Conjugated block copolymer photovoltaics with near 3% efficiency through microphase separation.

Organic electronic materials have the potential to impact almost every aspect of modern life including how we access information, light our homes, and power personal electronics. Nevertheless, weak intermolecular interactions and disorder at junctions of different organic materials limit the performance and stability of organic interfaces and hence the applicability of organic semiconductors to electronic devices. Here, we demonstrate control of donor-acceptor heterojunctions through microphase-separated conjugated block copolymers. When utilized as the active layer of photovoltaic cells, block copolymer-based devices demonstrate efficient photoconversion well beyond devices composed of homopolymer blends. The 3% block copolymer device efficiencies are achieved without the use of a fullerene acceptor. X-ray scattering results reveal that the remarkable performance of block copolymer solar cells is due to self-assembly into mesoscale lamellar morphologies with primarily face-on crystallite orientations. Conjugated block copolymers thus provide a pathway to enhance performance in excitonic solar cells through control of donor-acceptor interfaces.

Phase Diagrams of Alloys and Their Hydrides via On-Lattice Graph Neural Networks and Limited Training Data.

Efficient prediction of sampling-intensive thermodynamic properties is needed to evaluate material performance and permit high-throughput materials modeling for a diverse array of technology applications. To alleviate the prohibitive computational expense of high-throughput configurational sampling with density functional theory (DFT), surrogate modeling strategies like cluster expansion are many orders of magnitude more efficient but can be difficult to construct in systems with high compositional complexity. We therefore employ minimal-complexity graph neural network models that accurately predict and can even extrapolate to out-of-train distribution formation energies of DFT-relaxed structures from an ideal (unrelaxed) crystallographic representation. This enables the large-scale sampling necessary for various thermodynamic property predictions that may otherwise be intractable and can be achieved with small training data sets. Two exemplars, optimizing the thermodynamic stability of low-density high-entropy alloys and modulating the plateau pressure of hydrogen in metal alloys, demonstrate the power of this approach, which can be extended to a variety of materials discovery and modeling problems.

Tuning the Spin Transition and Carrier Type in Rare-Earth Cobaltates via Compositional Complexity.

There is growing interest in material candidates with properties that can be engineered beyond traditional design limits. Compositionally complex oxides (CCO), often called high entropy oxides, are excellent candidates, wherein a lattice site shares more than four cations, forming single-phase solid solutions with unique properties. However, the nature of compositional complexity in dictating properties remains unclear, with characteristics that are difficult to calculate from first principles. Here, compositional complexity is demonstrated as a tunable parameter in a spin-transition oxide semiconductor La1- x(Nd, Sm, Gd, Y)x/4CoO3, by varying the population x of rare earth cations over 0.00≤ x≤ 0.80. Across the series, increasing complexity is revealed to systematically improve crystallinity, increase the amount of electron versus hole carriers, and tune the spin transition temperature and on-off ratio. At high a population (x = 0.8), Seebeck measurements indicate a crossover from hole-majority to electron-majority conduction without the introduction of conventional electron donors, and tunable complexity is proposed as new method to dope semiconductors. First principles calculations combined with angle resolved photoemission reveal an unconventional doping mechanism of lattice distortions leading to asymmetric hole localization over electrons. Thus, tunable complexity is demonstrated as a facile knob to improve crystallinity, tune electronic transitions, and to dope semiconductors beyond traditional means.

Defect graph neural networks for materials discovery in high-temperature clean-energy applications.

We present a graph neural network approach that fully automates the prediction of defect formation enthalpies for any crystallographic site from the ideal crystal structure, without the need to create defected atomic structure models as input. Here we used density functional theory reference data for vacancy defects in oxides, to train a defect graph neural network (dGNN) model that replaces the density functional theory supercell relaxations otherwise required for each symmetrically unique crystal site. Interfaced with thermodynamic calculations of reduction entropies and associated free energies, the dGNN model is applied to the screening of oxides in the Materials Project database, connecting the zero-kelvin defect enthalpies to high-temperature process conditions relevant for solar thermochemical hydrogen production and other energy applications. The dGNN approach is applicable to arbitrary structures with an accuracy limited principally by the amount and diversity of the training data, and it is generalizable to other defect types and advanced graph convolution architectures. It will help to tackle future materials discovery problems in clean energy and beyond.