Dilworth Y. Parkinson

Behavior contains rich structure across many timescales. However, there is a dearth of methods to identify relevant components, especially over the longer periods required for learning and decision-making. Inspired by the goals and techniques of genome-wide association studies, we present a data-driven method—the choice-wide behavioral association study: CBAS—that systematically identifies such behavioral features. CBAS extracts sequences of actions or choices that either significantly differ between groups or correlate with a covariate of interest, using a powerful resampling-based method to correct for multiple comparisons. We illustrate CBAS through application to tasks performed by flies, rats, and humans, showing that it provides unexpected and interpretable information in each case.

This work focuses on an electrolyte obtained by infusing linear poly(pentyl malonate) (PPM) chains into cross-linked poly(ethylene oxide) (PEO). A lithium salt is added to enable ion transport. Obtaining this electrolyte is only possible because PPM is miscible with PEO irrespective of salt concentration. The properties of this electrolyte are compared against two benchmarks: an electrolyte obtained by infusing linear PEO into cross-linked PEO and cross-linked PEO with no added linear polymer. Measurements of storage moduli and ion transport efficacies are found to be similar across the three electrolytes, suggesting similar stability behavior against lithium metal anodes, based on current theoretical models. Instead, the PPM infused electrolyte outperforms the other two electrolytes. In the cycling tests, the current supported by the PPM infused electrolyte was a factor of 20 higher than that of the PEO infused electrolyte and a factor of 4 higher than that of the cross-linked PEO electrolyte. In the unidirectional plating experiments, the other two electrolytes fail after about 4 h or less, while the PPM containing electrolyte did not fail during the 140 h testing period. Insight into the nature of failure modes was obtained using synchrotron hard X-ray microtomography.

Synchrotron beamlines generate high-resolution volumetric datasets that reveal detailed sample microstructure, but converting raw scans into scientific insight remains computationally demanding. We present an automated AI-driven high-performance computing (HPC) pipeline for near real-time segmentation and analysis of micro-CT scans collected at Beamline 8.3.2 at the Advanced Light Source (ALS). After acquisition, data are automatically transferred to HPC systems at the National Energy Research Scientific Computing Center (NERSC) and the Argonne Leadership Computing Facility (ALCF) for distributed tomographic reconstruction, followed by large-scale segmentation using two fine-tuned state-of-the-art foundation models, Segment Anything Model 3 (SAM3) and a Self-Distillation with No Labels (DINO)-based segmentation model, DINOv3-Seg. Their results are aggregated to produce final segmentation masks. Orchestrated with Prefect, it achieves a scan-to-discovery turnaround time of approximately 20 minutes. The pipeline generalizes across diverse micro-CT experiments; here, we applied it to examine xylem vessels in grapevine petiole datasets, revealing structural changes for in situ plant physiology studies.

The densification of solid electrolyte (SE) materials is crucial for improving the performance and stability of all‐solid‐state batteries (ASSBs). In this study, the role of granular creep in SE densification is investigated using numerical simulations and experimental validation on Li ₆ PS ₅ Cl separators. Using discrete element method simulations, we analyze the influence of strain rate and cohesion on force chain evolution, particle rearrangement, and porosity reduction. Our results indicate that low strain rates promote granular creep, allowing for gradual particle reorientation and stress relaxation, leading to higher packing density and lower residual porosity. To validate these findings, we also performed experiments at different strain rates, where X‐ray computed tomography and scanning electron microscopy confirm that low strain rates produce a more homogeneous microstructure. Furthermore, critical current density (CCD) tests on lithium symmetric cells reveal that samples processed at the lowest strain rate exhibit a CCD of mA cm ⁻² , three times higher than samples processed at faster strain rates, highlighting the direct correlation between granular creep, densification, and ionic transport enhancement. These findings underscore the importance of strain‐rate‐controlled processing in optimizing SE microstructure, mechanical, and electrochemical performance, offering insights into the fabrication of high‐density, high‐performance separators for next‐generation ASSBs.

Solid-state synthesis involves a web of coupled chemical reactions and physical changes that unfold across multiple scales. Efforts to fine-tune its parameters have historically followed heuristic, trial-driven workflows that demand significant time and resources. In this study, we aimed to open this black box by employing multiscale in situ synchrotron imaging and diffraction. Using LiNi_0.5Mn_0.3Co_0.2O₂ battery positive electrode material as a model system and Ba-based sintering aids, we reveal dopant segregation, intergranular mass transport, and porosity evolution as key drivers of single-crystalline particle formation. Notably, we uncovered a dynamic competition between particle-level grain coalescence and atomic-scale cation disordering, both of which are thermally activated yet have opposing impacts on battery performance. These findings highlight the coupled, multiscale nature of structure development and offer a mechanistic basis for optimizing the solid-state synthesis process. This framework provides a path toward more controlled, efficient, and scalable production of high-performance battery positive electrode materials.