Conformational Sampling for Transition State Searches on a Computational Budget

Transition state searches are the basis for characterizing reaction mechanisms and activation energies, and are thus central to myriad chemical applications. Nevertheless, common search algorithms are sensitive to molecular conformation and the conformational space of even medium-sized reacting systems are too complex to explore with brute force. Here we show that it is possible to train a classifier to learn the features of conformers that conduce successful transition state searches, such that optimal conformers can be down-selected before incurring the cost of a high-level transition state search. To this end, we have benchmarked the use of a modern conformational generation algorithm with our reaction prediction methodology, Yet Another Reaction Program (YARP), for reaction prediction tasks. We demonstrate that neglecting conformer contributions leads to qualitatively incorrect activation energy estimations for a broad range of reactions, whereas a simple random forest classifier can be used to reliably down-select low-barrier conformers. We also compare the relative advantage of performing conformational sampling on reactant, product, and putative transition state geometries. The robust performance of this relatively simple machine learning classifier mitigates cost as a factor when implementing conformational sampling into contemporary reaction prediction workflows.

Calculation of Equilibrium Composition of Complex Thermodynamic Systems using Julia Language and Ipopt Library

The article considers the possibility of using the Ipopt optimization package for the calculating the phase and equilibrium compositions of a multicomponent heterogeneous thermodynamic system. Two functions are presented for calculating the equilibrium composition and properties of complex thermodynamic systems, written in the Julia programming language. These functions are the key ones in the program integrated with the IVTANTERMO database on thermodynamic properties of individual substances and used for conducting test calculations. The test calculations showed that Ipopt package allows determining the phase and chemical compositions of simple and complex thermodynamic systems with a fairly high speed. Using the JuMP modeling language significantly simplifies the preparation of the initial data for the Ipopt package, therefore the functions presented in this article are very compact. It is shown how the Ipopt package can be used when the temperature of the thermodynamic system is unknown. The approach proposed in this work is applicable both for analyzing the equilibrium of individual chemical reactions and for calculating the equilibrium composition of complex chemically reacting systems. The simplicity of the proposed functions allows their easy integrating into application programs, embedding them into more complex applications, using them in combination with more complex models (real gas, nonideal solutions, constrained equilibria), and, if necessary, modifying them. It should be noted that the versatility of the JuMP modeling language makes it possible to replace the Ipopt package with another one without significant modification of the program text

Performance Indicators for Benchmarking Solar Thermochemical Fuel Processes and Reactors

Concentrated solar energy offers a source for renewable high-temperature process heat that can be used to efficiently drive endothermic chemical processes, converting the entire spectrum of solar radiation into chemical energy. In particular, solar-driven thermochemical processes for the production of fuels include reforming of methane and other hydrocarbons, gasification of biomass, coal, and other carbonaceous feedstock, and metal oxide redox cycles for splitting H2O and CO2. A notable issue in the development of these processes and their associated solar reactors is the lack of consistent reporting methods for experimental demonstrations and modelling studies, which complicates the benchmarking of the corresponding technologies. In this work we formulate dimensionless performance indicators based on mass and energy balances of such reacting systems, namely: energy efficiency, conversion extent, selectivity, and yield. Examples are outlined for the generic processes mention above. We then provide guidelines for reporting on such processes and reactors and suggest performance benchmarking on four key criteria: energy efficiency, conversion extent, product selectivity, and performance stability.

Numerical studies of higher order tau-leaping methods

Gillespie's algorithm, also known as the Stochastic Simulation Algorithm (SSA), is an exact simulation method for the Chemical Master Equation model of well-stirred biochemical systems. However, this method is computationally intensive when some fast reactions are present in the system. The tau-leap scheme developed by Gillespie can speed up the stochastic simulation of these biochemically reacting systems with negligible loss in accuracy. A number of tau-leaping methods were proposed, including the explicit tau-leaping and the implicit tau-leaping strategies. Nonetheless, these schemes have low order of accuracy. In this thesis, we investigate tau-leap strategies which achieve high accuracy at reduced computational cost. These strategies are tested on several biochemical systems of practical interest.

Numerical studies of higher order tau-leaping methods

Gillespie's algorithm, also known as the Stochastic Simulation Algorithm (SSA), is an exact simulation method for the Chemical Master Equation model of well-stirred biochemical systems. However, this method is computationally intensive when some fast reactions are present in the system. The tau-leap scheme developed by Gillespie can speed up the stochastic simulation of these biochemically reacting systems with negligible loss in accuracy. A number of tau-leaping methods were proposed, including the explicit tau-leaping and the implicit tau-leaping strategies. Nonetheless, these schemes have low order of accuracy. In this thesis, we investigate tau-leap strategies which achieve high accuracy at reduced computational cost. These strategies are tested on several biochemical systems of practical interest.