Obed Issakah | Searcharxiv

arXiv2026

Graph-Based Kirchhoff Modeling of Non-Ohmic Electron Transport in Self-Assembled Nanonecklace Networks

Gold nanonecklace networks exhibit non-Ohmic current--voltage I-V characteristics with a sharp conduction threshold, making them promising for sensing and transistor-like devices. However, their conduction has been understood largely empirically, because experiments cannot see how conduction evolves inside the network. To address this, we develop a graph-based Kirchhoff modeling framework that resolves these internal processes, tracking the potential and current of every junction. Although every active element conducts linearly, the model reproduces the non-Ohmic response $I \propto (V-V_T)^ζ$. It also captures the experimentally observed effects of junction properties, network density, and voids, which lends confidence to its predicted internal states. These states show that the non-Ohmic response emerges from voltage-driven percolation. Conduction is governed by how junctions are connected rather than by how many are active, with the current scaling approximately as $N^{2.3}$ while the number of active junctions grows only linearly. Contrary to the expectation that junction heterogeneity strengthens the nonlinear response, homogeneous junctions in a well-connected network produce the most abrupt and reproducible turn-on. Thus, each structural parameter acts as a distinct control knob. The mean junction characteristics set the threshold, junction uniformity sets the switching sharpness, and network density and voids set the connectivity and current. These relationships provide design rules for tailoring nanonecklace networks to a desired conduction behavior through synthesis and fabrication.

cond-mat.mes-hall↗

arXiv2026

Functional-Group Knowledge Is Structured in Language Models Before It Is Spoken

Language models serve as the reasoning core of agents for chemistry and materials science, where they must interpret molecular structure before planning syntheses or proposing materials. Yet their outputs show only what they say, not what they know. Given a prompt such as ``The primary functional group in 1-Fluorobutane (CCCCF) is'', a small general-purpose model may answer ``the alkyne group'' instead of an alkyl fluoride. The output alone cannot tell whether the model failed to recognize the fluorine or recognized it and failed to say so. We address this by reading the internal representations of five open-source models on prompts covering 92 molecules and 20 functional groups. Even without chemistry training, the models organize functional groups by chemical family, and they encode chemical differences as consistent directions. The shift from thioether ($\mathrm{R{-}S{-}R'}$) to thiol ($\mathrm{R{-}SH}$), added to ether ($\mathrm{R{-}O{-}R'}$), lands nearest alcohol ($\mathrm{R{-}OH}$). This structure emerges abruptly in the middle layers, where representations shift from tracking the wording of the prompt to tracking the functional group. The base model names the correct functional group in only 58.5\% of prompts, the lowest of the five, yet it encodes functional-group families and preserves chemically meaningful patterns. Its errors are mostly failures to answer rather than wrong answers. Functional-group knowledge is structured in the model before it is spoken. Chemistry fine-tuning reorganizes it, making functional groups rather than prompt wording dominate the representation. These results suggest that evaluating chemical language models through outputs alone can underestimate the chemical information they encode. Representation analysis can distinguish a lack of knowledge from a failure to express it, which matters when choosing or adapting models for chemical agents.

cs.AI↗