Machine-learning Prediction of Infrared Spectra of Interstellar Polycyclic Aromatic Hydrocarbons

@article{Kovcs2020MachinelearningPO,
  title={Machine-learning Prediction of Infrared Spectra of Interstellar Polycyclic Aromatic Hydrocarbons},
  author={P{\'e}ter Kov{\'a}cs and Xiaosi Zhu and Jes{\'u}s Carrete and Georg K. H. Madsen and Zhao Wang},
  journal={arXiv: Astrophysics of Galaxies},
  year={2020}
}
  • P. Kovács, Xiaosi Zhu, +2 authors Zhao Wang
  • Published 19 October 2020
  • Physics
  • arXiv: Astrophysics of Galaxies
We design and train a neural network (NN) model to efficiently predict the infrared spectra of interstellar polycyclic aromatic hydrocarbons (PAHs) with a computational cost many orders of magnitude lower than what a first-principles calculation would demand. The input to the NN is based on the Morgan fingerprints extracted from the skeletal formulas of the molecules and does not require precise geometrical information such as interatomic distances. The model shows excellent predictive skill… Expand
5 Citations

Figures from this paper

Machine-learning Interpretation of the Correlation between Infrared Emission Features of Interstellar Polycyclic Aromatic Hydrocarbons
Supervised machine-learning models are trained with various molecular descriptors to predict infrared (IR) emission spectra of interstellar polycyclic aromatic hydrocarbons. We demonstrate that aExpand
Infrared spectra of neutral polycyclic aromatic hydrocarbons based on machine learning potential energy surface and dipole mapping
The interest in polycyclic aromatic hydrocarbons (PAH) spans numerous fields, and infrared (IR) spectroscopy is usually the method of choice to disentangle their molecular structure. In order toExpand
Exploring machine learning methods for absolute configuration determination with vibrational circular dichroism.
TLDR
Increasing the sampling interval proves to be the best method to lower the dimensionality of the input data, thereby decreasing the computational cost associated with the training of the models. Expand
Predicting Infrared Spectra with Message Passing Neural Networks
TLDR
This work serves the dual purpose of providing a trained general-purpose model for the prediction of IR spectra with ease and providing the Chemprop-IR software framework for the training of new models. Expand
Computational Infrared Spectroscopy of 958 Phosphorus-Bearing Molecules
Phosphine is now well-established as a biosignature, which has risen to prominence with its recent tentative detection on Venus. To follow up this discovery and related future exoplanet biosignatureExpand

References

SHOWING 1-10 OF 113 REFERENCES
Effect of molecular structure on the infrared signatures of astronomically relevant PAHs
Emission bands from polycyclic aromatic hydrocarbons (PAHs) dominate the mid-infrared spectra of a wide variety of astronomical sources, encompassing nearly all stages of stellar evolution. DespiteExpand
Simulation of infrared spectra using artificial neural networks based on semiempirical and empirical data
Abstract Two different methods using artificial neural networks of the backpropagation type have been applied to simulate infrared spectra of organic CHO compounds. Their performance is compared. TheExpand
Supervised machine learning for analysing spectra of exoplanetary atmospheres
TLDR
A method of atmospheric retrieval for exoplanets that uses supervised ‘random forest’ machine learning, less time-consuming than standard techniques, is presented and results are obtained consistent with standard atmospheric retrievals. Expand
Deep Learning Spectroscopy: Neural Networks for Molecular Excitation Spectra
TLDR
Deep learning methods for the prediction of molecular excitation spectra are presented and the learning quality improves significantly for the convolutional neural network and reaches its best performance for the deep tensor neural network. Expand
Formation of Interstellar Complex Polycyclic Aromatic Hydrocarbons: Insights from Molecular Dynamics Simulations of Dehydrogenated Benzene
Small organic molecules are thought to provide building blocks for the formation of complex interstellar polycyclic aromatic hydrocarbons (PAHs). However, the underlying chemical mechanisms remainExpand
The NASA Ames Polycyclic Aromatic Hydrocarbon Infrared Spectroscopic Database: The Computed Spectra
The astronomical emission features, formerly known as the unidentified infrared bands, are now commonly ascribed to polycyclic aromatic hydrocarbons (PAHs). The laboratory experiments andExpand
Interstellar Polycyclic Aromatic Hydrocarbon Molecules
Large polycyclic aromatic hydrocarbon (PAH) molecules carry the infrared (IR) emission features that dominate the spectra of most galactic and extragalactic sources. This review surveys the observedExpand
High-Resolution IR Absorption Spectroscopy of Polycyclic Aromatic Hydrocarbons: The Realm of Anharmonicity
We report on an experimental and theoretical investigation of the importance of anharmonicity in the 3 micrometers CH stretching region of Polycyclic Aromatic Hydrocarbon (PAH) molecules. We presentExpand
Spitzer’s perspective of polycyclic aromatic hydrocarbons in galaxies
Polycyclic aromatic hydrocarbon (PAH) molecules are abundant and widespread throughout the Universe, as revealed by their distinctive set of emission bands at 3.3, 6.2, 7.7, 8.6, 11.3 and 12.7 μm,Expand
Detection of the aromatic molecule benzonitrile (c-C6H5CN) in the interstellar medium
TLDR
The discovery of benzonitrile (c-C6H5CN), one of the simplest nitrogen-bearing aromatic molecules, in the interstellar medium is presented, providing a chemical link to the carriers of the unidentified infrared bands. Expand
...
1
2
3
4
5
...