Engineering topics
Zeng, Zheng
Publications and source records attributed to Zeng, Zheng.
Stable Low-Current Electrodeposition of α-MnO 2 on Superaligned Electrospun Carbon Nanofibers for High-Performance Energy Storage
Metal oxide/carbonaceous nanomaterials are promising candidates for energy-storage applications. Yet, inhomogeneous mass and charge transfer across the electrode/electrolyte interface due to unstable metal oxide/carbonaceous nanomaterial synthesis limit their performance in supercapacitors. Here, it is shown that the above problems can be mitigated through stable low-current electrodeposition of MnO 2 on superaligned electrospun carbon nanofibers (ECNFs). The key to this method is coupling a self-designed four steel poles collector for aligned ECNFs and a constant lowcurrent (40 μA) electrodeposition technique to form a uniform Na + -induced α-MnO 2 film which proceeds by a time-dependent growth mechanism involving cluster-“kebab” structures and ending with a compact, uniform MnO 2 film for high-performance energy storage.
A fluorescence-electrochemical study of carbon nanodots (CNDs) in bio- and photoelectronic applications and energy gap investigation
Carbon nanodots (CNDs) have attracted significant attention due to their superior solubility, biocompatibility, tunable photoluminescence, and opto-electronic properties. Our work describes a new fluorescence-based spectroelectrochemistry approach to simultaneously study the photoluminescence and wavelength dependent photocurrent of microwave synthesized CNDs. The fluorescence of CNDs shows selective quenching upon a reversible redox couple, ferricyanide/ferrocyanide, reaction during cyclic voltammetry. The CND modified gold slide electrode demonstrates wavelength dependent photocurrent generation during the fluorescence-electrochemical study, suggesting the potential application of CNDs in photoelectronics. UV-Vis absorption and electrochemistry are used to quantify the energy gap of the CNDs, and then to calibrate a Hückel model for CNDs’ electronic energy levels. The Hückel (or tight binding) model treatment of an individual CND as a molecule combines the conjugated π states (C[double bond, length as m-dash]C) with the functional groups (C[double bond, length as m-dash]O, C–O, and COOH) associated with the surface electronic states. This experimental and theoretical investigation of CNDs provides a new perspective on the optoelectronic properties of CNDs and should aid in their development for practical use in biomedicine, chemical sensing, and photoelectric devices.
Improving Gas Flooding Efficiency
This study focuses on laboratory studies with related analytical and numerical models, as well as work with operators for field tests to enhance our understanding of and capabilities for more efficient enhanced oil recovery (EOR). Much of the work has been performed at reservoir conditions. This includes a bubble chamber and several core flood apparatus developed or modified to measure interfacial tension (IFT), critical micelle concentration (CMC), foam durability, surfactant sorption at reservoir conditions, and pressure and temperature effects on foam systems.Carbon dioxide and N{sub 2} systems have been considered, under both miscible and immiscible conditions. The injection of CO2 into brine-saturated sandstone and carbonate core results in brine saturation reduction in the range of 62 to 82% brine in the tests presented in this paper. In each test, over 90% of the reduction occurred with less than 0.5 PV of CO{sub 2} injected, with very little additional brine production after 0.5 PV of CO{sub 2} injected. Adsorption of all considered surfactant is a significant problem. Most of the effect is reversible, but the amount required for foaming is large in terms of volume and cost for all considered surfactants. Some foams increase resistance to the value beyond what is practical in the reservoir. Sandstone, limestone, and dolomite core samples were tested. Dissolution of reservoir rock and/or cement, especially carbonates, under acid conditions of CO2 injection is a potential problem in CO2 injection into geological formations. Another potential change in reservoir injectivity and productivity will be the precipitation of dissolved carbonates as the brine flows and pressure decreases. The results of this report provide methods for determining surfactant sorption and can be used to aid in the determination of surfactant requirements for reservoir use in a CO{sub 2}-foam flood for mobility control. It also provides data to be used to determine rock permeability changes during CO{sub 2} flooding due to saturation changes, dissolution, and precipitation.
Pseudogradient Training For A Class Of Neural Networks
Developmental second-order recurrent neural networks of special type modified to enhance stability in face of inputs beyond range of inputs on which trained. Second-order recurrent neural networks contain product feedback units and can be trained, by use of example inputs and outputs, to act as finite-state automatons. Particular second-order recurrent neural networks in question learn grammars in sense they are trained to generate binary responses to input training sequences of ones and zeros, each sequence being marked "legal" or "illegal" according to grammar to be learned.
Discrete Recurrent Neural Networks as Pushdown Automata
In this paper, we describe a new discrete recurrent model with discrete external stacks for learning context-free grammars (or pushdown automata).
Discrete Recurrent Neural Networks for Grammatical Inference
Recurrent neural networks have recently been shown to have the ability to learn regular and context-free grammars from examples. We show that while conventional analog recurrent networks try to form clusters in activation space to represent discrete states of the grammars during learning, and can be successful in doing so, the clusters so formed tend to become unstable as longer and longer test input strings are presented to the network.