Catalysts informatics: paradigm shift towards data-driven catalyst design

Abstract

Catalysts informatics is proposed as an alternative way to design and understand the catalysts. The idea of catalysts informatics is to design the catalysts from trends and pattern found in...

Achieving Digital Catalysis: Strategies for Data Acquisition, Storage and Use

“…This data is the core data in the field, [41] and consists primarily of chemical analytics (e.g., concentration of compounds in the reacting phase determined by techniques such as different types of chromatography, thermal conductivity, mass spectrometry, or infrared detectors) coupled to time, which is then transformed to Figures of merits such as selectivity to a product as a function of the conversion of a substrate or a reaction rate as a function of the reaction time or the temperature. In high‐throughput experiments, which are often applied in catalysis, these data are generated in parallel for several or a large number of catalysts partly at the expense of accuracy, but reliable enough to achieve a catalyst ranking [5f,h, 8d] …”

“…Many different algorithms are available that vary in the size of the dataset required, in the precision of predictions, in complexity and in whether there is a gain in understanding of the underlying principles associated with the prediction [5a, 46] . Neural networks and ensemble‐based methods, for example, offer little insight, [46, 47] but can help bridge the gap between different time and length scales of typical DFT calculations and reactor studies and enable fast screening in catalysts (materials) informatics [5h] . Interpretable ML algorithms, on the other hand, can reveal the physical laws underlying catalyst's properties and enable further development of catalyst systems based on hypotheses that follow from this analysis, which can accelerate catalyst discovery in a more sustainable manner [5a, 34b] .…”

“…These transparent methods in catalysis informatics [5i] include, for example, tree‐based classification and regression methods [5a, 46] . Active learning algorithms using Bayesian optimization techniques have been developed to rapidly identify candidates with tailored properties for further validation either through experimental synthesis or more elaborate computations in research campaigns [5h, 46] …”

“…As outlined above, classical approaches to data acquisition and storage in catalysis need to be revised [5h–i, 8a,d, 84] . Therefore, in this section, we discuss practical aspects and already existing solutions for important components of a local research data infrastructure in catalysis.…”

MATEO: InterMolecular α-Amidoalkylation Theoretical Enantioselectivity Optimization. Online Tool for Selection and Design of Chiral Catalysts and Products

“…Unfortunately, these techniques are less useful when it is necessary a fast scanning/optimization of new CPA catalysts for large libraries of reactions with diverse substrates, nucleophiles, products, and conditions (temperature, time, catalyst load, etc.). Cheminformatics methods relying upon Artificial Intelligence/Machine Learning (AI/ML) algorithms could help to speed up the discovery of new molecules [7][8][9] and also in the design new chiral catalysts and products without engaging in a long term, empirical or quantum investigation [10][11][12][13]. Therefore, there is a need to develop fast-track computational tools able to predict the enantiomeric excess saving time and experimental resources.…”