Igor Sikorsky Kyiv Polytechnic Institute

Hydrophobic deep eutectic solvents (HDES) are emerging as sustainable and tunable alternatives to conventional organic solvents due to their low toxicity, environmental compatibility, and adjustable physicochemical properties. This study investigated the influence of hydrogen-bond donor type on the structure, surface properties, antibacterial activity, and extraction performance of menthol-based HDES, assessing their potential for environmental remediation. Five HDES were synthesized using levulinic acid, acetic acid, lactic acid, oleic acid, and linalool as hydrogen-bond donors and characterized by nuclear magnetic resonance and infrared spectroscopy. Quantum-chemical calculations predicted donor reactivity and hydrogen-bonding interactions. Surface properties, including interfacial tension, contact angles, drop volume, and surface energy components, were measured, and polarity was determined using a solvatochromic probe. Antibacterial activity was tested against Gram-positive and Gram-negative bacteria and Candida species. Extraction efficiency of methylene blue and diclofenac from aqueous solutions was evaluated to assess pollutant removal. The HDES exhibited low interfacial tension and dispersive-dominated surface energy, confirming their hydrophobic nature. Polarity was tunable according to hydrogen-bond donor type, with menthol–linalool HDES showing the strongest antibacterial activity. Extraction efficiencies ranged from 65 to 84

Large language models are rapidly being deployed as AI tutors, yet current evaluation paradigms assess problem-solving accuracy and generic safety in isolation, failing to capture whether a model is simultaneously pedagogically effective and safe across student-tutor interaction. We argue that tutoring safety is fundamentally different from conventional LLM safety: the primary risk is not toxic content but the quiet erosion of learning through answer over-disclosure, misconception reinforcement, and the abdication of scaffolding. To systematically study this failure mode, we introduce SafeTutors, a benchmark that jointly evaluates safety and pedagogy across mathematics, physics, and chemistry. SafeTutors is organized around a theoretically grounded risk taxonomy comprising 11 harm dimensions and 48 sub-risks drawn from learning-science literature. We uncover that all models show broad harm; scale doesn't reliably help; and multi-turn dialogue worsens behavior, with pedagogical failures rising from 17.7

Iksanov and Pilipenko (2023) defined a skew stable Lévy process as a scaling limit of a sequence of perturbed at 0 symmetric stable Lévy processes (continuous-time processes). Here, we provide a simpler construction of the skew stable Lévy process as a scaling limit of a sequence of perturbed at 0 standard random walks (random sequences).

In recent years, RAPTOR based algorithms have been considered the state-of-the-art for path-finding with unlimited transfers without preprocessing. However, this status largely stems from the evolution of routing research, where Dijkstra-based solutions were superseded by timetable-based algorithms without a systematic comparison. In this work, we revisit classical Dijkstra-based approaches for public transit routing with unlimited transfers and demonstrate that Time-Dependent Dijkstra (TD-Dijkstra) outperforms MR. However, efficient TD-Dijkstra implementations rely on filtering dominated connections during preprocessing, which assumes passengers can always switch to a faster connection. We show that this filtering is unsound when stops have buffer times, as it cannot distinguish between seated passengers who may continue without waiting and transferring passengers who must respect the buffer. To address this limitation, we introduce Transfer Aware Dijkstra (TAD), a modification that scans entire trip sequences rather than individual edges, correctly handling buffer times while maintaining performance advantages over MR. Our experiments on London and Switzerland networks show that we can achieve a greater than two time speed-up over MR while producing optimal results on both networks with and without buffer times.

Periodic density structures (PDS) are trains of plasma enhancements carried into the heliosphere, but their repeated timing does not by itself identify where the cadence is set. We test whether PDS selected independently in the outer corona can be traced back to a low-coronal clock, and whether that organization survives through the EUVI–COR1–COR2 observing chain. We analyze 12 STEREO-A/SECCHI events on 2008 January 11–14 along a nonradial path. EUVI 171 A at 1.10–1.20 R_sun shows an unusual concentration of power in the predeclared 80–130 minute band relative to filter-matched red-noise controls, while the ensemble is most strongly organized in upper COR1 at 2.5–3.0 R_sun. These signatures identify a low-coronal modulation candidate and an intermediate-height organization domain, but they do not form a unique phase-preserving clock extending into COR2. Instead, the more persistent observable is spatial order. Events 9 and 12 retain expansion-stable outward ordering. In polar r–theta maps, intersections of oppositely inclined, expansion-aware matched-filter ridge supports form repeated X-/diamond-like patterns. The morphology is compatible with stationary or quasi-stationary shock-cell processing, although the brightness diagnostics are insufficient to establish a stationary MHD shock branch. The observations therefore favor an intermittent source–gate–transfer picture: a low-coronal cadence may modulate plasma release, while event-dependent propagation progressively destroys exact phase coherence. The outward ordering of the density structures can survive, providing a more persistent signature of the source-to-wind transfer than phase locking itself.