¹³C NMR Chemical Shift Prediction of sp² Carbon Atoms in Acyclic Alkenes Using Neural Networks

Abstract

The 13 C NMR chemical shift of sp 2 carbon atoms in acyclic alkenes was estimated with multilayer feedforward artificial neural networks (ANNs) and multilinear regression (MLR), using as structural descriptors a vector made of 12 components encoding the environment of the resonating carbon atom. The neural network quantitative model provides better results than the MLR model calibrated with the same data. The predictive ability of both the ANN and MLR models was tested by the leave-20%-out (L20%O) cross-validation method, demonstrating the superior performance of the neural model. The number of neurons in the hidden layer was varied between 2 and 7, and three activation functions were tested in the neural model: the hyperbolic tangent or a bell-shaped function for the hidden layer and a linear or a hyperbolic tangent function for the output layer. All four combinations of activation functions give close results in the calibration of the ANN model, while for the prediction a linear output function performs better than a hyperbolic tangent one, but from a statistical point of view one could not choose a particular combination against the others. For the ANNs with four neurons in the hidden layer, the standard deviation for calibration ranges between 0.59 and 0.63 ppm, while for prediction it lies between 0.89 and 1.07 ppm. We propose a parallel use of the four ANNs for the prediction of unknown shifts, because the mean of the four predictions exhibit a smaller number of outliers with lower residuals. The present model is compared with three additive schemes for the calculation of the sp 2 13 C NMR chemical shifts, and the statistical analysis of the results demonstrates that the ANN model gives better predictions than the classical ones.

Carbonization of covalent triazine-based frameworks via ionic liquid induction

“…The formation of carbon in these last three samples was further confirmed by the presence of a peak located at 127.4 ppm, typical attribution of carbon-carbon double bond 39 formed by carbonization of adamantine indicating that the major components of KPOP-6c, KPOP-6d, and KPOP-6e are amorphous porous carbon. The conclusion obtained from solid-state NMR was also supported by IR and EA.…”

13C NMR Chemical Shift Prediction of the sp3 Carbon Atoms in the α Position Relative to the Double Bond in Acyclic Alkenes

“…1 The use of ANNs in spectra interpretation and structure elucidation is 2-fold, i.e., either for classification (recognition of structural characteristics from IR [2][3][4][5][6][7][8][9][10][11] or MS spectra, [12][13][14][15] joint IR-13 C-NMR spectra 1 or IR-MS spectra 16 ) or for a quantitative prediction of a certain atomic property (the chemical shift in 13 C NMR spectra). [17][18][19][20][21][22][23][24][25][26][27][28] In a previous paper 28 we have estimated the 13 C NMR chemical shift of sp 2 carbon atoms in acyclic alkenes with MultiLinear Regression (MLR) and MultiLayer Feedforward (MLF) ANN models, using as structural descriptor of the environment of the resonating carbon a Topo-Stereochemical Code (TSC) with 12 components allowing for a unique description of the topo-stereochemical location of the carbon atoms around the double bond. The study investigated the 13 C NMR chemical shift of 130 acyclic alkenes with 244 structurally unique sp 2 carbon atoms.…”

Prediction of carbon-13 NMR chemical shift of alkanes with rooted path vector

“…1) as an example, three carbons C 1 , C 7 and C 8 in our case are chemically equivalent while the other five carbons, C 2 , C 3 , C 4 , C 5 , C 6 are all chemically non-equivalent, corresponding six values of chemically shift. The procedure of creating a rooted path vector for the total chemically non-equivalent carbon atoms in the examined molecular graph is illustrated as follows: for each atom, no.1, no.7 or no.8: the number of the path length 1 is 1, P 1 = 1; the number of the path length 2 is 3, P 2 = 3; the number of the path length 3 is 2, P 3 = 2; the number of the path length 4 is 1, P 4 = 1; the number of the path length more than 4 are all 0, P i = 0 (i = 5, 6,7,8,9). Therefore, the rooted path vector is written as p 1 = p 7 = p 8 = (1, 3, 2, 1, 0, 0, 0, 0, 0) 0 .…”

“…Wiener approach was also applied to molecules containing heteroatoms. And recently, many mathematical, chemometric, statistic methods predicting 13 C NMR chemical shifts (CS) of organic compounds have been developed by means of artificial neural network [6][7][8][9][10][11][12][13][14][15] algorithm and/or multiple linear regression method [16]. Neural networks were also used to predict a group of alkanes [17][18][19].…”