Engineering Papers⌕ Search

Engineering topics

Chen, Zitao

Publications and source records attributed to Chen, Zitao.

4 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Extraordinary Thermal Stability and Sinter Resistance of Sub-2 nm Platinum Nanoparticles Anchored to a Carbon Support by Selenium

Nanoparticle sintering has long been a major challenge in developing catalytic systems for use at elevated temperatures. Here we report an in situ electron microscopy study of the extraordinary sinter resistance of a catalytic system comprised of sub-2 nm Pt nanoparticles on a Se-decorated carbon support. When heated to 700 °C, the average size of the Pt nanoparticles only increased from 1.6 to 2.2 nm, while the crystal structure, together with the {111} and {100} facets, of the Pt nanoparticles was well retained. Our electron microscopy analyses suggested that the superior resistance against sintering originated from the Pt–Se interaction. Confirmed by energy-dispersive X-ray elemental mapping and electron energy loss spectra, the Se atoms surrounding the Pt nanoparticles could survive the heating. This work not only offers an understanding of the physics behind the thermal behavior of this catalytic material but also sheds light on the future development of sinter-resistant catalytic systems.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Fault Injection for TensorFlow Applications

As machine learning (ML) has seen increasing adoption in safety-critical domains (e.g., autonomous vehicles), the reliability of ML systems has also grown in importance. While prior studies have proposed techniques to enable efficient error-resilience (e.g., selective instruction duplication), a fundamental requirement for realizing these techniques is a detailed understanding of the application’s resilience. In this work, we present TensorFI 1 and TensorFI 2, high-level fault injection (FI) frameworks for TensorFlow-based applications. TensorFI 1 and 2 are able to inject both hardware and software faults in any general TensorFlow 1 and 2 program respectively. Both are configurable FI tools that are flexible, easy to use, and portable. They can be integrated into existing TensorFlow programs to assess their resilience for different fault types (e.g., bit-flips in particular operations or layers). We use the TensorFI 1 and TensorFI 2 to evaluate the resilience of 12 and 10 ML programs written in TensorFlow, including DNNs used in the autonomous vehicle domain. The results give us insights into why some of the models are more resilient. We also measure the performance overheads of the two injectors, and present 4 case studies, two for each tool, to demonstrate their utility.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Synthesis and Characterization of Pt‐Ag Icosahedral Nanocages with Enhanced Catalytic Activity toward Oxygen Reduction

Abstract There is an urgent need to develop cost‐effective electrocatalysts based on Pt for a broad spectrum of applications, including those vital to the operation of fuel cells. Hollowing out the interior of Pt nanocrystals offers a simple and viable strategy for maximizing the utilization efficiency of this precious metal while enhancing the electrocatalytic performance. Herein, we report the synthesis and electrocatalytic evaluation of Pt−Ag icosahedral nanocages with an average wall thickness of 1.6 nm. The Pt atoms are coated on the surface of Ag icosahedral seeds, leading to the formation of Ag@Pt nL core‐shell icosahedral nanocrystals with tunable shell thicknesses. The core‐shell nanocrystals are then converted to icosahedral nanocages by selectively etching away the Ag in the core. The as‐obtained nanocages with a composition of Pt 4.5 Ag exhibit an almost 3‐fold enhancement in specific activity toward oxygen reduction relative to the commercial Pt/C in acid media.

Wang, Wenxia↗