Atomic Hourglass and Thermometer Based on Diffusion of a Mobile Dopant in VO2.

Transformations between different atomic configurations of a material oftentimes bring about dramatic changes in functional properties as a result of the simultaneous alteration of both atomistic and electronic structure. Transformation barriers between polytypes can be tuned through compositional modification, generally in an immutable manner. Continuous, stimulus-driven modulation of phase stabilities remains a significant challenge. Utilizing the metal-insulator transition of VO2, we exemplify that mobile dopants weakly coupled to the crystal lattice provide a means of imbuing a reversible and dynamical modulation of the phase transformation. Remarkably, we observe a time- and temperature-dependent evolution of the relative phase stabilities of the M1 and R phases of VO2 in an "hourglass" fashion through the relaxation of interstitial boron species, corresponding to a 50 °C modulation of the transition temperature achieved within the same compound. The material functions as both a chronometer and a thermometer and is "reset" by the phase transition. Materials possessing memory of thermal history hold promise for applications such as neuromorphic computing, atomic clocks, thermometry, and sensing.

Conductive filament shape in HfO2 electrochemical metallization cells under a range of forming voltages.

The development of neuromorphic computing architectures based on two terminal filamentary resistance switching devices is limited in part by the high degree of variability in resistance states and switching voltages. Because of the large role filament shape plays in directing thermal and electric fields around the filament (and thus switching parameters), unambiguous knowledge of filament morphology resulting from direct characterization of filament shape is essential to solve critical ongoing challenges of device switching variability. Here, we have utilized a conductive atomic force microscopy scalpel technique to simultaneously scribe through a polycrystalline dielectric layer in formed Cu/HfO2/p+Si electrochemical metallization cell devices. Filament tomograms reveal that when conductive filaments are formed at typical bias conditions (4 V, 100 µA), a variety of filament shapes result, which deviate from the inverse conical shape predicted by the phenomenological electrochemical model. Furthermore, the observation of an increasing spectrum of damage which scales with forming voltage (associated with compliance current overshoot), and which is uncorrelated with electric field or oxide microstructure, supports the role of thermal pulses in expanding filaments, leading to irreversible dielectric breakdown structures at the extreme. Overall, these findings suggest that the original conductive filament shape can be highly varied as a result of thermally driven expansion from joule heating during the forming step, which is not explicitly accounted for in the widely accepted electrochemical model.

Microstructure dependent filament forming kinetics in HfO2 programmable metallization cells.

Variability remains the principal concern for commercialization of HfO2 based resistance switching devices. Here, we investigate the role of thermal processing conditions on internal structure of atomic layer deposited HfO2 thin films, and the impact of that structure on filament forming kinetics of p+ Si/HfO2/Cu and TiN/HfO2/Cu devices. Regardless of bias polarity or electrode metal, filament formation times are at least one order of magnitude shorter in polycrystalline than in amorphous films, which we attribute to the presence of fast ion migration along grain boundaries. Within polycrystalline films, filament formation times are correlated with degree of crystalline orientation. Inter-device variability in forming time is roughly equivalent across HfO2 film processing conditions. The kinetics of filament forming are shown to be highly dependent on HfO2 microstructure, with possible implications for the inter-device variability of subsequent switching cycles.

Microstructure defines the electroconductive and mechanical performance of plant-derived renewable carbon fiber.

A plant-derived lignin polymer has been sought-after as a low-cost carbon fiber (CF) precursor, but the underlying mechanisms defining CF performances are still elusive. This study revealed that both the electroconductive and mechanical performances of lignin-based CF were synergistically improved by enhancing the microstructures through modifying the lignin chemistry, which paved a pathway to holistically improve the lignin CF quality.