To the Editors-Electrochemical conversion of dinitrogen to ammonia in aqueous electrolyte solutions at high selectivity and rate requires a catalyst with unique properties. In particular, the surface of such a catalyst should exhibit higher affinity for nitrogen, as opposed to hydrogen, to facilitate the nitrogen reduction reaction (NRR) and suppress the hydrogen evolution reaction (HER). Detailed density functional theory (DFT) studies by Skúlason and co-workers 3 suggest that this is hardly possible with most metals traditionally considered in electrocatalytic research. One metal not included in the analysis in ref. 3 and long known to be a very poor HER catalyst 4 is bismuth, which was very recently introduced by Hao et al. 1 as an efficient NRR catalyst in their report in Nature Catalysis. However, DFT data reported in the same paper clearly indicate that N2 adsorption on various bismuth surfaces is highly thermodynamically unfavourable with an essentially insurmountable energy barrier of more than 2.7 eV. Nevertheless, despite this contradiction, the highly impressive experimental data on the Bicatalysed NRR reported by Hao et al. 1 present the best performances reported to date and therefore would represent important progress in this field. This prompted us to attempt to reproduce these experimental results. Details of our materials and equipment are provided

The challenges of developing neuromorphic vision systems inspired by the human eye come not only from how to recreate the flexibility, sophistication, and adaptability of animal systems, but also how to do so with computational efficiency and elegance. Similar to biological systems, these neuromorphic circuits integrate functions of image sensing, memory and processing into the device, and process continuous analog brightness signal in real-time. High-integration, flexibility and ultra-sensitivity are essential for practical artificial vision systems that attempt to emulate biological processing. Here, we present a flexible optoelectronic sensor array of 1024 pixels using a combination of carbon nanotubes and perovskite quantum dots as active materials for an efficient neuromorphic vision system. The device has an extraordinary sensitivity to light with a responsivity of 5.1 × 107 A/W and a specific detectivity of 2 × 1016 Jones, and demonstrates neuromorphic reinforcement learning by training the sensor array with a weak light pulse of 1 μW/cm2.

Photocatalytic nitrogen fixation reaction can harvest the solar energy to convert the abundant but inert N2 into NH3. Here, utilizing metal–organic framework (MOF) membranes as the ideal assembly of nanoreactors to disperse and confine gold nanoparticles (AuNPs), we realize the direct plasmonic photocatalytic nitrogen fixation under ambient conditions. Upon visible irradiation, the hot electrons generated on the AuNPs can be directly injected into the N2 molecules adsorbed on Au surfaces. Such N2 molecules can be additionally activated by the strong but evanescently localized surface plasmon resonance field, resulting in a supralinear intensity dependence of the ammonia evolution rate with much higher apparent quantum efficiency and lower apparent activation energy under stronger irradiation. Moreover, the gas-permeable Au@MOF membranes, consisting of numerous interconnected nanoreactors, can ensure the dispersity and stability of AuNPs, further facilitate the mass transfer of N2 molecules and (hydrated) protons, and boost the plasmonic photocatalytic reactions at the designed gas–membrane–solution interface. As a result, an ammonia evolution rate of 18.9 mmol gAu –1 h–1 was achieved under visible light (>400 nm, 100 mW cm–2) with an apparent quantum efficiency of 1.54% at 520 nm.

The demand for sustainable energy has motivated the development of artificial photosynthesis. Yet the catalyst and reaction interface designs for directly fixing permanent gases (e.g. CO2, O2, N2) into liquid fuels are still challenged by slow mass transfer and sluggish catalytic kinetics at the gas-liquid-solid boundary. Here, we report that gas-permeable metal-organic framework (MOF) membranes can modify the electronic structures and catalytic properties of metal single-atoms (SAs) to promote the diffusion, activation, and reduction of gas molecules (e.g. CO2, O2) and produce liquid fuels under visible light and mild conditions. With Ir SAs as active centers, the defect-engineered MOF (e.g. activated NH2-UiO-66) particles can reduce CO2 to HCOOH with an apparent quantum efficiency (AQE) of 2.51% at 420 nm on the gas-liquid-solid reaction interface. With promoted gas diffusion at the porous gas-solid interfaces, the gas-permeable SA/MOF membranes can directly convert humid CO2 gas into HCOOH with a near-unity selectivity and a significantly increased AQE of 15.76% at 420 nm. A similar strategy can be applied to the photocatalytic O2-to-H2O2 conversions, suggesting the wide applicability of our catalyst and reaction interface designs.

Assisted by potassium cations, Au decorated P25 nanopaticles can produce ammonia from nitrogen and water with high efficiency under ambient conditions in the presence of sunlight.