Mending cracks atom-by-atom in rutile TiO2 with electron beam radiolysis.

Rich electron-matter interactions fundamentally enable electron probe studies of materials such as scanning transmission electron microscopy (STEM). Inelastic interactions often result in structural modifications of the material, ultimately limiting the quality of electron probe measurements. However, atomistic mechanisms of inelastic-scattering-driven transformations are difficult to characterize. Here, we report direct visualization of radiolysis-driven restructuring of rutile TiO2 under electron beam irradiation. Using annular dark field imaging and electron energy-loss spectroscopy signals, STEM probes revealed the progressive filling of atomically sharp nanometer-wide cracks with striking atomic resolution detail. STEM probes of varying beam energy and precisely controlled electron dose were found to constructively restructure rutile TiO2 according to a quantified radiolytic mechanism. Based on direct experimental observation, a "two-step rolling" model of mobile octahedral building blocks enabling radiolysis-driven atomic migration is introduced. Such controlled electron beam-induced radiolytic restructuring can be used to engineer novel nanostructures atom-by-atom.

Engineering metal oxidation using epitaxial strain.

The oxides of platinum group metals are promising for future electronics and spintronics due to the delicate interplay of spin-orbit coupling and electron correlation energies. However, their synthesis as thin films remains challenging due to their low vapour pressures and low oxidation potentials. Here we show how epitaxial strain can be used as a control knob to enhance metal oxidation. Using Ir as an example, we demonstrate the use of epitaxial strain in engineering its oxidation chemistry, enabling phase-pure Ir or IrO2 films despite using identical growth conditions. The observations are explained using a density-functional-theory-based modified formation enthalpy framework, which highlights the important role of metal-substrate epitaxial strain in governing the oxide formation enthalpy. We also validate the generality of this principle by demonstrating epitaxial strain effect on Ru oxidation. The IrO2 films studied in our work further revealed quantum oscillations, attesting to the excellent film quality. The epitaxial strain approach we present could enable growth of oxide films of hard-to-oxidize elements using strain engineering.

Platinum Graphene Catalytic Condenser for Millisecond Programmable Metal Surfaces.

Accelerating catalytic chemistry and tuning surface reactions require precise control of the electron density of metal atoms. In this work, nanoclusters of platinum were supported on a graphene sheet within a catalytic condenser device that facilitated electron or hole accumulation in the platinum active sites with negative or positive applied potential, respectively. The catalytic condenser was fabricated by depositing on top of a p-type Si wafer an amorphous HfO2 dielectric (70 nm), on which was placed the active layer of 2-4 nm platinum nanoclusters on graphene. A potential of ±6 V applied to the Pt/graphene layer relative to the silicon electrode moved electrons into or out of the active sites of Pt, attaining charge densities more than 1% of an electron or hole per surface Pt atom. At a level of charge condensation of ±10% of an electron per surface atom, the binding energy of carbon monoxide to a Pt(111) surface was computed via density functional theory to change 24 kJ mol-1 (0.25 eV), which was consistent with the range of carbon monoxide binding energies determined from temperature-programmed desorption (ΔBECO of 20 ± 1 kJ mol-1 or 0.19 eV) and equilibrium surface coverage measurements (ΔBECO of 14 ± 1 kJ mol-1 or 0.14 eV). Impedance spectroscopy indicated that Pt/graphene condensers with potentials oscillating at 3000 Hz exhibited negligible loss in capacitance and charge accumulation, enabling programmable surface conditions at amplitudes and frequencies necessary to achieve catalytic resonance.

Alumina Graphene Catalytic Condenser for Programmable Solid Acids.

Precise control of electron density at catalyst active sites enables regulation of surface chemistry for the optimal rate and selectivity to products. Here, an ultrathin catalytic film of amorphous alumina (4 nm) was integrated into a catalytic condenser device that enabled tunable electron depletion from the alumina active layer and correspondingly stronger Lewis acidity. The catalytic condenser had the following structure: amorphous alumina/graphene/HfO2 dielectric (70 nm)/p-type Si. Application of positive voltages up to +3 V between graphene and the p-type Si resulted in electrons flowing out of the alumina; positive charge accumulated in the catalyst. Temperature-programmed surface reaction of thermocatalytic isopropanol (IPA) dehydration to propene on the charged alumina surface revealed a shift in the propene formation peak temperature of up to ΔT peak∼50 °C relative to the uncharged film, consistent with a 16 kJ mol-1 (0.17 eV) reduction in the apparent activation energy. Electrical characterization of the thin amorphous alumina film by ultraviolet photoelectron spectroscopy and scanning tunneling microscopy indicates that the film is a defective semiconductor with an appreciable density of in-gap electronic states. Density functional theory calculations of IPA binding on the pentacoordinate aluminum active sites indicate significant binding energy changes (ΔBE) up to 60 kJ mol-1 (0.62 eV) for 0.125 e- depletion per active site, supporting the experimental findings. Overall, the results indicate that continuous and fast electronic control of thermocatalysis can be achieved with the catalytic condenser device.

Ohmic-Contact-Gated Carbon Nanotube Transistors for High-Performance Analog Amplifiers.

The growing demand for ubiquitous data collection has driven the development of sensing technologies with local data processing. As a result, solution-processed semiconductors are widely employed due to their compatibility with low-cost additive manufacturing on a wide range of substrates. However, to fully realize their potential in sensing applications, high-performance scalable analog amplifiers must be realized. Here, ohmic-contact-gated transistors (OCGTs) based on solution-processed semiconducting single-walled carbon nanotubes are introduced to address this unmet need. This new device concept enables output current saturation in the short-channel limit without compromising output current drive. The resulting OCGTs are used in common-source amplifiers to achieve the highest width-normalized output current (≈30 µA µm-1 ) and length-scaled signal gain (≈230 µm-1 ) to date for solution-processed semiconductors. The utility of these amplifiers for emerging sensing technologies is demonstrated by the amplification of complex millivolt-scale analog biological signals including the outputs of electromyography, photoplethysmogram, and accelerometer sensors. Since the OCGT design is compatible with other solution-processed semiconducting materials, this work establishes a general route to high-performance, solution-processed analog electronics.

Reconfigurable MoS2 Memtransistors for Continuous Learning in Spiking Neural Networks.

Artificial intelligence and machine learning are growing computing paradigms, but current algorithms incur undesirable energy costs on conventional hardware platforms, thus motivating the exploration of more efficient neuromorphic architectures. Toward this end, we introduce here a memtransistor with gate-tunable dynamic learning behavior. By fabricating memtransistors from monolayer MoS2 grown on sapphire, the relative importance of the vertical field effect from the gate is enhanced, thereby heightening reconfigurability of the device response. Inspired by biological systems, gate pulses are used to modulate potentiation and depression, resulting in diverse learning curves and simplified spike-timing-dependent plasticity that facilitate unsupervised learning in simulated spiking neural networks. This capability also enables continuous learning, which is a previously underexplored cognitive concept in neuromorphic computing. Overall, this work demonstrates that the reconfigurability of memtransistors provides unique hardware accelerator opportunities for energy efficient artificial intelligence and machine learning.

Spiking neurons from tunable Gaussian heterojunction transistors.

Spiking neural networks exploit spatiotemporal processing, spiking sparsity, and high interneuron bandwidth to maximize the energy efficiency of neuromorphic computing. While conventional silicon-based technology can be used in this context, the resulting neuron-synapse circuits require multiple transistors and complicated layouts that limit integration density. Here, we demonstrate unprecedented electrostatic control of dual-gated Gaussian heterojunction transistors for simplified spiking neuron implementation. These devices employ wafer-scale mixed-dimensional van der Waals heterojunctions consisting of chemical vapor deposited monolayer molybdenum disulfide and solution-processed semiconducting single-walled carbon nanotubes to emulate the spike-generating ion channels in biological neurons. Circuits based on these dual-gated Gaussian devices enable a variety of biological spiking responses including phasic spiking, delayed spiking, and tonic bursting. In addition to neuromorphic computing, the tunable Gaussian response has significant implications for a range of other applications including telecommunications, computer vision, and natural language processing.