Search

This work summarizes how catalysts informatics plays a role in catalyst design.

Unique 30 types of multi-component La2O3-based catalysts for oxidative coupling of methane were discovered in 75 types of selected catalysts based on high-throughput screening and literature datasets with multi-output machine learning approaches.

Reaction networks of hydrocarbons are explored using first principles calculations, data science, and experiments.

Identification of catalysts is a difficult matter as catalytic activities involve a vast number of complex features that each catalyst possesses. Here, catalysis gene expression profiling is proposed from unique features discovered in catalyst data collected by high-throughput experiments as an alternative way of representing the catalysts. Combining constructed catalyst gene sequences with hierarchical clustering results in catalyst gene expression profiling where natural language processing is used to identify similar catalysts based on edit distance. In addition, catalysts with similar properties are designed by modifying catalyst genes where the designed catalysts are experimentally confirmed to have catalytic activities that are associated with their catalyst gene sequences. Thus, the proposed method of catalyst gene expressions allows for a novel way of describing catalysts that allows for similarities in catalysts and catalytic activity to be easily recognized while enabling the ability to design new catalysts based on manipulating chemical elements of catalysts with similar catalyst gene sequences.

Direct design of low temperature oxidative coupling of methane (OCM) catalysts is proposed via machine learning and data mining. 58 OCM catalysts are experimentally synthesized and evaluated. The collected 58 sets of data are then classified by unsupervised machine learning in a multi-dimensional space where an active catalyst group for low temperature OCM is identified. Data mining then identifies the physical rule within the group. Catalysts satisfying such a physical rule are designed where 2 undiscovered low temperature OCM catalysts are found and experimentally validated. Thus, machine learning and data mining reveal the hidden physical rule behind the catalysis leading to the direct design of catalysts. Hence, machine learning and data mining open up the insight on a powerful strategy for designing catalysts.

An innovative web-based integrated catalyst informatics platform, Catalyst Acquisition by Data Science (CADS), is developed for use towards the discovery and design of catalysts. The platform provides three main functionalities: a repository for data sharing and publishing, an analytic workspace for exploratory visual analysis, and catalyst property prediction tools with pretrained machine learning models. Access to such a platform helps decrease barriers to entry faced by researchers in catalytic chemistry when attempting to apply catalyst informatics towards data by providing analytical and visualization tools that can be simultaneously applied and easily accessed within a central space, thereby helping the advancement of catalyst informatics. The developed platform allows researchers to upload and collect data onto the platform and conduct data analysis using a system of linked workspaces consisting of interactive visualization tools and machine learning tools that simultaneously update according to the researchers' actions in real time. The platform also provides a space for collaboration where researchers can choose to publish their uploaded data and resulting analyses to the platform for collaborations with other users and groups. As an example, CADS is applied towards oxidative coupling of methane (OCM) data where use of the platform tools reveals underlying patterns and trends that are otherwise hidden within the original data. Thus, the proposed platform contributes towards the advancement of catalyst informatics for both specialists and non-specialists.

Catalysis research is on the verge of experiencing a paradigm shift regarding how catalysts are designed and characterized due to the rise of catalyst informatics. The details of catalyst informatics are reviewed where the following three key concepts are proposed: catalyst data, catalyst data to catalyst design via data science, and catalyst platform. Additionally, progress and opportunities within catalyst informatics are explored and introduced. If the field of catalyst informatics grows in the appropriate manner, the methods and approaches taken for catalyst design would be fundamentally altered, leading towards great advancement within catalysis research.

There is not much doubt that the progress of information and communication technologies, the computerization of all areas of life, and the engagement of increasingly more human beings in the usage of computerized gadgets results in an enormous growth of data available. The data available bear potential for solving urgent problems such as, e.g., forecasting of the spreading of diseases and related prevention, the estimation of the impact of forthcoming disasters like sinkholes, earthquakes, and tsunamies and the preparation of adequate measures, or the development of more precise weather forecasts, to name just a few. Data need to be analyzed. There is a manifold of methodologies and tools to support human exploration. How to do this is treated as an art. But scenarios of data analysis, visualization, and exploration are not yet considered. The present work is intended to fill the gap and to contribute to a paradigmatic shift from the art to a science.

Many vendors or producers of consumer electronics provide web services for collecting data from consumer devices or accessing it from web applications. In this paper, we develop an approach for supporting users of consumer devices in building customized applications on demand. We use Webble technology, a component-based middleware system, as a base technology for distribution of visual components. Users can reuse web services that are connected to data or devices by wrapping those web services as visual components. We demonstrate the potential of the approach with an application example with real world consumer web services.