These are preliminary reports that have not been peer-reviewed. They should not be regarded as conclusive, guide clinical practice/health-related behavior, or be reported in news media as established information. For more information, please see our FAQs.
High-Throughput Assessment of Hypothetical Zeolite Materials for Their Synthesizability and Industrial Deployability
preprintsubmitted on 26.02.2019, 13:38 and posted on 26.02.2019, 17:10 by Nils Zimmermann, Jose Luis Salcedo Perez, Maciej Haranczyk
Zeolites are important microporous framework materials, where 200+ structures are known to exist and many millions so-called hypothetical materials can be computationally created. Here, we screen the “Deem” database of hypothetical zeolite structures to find experimentally feasible and industrially relevant materials. We use established and existing criteria and structure descriptors (lattice energy, local interatomic distances, TTT angles), and we develop new criteria which are based on 5-th neighbor distances to T-atoms, tetrahedral order parameters (or, tetrahedrality), and porosity and channel dimensionality. Our filter funnel for screening the most attractive zeolite materials that we construct consists of 9 different types of criteria and a total of 53 subcriteria. The funnel reduces the pool of candidate materials from initially >300,000 to 24 and 11, respectively, depending on the channel dimensionality constraint applied (2- and 3-dimensional vs only 3-dimensional channels).
We find that it is critically important to define longer range and more stringent criteria such as the new 5-th neighbor distances to T-atoms and the tetrahedrality descriptor in order to succeed in reducing the huge pool of candidates to a manageable number. Apart from one experimentally achieved structure (SSF), all other candidates are hypothetical frameworks, thus, representing most valuable targets for synthesis and application. Detailed analysis of the screening data allowed us to also propose an exciting future direction how such screening studies as ours could be improved and how framework generating algorithms could be competitively optimized.