11563 research outputs found
Sort by
Investigating cloud-cloud collisions and hub-filament systems: pathways to massive star formation
No full text
Impact of thermally activated ionic dynamics on the trap-mediated current-voltage characteristics of a mixed-halide hybrid perovskite
No full textOrganic-inorganic metal halide perovskites (OIMHPs) are at the forefront of leading energy research. Therefore, it is pivotal to understand the effect of operating conditions like temperature, humidity, light exposure, etc. on these materials. The transient ionic dynamics and its effect on the steady-state J-V characteristics of an OIMHP, viz. FAPbBr2I, having a mixed halide composition, were investigated by temperature-dependent dielectric spectroscopy and temperature-dependent space charge limited current (SCLC) measurements in the temperature range of 305-454 K. The contribution of the resistance and capacitance of grains and grain boundaries to the total impedance at different temperatures has been interpreted by analyzing the Bode plots using the Maxwell-Wagner equivalent circuit model. The AC conductivity spectra demonstrate different behaviors in two different temperature regimes. In the low-temperature (LT) regime (323-381 K), the temperature response of ionic conductivity is only dependent on hopping frequency (the ionic carrier concentration factor being temperature-independent), leading to almost similar activation energies of ionic conduction (Ea) and hopping migration (Em), where Ea = Em = 0.30 ± 0.05 eV. However, in the high temperature (HT) regime (395-454 K), we observed a difference in Ea (0.74 ± 0.05 eV) and Em (0.50 ± 0.05 eV) values, which is attributed to the activation energy of mobile charge carrier formation (Ef = Ea − Em = 0.24 ± 0.05 eV). We propose that the trapped ions in the LT regime are now released by overcoming the barrier Ef in the HT regime, leading to a substantial increase in the mobile ion concentration. Furthermore, we have unveiled the effect of these mobile ions and trapped carriers on the J-V characteristics in both temperature regimes by analyzing the temperature-dependent SCLC J-V characteristics in the Ag/FAPbBr2I/Ag device configuration. The AC conductivity and electric modulus loss spectra scale to different master curves in the LT and HT regimes, further corroborating the observed thermally activated interplay of ionic conduction and hopping migration. The key findings of this work stimulate more such fundamental investigations of electrical transport in mixed halide OIMHPs and establish their potential in various energy storage applications like batteries, integrated PV-battery/supercapacitor systems, and others
Understanding polymorphic behaviour of L-glutamic acid using combined cooling and anti-solvent crystallization using in-situ Raman spectroscopy and MD simulations
No full textSpacers and Mismatches-Based Algorithm (SMA): G Domain Prediction and Application in Human Biology and Human Health
No full textGuanine nucleotide-binding proteins (G proteins) are seen to play an important role in cellular processes such as sensual and visual perception, protein synthesis, vesicular transport, cell growth and differentiation to name a few. The ability of G proteins to function as molecular switches that cycle between a GTP bound (Active) and GDP bound (Inactive) states mediates their cellular functions. A structurally and mechanistically conserved GTP-binding domain (G domain) allows GTP binding to G proteins. The G domain comprises five adjacent consensus motifs called G boxes, which are separated by amino acid spacers of different lengths. Mutations in the G domain coding region of G proteins lead to heritable disorders. Changes in expression of G proteins lead to physiological or neurodegenerative disorders and their treatment. Accurate determination of the G domain sequence in G proteins is of utmost importance for their use in the identification of diseasecausing mutations, identification of mechanism of disorders and exploring their use as therapeutic proteins. In the current study, we introduce a Spacers and Mismatch Algorithm (SMA) which can predict G domains in a given protein sequence, based on user-specified constraints. SMA can be implemented via a web-based server at https://labs.iitgn.ac.in/datascience/gboxes/
Robust and High-Performance Digital In-Memory Computing in 5T Gain Cell Embedded DRAM
No full textThis paper proposes a novel 5T Dual Port Gain Cell (5T-DPGC) embedded DRAM design for digital in-memory computing (IMC). The proposed 5T-DPGC-eDRAM cell enables the execution of N-input logical operations (specifically NAND/NOR) within the IMC without requiring any additional peripherals. Further, it allows for MAC operations using an extra compute circuit. The circuit implementation was carried out using CMOS 28 nm technology. The proposed cell gives an area benefit of 25% compared to standard 6T-SRAM. To evaluate the performance, we performed a post-layout simulation of 16 kB (128 × 128) 5T-DPGC-eDRAM array, and it achieves a throughput of 224 GOPS. Moreover, it demonstrated an energy efficiency of 8.21 TOPS/W at 0.65 V and 0.25 GHz, which is 1.61× better than the SOTA 6T-SRAM-based digital IMC for 8-bit addition. Additionally, our design achieved a frequency of 1.75 GHz at 0.9 V. In terms of area efficiency, our approach reached 1.98 TOPS/mm2, surpassing the SOTA 6T-SRAM-based digital IMC by 1.33×
PolyGuard: a multilingual safety moderation tool for 17 languages
No full textTruly multilingual safety moderation efforts for Large Language Models (LLMs) have been hindered by a narrow focus on a small set of languages (e.g., English, Chinese) as well as a limited scope of safety definition, resulting in significant gaps in moderation capabilities. To bridge these gaps, we release POLYGUARD, a new state-of-the-art multilingual safety model for safeguarding LLM generations, and the corresponding training and evaluation datasets. POLYGUARD is trained on POLYGUARDMIX, the largest multilingual safety training corpus to date containing 1.91M samples across 17 languages (e.g., Chinese, Czech, English, Hindi). We also introduce POLYGUARDPROMPTS, a high quality multilingual benchmark with 29K samples for the evaluation of safety guardrails. Created by combining naturally occurring multilingual human-LLM interactions and human-verified machine translations of an English-only safety dataset (WildGuardMix; Han et al., 2024), our datasets contain prompt-output pairs with labels of prompt harmfulness, response harmfulness, and response refusal. Through extensive evaluations across multiple safety and toxicity benchmarks, we demonstrate that POLYGUARD outperforms existing state-of-the-art open-weight and commercial safety classifiers by 5.5%. Our contributions advance efforts toward safer multilingual LLMs for all global users
Reaction mechanism and micro-structural analysis of ground granulated blast furnace slag activated with lithium hydroxide
No full textMeasuring Student Satisfaction: A Study of MOOC Videos Created with RightScripts
No full textThe rise of MOOCs has transformed higher education, increasing demand for high-quality, learner-centric content. RightScripts, a tool for creating educational video scripts, was developed to help teachers produce effective MOOC videos. This study evaluates RightScripts’ effectiveness in creating instructionally sound, learner-focused scripts, with a focus on student satisfaction. Initially tested by educational technology experts, RightScripts underwent refinements before usability testing by 143 teachers yielded high satisfaction. Teachers then used it to script a B2B marketing MOOC on the SWAYAM platform, with student feedback gathered from 158 respondents. The analysis showed high scores for clarity, objectives, and reflection opportunities, with a Cronbach's alpha of 0.9177 for reliability. Qualitative responses praised instructional quality, highlighting RightScripts’ value in enhancing MOOC video quality and supporting educators