity control for applications including null-depth balancing in interferometric PICs. Future work will extend these concepts to lithographically fabricated chalcogenide glass platforms operating in the mid-infrared, enabling compact photonic beam combiners for future ground- and space-based nulling interferometers targeting Earth-like
exoplanets.
Read the full abstract ▾
All authors Harry-Dean Kenchington Goldsmith, Nemanja Jovanovic, Anusha Pai Asnodkar, Sanny Ahmed, Elsa Huby, Sylvestre Lacour, Michael Fitzgerald, Yoo Jung Kim, Pierre Labeye, Nicolas Dunoyer, Michael Ireland, Stephen Madden
Chemical Physics · Cross-listed from quant-ph
Multiconfigurational Mixed Quantum-Classical Approach for Correlated Many-Body Dynamics
Pritha Ghosh, Rajanya Sarkar and Arkajit Mandal
In this work, we introduce a multiconfigurational mixed
quantum-classical many-body approach for simulating the finite-temperature correlated multi-
exciton dynamics in the presence of
phonon-induced static and dynamic disorder. In this mixed
quantum-classical approach, the excitonic subsystem is described using a multiconfigurational wavefunction that extends beyond the mean-field limit, while the
phonons are evolved quasi-classically. Using this approach, we simulate a multi-excitonic dissipative system and show how the interplay between
phonon-induced dynamic disorder and
exciton-
exciton many-body interactions determines excitation-dependent excitonic transport and spatial correlations. Our results show that while the mean-field approach produces semi-quantitatively accurate diffusive dynamics, it does not capture the spatial correlations as expected. We find that a mean-field path approximation, where we generate pre-computed trajectories using our mean-field mixed
quantum-classical approach and then perform multiconfigurational dynamics, can reproduce the spatial correlations to a good accuracy, positioning this approach as an efficient method for capturing spatial correlations in complex systems.
Read the full abstract ▾
Biological Physics · Cross-listed from cond-mat.soft
The Polymer Physics of Kinetoplast DNA as a Polymerised Membrane
Takahiro Sakaue and Davide Michieletto
We analyze the conformational and dynamical properties of the kinetoplast DNA (kDNA), a massive sheet-like structure made from thousands of circular DNA molecules, found in the mitochondrion of certain parasites. The connectivity between circular DNA molecules is achieved by topological linking, hence, the kDNA may be regarded as a naturally occurring two-dimensional version of Olympic gels, whose physical properties are yet to be understood. We propose that the basic aspects in the large scale behaviors of kDNA could be described by the physics of polymerized membrane. Our analysis indicates the relevance of the hydrodynamic interactions in the dynamics of kDNA in aqueous solution. We demonstrate that the predicted dynamical scaling scenario captures various experimental data recently obtained from in vitro imaging experiments in a unified manner. We also provide an estimate for the in-plane elastic modulus of kDNA, whose magnitude agrees well with recent measurements.
Read the full abstract ▾
Chemical Physics · Cross-listed from cond-mat.mtrl-sci
Tailoring the Frequency-Dependent Optical Response of Hematite through Mono- and Co-Doping: A First-Principles Study
Abdul Ahad Mamun and Muhammad Anisuzzaman Talukder
Understanding the effects of doping on the
crystal structure and optical properties of
semiconductor materials is crucial for advancing next-generation
semiconductor and photonic technologies. Although various studies have focused on doped hematite ($\alpha$-Fe$_2$O$_3$), the relationship between dynamical stability and optical properties remains insufficiently explored. This study presents a comprehensive first-principles investigation that simultaneously evaluates the
phonon dispersion characteristics and frequency-dependent optical response of B-doped, Y-doped, and (B, Y)-co-doped $\alpha$-Fe$_2$O$_3$, providing deeper insights into the underlying mechanisms. We examined the finite-temperature vibrational properties, dielectric function, and optical characteristics to comprehend the
lattice dynamics and light-matter interactions under electromagnetic radiation. Vibrational thermodynamics reveal that pristine and Y-doped hematite maintain dynamic stability, while B-doped hematite exhibits imaginary
phonon modes indicating
lattice instability due to distortions in the Fe–O framework. Notably, Y co-doping with B helps suppress these soft modes, restoring structural stability through
lattice relaxation and improved interatomic forces. B doping enhances low-energy absorption by introducing additional states in the valence band, while Y doping alters orbital hybridization, leading to a broader dispersion. In the optical regime, doped hematite displays dominant interband transitions below $2$ eV and strong absorption between 1.80 eV and 4 eV. The (B, Y) co-doping combines the low-energy benefits with an improved optical response profile. In summary, doping significantly enhances
lattice vibrations, light-matter interactions, and optical responses, providing an effective strategy for tailoring hematite for diverse applications in photoactive, optoelectronic, and photonic technologies.
Read the full abstract ▾
Plasma Physics
Comparative qualification of advanced plasma-facing materials for fusion pilot plants through public- and private-sector experiments in DIII-D
Florian Effenberg, Jonathan D. Coburn, Luca Cappelli et al. (+44 more)
A coordinated DIII-D campaign exposed and comparatively assessed 44 advanced
plasma-facing materials from 12 institutions, including four public-private fusion partnerships, to support fusion pilot plant wall and divertor material down-selection. Samples were exposed using the Divertor Materials Evaluation System (DiMES) under Ohmic, L-mode, and H-mode conditions with edge-localized modes, at 0.2-2.5 MW m$^{-2}$ on flush geometries and 10-15 MW m$^{-2}$ on 10$^{\circ}$ angled geometries. Engineered tungsten architectures retained integrity; long-fiber Wf/W showed the clearest crack-arrest behavior. W-Re and K-doped W showed near-ITER-W-like responses, while additively manufactured W-Ta showed heat-flux-sensitive mass losses of 0.64 mg for the flat sample and 2.19-2.87 mg for angled samples. After irradiation to 0.3 dpa at 550$^{\circ}$C,
neutron-irradiated ITER-grade W retained 2.8 times more deuterium than pristine W, while TiB$_2$ showed the lowest D$_2$ release in the Ohmic set. VTaHfMo was the most stable refractory multi-principal-element
alloy. NbC and (Nb$_{0.5}$Ta$_{0.5}$)C retained integrity with 0.02-0.03 mg mass loss, whereas ZrC lost 7 mg. CVD SiC retained macroscopic integrity but exhibited an effective Si erosion yield of 0.5, about 5-10 times above prior DIII-D trends. Renewable boron pebble rods underwent controlled recession; 13% of released boron was ionized near the outer strike point and up to 50% was recovered locally. Initial in-situ chromium gross-erosion measurements yielded values of order $10^{-2}$. Together, these results provide cross-material
benchmarks for fusion pilot plant down-selection and future AI/ML-assisted
plasma-facing-material development.
Read the full abstract ▾
All authors Florian Effenberg, Jonathan D. Coburn, Luca Cappelli, Simon D. Corah, Amoolya Grandhi, Jerome Guterl, Charlie Hirst, Mike Jackson, Dylan A. Kohler, Rob Kolasinski, Erick Martinez-Loran, Ria Meston, Lauren Nuckols, Angelica Ottaviano, Zana Popovic, Sergey Tsurkan, Tessa Van Volkenburg, Daniel Velazquez, Aaliyah Zuniga, Arunodaya Bhattacharya, Jose Boedo, Kent Christian, Giacomo Dose, Eric Hollmann, Mykola Ialovega, Florian M. Laggner, Samara Levine, Xavier Navarro, Andrew Shone, Lance L. Snead, David Sprouster, Jason Trelewicz, Jayson Barr, Ryan Hood, Charlie Lasnier, Yan-Ru Lin, Vincent Mazon, Jun Ren, Gilson Ronchi, Dmitry Rudakov, Dinh Truong, Cedric Tsui, Shawn Zamperini, Weicheng Zhong, Tyler Abrams, Anthony W. Leonard, Roberto Maurizio