Explore the vast range of titles available.

Cavitation has a potential effect on the performance of full circle journal bearings. This paper studied the effects of cavitation on steadily loaded journal bearings, with the purpose of analyzing the necessity of adopting a mass-conserving model for ordinary journal bearings. The Christopherson’s method and Elrod cavitation algorithm were implemented to represent the non-mass-conserving Reynolds model and the mass-conserving Jakobsson-Floberg-Olsson (JFO) theory, respectively. The difference in the oil film reformation boundaries predicted by the two methods was focused on. The typical performance parameters including oil film pressure, load-carrying capacity, attitude angle, friction force, and leakage were comprehensively compared. The results show that the load-carrying capacity is improved by the decrease in cavitation pressure, and the effect is significant in lightly loaded cavitated bearings. In non-cavitated cases and the cavitated cases with intermediate and heavy loads, the difference between the Reynolds model and the JFO theory can be effectively ignored, but the accuracy of the leakage predicted using the Reynolds model should be carefully evaluated.

Chemical looping biomass gasification is a novel technology converting biomass into syngas, and the selection of oxygen carrier is key for efficient tar conversion. The performance of LaFe1-xNix as a robust catalytic oxygen carrier was investigated in the chemical looping conversion of toluene (tar model compound) into syngas in a fixed bed. LaM (M = Fe, Ni, Mn, Co, and Cu) was initially compared to evaluate the effect of transition metal on toluene conversion. LaFe (partial oxidation) and LaNi (catalytic pyrolysis) exhibited better performance in promoting syngas production than other oxygen carriers. Therefore, Ni-substituted ferrite LaFe1-xNix (x = 0, 0.2, 0.4, 0.6, 0.8 and 1) was further developed. The effects of Ni-substitution, steam/carbon ratio (S/C), and temperature on toluene conversion into C1 and H2 were evaluated. Results showed that the synergistic effect of Fe and Ni promoted toluene conversion, improving H2 yield yet with serious carbon deposition. Steam addition promoted toluene steam reforming and carbon gasification. With S/C increasing from 0.8 to 2.0, the C1 and H2 yield increased from 73.9% to 97.5% and from 197.7% to 269.6%, respectively. The elevated temperature favored toluene conversion and C1 yield. LaFe0.6Ni0.4 exhibited strong reactivity stability during toluene conversion at S/C = 1.6 and 900 °C.

Chemical looping gasification (CLG) is an effective technology for efficient utilization of coal, biomass and other fuels. In this work, the detailed mechanism of catalytic decomposition during CLG for toluene, a tar model compound, was studied by using reactive force field molecular dynamics (ReaxFF MD) method. Results show that toluene hardly decomposes at temperature lower than 2000 K. Improving temperature could significantly improve decomposition efficiency but also enhances the polymerization to produce PAHs and soot precursor, with largest molecule weight of 2175 (C 177 H 51 , 3000 K, 400 ps). Fe 2 O 3 nanocluster, as oxygen carrier, could improve the decomposition efficiency of toluene and reduce the decomposition temperature. At 2000 K and 200 ps, the catalytic conversion of toluene reaches 60%. A large amount of H 2 , CO, C 2 H 2 and other small molecular gases are generated during the catalytic decomposition of toluene. At 3000 K, the yield of H 2 , CO and C 2 H 2 reached 132 %mole, 117 %mole and 40 %mole of toluene, respectively. Meanwhile, polymerization reactions are largely inhibited by Fe 2 O 3 nanocluster and the largest molecule is C 20 H 9 O, the weight of which is much lower than soot precursor in thermal decomposition. Kinetic results show that the activated energy of catalytic decomposition is about 74 kJ/mole, which is much lower than thermal decomposition (382 kJ/mole). Detailed reaction mechanism reveals that lattice oxygen on Fe 2 O 3 nanocluster act as the active sites, which enhance the decomposition of toluene. Graphical Abstract

Chemical looping gasification (CLG) is an effective technology for efficient utilization of coal, biomass and other fuels. In this work, the detailed mechanism of catalytic decomposition during CLG for toluene, a tar model compound, was studied by using reactive force field molecular dynamics (ReaxFF MD) method. Results show that toluene hardly decomposes at temperature lower than 2000 K. Improving temperature could significantly improve decomposition efficiency but also enhances the polymerization to produce PAHs and soot precursor, with largest molecule weight of 2175 (C.sub.177H.sub.51, 3000 K, 400 ps). Fe.sub.2O.sub.3 nanocluster, as oxygen carrier, could improve the decomposition efficiency of toluene and reduce the decomposition temperature. At 2000 K and 200 ps, the catalytic conversion of toluene reaches 60%. A large amount of H.sub.2, CO, C.sub.2H.sub.2 and other small molecular gases are generated during the catalytic decomposition of toluene. At 3000 K, the yield of H.sub.2, CO and C.sub.2H.sub.2 reached 132 %mole, 117 %mole and 40 %mole of toluene, respectively. Meanwhile, polymerization reactions are largely inhibited by Fe.sub.2O.sub.3 nanocluster and the largest molecule is C.sub.20H.sub.9O, the weight of which is much lower than soot precursor in thermal decomposition. Kinetic results show that the activated energy of catalytic decomposition is about 74 kJ/mole, which is much lower than thermal decomposition (382 kJ/mole). Detailed reaction mechanism reveals that lattice oxygen on Fe.sub.2O.sub.3 nanocluster act as the active sites, which enhance the decomposition of toluene. Graphical

Chemical looping gasification (CLG) is an effective technology for efficient utilization of coal, biomass and other fuels. In this work, the detailed mechanism of catalytic decomposition during CLG for toluene, a tar model compound, was studied by using reactive force field molecular dynamics (ReaxFF MD) method. Results show that toluene hardly decomposes at temperature lower than 2000 K. Improving temperature could significantly improve decomposition efficiency but also enhances the polymerization to produce PAHs and soot precursor, with largest molecule weight of 2175 (C.sub.177H.sub.51, 3000 K, 400 ps). Fe.sub.2O.sub.3 nanocluster, as oxygen carrier, could improve the decomposition efficiency of toluene and reduce the decomposition temperature. At 2000 K and 200 ps, the catalytic conversion of toluene reaches 60%. A large amount of H.sub.2, CO, C.sub.2H.sub.2 and other small molecular gases are generated during the catalytic decomposition of toluene. At 3000 K, the yield of H.sub.2, CO and C.sub.2H.sub.2 reached 132 %mole, 117 %mole and 40 %mole of toluene, respectively. Meanwhile, polymerization reactions are largely inhibited by Fe.sub.2O.sub.3 nanocluster and the largest molecule is C.sub.20H.sub.9O, the weight of which is much lower than soot precursor in thermal decomposition. Kinetic results show that the activated energy of catalytic decomposition is about 74 kJ/mole, which is much lower than thermal decomposition (382 kJ/mole). Detailed reaction mechanism reveals that lattice oxygen on Fe.sub.2O.sub.3 nanocluster act as the active sites, which enhance the decomposition of toluene.

Large language models (LLMs) have brought exciting new advances to mobile UI agents, a long-standing research field that aims to complete arbitrary natural language tasks through mobile UI interactions. However, existing UI agents usually demand high reasoning capabilities of powerful large models that are difficult to be deployed locally on end-users' devices, which raises huge concerns about user privacy and centralized serving cost. One way to reduce the required model size is to customize a smaller domain-specific model with high-quality training data, e.g. large-scale human demonstrations of diverse types of apps and tasks, while such datasets are extremely difficult to obtain. Inspired by the remarkable coding abilities of recent small language models (SLMs), we propose to convert the UI task automation problem to a code generation problem, which can be effectively solved by an on-device SLM and efficiently executed with an on-device code interpreter. Unlike normal coding tasks that can be extensively pretrained with public datasets, generating UI automation code is challenging due to the diversity, complexity, and variability of target apps. Therefore, we adopt a document-centered approach that automatically builds fine-grained API documentation for each app and generates diverse task samples based on this documentation. By guiding the agent with the synthetic documents and task samples, it learns to generate precise and efficient scripts to complete unseen tasks. Based on detailed comparisons with state-of-the-art mobile UI agents, our approach effectively improves the mobile task automation with significantly higher success rates and lower latency/token consumption. Code will be open-sourced.

Mobile task automation is an attractive technique that aims to enable voice-based hands-free user interaction with smartphones. However, existing approaches suffer from poor scalability due to the limited language understanding ability and the non-trivial manual efforts required from developers or end-users. The recent advance of large language models (LLMs) in language understanding and reasoning inspires us to rethink the problem from a model-centric perspective, where task preparation, comprehension, and execution are handled by a unified language model. In this work, we introduce AutoDroid, a mobile task automation system capable of handling arbitrary tasks on any Android application without manual efforts. The key insight is to combine the commonsense knowledge of LLMs and domain-specific knowledge of apps through automated dynamic analysis. The main components include a functionality-aware UI representation method that bridges the UI with the LLM, exploration-based memory injection techniques that augment the app-specific domain knowledge of LLM, and a multi-granularity query optimization module that reduces the cost of model inference. We integrate AutoDroid with off-the-shelf LLMs including online GPT-4/GPT-3.5 and on-device Vicuna, and evaluate its performance on a new benchmark for memory-augmented Android task automation with 158 common tasks. The results demonstrated that AutoDroid is able to precisely generate actions with an accuracy of 90.9%, and complete tasks with a success rate of 71.3%, outperforming the GPT-4-powered baselines by 36.4% and 39.7%. The demo, benchmark suites, and source code of AutoDroid will be released at url{https://autodroid-sys.github.io/}.

Energy transferred via thermal radiation between two surfaces separated by nanometer distances can be much larger than the blackbody limit. However, realizing a scalable platform that utilizes this near-field energy exchange mechanism to generate electricity remains a challenge. Here, we present a fully integrated, reconfigurable and scalable platform operating in the near-field regime that performs controlled heat extraction and energy recycling. Our platform relies on an integrated nano-electromechanical system that enables precise positioning of a thermal emitter within nanometer distances from a room-temperature germanium photodetector to form a thermo-photovoltaic cell. We demonstrate over an order of magnitude enhancement of power generation ( P gen  ~ 1.25 μWcm −2 ) in our thermo-photovoltaic cell by actively tuning the gap between a hot-emitter ( T E  ~ 880 K) and the cold photodetector ( T D  ~ 300 K) from ~ 500 nm down to ~ 100 nm. Our nano-electromechanical system consumes negligible tuning power ( P gen / P NEMS ~ 10 4 ) and relies on scalable silicon-based process technologies. Designing a scalable platform to generate electricity from the energy exchange mechanism between two surfaces separated by nanometer distances remains a challenge. Here, the authors demonstrate reconfigurable, scalable and fully integrated near-field thermo-photovoltaics for on-demand heat recycling.

We study the relation between angular spectral absorptivity and emissivity for any thermal emitter, which consists of any linear media that can be dispersive, inhomogeneous, bianisotropic, or nonreciprocal. First, we establish an adjoint Kirchhoff’s law for mutually adjoint emitters. This law is based on generalized reciprocity and is a natural generalization of conventional Kirchhoff’s law for reciprocal emitters. Using this law, we derive all the relations between absorptivity and emissivity for an arbitrary thermal emitter We reveal that such relations are determined by the symmetries of the system, which are characterized by a Shubnikov point group. We classify all thermal emitters based on their symmetries using the known list of all three-dimensional Shubnikov point groups. Each class possesses its own set of laws that relates the absorptivity and emissivity. We numerically verify our theory for all three types of Shubnikov point groups: Gray groups, colorless groups, and black and white groups. We also verify the theory for both planar and nonplanar structures with single or multiple diffraction channels. Our theory provides a theoretical foundation for further exploration of thermal radiation in general media.

Direct current (DC) converters play an essential role in electronic circuits. Conventional high-efficiency DC voltage converters, especially step-up type, rely on switching operation, where energy is periodically stored within and released from inductors and/or capacitors connected in a variety of circuit topologies. Since these energy storage components, especially inductors, are fundamentally difficult to scale down, miniaturization of switching converters proves challenging. Furthermore, the resulting switching currents produce significant electromagnetic noise. To overcome the limitations of switching converters, photonic transformers, where voltage conversion is achieved through light emission and detection processes, have been demonstrated. However, the demonstrated efficiency is significantly below that of the switching converter. Here we perform a detailed balance analysis and show that with a monolithically integrated design that enables efficient photon transport, the photonic transformer can operate with a near-unity conversion efficiency and high voltage conversion ratio. We validate the theory with a transformer constructed with off-the-shelf discrete components. Our experiment showcases near noiseless operation and a voltage conversion ratio that is significantly higher than obtained in previous photonic transformers. Our findings point to the possibility of a high-performance optical solution to miniaturizing DC power converters and improving the electromagnetic compatibility and quality of electrical power. Conventional DC-DC converters rely on switching operations and energy storing components which face both noise and scaling difficulties. Here, the authors present an alternative design for a DC-to-DC converter based on closely coupled LEDs and photovoltaic cells, which exhibits high efficiency, low noise, and miniaturizability.