scholarly journals Genereting Transition States of Isomerization Reactions with Deep Learning

2020 ◽  
Author(s):  
Lagnajit Pattanaik ◽  
John Ingraham ◽  
Colin Grambow ◽  
William H. Green
Keyword(s):  
Transition State ◽  
Transition States ◽  
Three Dimensional ◽  
Ground Truth ◽  
Distance Matrix ◽  
Quality Data ◽  
Quantum Chemistry Calculations ◽  
Quantitative Understanding ◽  
Speed Accuracy ◽  
Transition State Structures

Lack of quality data and difficulty generating these data hinder quantitative understanding of reaction kinetics. Specifically, conventional methods to generate transition state structures are deficient in speed, accuracy, or scope. We describe a novel method to generate three-dimensional transition state structures for isomerization reactions using reactant and product geometries. Our approach relies on a graph neural network to predict the transition state distance matrix and a least squares optimization to reconstruct the coordinates based on which entries of the distance matrix the model perceives to be important. We feed the structures generated by our algorithm through a rigorous quantum mechanics workflow to ensure the predicted transition state corresponds to the ground truth reactant and product. In both generating viable geometries and predicting accurate transition states, our method achieves excellent results. We envision workflows like this, which combine neural networks and quantum chemistry calculations, will become the preferred methods for computing chemical reactions.

scholarly journals Generating Transition States of Isomerization Reactions with Deep Learning

2020 ◽  
Author(s):  
Lagnajit Pattanaik ◽  
John Ingraham ◽  
Colin Grambow ◽  
William H. Green
Keyword(s):  
Transition State ◽  
Transition States ◽  
Three Dimensional ◽  
Ground Truth ◽  
Distance Matrix ◽  
Quality Data ◽  
Quantum Chemistry Calculations ◽  
Quantitative Understanding ◽  
Speed Accuracy ◽  
Transition State Structures

Lack of quality data and difficulty generating these data hinder quantitative understanding of reaction kinetics. Specifically, conventional methods to generate transition state structures are deficient in speed, accuracy, or scope. We describe a novel method to generate three-dimensional transition state structures for isomerization reactions using reactant and product geometries. Our approach relies on a graph neural network to predict the transition state distance matrix and a least squares optimization to reconstruct the coordinates based on which entries of the distance matrix the model perceives to be important. We feed the structures generated by our algorithm through a rigorous quantum mechanics workflow to ensure the predicted transition state corresponds to the ground truth reactant and product. In both generating viable geometries and predicting accurate transition states, our method achieves excellent results. We envision workflows like this, which combine neural networks and quantum chemistry calculations, will become the preferred methods for computing chemical reactions.

scholarly journals Generating Transition States of Isomerization Reactions with Deep Learning

2020 ◽  
Author(s):  
Lagnajit Pattanaik ◽  
John Ingraham ◽  
Colin Grambow ◽  
William H. Green
Keyword(s):  
Transition State ◽  
Transition States ◽  
Three Dimensional ◽  
Ground Truth ◽  
Distance Matrix ◽  
Quality Data ◽  
Quantum Chemistry Calculations ◽  
Quantitative Understanding ◽  
Speed Accuracy ◽  
Transition State Structures

Lack of quality data and difficulty generating these data hinder quantitative understanding of reaction kinetics. Specifically, conventional methods to generate transition state structures are deficient in speed, accuracy, or scope. We describe a novel method to generate three-dimensional transition state structures for isomerization reactions using reactant and product geometries. Our approach relies on a graph neural network to predict the transition state distance matrix and a least squares optimization to reconstruct the coordinates based on which entries of the distance matrix the model perceives to be important. We feed the structures generated by our algorithm through a rigorous quantum mechanics workflow to ensure the predicted transition state corresponds to the ground truth reactant and product. In both generating viable geometries and predicting accurate transition states, our method achieves excellent results. We envision workflows like this, which combine neural networks and quantum chemistry calculations, will become the preferred methods for computing chemical reactions.

scholarly journals TSNET: PREDICTING TRANSITION STATE STRUCTURES WITH TENSOR FIELD NETWORKS AND TRANSFER LEARNING

2021 ◽  
Author(s):  
Riley Jackson ◽  
Wenyuan Zhang ◽  
Jason Pearson
Keyword(s):  
Transition State ◽  
Reaction Kinetics ◽  
Transfer Learning ◽  
Protein Function ◽  
Tensor Field ◽  
Transition States ◽  
Catalyst Design ◽  
Molecular Structures ◽  
Transition State Structures

Transition states are among the most important molecular structures in chemistry, critical to a variety of fields such as reaction kinetics, catalyst design, and the study of protein function. However,...

Immutable and Variable Transition State Structures. A Reply

1979 ◽  
Vol 32 (11) ◽  
pp. 2361 ◽  
Author(s):  
DJ McLennan ◽  
PL Martin
Keyword(s):  
Transition State ◽  
Transition States ◽  
Solvent Mixtures ◽  
Water Solvent ◽  
Recent Suggestion ◽  
Detailed Treatment ◽  
Theoretical Reasoning ◽  
State Charge ◽  
Transition State Structures

The factual basis of, and the theoretical reasoning behind, a recent suggestion by Poh that transition state structures may or may not vary according to the severity of substituent change are critically examined in terms of currently held views. Both aspects of Poh's paper are found to be defective. Some results on the rates of solvolysis of diarylmethyl p-nitrobenzoates in ethanol/water solvent mixtures are considered. A superficial examination of the Hammett p and Winstein-Grunwald m parameters indicates that Poh's suggestion of immutable transition states is exemplified by this work, but a more detailed treatment shows that p and m are not necessarily functions of transition state charge separations alone.

Elimination-Reactions of Stilbene Dibromides. Debromination by Cyanide, Chloride, Iodide or 4-Nitrothiophenoxide Ions in Dimethylformamide

1986 ◽  
Vol 39 (4) ◽  
pp. 677 ◽  
Author(s):  
J Avraamides
Keyword(s):  
Transition State ◽  
Transition States ◽  
Product Distribution ◽  
Nucleophilic Attack ◽  
Distribution Data ◽  
Elimination Reactions ◽  
Transition State Structures

Transition state structures for debromination reactions of a series of para-substituted stilbene dibromides were evaluated from kinetic and product distribution data for the nucleophiles chloride, cyanide, iodide and 4-nitrothiophenoxide in dimethylformamide . The debrominations appear to utilize transition states in which nucleophilic attack is at bromine despite the strong carbon- nucleophilicity of some of the bases. Products are largely those derived from anti- debromination with all nucleophiles and reactivity is in the order 4-nitrothiophenoxide > cyanide > iodide > chloride.

scholarly journals On the 3D → 2D Isomerization of Hexaborane(12)

Chemistry ◽  
2021 ◽  
Vol 3 (1) ◽  
pp. 28-38
Author(s):  
Josep M. Oliva-Enrich ◽  
Ibon Alkorta ◽  
José Elguero ◽  
Maxime Ferrer ◽  
José I. Burgos
Keyword(s):  
Potential Energy ◽  
Potential Energy Surface ◽  
Energy Surface ◽  
Transition States ◽  
Three Dimensional ◽  
Reaction Coordinate ◽  
Intrinsic Reaction Coordinate ◽  
Limiting Factor ◽  
Two Dimensional ◽  
Axis Of Rotation

By following the intrinsic reaction coordinate connecting transition states with energy minima on the potential energy surface, we have determined the reaction steps connecting three-dimensional hexaborane(12) with unknown planar two-dimensional hexaborane(12). In an effort to predict the potential synthesis of finite planar borane molecules, we found that the reaction limiting factor stems from the breaking of the central boron-boron bond perpendicular to the C2 axis of rotation in three-dimensional hexaborane(12).

scholarly journals Assessing biases, relaxing moralism: On ground-truthing practices in machine learning design and application

2021 ◽  
Vol 8 (1) ◽  
pp. 205395172110135
Author(s):  
Florian Jaton
Keyword(s):  
Machine Learning ◽  
William James ◽  
A Priori ◽  
Learning Algorithms ◽  
Three Dimensional ◽  
Ground Truth ◽  
Machine Learning Algorithms ◽  
Ground Truthing ◽  
Set Up ◽  
The Moment

This theoretical paper considers the morality of machine learning algorithms and systems in the light of the biases that ground their correctness. It begins by presenting biases not as a priori negative entities but as contingent external referents—often gathered in benchmarked repositories called ground-truth datasets—that define what needs to be learned and allow for performance measures. I then argue that ground-truth datasets and their concomitant practices—that fundamentally involve establishing biases to enable learning procedures—can be described by their respective morality, here defined as the more or less accounted experience of hesitation when faced with what pragmatist philosopher William James called “genuine options”—that is, choices to be made in the heat of the moment that engage different possible futures. I then stress three constitutive dimensions of this pragmatist morality, as far as ground-truthing practices are concerned: (I) the definition of the problem to be solved (problematization), (II) the identification of the data to be collected and set up (databasing), and (III) the qualification of the targets to be learned (labeling). I finally suggest that this three-dimensional conceptual space can be used to map machine learning algorithmic projects in terms of the morality of their respective and constitutive ground-truthing practices. Such techno-moral graphs may, in turn, serve as equipment for greater governance of machine learning algorithms and systems.

Volumetric Velocimetry Measurements of Film Cooling Jets

2018 ◽  
Vol 141 (3) ◽  
Author(s):  
Artur Joao Carvalho Figueiredo ◽  
Robin Jones ◽  
Oliver J. Pountney ◽  
James A. Scobie ◽  
Gary D. Lock ◽  
...  
Keyword(s):  
Film Cooling ◽  
Momentum Flux ◽  
Three Dimensional ◽  
Flux Ratio ◽  
Quality Data ◽  
Data Set ◽  
Momentum Flux Ratio ◽  
Jet In Crossflow ◽  
Lift Off ◽  
Induced Velocity

This paper presents volumetric velocimetry (VV) measurements for a jet in crossflow that is representative of film cooling. VV employs particle tracking to nonintrusively extract all three components of velocity in a three-dimensional volume. This is its first use in a film-cooling context. The primary research objective was to develop this novel measurement technique for turbomachinery applications, while collecting a high-quality data set that can improve the understanding of the flow structure of the cooling jet. A new facility was designed and manufactured for this study with emphasis on optical access and controlled boundary conditions. For a range of momentum flux ratios from 0.65 to 6.5, the measurements clearly show the penetration of the cooling jet into the freestream, the formation of kidney-shaped vortices, and entrainment of main flow into the jet. The results are compared to published studies using different experimental techniques, with good agreement. Further quantitative analysis of the location of the kidney vortices demonstrates their lift off from the wall and increasing lateral separation with increasing momentum flux ratio. The lateral divergence correlates very well with the self-induced velocity created by the wall–vortex interaction. Circulation measurements quantify the initial roll up and decay of the kidney vortices and show that the point of maximum circulation moves downstream with increasing momentum flux ratio. The potential for nonintrusive VV measurements in turbomachinery flow has been clearly demonstrated.

Sign in / Sign up

Export Citation Format

Share Document