Kavli Affiliate: Kristin A. Persson
| First 5 Authors: Matthew J. McDermott, Brennan C. McBride, Corlyn Regier, Gia Thinh Tran, Yu Chen
| Summary:
Synthesis is a major challenge in the discovery of new inorganic materials.
There is currently limited theoretical rationale for planning optimal
solid-state synthesis procedures that selectively yield desired targets with
minimal impurities. Using an interface reaction model, we propose two
selectivity metrics — primary and secondary competition — to measure the
favorability of target and impurity phase formation in solid-state reactions.
We first apply these metrics to assess the selectivity of 3,520 solid-state
synthesis recipes extracted from the scientific literature, comparing the
thermodynamic optimality of various approaches to popular targets (e.g.,
LiMn$_2$O$_4$, LiFePO$_4$) under our framework. Secondly, we implement these
metrics in a data-driven synthesis planning workflow and test its use in
suggesting synthetic routes to barium titanate (BaTiO$_3$). From an 18-element
chemical reaction network created with first-principles materials thermodynamic
data from the Materials Project, we identify 82,985 synthesis reactions for
BaTiO$_3$ and select nine to test experimentally, characterizing their reaction
pathways using ex post facto synchrotron powder X-ray diffraction over a wide
range of temperatures. Analysis of experimental results indicates that our
proposed selectivity metrics correlate with observed target and impurity
formation. We identify two favorably predicted reactions with unconventional
BaS/BaCl$_2$ and Na$_2$TiO$_3$ precursors that yield BaTiO$_3$ more rapidly and
with fewer impurities than conventional solid-state approaches. Altogether, our
framework serves as a foundation for the predictive synthesis of inorganic
materials, facilitating the optimization of existing synthesis approaches and
the design of syntheses to novel materials, including those that cannot be
easily made from conventional "off-the-shelf" precursors.
| Search Query: ArXiv Query: search_query=au:”Kristin A. Persson”&id_list=&start=0&max_results=3