No abstract

highly dependent on the adsorption model of oxygen molecules (O 2 ) on the surface of catalysts. [2] The side-on adsorption with a "*O-O*" configuration (* presents the active site) is conducive to weakening the O-O bond for reduction of O 2 into H 2 O via a four-electron (4e) ORR pathway, [3] while the end-on configuration formed by a solo oxygen atom coordinated on a single active site ("*OOH" intermediate) facilitates to selectively catalyzing oxygen to generate H 2 O 2 via a two-electron (2e) ORR pathway. [4] To realize 2e oxygen electroreduction, various strategies including alloying, [5] chemical functionalization, [6] downsizing, [7] and single-atom engineering [8] have been developed to regulate the physicochemical properties of the catalysts. Though substantial progress has been made, the activity and durability of reported works still cannot compete with the demand of the practical application. [9] The cation vacancy engineering strategy could be an effective approach to develop high-performance catalysts for the electrocatalytic synthesis of H 2 O 2 owing to the following merits: creating cation vacancy on host materials can prolong the distance or spacing of the active sites, thereby leading to the formation of *OOH adsorption favorable; [4a] the charge density between active sites and adjacent coordination atoms will be redistributed, which optimizes the Electrocatalytic hydrogen peroxide (H 2 O 2 ) synthesis via the two-electron oxygen reduction reaction (2e ORR) pathway is becoming increasingly important due to the green production process. Here, cationic vacancies on nickel phosphide, as a proof-of-concept to regulate the catalyst's physicochemical properties, are introduced for efficient H 2 O 2 electrosynthesis. The as-fabricated Ni cationic vacancies (V Ni )-enriched Ni 2−x P-V Ni electrocatalyst exhibits remarkable 2e ORR performance with H 2 O 2 molar fraction of >95% and Faradaic efficiencies of >90% in all pH conditions under a wide range of applied potentials. Impressively, the as-created V Ni possesses superb longterm durability for over 50 h, suppassing all the recently reported catalysts for H 2 O 2 electrosynthesis. Operando X-ray absorption near-edge spectroscopy (XANES) and synchrotron Fourier transform infrared (SR-FTIR) combining theoretical calculations reveal that the excellent catalytic performance originates from the V Ni -induced geometric and electronic structural optimization, thus promoting oxygen adsorption to the 2e ORR favored "end-on" configuration. It is believed that the demonstrated cation vacancy engineering is an effective strategy toward creating active heterogeneous catalysts with atomic precision.

A metallic trench Fabry-Perot resonator achieves room temperature lasing of surface plasmons with a record-narrow linewidth.

In many settings, it is important that a model be capable of providing reasons for its predictions (i.e., the model must be interpretable). However, the model's reasoning may not conform with well-established knowledge. In such cases, while interpretable, the model lacks credibility. In this work, we formally define credibility in the linear setting and focus on techniques for learning models that are both accurate and credible. In particular, we propose a regularization penalty, expert yielded estimates (EYE), that incorporates expert knowledge about well-known relationships among covariates and the outcome of interest. We give both theoretical and empirical results comparing our proposed method to several other regularization techniques. Across a range of settings, experiments on both synthetic and real data show that models learned using the EYE penalty are significantly more credible than those learned using other penalties. Applied to two large-scale patient risk stratification task, our proposed technique results in a model whose top features overlap significantly with known clinical risk factors, while still achieving good predictive performance.

Optical multi-layer thin films are widely used in optical and energy applications requiring photonic designs. Engineers often design such structures based on their physical intuition. However, solely relying on human experts can be time-consuming and may lead to sub-optimal designs, especially when the design space is large. In this work, we frame the multi-layer optical design task as a sequence generation problem. A deep sequence generation network is proposed for efficiently generating optical layer sequences. We train the deep sequence generation network with proximal policy optimization to generate multi-layer structures with desired properties. The proposed method is applied to two energy applications. Our algorithm successfully discovered high-performance designs, outperforming structures designed by human experts in task 1, and a state-of-the-art memetic algorithm in task 2.