11 February 2025 in IEEE Transactions on Computers
Physical phenomena such as bond breaking and phase transitions require molecular dynamics (MD) with ab initio accuracy, involving up to billions of atoms and over nanosecond timescales. Previous state-of-the-art work has demonstrated that neural network molecular dynamics (NNMD) like deep potential molecular dynamics (DeePMD), can successfully extend the temporal and spatial scales of MD with ab initio accuracy on both ARM and GPU platforms. However, the DeePMD-kit package is currently unable to fully exploit the computational potential of the new Sunway supercomputer due to its unique many-core architecture, memory hierarchy, and low precision capability. In this paper, we re-design the DeePMD-kit to harness the massive computing power of the new Sunway, enabling the MD with over ten billion atoms. We first design a large-scale parallelization scheme to exploit the massive parallelism of the new Sunway. Then we devise specialized optimizations for the time-consuming operators. Finally, we design a novel mixed precision method for DeePMD-kit customized operators to leverage the low precision computing power of the new Sunway. The optimized DeePMD-kit achieves 67.6 / 56.5 × speedup for water / copper systems on the new Sunway. Meanwhile, it can perform 29 billion atoms simulation for the water system on 35 million cores (i.e., 90,000 computing nodes, around 84% of the whole supercomputer) with a peak performance of 57.1 PFLOPs, which is 7.9× bigger and 1.2× faster than state-of-the-art results. This paves the way for investigating more realistic scenarios, such as studying the mechanical properties of metals, semiconductor devices, batteries, and other materials and physical systems.
18 July 2023 in Computational Materials Science
1 December 2024 in AIP Advances
Humidity sensors are integral to modern technology and extensively utilize Ti3C2Tx MXene for its superior sensitivity and accuracy. This material’s effectiveness stems from its abundant hydrophilic functional groups and high electrical conductivity. In this study, we enhanced the responsiveness to relative humidity changes by incorporating bentonite, a 2D clay mineral known for its robust water adsorption capacity and ionic conductivity, attributable to its distinctive structure. Our analysis of MXene/bentonite thin films through sheet resistance measurements indicated a clear inverse relationship in resistance changes above and below a specific MXene-to-bentonite ratio. Impedance analysis further revealed that the dominant conduction mechanisms within these films vary according to the mix ratio. In fabricating conductive-type humidity sensors from MXene/bentonite films, we observed a remarkable increase in humidity sensitivity. Particularly, MXene rich film (33 wt. % MXene) demonstrated a sensitivity of ∼58.1 times greater than MXene film, and bentonite rich film (10 wt. % MXene) demonstrated about 5000 times greater than MXene film. This study not only underscores the potential of MXene/bentonite composites in humidity sensing but also highlights the critical role of composition in optimizing sensor performance.
8 November 2011 in Physica Status Solidi (b)
1 August 2025 in Review of Scientific Instruments
The thermal transport properties of materials are critical in the thermal design of electronic devices, particularly as these designs increase in complexity. Modern devices frequently incorporate materials with low thermal conductivity, such as insulators and flexible substances used in wearable technology. Frequency-domain thermoreflectance (FDTR) is a common method for measuring thermal conductivity due to its simplicity and accuracy. However, challenges arise when measuring materials with low thermal conductivity, as the sensitivity of the measurement diminishes. This reduction in sensitivity is primarily due to significant lateral heat transfer within the metal transducer on the sample’s surface, complicating the precise determination of thermal conductivity. This study introduces a method to enhance measurement precision by minimizing heat transfer within the transducer. This approach involves selectively patterning the transducer and positioning it near the laser irradiation point to curtail internal heat transfer. This strategy ensures that heat transfer within the sample predominates, thereby augmenting the sensitivity of thermal conductivity measurements. We employed a finite element model to evaluate the effectiveness of this technique and established that transducer patterning notably diminishes the uncertainty in measured thermal conductivities. Validation was achieved by measuring the thermal conductivity of glass [1.216 ± 0.056 W/(m K)]. This measurement is consistent with anticipated values for glass conductivity and demonstrates a reduction in uncertainty to less than one-fourth of that observed in conventional FDTR measurements. Such improvements highlight the method’s capability to yield precise thermal conductivity measurements for low-conductivity materials. This advancement addresses the growing imperative for advanced thermal design in modern electronic devices.