scholarly journals The Role of Structural Representation in the Performance of a Deep Neural Network for X-ray Spectroscopy

Molecules ◽  
2020 ◽  
Vol 25 (11) ◽  
pp. 2715
Author(s):  
Marwah M.M. Madkhali ◽  
Conor D. Rankine ◽  
Thomas J. Penfold
Keyword(s):  
Neural Network ◽  
Deep Neural Network ◽  
Mean Squared Error ◽  
Chemical Space ◽  
Theoretical Calculations ◽  
Local Environment ◽  
Structural Representation ◽  
Edge Structure ◽  
X Ray ◽  
X Ray Absorption

An important consideration when developing a deep neural network (DNN) for the prediction of molecular properties is the representation of the chemical space. Herein we explore the effect of the representation on the performance of our DNN engineered to predict Fe K-edge X-ray absorption near-edge structure (XANES) spectra, and address the question: How important is the choice of representation for the local environment around an arbitrary Fe absorption site? Using two popular representations of chemical space—the Coulomb matrix (CM) and pair-distribution/radial distribution curve (RDC)—we investigate the effect that the choice of representation has on the performance of our DNN. While CM and RDC featurisation are demonstrably robust descriptors, it is possible to obtain a smaller mean squared error (MSE) between the target and estimated XANES spectra when using RDC featurisation, and converge to this state a) faster and b) using fewer data samples. This is advantageous for future extension of our DNN to other X-ray absorption edges, and for reoptimisation of our DNN to reproduce results from higher levels of theory. In the latter case, dataset sizes will be limited more strongly by the resource-intensive nature of the underlying theoretical calculations.

scholarly journals A Deep Neural Network for the Rapid Prediction of X-ray Absorption Spectra

2020 ◽  
Author(s):  
Conor Rankine ◽  
Marwah Madkhali ◽  
Thomas Penfold
Keyword(s):  
Neural Network ◽  
Absorption Spectra ◽  
Deep Neural Network ◽  
Theoretical Calculations ◽  
End Users ◽  
Supervised Machine Learning ◽  
Strong Impact ◽  
Edge Structure ◽  
X Ray ◽  
X Ray Absorption

<p>X-ray spectroscopy delivers strong impact across the physical and biological sciences by providing end-users with highly-detailed information about the electronic and geometric structure of matter. To decode this information in challenging cases, e.g. <i>in operando</i> catalysts, batteries, and temporally-evolving systems, advanced theoretical calculations are necessary. The complexity and resource requirements often render these out of reach for end-users, and therefore data are often not interpreted exhaustively, leaving a wealth of valuable information unexploited. In this paper, we introduce supervised machine learning of X-ray absorption spectra, by developing a deep neural network (DNN) that is able to estimate Fe K-edge X-ray absorption near-edge structure spectra in less </p><p>than a second with no input beyond geometric information about the local environment of the absorption site. We predict peak positions with sub-eV accuracy and peak intensities with errors over an order of magnitude smaller than the spectral variations that the model is engineered to capture. The performance of the DNN is promising, as illustrated by its application to the structural refinement of iron(II)tris(bipyridine) and nitrosylmyoglobin, but also highlights areas for which future developments should focus.</p>

scholarly journals A Deep Neural Network for the Rapid Prediction of X-ray Absorption Spectra

2020 ◽  
Author(s):  
Conor Rankine ◽  
Marwah Madkhali ◽  
Thomas Penfold
Keyword(s):  
Neural Network ◽  
Absorption Spectra ◽  
Deep Neural Network ◽  
Theoretical Calculations ◽  
End Users ◽  
Supervised Machine Learning ◽  
Strong Impact ◽  
Edge Structure ◽  
X Ray ◽  
X Ray Absorption

<p>X-ray spectroscopy delivers strong impact across the physical and biological sciences by providing end-users with highly-detailed information about the electronic and geometric structure of matter. To decode this information in challenging cases, e.g. <i>in operando</i> catalysts, batteries, and temporally-evolving systems, advanced theoretical calculations are necessary. The complexity and resource requirements often render these out of reach for end-users, and therefore data are often not interpreted exhaustively, leaving a wealth of valuable information unexploited. In this paper, we introduce supervised machine learning of X-ray absorption spectra, by developing a deep neural network (DNN) that is able to estimate Fe K-edge X-ray absorption near-edge structure spectra in less </p><p>than a second with no input beyond geometric information about the local environment of the absorption site. We predict peak positions with sub-eV accuracy and peak intensities with errors over an order of magnitude smaller than the spectral variations that the model is engineered to capture. The performance of the DNN is promising, as illustrated by its application to the structural refinement of iron(II)tris(bipyridine) and nitrosylmyoglobin, but also highlights areas for which future developments should focus.</p>

scholarly journals A Deep Neural Network for the Rapid Prediction of X-ray Absorption Spectra

2020 ◽  
Vol 124 (21) ◽  
pp. 4263-4270 ◽  
Author(s):  
C. D. Rankine ◽  
M. M. M. Madkhali ◽  
T. J. Penfold
Keyword(s):  
Neural Network ◽  
Absorption Spectra ◽  
Deep Neural Network ◽  
X Ray ◽  
Rapid Prediction ◽  
X Ray Absorption

scholarly journals Neural network assisted analysis of bimetallic nanocatalysts using X-ray absorption near edge structure spectroscopy

2020 ◽  
Vol 22 (34) ◽  
pp. 18902-18910 ◽  
Author(s):  
Nicholas Marcella ◽  
Yang Liu ◽  
Janis Timoshenko ◽  
Erjia Guan ◽  
Mathilde Luneau ◽  
...  
Keyword(s):  
Neural Network ◽  
Local Structure ◽  
Bimetallic Nanoparticles ◽  
Edge Structure ◽  
X Ray ◽  
Coordination Numbers ◽  
Rich Information ◽  
Bimetallic Nanocatalysts ◽  
Assisted Analysis ◽  
X Ray Absorption

Trained neural networks are used to extract the first partial coordination numbers from XANES spectra. In bimetallic nanoparticles, the four local structure descriptors provide rich information on structural motifs.

scholarly journals Characterization of Pressure-induced Phase Transition on [Co(bpy)3](NO3)2·3H2O

2014 ◽  
Vol 70 (a1) ◽  
pp. C57-C57
Author(s):  
Ya-Wen Lee ◽  
Yu-Chun Chuang ◽  
Jyh-Fu Lee ◽  
Chi-Rung Lee ◽  
Chih-Ming Lin ◽  
...  
Keyword(s):  
Phase Transition ◽  
Raman Spectroscopy ◽  
Density Functional ◽  
Theoretical Calculations ◽  
Spin Transition ◽  
Xrd Analysis ◽  
External Pressure ◽  
Edge Structure ◽  
X Ray ◽  
X Ray Absorption

The pressure-induced phase transition study of high-spin (HS) compound, [Co(bpy)3](NO3)2·3H2O (bpy = 2,2'-bipyridine), is characterized by powder x-ray diffraction (XRD), x-ray absorption spectroscopy (XAS), Raman spectroscopy, and theoretical calculations. The results indicate that the HS ground state t2g5eg2 on Co(II) is gradually transformed to low-spin (LS) state with configuration t2g6eg1 . This phase transition behavior is similar to the thermal-induced spin crossover phenomenon once it is incorporated into certain framework. In this study, we put the compound into diamond anvil cell and applied physical pressure to replace the framework effect. To analyze the x-ray absorption near edge structure (XANES) and Raman spectroscopy, the finite difference method for near-edge structure (FDMNES) and density functional theory (DFT) calculations are applied to illustrate the experimental spectroscopies, respectively. In XANES results, an intersection point around 7756.33 eV beyond 1.73 GPa is assigned as the critical point between HS and LS state. The extended x-ray absorption fine structure (EXAFS) analysis indicates that the averaged Co-N bond lengths is 2.127(7) Å at HS state and decreased to 1.950(4) Å at LS state. Based on XRD analysis, the external pressure reduces the hexagonal cell constants from a = 13.77(3) Å and c = 21.71(3) Å to a = 13.37(5) Å and c = 21.11(1) Å. According to those experimental results, the mechanism of such pressure-induce spin transition can be interpreted as the enhancement of intermolecular interaction by increasing the external pressure.

scholarly journals The local environment of metal sites in intracellular granules investigated by using X-ray-absorption spectroscopy

1984 ◽  
Vol 221 (3) ◽  
pp. 855-868 ◽  
Author(s):  
G N Greaves ◽  
K Simkiss ◽  
M Taylor ◽  
N Binsted
Keyword(s):  
Absorption Spectroscopy ◽  
Local Environment ◽  
Nearest Neighbour ◽  
Model Compounds ◽  
Edge Structure ◽  
X Ray ◽  
X Ray Absorption ◽  
Oxygen Bond ◽  
Phosphate Deposits ◽  
Phosphate Groups

We report the use of X-ray-absorption spectroscopy (x.a.s.) to study the local atomic environment of cations in intracellular granules from the hepatopancreas of Helix aspersa. Both the calcium K-edge in these concretions and the manganese K-edge in doped specimens were measured. Electron-microprobe measurements confirm that the introduced Mn2+ is concentrated in irregular growths on the surfaces of the granules. The near-edge structure (x.a.n.e.s.) of calcium is similar to that of manganese, indicating that the oxygen-co-ordination spheres of both cations share a similar symmetry. From the extended structure (e.x.a.f.s.) the metal-oxygen bond lengths of 0.230 nm (2.30A) for Ca-O and 0.218 nm (2.18A) for Mn-O [+/- 0.004 nm (0.04A)] were determined, reference being made to a variety of model compounds. The low density of the granules (2.07 g/cm3), together with the local atomic distribution, suggest an open hydrated structure for these phosphate deposits. Detailed analysis of the distribution of nearest-neighbour oxygen atoms demonstrates that this is asymmetric and considerably broader for Ca2+ than for Mn2+. Compared with the model compounds, the Ca2+ environment in the granules is similar to that observed in Ca2P2O7. I.r. spectra indicate the presence of condensed phosphate groups in the granules, with the strong possibility these are pyrophosphate (P2O7(4-) groups.

Comparative analysis of XANES and EXAFS for local structural characterization of disordered metal oxides

2021 ◽  
Vol 28 (5) ◽  
Author(s):  
Junying Li ◽  
Yuanyuan Li ◽  
Prahlad K. Routh ◽  
Evgeniy Makagon ◽  
Igor Lubomirsky ◽  
...  
Keyword(s):  
Small Sample Size ◽  
Functional Materials ◽  
Local Environment ◽  
Structural Disorder ◽  
Active Species ◽  
Small Sample ◽  
Edge Structure ◽  
X Ray ◽  
X Ray Absorption ◽  
Xanes Analysis

In functional materials, the local environment around active species that may contain just a few nearest-neighboring atomic shells often changes in response to external conditions. Strong disorder in the local environment poses a challenge to commonly used extended X-ray absorption fine structure (EXAFS) analysis. Furthermore, the dilute concentrations of absorbing atoms, small sample size and the constraints of the experimental setup often limit the utility of EXAFS for structural analysis. X-ray absorption near-edge structure (XANES) has been established as a good alternative method to provide local electronic and geometric information of materials. The pre-edge region in the XANES spectra of metal compounds is a useful but relatively under-utilized resource of information of the chemical composition and structural disorder in nano-materials. This study explores two examples of materials in which the transition metal environment is either relatively symmetric or strongly asymmetric. In the former case, EXAFS results agree with those obtained from the pre-edge XANES analysis, whereas in the latter case they are in a seeming contradiction. The two observations are reconciled by revisiting the limitations of EXAFS in the case of a strong, asymmetric bond length disorder, expected for mixed-valence oxides, and emphasize the utility of the pre-edge XANES analysis for detecting local heterogeneities in structural and compositional motifs.

Interplay between Local Electronic Structure, Crystalline Structure and Magnetic Ordering in Intermetallic Compounds Ce2Fe17−xMnx

2012 ◽  
Vol 190 ◽  
pp. 251-254 ◽  
Author(s):  
Alexander A. Yaroslavtsev ◽  
Alexey P. Menushenkov ◽  
Olga V. Grishina ◽  
Roman V. Chernikov ◽  
Anatoly G. Kuchin
Keyword(s):  
Crystalline Structure ◽  
Magnetic Ordering ◽  
Local Environment ◽  
Edge Structure ◽  
X Ray ◽  
Local Electronic Structure ◽  
Magnetic States ◽  
First Time ◽  
X Ray Absorption ◽  
Mn Concentration

For the first time the rearrangement of cerium local environment in Ce2Fe17-xMnx(x = 0; 1; 2) intermetallics vs. Mn concentration and temperature was investigated by thespectroscopy of extended X-ray absorption fine structure (EXAFS) above K-Ce absorptionedge. At the same time under similar conditions by the spectroscopy of X-ray absorption near-edge structure (XANES) above L3-Ce absorption edge the valence state of Ce was studied. Thecorrelation is found between changes in local electronic and crystalline structure observed inCe2Fe17-xMnx and the types of magnetic states in these compounds.

Sign in / Sign up

Export Citation Format

Share Document