Abstract:Reaction mechanisms are at the core of understanding reaction systems and designing high-performance catalysts. A complex reaction system often involves various species and elementary reactions, posing a great challenge to determining the reaction mechanism. Here, we proposed a scheme to automatically generate reaction intermediates and elementary reactions to construct a complete reaction network represented by graph theory and employed a depth first search algorithm in the scheme to prune the reaction network to reduce the complexity of the network. With this scheme, microkinetic simulations of CO2 hydrogenation on Pd2Cu using the barriers predicted with the linear thermodynamics-kinetics relations were performed on the network to determine the mechanism and rate- and selectivity-controlling steps of CO2 hydrogenation to ethanol and methanol. Analysis shows that the simulated selectivity of ethanol and methanol agrees well with the experimental results. CO2 + H -> COOH is the rate-controlling step, and CHOH + H -> CH + H2O, CH2OH + H -> CH2 + H2O, and CH2OH + H -> CH3OH dominate the ethanol selectivity. Both ethanol and methanol are generated via multiple reaction pathway mechanisms. Investigations of the pruned networks show that quantitatively correct results can be obtained from the pruned or pseudocomplete reaction network, as long as the key pathways are embodied in the network. 94% ethanol selectivity of the complete network can be obtained with the pruned network composed of 60 elementary steps, compared to 176 steps of the complete network. The present work articulates graph theory representation, depth first search algorithm, linear thermodynamics-kinetics relations, and microkinetic simulations to approach complicated heterogeneous reaction systems and exemplifies their comprehensive roles in exploring complex reaction networks.

Extraction of Important Reaction Pathways for Complex Reaction Network Based on Community Detection Algorithm

Reaction network simplification and key routes extraction for steam cracking process

Kinetic Modeling with Automatic Reaction Network Generator, an Application to Naphtha Steam Cracking

Study on Complex Reaction Network and Kinetic Model of a Methanol to Propene Process

An Ingenious Characterization of Reaction Network Using Sub-Network Reconstruction.

Analyzing And Modeling Ethylene Cracking Process With Complex Networks Approach

Emergence of Heterogeneous Structures in Chemical Reaction-Diffusion Networks.

General Reaction Network Exploration Scheme Based on Graph Theory Representation and Depth First Search Applied to CO2 Hydrogenation on Pd2Cu Catalyst

Community Reaction Network Reduction for Constructing a Coarse-Grained Representation of Combustion Reaction Mechanisms

ReacNetGen: an Automatic Reaction Network Generator for Reactive Molecular Dynamic Simulations

ReacNetGenerator: an Automatic Reaction Network Generator for Reactive Molecular Dynamic Simulations

Automated Exploration of Reaction Network and Mechanism via Meta-dynamics Nanoreactor

Deep reaction network exploration at a heterogeneous catalytic interface

Molecular-level Reaction Network in Delayed Coking Process Based on Structure-Oriented Lumping

Context-Driven Exploration of Complex Chemical Reaction Networks

Autonomous Discovery of Unknown Reaction Pathways from Data by Chemical Reaction Neural Network

Exploring catalytic reaction networks with machine learning

Ideation and Evaluation of Novel Multicomponent Reactions via Mechanistic Network Analysis and Automation

Heuristics-Guided Exploration of Reaction Mechanisms