arXiv subjects

Tanveer Hussain

Publications and source records attributed to Tanveer Hussain.

At least 19 recordsLinked to original sources

arXiv2026

Generative Late-Interaction Embeddings For Visual Document Retrieval

Late-interaction retrieval is the state-of-the-art for visual document search, but it pays for its accuracy in storage. Existing compression methods retain a subset or local average of the N~1,000 vectors per page. Under aggressive storage budgets, however, these methods degrade sharply, and alternatives require retraining the encoder. Investigating this degradation across three encoders, we found two consistent properties: the vectors lie exactly on the unit sphere and concentrate near a manifold of intrinsic dimension five to six. This geometry yields two insights. First, standard k-means centroids fall inside the sphere, causing systematic underestimation of MaxSim scores. Normalizing them to the surface is a free correction worth up to +0.093 nDCG@5 over raw centroids. Second, because the page manifold has few degrees of freedom, the full set of vectors can be regenerated from only a few. To this end, we introduce Generative Late-Interaction Embeddings (GLIE): k << N vectors per page learned from the normalized centroids to serve as both a lightweight index and a basis for regenerating the page's full embedding set. At query time, search runs exclusively on these k vectors, and a decoder expands only the top candidates back to all N vectors for exact rescoring. At four vectors per page on ViDoRe v1, GLIE retains nearly 80% of the uncompressed system's nDCG@5, against 70% for the best prior post-hoc method. These results use a 415K-parameter network fitted in under three GPU-minutes on just a thousand training pages. At a matched training budget, fine-tuning the encoder does not reach even the training-free stage of GLIE, and the full system beats it at every budget. These patterns hold across a second encoder and ViDoRe v2. By reconstructing evidence on demand rather than sampling it, GLIE opens a new axis for storage-efficient retrieval, with the decoder as its main design surface.

cs.IR↗

arXiv2026

Light alkali metal functionalized two-dimensional C5N monolayers for enhanced hydrogen storage

This work presents a density functional theory (DFT) investigation of a two-dimensional (2D) C5N monolayer functionalized with Li, Na, and K for hydrogen storage. Pristine C5N exhibits weak H2 adsorption, while alkali-metal functionalization significantly enhances its storage capability. The C5N monolayer can stably accommodate up to six metal dopants, with binding energies stronger than the corresponding cohesive energies, indicating resistance to metal aggregation. Ab initio molecular dynamics simulations further confirm the thermal stability of the functionalized systems at 300 K. Charge transfer from the metal dopants to C5N enhances polarization and strengthens H2 adsorption. Each dopant can adsorb up to eight H2 molecules, yielding a maximum of 48 H2 molecules per unit cell and gravimetric storage capacities of 9.42, 8.61, and 7.93 wt% for Li-, Na-, and K-functionalized C5N, respectively. The average H2 adsorption energies of -0.16 to -0.17 eV/H2 indicate moderate interactions suitable for reversible storage. Thermodynamic analysis further demonstrates favourable H2 adsorption/desorption under practical operating conditions, while desorption-temperature, recovery-time, and volumetric analyses support the potential reversibility and storage performance of these systems. Overall, alkali-metal-functionalized C5N emerges as a promising 2D material for efficient and reversible H2 storage.

cond-mat.mtrl-sci↗

arXiv2026

Geometry-Aware Superpixel Graph Transformer with Metadata for Skin Lesion Classification

Automated skin cancer classification from dermoscopic images remains challenging due to heterogeneous lesion structure, strong intra-class variability, and subtle visual differences between benign and malignant cases. Existing CNN/ViT pipelines typically rely on global or patch-level features and often combine patient metadata via late fusion, which limits spatially grounded multimodal reasoning. We present a novel region-based graph learning framework that explicitly models lesions as graphs of spatially coherent superpixel regions represented as frozen CNN features. To capture fine-grained lesion arrangements, we encode inter-regional geometry as edge attributes and introduce a dedicated metadata context node connected to all regions, providing structured integration of demographic/clinical variables within the same relational space. Node representations are updated using our edge-aware graph transformer followed by attention-driven propagation, and a final graph-level embedding for benign-malignant classification. Experiments on four public benchmarks demonstrate that explicit region-level relational modeling and graph-native multimodal fusion yield consistent gains over the state-of-the-art. Consequently, we establish a new graph-centric perspective in which CNN features are modeled as relational nodes and improved through contextual integration, yielding more expressive and robust classifications.

cs.CV↗

arXiv2026

N-Graphdiyne as a Tunable Platform for Stabilizing Light Metals toward High-Capacity Reversible Hydrogen Storage

Hydrogen (H2) is a promising carbon-neutral energy carrier. However, its deployment is limited by the lack of lightweight, reversible storage media that operate under practical conditions. Here, we establish nitrogen-doped graphdiyne (N-GDY) as a programmable two-dimensional platform for stabilizing dispersed light-metal dopants and enabling high-capacity physisorption of molecular H2. The computational package involves density functional theory (DFT) combined with ab initio molecular dynamics (AIMD) and Langmuir-based statistical thermodynamic modeling. The results revealed that N-sites of N-GDY bind up to five Li, Na, K, and Ca atoms per primitive cell with binding energies of -2.27, -1.57, -1.80, and -2.13 eV, respectively, exceeding their respective bulk cohesive energies. AIMD simulations at 400 K further confirm the structural robustness of the decorated frameworks and the absence of metal aggregation. The polarised metal centres activate reversible H2 adsorption through electrostatic and dispersion interactions, with average adsorption energies falling within the optimal window (-0.15 to -0.35 eV per H2). Sequential adsorption analysis reveals uptake of up to 25 H2 molecules per primitive cell, achieving intrinsic gravimetric capacities of 13.08, 10.82, 9.23, and 9.15 wt% for Li-, Na-, K-, and Ca-functionalized systems, respectively. Thermodynamic analysis indicates favorable adsorption-desorption behavior under near-ambient conditions, with Li- and Ca-functionalized systems exceeding the 6.5 wt% U.S. Department of Energy's ultimate system-level target when considering intrinsic material capacity. These results identify N-GDY as a chemically tunable scaffold for dispersing lightweight metals and provide a mechanistic design strategy for achieving high-capacity, reversible hydrogen storage in porous two-dimensional materials.

cond-mat.mtrl-sci↗

arXiv2026

GridProbe: Posterior-Probing for Adaptive Test-Time Compute in Long-Video VLMs

Long-video understanding in VLMs is bottlenecked by a single monolithic forward pass over thousands of frames at quadratic attention cost. A common mitigation is to first select a small subset of informative frames before the forward pass; common for training-free selectors via auxiliary encoder-space similarities. Such signals are capped by contrastive pretraining, which usually fails on reasoning-heavy queries (negation, cross-frame counting, holistic summarization). We propose GridProbe, an efficient training-free posterior-probing inference paradigm that scores evidence in answer space using a frozen VLM's own reasoning and then selects question-relevant frames adaptively, resulting in sub-quadratic attention cost with little to no accuracy loss. We arrange frames on a $K{\times}K$ grid and run lightweight row R and column C probes, where each probe reads its peak posterior as a query-conditioned confidence. The outer product of R and C yields an interpretable importance map whose skewness and kurtosis drive Shape-Adaptive Selection, a closed-form rule that reliably replaces the fixed frame budget $M$ with a per-question $M_{\mathrm{eff}}$. We show empirically that $M_{\mathrm{eff}}$ tracks intrinsic question difficulty without ever seeing the answer, a sign of test-time adaptive compute. On Video-MME-v2, GridProbe matches the monolithic baseline within $1.6$ pp Avg Acc at $3.36\times$ TFLOPs reduction, while on LongVideoBench it Pareto-dominates the baseline ($+0.9$ pp at $0.35\times$ compute). Because the selector and QA models can be decoupled, pairing a small 2B selector with a stronger 4B or 8B QA is strictly Pareto-dominant over the 2B monolithic baseline (up to $+4.0$ pp at $0.52\times$ compute, on average), with no retraining. Finally, the interpretability of the importance maps opens future avenues for behavioral diagnostics, grounding, and frame-selection distillation.

cs.CV↗

arXiv2026

Defect-Engineered Beryllium Dinitride (BeN2) Monolayer with Light-Metal Decoration for Reversible High-Capacity Hydrogen Storage

Hydrogen (H2) possesses the highest gravimetric energy density of any chemical fuel and is the most abundant element in the universe. However, its extremely low volumetric energy density at standard conditions imposes a fundamental materials challenge for safe, efficient, and reversible storage. Here, we report a defect-engineered 2D beryllium dinitride (BeN2) monolayer that enables stable light-metal functionalization for high-capacity H2 storage. A 2 x 2 supercell containing two intrinsic beryllium vacancies accommodates four Li, Na, and K atoms without clustering, exhibiting strong average metal-vacancy binding energies of -3.80, -2.94, and -3.18 eV, respectively. Ab initio molecular dynamics simulations at 400 K confirm the thermal stability of the metal-decorated frameworks and the suppression of metal aggregation. The vacancy-stabilized alkali-metal centers generate localized charge polarization that facilitates the adsorption of up to 20 H2 molecules per supercell, with average adsorption energies of -0.182 eV (Li), -0.191 eV (Na), and -0.171 eV (K), making the adsorption reversible under near-ambient conditions. The corresponding gravimetric H2 storage capacities reach 11.64, 9.82, and 8.49 wt percent, respectively, significantly exceeding the US Department of Energy (DOE) ultimate target of 6.50 wt percent. Moreover, thermodynamic analysis further confirms favorable adsorption-desorption behavior within practical operating windows. These results establish vacancy-defected light-metal decorated BeN2 as a viable design strategy for high-density, reversible H2 storage, providing a scalable framework for engineering polar lightweight materials for energy storage applications.

cond-mat.mtrl-sci↗

arXiv2026

VideoAtlas: Navigating Long-Form Video in Logarithmic Compute

Extending language models to video introduces two challenges: representation, where existing methods rely on lossy approximations, and long-context, where caption- or agent-based pipelines collapse video into text and lose visual fidelity. To overcome this, we introduce \textbf{VideoAtlas}, a task-agnostic environment to represent video as a hierarchical grid that is simultaneously lossless, navigable, scalable, caption- and preprocessing-free. An overview of the video is available at a glance, and any region can be recursively zoomed into, with the same visual representation used uniformly for the video, intermediate investigations, and the agent's memory, eliminating lossy text conversion end-to-end. This hierarchical structure ensures access depth grows only logarithmically with video length. For long-context, Recursive Language Models (RLMs) recently offered a powerful solution for long text, but extending them to visual domain requires a structured environment to recurse into, which \textbf{VideoAtlas} provides. \textbf{VideoAtlas} as a Markov Decision Process unlocks Video-RLM: a parallel Master-Worker architecture where a Master coordinates global exploration while Workers concurrently drill into assigned regions to accumulate lossless visual evidence. We demonstrate three key findings: (1)~logarithmic compute growth with video duration, further amplified by a 30-60\% multimodal cache hit rate arising from the grid's structural reuse. (2)~environment budgeting, where bounding the maximum exploration depth provides a principled compute-accuracy hyperparameter. (3)~emergent adaptive compute allocation that scales with question granularity. When scaling from 1-hour to 10-hour benchmarks, Video-RLM remains the most duration-robust method with minimal accuracy degradation, demonstrating that structured environment navigation is a viable and scalable paradigm for video understanding.

cs.CV↗

arXiv2025

CrSe_2 and CrTe_2 Monolayers as Efficient Air Pollutants Nanosensors

Nanosensors are critical in environmental monitoring, industrial safety, and public health by detecting specific hazardous gases like CO, NO, SO_2, and CH_4 at trace levels. This study uses density functional theory (DFT) calculations to examine the gas-sensing capabilities of chromium diselenide (CrSe_2) and chromium ditelluride (CrTe_2) monolayers through their structural and electronic responses to gas adsorption. Adsorption energy analysis shows that Te vacancy-induced CrTe_2 (VTe-CrTe_2) exhibits the strongest binding with energies of -1.52, -1.79, and -1.61 eV for CO, NO, and SO_2, respectively. Similarly, CrSe_2 has its values of -1.13, -1.17, -0.90, and -1.12 eV for CO, NO, SO_2, and CH_2, respectively, indicating suitability for reversible sensing. This study also investigates how substitutional doping of Ge, Sb, and Sn influences the sensing mechanism of CrSe_2 and CrTe_2 monolayers. Density of states (DOS) analysis highlights notable electronic changes around the Fermi level, especially in VTe-CrTe_2 and Sb/Sn-doped CrTe_2, confirming their enhanced sensing abilities. Charge density difference analysis shows significant charge redistribution, with CrTe_2 experiencing stronger charge transfer effects than CrSe_2. Variations in electrostatic potential and work function further demonstrate the higher sensitivity of CrTe_2, particularly in its defective and doped forms, confirming its status as a superior material for gas sensing applications.

cond-mat.mtrl-sci↗

arXiv2025

Deep Multi-modal Breast Cancer Detection Network

Automated breast cancer detection via computer vision techniques is challenging due to the complex nature of breast tissue, the subtle appearance of cancerous lesions, and variations in breast density. Mainstream techniques primarily focus on visual cues, overlooking complementary patient-specific textual features that are equally important and can enhance diagnostic accuracy. To address this gap, we introduce Multi-modal Cancer Detection Network (MMDCNet) that integrates visual cues with clinical data to improve breast cancer detection. Our approach processes medical images using computer vision techniques while structured patient metadata patterns are learned through a custom fully connected network. The extracted features are fused to form a comprehensive representation, allowing the model to leverage both visual and clinical information. The final classifier is trained based on the joint features embedding space of visual and clinical cues and experiments prove enhanced performance, improving accuracy from 79.38\% to 90.87\% on a Mini-DDSM dataset. Additionally, our approach achieves 97.05\% accuracy on an image-only dataset, highlighting the robustness and effectiveness of visual feature extraction. These findings emphasise the potential of multi-modal learning in medical diagnostics, paving the way for future research on optimising data integration strategies and refining AI-driven clinical decision support systems.

q-bio.QM↗

arXiv2025

Vote-in-Context: Turning VLMs into Zero-Shot Rank Fusers

In the retrieval domain, candidates' fusion from heterogeneous retrievers is a long-standing challenge, particularly for complex, multi-modal data such as videos. While typical fusion techniques are training-free, they rely solely on rank or score signals, disregarding candidates' representations. This work introduces Vote-in-Context (ViC), a generalized, training-free framework that re-thinks list-wise reranking and fusion as a zero-shot reasoning task for a Vision-Language Model (VLM). The core insight is to serialize both content evidence and retriever metadata directly within the VLM's prompt, allowing the model to adaptively weigh retriever consensus against visual-linguistic content. We demonstrate the generality of this framework by applying it to the challenging domain of cross-modal video retrieval. To this end, we introduce the S-Grid, a compact serialization map that represents each video as an image grid, optionally paired with subtitles to enable list-wise reasoning over video candidates. ViC is evaluated both as a single-list reranker, where it dramatically improves the precision of individual retrievers, and as an ensemble fuser, where it consistently outperforms strong baselines like CombSUM. Across video retrieval benchmarks including ActivityNet and VATEX, the framework establishes new state-of-the-art zero-shot retrieval performance, demonstrating its effectiveness in handling complex visual and temporal signals alongside text. In zero-shot settings, ViC achieves Recall@1 scores of 87.1% (t2v) / 89.0% (v2t) on MSR-VTT and 99.6% (v2t) on VATEX, representing massive gains of up to +40 Recall@1 over previous state-of-the-art baselines. We present ViC as a simple, reproducible, and highly effective recipe for turning modern VLMs into powerful zero-shot rerankers and fusers. Code and resources are publicly available at: https://github.com/mohammad2012191/ViC

cs.CV↗

arXiv2025

AutoArabic: A Three-Stage Framework for Localizing Video-Text Retrieval Benchmarks

Video-to-text and text-to-video retrieval are dominated by English benchmarks (e.g. DiDeMo, MSR-VTT) and recent multilingual corpora (e.g. RUDDER), yet Arabic remains underserved, lacking localized evaluation metrics. We introduce a three-stage framework, AutoArabic, utilizing state-of-the-art large language models (LLMs) to translate non-Arabic benchmarks into Modern Standard Arabic, reducing the manual revision required by nearly fourfold. The framework incorporates an error detection module that automatically flags potential translation errors with 97% accuracy. Applying the framework to DiDeMo, a video retrieval benchmark produces DiDeMo-AR, an Arabic variant with 40,144 fluent Arabic descriptions. An analysis of the translation errors is provided and organized into an insightful taxonomy to guide future Arabic localization efforts. We train a CLIP-style baseline with identical hyperparameters on the Arabic and English variants of the benchmark, finding a moderate performance gap (about 3 percentage points at Recall@1), indicating that Arabic localization preserves benchmark difficulty. We evaluate three post-editing budgets (zero/ flagged-only/ full) and find that performance improves monotonically with more post-editing, while the raw LLM output (zero-budget) remains usable. To ensure reproducibility to other languages, we made the code available at https://github.com/Tahaalshatiri/AutoArabic.

cs.CV↗

arXiv2025

Multimodal Lengthy Videos Retrieval Framework and Evaluation Metric

Precise video retrieval requires multi-modal correlations to handle unseen vocabulary and scenes, becoming more complex for lengthy videos where models must perform effectively without prior training on a specific dataset. We introduce a unified framework that combines a visual matching stream and an aural matching stream with a unique subtitles-based video segmentation approach. Additionally, the aural stream includes a complementary audio-based two-stage retrieval mechanism that enhances performance on long-duration videos. Considering the complex nature of retrieval from lengthy videos and its corresponding evaluation, we introduce a new retrieval evaluation method specifically designed for long-video retrieval to support further research. We conducted experiments on the YouCook2 benchmark, showing promising retrieval performance.

cs.CV↗

arXiv2024

AllWeatherNet:Unified Image Enhancement for Autonomous Driving under Adverse Weather and Lowlight-conditions

Adverse conditions like snow, rain, nighttime, and fog, pose challenges for autonomous driving perception systems. Existing methods have limited effectiveness in improving essential computer vision tasks, such as semantic segmentation, and often focus on only one specific condition, such as removing rain or translating nighttime images into daytime ones. To address these limitations, we propose a method to improve the visual quality and clarity degraded by such adverse conditions. Our method, AllWeather-Net, utilizes a novel hierarchical architecture to enhance images across all adverse conditions. This architecture incorporates information at three semantic levels: scene, object, and texture, by discriminating patches at each level. Furthermore, we introduce a Scaled Illumination-aware Attention Mechanism (SIAM) that guides the learning towards road elements critical for autonomous driving perception. SIAM exhibits robustness, remaining unaffected by changes in weather conditions or environmental scenes. AllWeather-Net effectively transforms images into normal weather and daytime scenes, demonstrating superior image enhancement results and subsequently enhancing the performance of semantic segmentation, with up to a 5.3% improvement in mIoU in the trained domain. We also show our model's generalization ability by applying it to unseen domains without re-training, achieving up to 3.9% mIoU improvement. Code can be accessed at: https://github.com/Jumponthemoon/AllWeatherNet.

cs.CV↗

arXiv2024

Vacancy-Induced Boron Nitride Monolayers as Multifunctional Materials for Metal Ion Batteries and Hydrogen Storage Applications

This study comprehensively examined the structural, electronic, electrochemical, and energy storage properties of boron-vacancy induced porous boron nitride monolayers (BN:VB) as multifunctional materials, anodes for MIBs and H2 storage applications. Our computational approaches, density functional theory (DFT), ab initio molecular dynamics (AIMD), and thermodynamic analysis, revealed exceptionally high energy and gravimetric densities for MIBs and H2 storage, respectively. We investigated the interactions of Li, Na, and K atoms on BN:VB, which strongly bonded with binding energies stronger than their bulk cohesive energies, which ensured structural stability and the absence of metal clustering. Electronic properties, analyzed through spin-polarized partial density of states (PDOS), band structure, and Bader charge analysis, revealed significant charge transfers from the metal atoms to BN:VB, enhancing the electronic conductivity of the latter. Theoretical specific capacities were calculated as 1821.53, 786.11, and 490.51 mA h/g for Li, Na, and K, respectively, which comfortably exceeded the conventional anodes, such as graphite. Average open-circuit voltages (OCVs) were found as 0.15, 0.25, and 0.32 V, for Li, Na, and K, respectively, indicating strong electrochemical stability. Diffusion studies showed lower barriers of 0.47, 0.08, and 0.60 eV for Li, Na, and K, respectively, with increased metal loadings, suggesting enhanced mobilities and charge/discharge rates. On the other side, the metal-functionalized BN:VB monolayers exhibited remarkably high H2 gravimetric capacities, supported by Langmuir adsorption model-based statistical thermodynamic analysis. Average adsorption energies of H2 on 4Li-, 4Na-, and 4K@BN:VB, were found in perfect range for practical storage applications.

cond-mat.mtrl-sci↗

arXiv2024

Light-Metal Functionalized Boron Monoxide Monolayers as Efficient Hydrogen Storage Material: Insights from DFT Simulations

Exceptionally high energy density by mass, natural abundance, widespread applications, and environmental friendliness make hydrogen (H2) a front-runner among clean energy options. However, the transition toward clean and renewable energy applications and the actualization of H2 economy require an efficient H2 storage medium. Material-based H2 storage is a viable option, as liquefaction and storage under pressure require ultra-low temperature (-253°C) and tremendously high pressure (700 atm), respectively. In this work, we highlight the exceptional H2 storage capabilities of recently synthesized boron monoxide (BO) monolayer functionalized with light metals (Li, Na, K, and Ca). Our computational approach, employing density functional theory (DFT), ab initio molecular dynamics (AIMD), and thermodynamic analysis, reveals promising results. We found that up to four metal dopants (Li, Na, K, and Ca) can be adsorbed onto BO monolayer with significantly strong binding energies. Importantly, these bindings surpass the cohesive counterparts of the parental metal bulks, consequently stabilizing the crystal integrities, as confirmed by AIMD simulations. Each metal dopant on BO efficiently adsorbs multiple H2 molecules through electrostatic and van der Waals interactions. Interestingly, the metal-functionalized BO monolayers exhibit exceptionally high H2 gravimetric capacities up to 11.75 wt%. These promising capacities exceed the 5.50 wt% target set by the US Department of Energy for 2025. Following the same guidelines, the average binding energy per H2 molecule is within the range of -0.17 to -0.32 eV. The adsorption and desorption of H2 under practical working conditions are investigated by Langmuir adsorption model based statistical thermodynamic analysis, further supporting the potential of metal-functionalized BO monolayers for material-based H2 storage applications.

cond-mat.mtrl-sci↗

arXiv2023

Open Source High Fidelity Modeling of a Type 5 Wind Turbine Drivetrain for Grid Integration

The increasing integration of renewable energy resources in evolving bulk power system (BPS) is impacting the system inertia. Type-5 wind turbine generation has the potential to behave like a traditional synchronous generator and can help improve system inertia. Hydraulic torque converter (TC) and gearbox with torque limiting feature are integral parts of a Type-5 wind turbine unit. High fidelity model of Type-5 wind turbine including these core components is not openly and widely available for grid integration and transient stability studies. This hinders appropriate assessment of Type-5 wind power plant's contribution to bulk grid resilience. This work develops a TC model based on those generally used in automobile's transmission system. Moreover, the concept of torsional coupling is leveraged to integrate the TC and gearbox system dynamics. The entire integrated model will be open sourced and publicly available for grid integration studies.

eess.SY↗

arXiv2023

Local and Global Contextual Features Fusion for Pedestrian Intention Prediction

Autonomous vehicles (AVs) are becoming an indispensable part of future transportation. However, safety challenges and lack of reliability limit their real-world deployment. Towards boosting the appearance of AVs on the roads, the interaction of AVs with pedestrians including "prediction of the pedestrian crossing intention" deserves extensive research. This is a highly challenging task as involves multiple non-linear parameters. In this direction, we extract and analyse spatio-temporal visual features of both pedestrian and traffic contexts. The pedestrian features include body pose and local context features that represent the pedestrian's behaviour. Additionally, to understand the global context, we utilise location, motion, and environmental information using scene parsing technology that represents the pedestrian's surroundings, and may affect the pedestrian's intention. Finally, these multi-modality features are intelligently fused for effective intention prediction learning. The experimental results of the proposed model on the JAAD dataset show a superior result on the combined AUC and F1-score compared to the state-of-the-art.

cs.CV↗

arXiv2023

Molybdenum Carbide MXenes as Efficient Nanosensors Towards Selected Chemical Warfare Agents

There has been budding demand for the fast, reliable, inexpensive, non-invasive, sensitive, and compact sensors with low power consumption in various fields, such as defence, chemical sensing, health care, and safe environment monitoring units. Particularly, an efficient detection of chemical warfare agents (CWAs) is of great importance for the safety and security of the humans. Inspired by this, we explored molybdenum carbide MXenes (Mo2CTx; Tx= O, F, S) as efficient sensors towards selected CWAs, such as arsine (AsH3), mustard gas (C4H8Cl2S), cyanogen chloride (NCCl), and phosgene (COCl2) both in aqueous and non-aqueous mediums. Our van der Waals corrected density functional theory (DFT) calculations reveal that the CWAs bind with Mo2CF2, and Mo2CS2 monolayers under strong chemisorption with binding energies in the range of -2.33 to -4.05 eV, whereas Mo2CO2 results in comparatively weak bindings of -0.29 to -0.58 eV. We further report the variations in the electronic properties, electrostatic potentials and work functions of Mo2CTx upon the adsorption of CWAs, which authenticate an efficient sensing mechanism. Statistical thermodynamic analysis is applied to explore the sensing properties of Mo2CTx at various of temperatures and pressures. We believe that our findings will pave the way to an innovative class of low-cost reusable sensors for the sensitive and selective detection of highly toxic CWAs in air as well as in aqueous media.

physics.comp-ph↗