The Quantitative Influence of Coordinated Halogen Atoms on the Catalytic Performance of Bisiminoacenaphthylnickel Complexes in Ethylene Polymerization.

In experiments, nickel bromine complexes usually show a better catalytic performance in ethylene polymerization compared to their nickel chlorine analogues. Therefore, the present modeling study has been performed to investigate the effect of coordinated halogen atoms on the catalytic performances of two bisiminoacenaphthyl nickel systems, namely, Ni-Br and Ni-Cl. By using the multiple linear regression analysis (MLRA), the catalytic activity can be well predicted by the descriptors of effective net charge (Qeff ) and bite angle (ß), with correlation coefficient R2 values over 0.91. Meanwhile, the molecular weights of polyethylene are predicted by the descriptors of Qeff and open cone angle (θ). The calculated contributions of each descriptor show that the electronic effect is the predominant factor in Ni-Br system, while the steric effect becomes the dominant factor in Ni-Cl system. The different determined effect is expected to the main reason for the different catalytic performance between two Ni systems.

Prediction of catalytic activities of bis(imino)pyridine metal complexes by machine learning.

This work demonstrates the potential of machine learning (ML) method to predict catalytic activity of transition metal complex precatalyst toward ethylene polymerization. For this purpose, 294 complexes and 15 molecular descriptors were selected to build the artificial neural network (ANN) model. The catalytic activity can be well predicted by the obtained ANN model, which was further validated by external complexes. Boruta algorithm was employed to explicitly decipher the importance of descriptors, illustrating the conjugated bond structure, and bulky substitutions are favorable for catalytic activity. The present work indicates that ML could give useful guidance for the new design of homogenous polyolefin catalyst.

Electrical Conductivity of Subsurface Ocean Analogue Solutions from Molecular Dynamics Simulations.

Investigating the habitability of ocean worlds is a priority of current and future NASA missions. The Europa Clipper mission will conduct approximately 50 flybys of Jupiter's moon Europa, returning a detailed portrait of its interior from the synthesis of data from its instrument suite. The magnetometer on board has the capability of decoupling Europa's induced magnetic field to high precision, and when these data are inverted, the electrical conductivity profile from the electrically conducting subsurface salty ocean may be constrained. To optimize the interpretation of magnetic induction data near ocean worlds and constrain salinity from electrical conductivity, accurate laboratory electrical conductivity data are needed under the conditions expected in their subsurface oceans. At the high-pressure, low-temperature (HPLT) conditions of icy worlds, comprehensive conductivity data sets are sparse or absent from either laboratory data or simulations. We conducted molecular dynamics simulations of candidate ocean compositions of aqueous NaCl under HPLT conditions at multiple concentrations. Our results predict electrical conductivity as a function of temperature, pressure, and composition, showing a decrease in conductivity as the pressure increases deeper into the interior of an icy moon. These data can guide laboratory experiments at conditions relevant to icy moons and can be used in tandem to forward-model the magnetic induction signals at ocean worlds and compare with future spacecraft data. We discuss implications for the Europa Clipper mission.

Machine Learning in Catalysis, From Proposal to Practicing.

Recently, machine learning (ML) methods have gained popularity and have performed as powerfully predictive tools in various areas of academic and industrious activities. In comparison, their application in catalysis has been underdeveloped. Relying on the rapid development of different algorithms and their implementation, it is the right timing to harvest the potential of ML in catalysis across academy and industry spectra. Herein, we discuss the current applications in the field of homogeneous and heterogeneous catalysis by using various ML approaches. To the best of our knowledge, modern statistical learning techniques will be a strong tool for computational optimization and discovery. This in turn will accurately extract the underlying mechanism in the model that converts readily available data and precatalysts into their promising and useful ones.