Lehigh University

About: Lehigh University is a education organization based out in Bethlehem, Pennsylvania, United States. It is known for research contribution in the topics: Catalysis & Fracture mechanics. The organization has 12684 authors who have published 26550 publications receiving 770061 citations.

Fast drop movements resulting from the phase change on a gradient surface.

Abstract: The movement of liquid drops on a surface with a radial surface tension gradient is described here. When saturated steam passes over a colder hydrophobic substrate, numerous water droplets nucleate and grow by coalescence with the surrounding drops. The merging droplets exhibit two-dimensional random motion somewhat like the Brownian movements of colloidal particles. When a surface tension gradient is designed into the substrate surface, the random movements of droplets are biased toward the more wettable side of the surface. Powered by the energies of coalescence and collimated by the forces of the chemical gradient, small drops (0.1 to 0.3 millimeter) display speeds that are hundreds to thousands of times faster than those of typical Marangoni flows. This effect has implications for passively enhancing heat transfer in heat exchangers and heat pipes.

Mode-Division Multiplexing Over 96 km of Few-Mode Fiber Using Coherent 6 $\,\times\,$ 6 MIMO Processing

Abstract: We report simultaneous transmission of six spatial and polarization modes, each carrying 40 Gb/s quadrature-phase-shift-keyed channels over 96 km of a low-differential group delay few-mode fiber. The channels are successfully recovered by offline DSP based on coherent detection and multiple-input multiple-output processing. A penalty of ;28 dB.

DeepXplore: Automated Whitebox Testing of Deep Learning Systems

Abstract: Deep learning (DL) systems are increasingly deployed in safety- and security-critical domains including self-driving cars and malware detection, where the correctness and predictability of a system's behavior for corner case inputs are of great importance Existing DL testing depends heavily on manually labeled data and therefore often fails to expose erroneous behaviors for rare inputs We design, implement, and evaluate DeepXplore, the first whitebox framework for systematically testing real-world DL systems First, we introduce neuron coverage for systematically measuring the parts of a DL system exercised by test inputs Next, we leverage multiple DL systems with similar functionality as cross-referencing oracles to avoid manual checking Finally, we demonstrate how finding inputs for DL systems that both trigger many differential behaviors and achieve high neuron coverage can be represented as a joint optimization problem and solved efficiently using gradient-based search techniques DeepXplore efficiently finds thousands of incorrect corner case behaviors (eg, self-driving cars crashing into guard rails and malware masquerading as benign software) in state-of-the-art DL models with thousands of neurons trained on five popular datasets including ImageNet and Udacity self-driving challenge data For all tested DL models, on average, DeepXplore generated one test input demonstrating incorrect behavior within one second while running only on a commodity laptop We further show that the test inputs generated by DeepXplore can also be used to retrain the corresponding DL model to improve the model's accuracy by up to 3%

Field assessment of nanoscale bimetallic particles for groundwater treatment.

Abstract: A field demonstration was performed in which nanoscale bimetallic (Fe/Pd) particles were gravity-fed into groundwater contaminated by trichloroethene and other chlorinated aliphatic hydrocarbons at a manufacturing site. With diameters on the order of 100−200 nm, the nanoparticles are uniquely suited to rapidly degrade redox-amenable contaminants and for optimal subsurface delivery and dispersion. Approximately 1.7 kg of the nanoparticles was fed into the test area over a 2-day period, resulting in minimal clogging of the injection well. The test area was located within a well-characterized region of the contaminant plume and included an injection well and three piezometer couplets spaced 1.5 m apart. Despite the low nanoparticle dosage, trichloroethene reduction efficiencies of up to 96% were observed over a 4-week monitoring period with the highest values observed at the injection well and adjacent piezometers. Data from the field assessment were consistent with the results of pre-injection laboratory st...