Search for: All records

Low-cost self-driving labs (SDLs) offer faster prototyping, low-risk hands-on experience, and a test bed for sophisticated experimental planning software which helps us develop state-of-the-art SDLs.

Bayesian optimization (BO) has been leveraged for guiding autonomous and high-throughput experiments in materials science. However, few have evaluated the efficiency of BO across a broad range of experimental materials domains. In this work, we quantify the performance of BO with a collection of surrogate model and acquisition function pairs across five diverse experimental materials systems. By defining acceleration and enhancement metrics for materials optimization objectives, we find that surrogate models such as Gaussian Process (GP) with anisotropic kernels and Random Forest (RF) have comparable performance in BO, and both outperform the commonly used GP with isotropic kernels. GP with anisotropic kernels has demonstrated the most robustness, yet RF is a close alternative and warrants more consideration because it is free from distribution assumptions, has smaller time complexity, and requires less effort in initial hyperparameter selection. We also raise awareness about the benefits of using GP with anisotropic kernels in future materials optimization campaigns.

The nanomaterial landscape is so vast that a high-throughput combinatorial approach is required to understand structure–function relationships. To address this challenge, an approach for the synthesis and screening of megalibraries of unique nanoscale features (>10,000,000) with tailorable location, size, and composition has been developed. Polymer pen lithography, a parallel lithographic technique, is combined with an ink spray-coating method to create pen arrays, where each pen has a different but deliberately chosen quantity and composition of ink. With this technique, gradients of Au-Cu bimetallic nanoparticles have been synthesized and then screened for activity by in situ Raman spectroscopy with respect to single-walled carbon nanotube (SWNT) growth. Au₃Cu, a composition not previously known to catalyze SWNT growth, has been identified as the most active composition.

A challenge in the synthesis of single‐wall carbon nanotubes (SWCNTs) is the lack of control over the formation and evolution of catalyst nanoparticles and the lack of control over their size or chirality. Here, zeolite MFI nanosheets (MFI‐Ns) are used to keep cobalt (Co) nanoparticles stable during prolonged annealing conditions. Environmental transmission electron microscopy (ETEM) shows that the MFI‐Ns can influence the size and shape of nanoparticles via particle/support registry, which leads to the preferential docking of nanoparticles to four or fewer pores and to the regulation of the SWCNT synthesis products. The resulting SWCNT population exhibits a narrow diameter distribution and SWCNTs of nearly all chiral angles, including sub‐nm zigzag (ZZ) and near‐ZZ tubes. Theoretical simulations reveal that the growth of these unfavorable tubes from unsupported catalysts leads to the rapid encapsulation of catalyst nanoparticles bearing them; their presence in the growth products suggests that the MFI‐Ns prevent nanoparticle encapsulation and prologue ZZ and near‐ZZ SWCNT growth. These results thus present a path forward for controlling nanoparticle formation and evolution, for achieving size‐ and shape‐selectivity at high temperature, and for controlling SWCNT synthesis.