Discover global chemical information.

Discover global chemical information — CAS numbers, compound properties, and discussion, all in one place.

chemarx

PREPRINT (not peer-reviewed). A Ni–Fe–Ga microwire showed a large shape-memory strain, but the number depends on how it was measured. Nulandaya and colleagues thermally cycled a Ni₅₀Fe₂₃Ga₂₇ wire between −150 and 150 °C at 20 °C per minute. They measured transformation strains of 14.0%, 13.3% and 13.2% under tensile stresses of 20, 50 and 100 MPa, respectively. A separate estimate from X-ray diffraction lattice parameters was about 15.4%. These results describe the tested microwire and protocol, not a performance guarantee for Ni–Fe–Ga devices generally. Find the paper on arXiv by its full title: Large transformation strain of Ni₅₀Fe₂₃Ga₂₇ microwires produced by the Taylor-Ulitovsky method.

chemarx

PREPRINT (not peer-reviewed). Handedness is not just a molecular property: calculations predict that mirror-image forms of the intermetallic WAl₂ would give opposite nonlinear Hall responses—sideways electrical responses to an applied field. Patel, Shah and Jha modeled both crystal structures; WAl₂ itself had been synthesized in earlier work. Their calculations also identify 12 Kramers–Weyl crossings, features of the electronic band structure whose predicted charges reverse between the mirror forms. The transport signals are predictions, not measured device performance, and depend on the modeled electronic structure. Find the paper on arXiv by its full title: Multicharged Kramers–Weyl Fermions, Quantum Geometric Hall Responses and de Haas–van Alphen Oscillations in WAl₂.

chemarx

PREPRINT (not peer-reviewed). Can measurements on a piece of rubber yield an interpretable equation for how it deforms? Abdolazizi, Linka and Cyron used a neural network within a finite-element model, then converted it into a closed-form energy equation. Using data from 2.5-mm-thick natural-rubber membranes, they report median displacement errors of 1–2% on one perforated test geometry and also tested a second hole pattern withheld from training. The paper shows why test coverage matters: a model trained on a single silicone-plate state did not resolve all deformation modes. These results are specific to the data and geometries tested, not a universal law for rubber. Find the paper on arXiv by its full title: Constitutive Kolmogorov-Arnold Networks in Finite Element Model Updating: Closed-Form Hyperelastic Strain Energies from Heterogeneous Experiments.

chemarx

PREPRINT (not peer-reviewed). Chiral molecules are not the only ones that can reshape electron spin at a gold surface. Kafle, Afugu and Liu used electronic-structure calculations to compare heptahelicene on gold with clean gold and with achiral coronene on gold. The coronene model also produced spin components pointing out of the surface: that feature alone is not a fingerprint of molecular handedness. Adsorption also changed the calculated local orbital angular momentum of gold-derived states. These are modeled equilibrium electron states, not measured photoelectron spin polarizations; actual measurements also depend on the photoemission process. Find the paper on arXiv by its full title: Spin and Orbital Angular Momentum Textures at Chiral Molecule–Metal Interfaces.

chemarx

PEER-REVIEWED RESEARCH. A rare-earth molybdate previously studied in powder form has now been grown as millimeter-sized single crystals. Sarikhani and colleagues report the structure and physical properties of Yb2MoO6 in the Journal of Alloys and Compounds. A diffuse-reflectance analysis estimated a 3.28 eV optical gap under a specific direct-transition model. Magnetic measurements showed no clear ordering transition down to 10 K; the reported negative Weiss temperature is consistent with possible antiferromagnetic correlations, not proof of magnetic order. Find the paper by its full title in the Journal of Alloys and Compounds or on arXiv: Single-Crystal Growth, Structure, and Physical Properties of Yb2MoO6.

ChemNews

M&A | Clariant and CD&R are reportedly in talks to acquire Ashland in an all-cash deal, Bloomberg News reported on Sunday. The talks, cited to people familiar with the matter, would pair Swiss specialty chemicals maker Clariant with private equity firm Clayton, Dubilier & Rice. Ashland, a US maker of additives and specialty ingredients for life sciences, personal care and coatings, has been exploring a sale since August after pressure from activist investors. A Clariant–CD&R consortium had already submitted initial bids in early October, according to ION Analytics, while rival interest from Evonik has since cooled, Bloomberg reported earlier this month. No price or terms have been disclosed, and the companies have not commented. Reuters said it could not immediately verify the report. A deal would combine two specialty portfolios with overlapping personal-care businesses — an overlap analysts have flagged as a likely antitrust consideration. Source: Bloomberg News via Reuters (22 November 2026)

chemarx

PREPRINT (not peer-reviewed). A universal atomistic model can gain rare-earth coverage and still struggle with a key surface-chemistry regime. Shintani and colleagues report that Matlantis PFP v9 extends r2SCAN training data to 96 elements and adds clusters and adsorption structures. For crystals containing newly supported elements, its formation-energy MAE was 93 meV/atom against experimental standard enthalpies, versus 81 meV/atom for crystals using previously supported elements. In a 36-system adsorption benchmark, the overall MAE was 354 meV, comparable to the reported DFT-PBE value of 333 meV—but all 23 chemisorption cases were overbound by 361 meV on average. On one A100 GPU, force evaluations were three to four times faster than SevenNet-Omni and handled five to ten times as many atoms in the tested systems. These are setup-specific validations: only clean low-index surfaces were tested, while charged and long-range systems remain outside the present architecture. Find the paper on arXiv by searching its full title: Technical Note on Matlantis PFP v9: Extending the Element and Domain Coverage of the r2SCAN Training Data.

ChemNews

Policy | Thailand to block exports of chemicals used in rare earth mining to Myanmar as arsenic hits 28× the safe limit in border rivers Thailand plans to stop exports of chemicals used in rare earth mining to Myanmar, Deputy Prime Minister Yodchanan Wongsawat said, as Bangkok moves to choke off the unregulated mining blamed for arsenic contamination of rivers flowing into northern Thailand. The ban targets rare earth production bound for Myanmar and other "relevant countries", Wongsawat said, without elaborating. Thailand, China and Myanmar plan talks on the issue on 24–26 November. Bangkok also plans to designate the Thai side of the Kok River a "pollution control zone". Arsenic levels of 28 times the safe limit were recorded in the Sai River in early November, and the government is piloting a sediment-trapping weir in Chiang Rai that simulations suggest could cut arsenic by up to 94.6%. The mining method at issue is in-situ leaching, which uses ammonium sulfate to extract heavy rare earths such as dysprosium and terbium. Myanmar is one of the world's largest producers of heavy rare earths. Source: Reuters / Bangkok Post (22 November 2026)

chemarx

PREPRINT (not peer-reviewed). How can a molecule’s rotation, vibration and nuclear spin share quantum correlations? Das and colleagues classify 21 types of symmetry-allowed coupling states across 75 molecular point groups. One is an unentangled product state; four involve a maximally entangled pair of sectors, and 16 involve all three. The classification uses a mathematical definition based on how mixed each individual sector’s quantum state is, including a trivial one-dimensional case—so not every entry is entangled. The authors illustrate the framework with molecules including water and propose a preparation protocol for CaOCH3. This is a theoretical classification and proposal, not an experimental demonstration. Find the paper by searching arXiv for its full title: Symmetry classification of hyperfine rovibrational molecular phase space.

chemarx

PEER-REVIEWED RESEARCH. Severe deformation usually scatters charge and heat more strongly, but this CrMnFeCoNi high-entropy alloy did the opposite in the tested setup. Hidalgo-Jiménez and colleagues used high-pressure torsion: after four turns, electrical conductivity rose from 0.953 to 1.003 MS m−1 and thermal conductivity from 10.3 to 14.4 W m−1 K−1. A Wiedemann–Franz analysis assigned the change mainly to an inferred phononic contribution, which increased by about 100%, versus about 5% for the electronic contribution. The authors hypothesize that deformation-induced atomic-scale disorder or short-range ordering reduces phonon scattering, but direct verification is still needed; the phonon term is model-dependent because it is obtained by subtraction. The result applies to this alloy and processing protocol, not all high-entropy alloys. Source: arXiv; Journal of Alloys and Compounds. Find the paper by searching the full title: Unusual increase in electrical and thermal conductivity of severely deformed CrMnFeCoNi high-entropy alloy: Roles of disordering and nanostructuring in electron and phonon transport.

chemx

China's Ministry of Ecology and Environment has agreed in principle to four environmental standards, including an air-emissions standard for the rubber-products industry. The others cover power plants and boilers, the automotive industry, and railway diesel-engine exhaust smoke. The ministry's announcement does not specify numerical emission limits or effective dates. For rubber producers, the next step is to check the final text before planning compliance changes. Source: Ministry of Ecology and Environment of China, 21 Nov 2026.

chemarx

PREPRINT (not peer-reviewed). A materials-search score can miss what researchers notice in a band diagram. Okabe and colleagues asked experts to compare crystal band structures, then learned their preferences to guide a crystal-diffusion model. Ten experts judged 200 pairs; their disagreements with a numerical flat-band score showed systematic patterns. In a separate test, one blinded author-rater preferred guided structures in 64.9% of 231 decisive comparisons. The learned ranking also highlighted three proposed crystals as computational candidates, assessed using electronic-structure calculations rather than laboratory synthesis or property measurements. The result shows a way to incorporate expert judgments, not proof that the candidates have useful properties. Find the paper on arXiv by searching the full title: Reward misalignment as an opportunity for materials discovery.

chemarx

PREPRINT (not peer-reviewed). Can language models turn chemical drawings into usable 3D structures? In the MolDraft benchmark, Xu and colleagues tested nine multimodal models on 300 drawings spanning natural products, hypervalent compounds, and coordination chemistry. Resolve rates ranged from 25.0% to 76.0%, but only 29 molecules passed the study’s checks for all nine models. Performance declined across total-atom bins, from 65.0% for 18–51 atoms to 38.0% for 108–202 atoms; literature structures averaged 41.1%, versus 68.4% for VSEPR-curated structures and 67.1% for natural products. The comparison uses one shared four-tool agent, one attempt per molecule, and a curated, deliberately demanding set; size and chemistry are confounded. The checks do not establish that the output is the correct conformer, and downstream usefulness was not tested. Find the paper on arXiv by searching the full title: Can Language Models Build Molecules in Three Dimensions?

chemarx

PREPRINT (not peer-reviewed). A machine-learned potential can look well behaved while still missing a competing phase’s free energy. Azadi and colleagues trained atomic-cluster-expansion potentials for five iron oxides across 140–260 GPa and 1000–11000 K. Their final k-point-trained models reached 2.0–5.5 meV/atom held-out energy errors, but an untrained NiAs-type FeO phase showed a first-order free-energy error of 57 meV/atom despite a low extrapolation grade. For retained start structures and tested liquids at matched electronic and ionic temperatures, errors were much smaller—within 3.4 meV/atom for liquids at 5000–9000 K. The lesson is methodological, not a universal phase-diagram guarantee: the calculations use non-spin-polarized PBEsol/CP2K labels, omit several phases, and the authors recommend validating every competing structure class. Find the paper on arXiv by searching the full title: “Machine-learned potentials for iron oxides at extreme conditions.”

chemarx

PREPRINT (not peer-reviewed). Which training compounds help a model predict materials containing an element it has never seen? Xiao and Tadano tested this with leave-one-element-out benchmarks across JARVIS, Materials Project and OQMD. A pretrained crystal-graph representation ranked source chemistries by how similarly their compounds moved through representation space. In OQMD, removing the six highest-ranked source subsets from an otherwise fixed 10,000-compound training sample increased formation-energy errors for targets with more than 40% of the held-out element by about 100–110 meV per atom and bandgap errors by 42–85 meV; removing six lower-ranked subsets changed them little. The ranking was useful across tested properties, regressors and datasets, but it did not explain everything: labeled coverage and regressor learning also mattered. This is a computational dataset-selection diagnostic, not a guarantee of experimental or universal chemical-transfer accuracy. Find the paper on arXiv by searching its full title: Materials representations organize transferable supervision for chemical extrapolation.

chemarx

PREPRINT (not peer-reviewed). Decades of powder X-ray diffraction curves are trapped as images in papers; Gao and Zheng built a hybrid vision-language-model and computer-vision pipeline to recover them. Applied to 3,954 covalent organic framework (COF) articles, it released 11,445 calibrated curves. Against hand digitization of a 60-panel sample, median peak-position difference was 0.0672° for 118 curves from 55 panels; 82.7% of operator peaks matched within 0.5° for 122 curves from 57 panels. The dataset also enabled a focused analysis: in 31 of 45 papers with usable synthesis-parameter ladders, the strongest low-angle peak became more prominent as reaction time or temperature increased—a possible, not direct, crystallinity signal. The result is not a universal accuracy guarantee: the corpus was incomplete, about 10% of curves in one comparison put the first peak more than 5° from the text value, and the paper flags model-derived, inset, shape-flagged, and axis-unconfirmed records. Find the paper by searching arXiv for its full title: Structured Extraction of Diffraction Data from Covalent Organic Frameworks.

chemarx

PREPRINT (not peer-reviewed). Can powder X-ray diffraction guide atomic-structure searches without a complete structural database entry? Lim and colleagues report XCAPE, a symmetry-constrained evolutionary search that combines diffraction agreement with energies from a universal machine-learning interatomic potential. With composition, unit-cell parameters, the number of formula units per cell, and space group supplied, it recovered 71 of 77 known structures (92.2%) from simulated, noise-free patterns, and all 12 reference structures in an experimental-pattern validation set. The boundary matters: the study’s unresolved candidates were generated from peak lists lacking full profiles, backgrounds, and peak shapes, so the authors say original profiles or new measurements are still needed for Rietveld refinement. Find the paper on arXiv by searching the full title: Physics-Guided Crystal Structure Elucidation from Powder X-Ray Diffraction with Universal Machine-Learning Interatomic Potentials.

chemarx

PREPRINT (not peer-reviewed). Different substrates and conditions can redirect which C–H bond of an indole is alkylated. Yang and colleagues used a chiral Rh(III) catalyst with nitroalkenes: one indole substrate in acetone gave the C2 product in 97% isolated yield, 92% ee, and >10:1 C2:C4 selectivity. Switching to another directing group and chlorobenzene at 45 °C for 48 hours favored C4 alkylation, giving 76% yield, 86% ee, and 4.5:1 C4:C2 selectivity. The scope was not uniform: one C2 example gave 29% yield, and a C4 substrate gave 52% yield with 2.2:1 selectivity. Find the paper on ChemRxiv by searching the full title: Rh(III)-Catalyzed Regiodivergent Enantioselective C2- and C4–H Alkylation of Indoles.

chemarx

PREPRINT (not peer-reviewed). Why do two related β-lactam antibiotics leave L1 metallo-β-lactamase in different product poses? Kalita and colleagues’ QM/MM study predicts that both ertapenem and faropenem begin with similar ring-opening steps, but ertapenem’s 1β-methyl group makes the rotation needed for the faropenem-like route much less favorable. Instead, rotation of ertapenem’s C6-carboxyl group offers a lower-barrier path to the (2S)-imine-Δ1 product in the binding mode seen crystallographically. In three 500 ns molecular-dynamics replicas, that ertapenem-derived imine also dissociated within 150 ns, whereas the faropenem-derived species stayed bound under the simulated conditions. These are model-dependent computational predictions, not direct measurements of enzyme kinetics or product residence times. Find the paper on ChemRxiv by searching the full title: QM/MM Simulations Reveal Differences in Hydrolysis Mechanisms of Ertapenem and Faropenem in L1 Metallo-β-Lactamase Leading to Different Product Binding Modes.

chemarx

PREPRINT (not peer-reviewed). A polymer support that is softer can still protect an oxide membrane better—if its interface with water and the oxide is engineered. Song, Choi and colleagues deconstructed poly(propylene carbonate) to lower its molecular weight, increasing surface polarity and surface energy. In their model system, a 15 kDa support kept an approximately 13 nm SrTiO3 membrane visibly crack-free after 200 minutes in water, over a centimeter-scale area; as-received PPC produced cracks and voids under the same exposure. The authors link the difference to suppressed water-induced interfacial blistering, while noting that the comparison covers their tested polymer range and release setup. They also transferred an approximately 30 nm La0.7Sr0.3MnO3 membrane onto SiO2/Si and measured preserved ferromagnetism and metallic transport. These are laboratory demonstrations, not proof of a universal transfer recipe. Find the paper on arXiv by searching its full title: Molecular engineering for crack-free transfer of single-crystalline oxide membranes onto silicon.

Loading…