arXiv subjects

Tienchong Chang

Publications and source records attributed to Tienchong Chang.

8 recordsLinked to original sources

arXiv2025

Modeling and Inverse Identification of Interfacial Heat Conduction in Finite Layer and Semi-Infinite Substrate Systems via a Physics-Guided Neural Framework

Heat transfer in semiconductor devices is dominated by chip and substrate assemblies, where heat generated within a finite chip layer dissipates into a semi-infinite substrate with much higher thermophysical properties. This mismatch produces steep interfacial temperature gradients, making the transient thermal response highly sensitive to the interface. Conventional numerical solvers require excessive discretization to resolve these dynamics, while physics-informed neural networks (PINNs) often exhibit unstable convergence and loss of physical consistency near the material interface. To address these challenges, we introduce HeatTransFormer, a physics-guided Transformer architecture for interface-dominated diffusion problems. The framework integrates physically informed spatiotemporal sampling, a Laplace-based activation emulating analytical diffusion solutions, and a mask-free attention mechanism supporting bidirectional spatiotemporal coupling. These components enable the model to resolve steep gradients, maintain physical consistency, and remain stable where PINNs typically fail. HeatTransFormer produces coherent temperature fields across the interface when applied to a finite layer and semi-infinite substrate configuration. Coupled with a physics-constrained inverse strategy, it further enables reliable identification of three unknown thermal properties simultaneously using only external measurements. Overall, this work demonstrates that physics-guided Transformer architectures provide a unified framework for forward and inverse modeling in interface-dominated thermal systems.

cs.LG↗

arXiv2023

Human body heat shapes the pattern of indoor disease transmission

Exhaled droplet and aerosol-mediated transmission of respiratory diseases, including SARS-CoV-2, is exacerbated in poorly ventilated environments where body heat-driven airflow prevails. Employing large-scale simulations, we reveal how the human body heat can potentially spread pathogenic species between occupants in a room. Morphological phase transition in airflow takes place as the distance between human heat sources is varied which shapes novel patterns of disease transmission: For sufficiently large distance, individual buoyant plume creates a natural barrier, forming a ``thermal armour'' that blocks suspension spread between occupants. However, for small distances, collective effect emerges and thermal plumes condense into super-structure, facilitating long-distance suspension transport via crossing between convection rolls. Our quantitative analysis demonstrates that infection risk increases significantly at critical distances due to collective behavior and phase transition. This highlights the importance of maintaining reasonable social distancing indoors to minimize viral particle transmission and offers new insights into the critical behavior of pathogen spread.

physics.flu-dyn↗

arXiv2018

Machine learning valence force field model

The valence force field (VFF) model is a concise physical interpretation of the atomic interaction in terms of the bond and angle variations in the explicit quadratic functional form, while the machine learning (ML) method is a flexible numerical approach to make predictions based on some pre-obtained training data without the need of any explicit functions. We propose a so-called ML-VFF model, by combining the clear physical essence of the VFF model and the numerical flexibility of the ML method. Instead of imposing any explicit functional forms for the atomic interaction, the ML-VFF model predicts the potential and force with the Gaussian regression approach. We take graphene as an example to illustrate the ability of the ML-VFF model to make accurate predictions with relatively low computational expenses. We also discuss some key advantages and drawbacks of the ML-VFF model.

physics.comp-ph↗

arXiv2017

Intrinsic energy conversion mechanism via telescopic extension and retraction of concentric carbon nanotubes

The conversion of other forms of energy into mechanical work through the geometrical extension and retraction of nanomaterials has a wide variety of potential applications, including for mimicking biomotors. Here, using molecular dynamic simulations, we demonstrate that there exists an intrinsic energy conversion mechanism between thermal energy and mechanical work in the telescopic motions of double-walled carbon nanotubes (DWCNTs). A DWCNT can inherently convert heat into mechanical work in its telescopic extension process, while convert mechanical energy into heat in its telescopic retraction process. These two processes are thermodynamically reversible. The underlying mechanism for this reversibility is that the entropy changes with the telescopic overlapping length of concentric individual tubes. We find also that the entropy effect enlarges with the decreasing intertube space of DWCNTs. As a result, the spontaneously telescopic motion of a condensed DWCNT can be switched to extrusion by rising the system temperature above a critical value. These findings are important for fundamentally understanding the mechanical behavior of concentric nanotubes, and may have general implications in the application of DWCNTs as linear motors in nanodevices.

cond-mat.mes-hall↗

arXiv2016

Tunable negative Poisson's ratio in hydrogenated graphene

We perform molecular dynamics simulations to investigate the effect of hydrogenation on the Poisson's ratio of graphene. It is found that the value for Poisson's ratio in graphene can be effectively tuned from positive to negative by varying the percentage of hydrogenation. Specifically, the Poisson's ratio decreases with the increase of the percentage of hydrogenation, and reaches a minimum value of -0.04 at the percentage of hydrogenation about 50%. The Poisson's ratio starts to increase with a further increase of the percentage of hydrogenation. The appearance of minimum negative Poisson's ratio in the hydrogenated graphene is attributed to the suppression of the hydrogenation-induced ripples during stretching of graphene. Our results demonstrate that hydrogenation is a valuable approach for the tuning of the Poisson's ratio from positive to negative in graphene.

cond-mat.mtrl-sci↗

arXiv2016

Intrinsic Negative Poisson's Ratio for Single-Layer Graphene

Negative Poisson's ratio (NPR) materials have drawn significant interest because the enhanced toughness, shear resistance and vibration absorption that typically are seen in auxetic materials may enable a range of novel applications. In this work, we report that single-layer graphene exhibits an intrinsic NPR, which is robust and independent of its size and temperature. The NPR arises due to the interplay between two intrinsic deformation pathways (one with positive Poisson's ratio, the other with NPR), which correspond to the bond stretching and angle bending interactions in graphene. We propose an energy-based deformation pathway criteria, which predicts that the pathway with NPR has lower energy and thus becomes the dominant deformation mode when graphene is stretched by a strain above 6%, resulting in the NPR phenomenon.

cond-mat.mtrl-sci↗

arXiv2015

Gas-like adhesion of two-dimensional materials onto solid surfaces

The adhesion of two-dimensional (2D) materials to other surfaces is so far believed to be a solid-solid mechanical contact. Here, we conduct both atomistic simulations and theoretical modeling to show that there exists a reversible conversion of energy between thermal and mechanical work in the attachment/detachment of 2D materials on/off a surface, indicating that 2D materials adhesion is fundamentally like gas adsorption rather than solid adhesion. We reveal that the underlying mechanism of this intriguing gas-like adhesion for 2D materials is the entropy difference between their freestanding and adhered states. Both the theoretical model and atomistic simulations predict that adhesion induced entropy difference increases with increasing adhesion energy and decreasing equilibrium binding distance. The present findings provide a fundamental guidance toward understanding the adhesion of 2D materials, which is important for designing 2D materials based devices and may have general implications for nanoscale efficient energy conversion.

cond-mat.mtrl-sci↗

arXiv2015

The Effects of Free Edge Interaction-Induced Knotting on the Buckling of Monolayer Graphene

Edge effects play an important role for many properties of graphene. While most works have focused on the effects from isolated free edges, we present a novel knotting phenomenon induced by the interactions between a pair of free edges in graphene, and investigate its effect on the buckling of monolayer graphene. Upon compression, the buckling of graphene starts gradually in the form of two buckling waves from the warped edges. The collision of these two buckling waves results in the creation of a knot structure in graphene. The knot structure enables the buckled graphene to exhibit two unique post-buckling characteristics. First, it induces a five-fold increase in graphene's mechanical stiffness during the buckling process. Second, the knotted structure enables graphene to exhibit a mechanically stable post-buckling regime over a large (3%) compressive strain regime, which is significantly larger than the critical buckling strain of about 0.5%. The combination of these two effects enables graphene to exhibit an unexpected post-buckling stability that has previously not been reported. We predict that numerical simulations or experiments should observe two distinct stress strain relations for the buckling of identical graphene samples, due to the characteristic randomness in the formation process of the knot structure.

cond-mat.mes-hall↗