- Research Article
- 10.1016/j.jfluchem.2026.110547
Rare octahedral coordination of cerium(IV) in the trinuclear fluorine complex [F3Ce(µ-Ph2PO2)3Ce(µ-Ph2PO2)3CeF3].3Me2SO
- Apr 01, 2026
- Journal of Fluorine Chemistry
- Evgeny G Ilin + 11 more +11
Publications from 2021 to 2026
Showing 10 of 1,558 papers
Rare octahedral coordination of cerium(IV) in the trinuclear fluorine complex [F3Ce(µ-Ph2PO2)3Ce(µ-Ph2PO2)3CeF3].3Me2SO
What Do Biological Foundation Models Compute? Sparse Autoencoders from Feature Recovery to Mechanistic Interpretability
Abstract Foundation models trained on protein and DNA sequences are increasingly deployed for variant interpretation, drug design, and gene regulation prediction, yet their internal representations remain opaque – limiting both biological insight and trust in model-guided decisions. Existing interpretation approaches establish what these models encode but cannot reveal how biological knowledge is internally organized and computed. Sparse autoencoders (SAEs) offer a complementary approach by decomposing model activations into interpretable features, each capturing a distinct biological concept. Over the past year, SAEs have been applied to protein language models, genomic language models, pathology vision transformers, single-cell foundation models, and protein structure generators. Here we provide a systematic review of sparse dictionary learning across biological foundation models. We find that independent studies using different architectures and evaluation strategies consistently recover features spanning biological scales – from secondary structure elements and functional domains in proteins to transcription factor binding sites and regulatory elements in genomes – providing convergent evidence that these models learn interpretable representations accessible through sparse decomposition. However, we identify a critical gap: validation relies almost exclusively on matching features against existing annotations, risking circularity when those annotations derive from the same sequence databases used for model training. We propose a three-level interpretability framework – representational, computational, and causal mechanistic – and argue that the field’s most distinctive opportunity lies in experimental validation through deep mutational scanning, massively parallel reporter assays, and structural characterization, which can establish whether these models have learned genuine biological mechanisms rather than training set statistics.
Read moreGrowth and investigation of mixed magnesium-zinc tungstate Zn Mg1-WO4 single crystals
Detection of Disease by Analysis of Volatile Organic Compounds: I. Detector Dogs
Behavior of Ultrafine-Grained Zn-1%Mg-0.1%Ca Alloy under Cyclic Loading
Tuning the Dimension: Comparative Insights into Bulk and Single-Layer MoS2
Molybdenum disulfide (MoS2), a prototypical two-dimensional transition metal dichalcogenide (TMD), has attracted substantial interest for its tunable properties and broad technological potential. This review presents a comprehensive comparison between monolayer and bulk MoS2, highlighting key differences in their structural, optical, electrical, mechanical, and bio-interactive properties. The transition from bulk to monolayer induces a transformation from an indirect to a direct bandgap, significantly enhancing photoluminescence, charge mobility, and surface sensitivity features critical for next-generation electronic, optoelectronic, and biosensing devices. We also explore state-of-the-art synthesis strategies, including top-down and bottom-up approaches, that enable the scalable production of high-quality monolayer MoS2. By synthesizing insights from recent theoretical and experimental advances, this article elucidates why monolayer MoS2 consistently outperforms its bulk counterpart in functional applications ranging from strain sensors and photodetectors to non-invasive biosensors for prenatal diagnostics. The review not only maps the current landscape but also outlines future directions, positioning monolayer MoS2 as a cornerstone material for advanced technologies.
Read moreComment on egusphere-2025-5041
<strong class="journal-contentHeaderColor">Abstract.</strong> This study presents lidar observations of long-range transported biomass burning aerosol (BBA) plumes from the exceptional 2023 Canadian wildfire season, recorded between May and September at the ATOLL observatory (France) and the GPI site (Russia). ATOLL operates a multi-wavelength Raman lidar with 3 polarization channels (355, 532 and 1064 nm) and a single fluorescence channel at 466 nm. GPI uses a fluorescence lidar with 5 broadband fluorescence channels excited by 355 nm. The dual-site dataset combines multi-wavelength elastic scattering and depolarization measurements with fluorescence observations, enabling a comprehensive characterization of BBA properties in the free troposphere (FT) and upper troposphere–lower stratosphere (UTLS). UTLS layers exhibit higher particle depolarization ratios, slightly lower lidar ratios, lower extinction- and backscatter-related Angström exponents, and a redshift in fluorescence spectral peaks. Cross-site comparisons show consistent fluorescence magnitudes and spectral shapes, highlighting the potential of coordinated multi-lidar fluorescence observations. Correlation analysis indicates that depolarization ratio, extinction-related Angström exponent, and fluorescence color ratio are moderately (r<sup>2</sup> ≈ 0.61–0.68) correlated with layer altitude, however, this correlation is not sufficient to confirm a solid altitude dependence. It is likely that altitude is an intermediate variable linked to other controlling factors such as injection height of the plume, in-layer temperature and the plume origin. In addition, we observed BBAs showing no clear hygroscopic growth at RH of 90 %–100 % and statistically low RH values in the detected nearly 100 layers, suggesting aged BBAs, which were typically considered as hygroscopic, may have limited water uptake capability.
Read moreComment on egusphere-2025-5856
<strong class="journal-contentHeaderColor">Abstract.</strong> During summer 2023, Greece experienced one of its most severe wildfire seasons in recent decades, with widespread fires across Evros, Rodopi, Attica, the Peloponnese, and several islands. This study investigates the aerosol optical and microphysical properties, as well as the impact on ground-level ultraviolet-B (UVB) radiation over Athens, focusing on two major wildfire episodes (18–21 July and 22–25 August). A synergistic approach was deployed, combining satellite imagery (MODIS), FLEXPART simulations, ground-based remote sensing, and in situ aerosol and radiation measurements. Elevated aerosol optical depths (AOD) up to 1.2, high fine-mode fractions (> 0.85), and Ångström exponents above 1.5 indicated a strong dominance of fine biomass burning aerosols. The Single scattering albedo (SSA) ranged from 0.85 to 0.98, showing enhanced absorption during biomass burning periods and weaker absorption when smoke was mixed with dust. At 320 nm, dust presence resulted in stronger absorption, with SSA below 0.8 for pure dust cases compared to smoke mixtures. Particle linear depolarization ratios (PLDR), varied between 0.03 and 0.20, with higher values (~0.10–0.20) reflecting the presence of non-spherical dust particles, and lower values (~0.03–0.08) indicating spherical smoke particles. Ground-level UVB irradiance decreased by up to 50 % during peak smoke episodes, highlighting strong aerosol radiative impacts. Concurrently, PM<sub>10</sub> and PM<sub>2.5</sub> concentrations increased to 94 and 49 µg m<sup>-3</sup>, respectively, while organic aerosols peaked at 22.77 µg m<sup>-3</sup>, consistent with intense fire activity. FLEXPART simulations confirmed long-range transport of smoke from active fire regions, with additional contributions from regional pollution and Saharan dust.
Read moreOptical anisotropy of 6-A graphene nanoribbons synthesized inside aligned nanotubes
Extracavity and intracavity mid-IR lasing in a ZnSe:Fe2+ crystal