Carbon-13 NMR spectral simulation

Abstract

Since its inception, NMR spectroscopy has become one of the most useful analytical techniques available to the chemist. It is especially useful for solving complex structure elucidation problems, because it yields detailed structural information about the surroundings of specific atoms within molecules. Chemical shifts, intensities, and multiplicities of NMR resonances are all helpful in structure elucidation. Many nuclei can be used in modern NMR analyses, including 'H, nR, I3C, 15N, ,70,19F, 29Si, andalP; however, the two most commonly used today are 1H and l3C.

Automated model selection for the simulation of carbon-13 nuclear magnetic resonance spectra of cyclopentanones and cycloheptanones

“…Four independent mean std error of est sets of models were evaluated: model set 1 included models developed for the atom subsets in scheme 1 when geometrical descriptors were not used; model set 1G comprised models generated for the atom subsets in scheme 1 when geometrical descriptors were used; model set 2 contained models for the atom subsets in scheme 2 when geometrical descriptors were not used; and model set 2G consisted of models developed for the atom subsets in scheme 2 when geometrical descriptors were used. 4 illustrates the results for model sets 1 and 1G, and Figure 5 displays the results for model sets 2 and 2G. Figures 4 and 5 show that the results improve upon the addition of geometrical descriptors for both subsetting schemes.…”

“…Empirical modeling techniques involve the development of linear mathematical models relating the chemical shifts to various numerical atom-based structural parameters (descriptors) (1). Such models have the form S = b0+ b^Xi + 62-^2 where S is the chemical shift of an individual carbon atom, the X¡s are the descriptor values, the b¡ s are the coefficients determined by multiple linear regression analysis, and d indicates the number descriptors used in the model.…”

“…In scanning electrochemical microscopy (SECM), a microvoltammetric tip electrode (with a microdisk geometry and a tip radius of 0.1-12.5 µ ) is rastered in close proximity to the substrate to be imaged in a solution containing an electroactive species (4)(5)(6). Previous reports from this laboratory described imaging of surfaces by employing the feedback 1 Current address: Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, CA 91125. mode, where the steady-state tip current, iT, controlled by an electrochemical reaction at the tip electrode, is a function of the solution composition, tip-substrate distance, d, and the nature of the substrate itself. The measurement of ¿r can thus provide information about the topography of the sample surface as well as its electrical and chemical properties; electrodes (e.g., minigrids and interdigitated electrode arrays) (6)(7)(8), polymer and oxide films on electrode (7,9), and biological materials (10) have been imaged.…”

Evaluation of automated methods for the selection of models for simulation of carbon-13 nuclear magnetic resonance spectra of keto steroids

“…The ability to determine when accurate chemical shift predictions can be obtained by using existing equations, and when new models are required, is critical to maximizing the overall utility of the current 13C NMR spectral simulation approach. Spectrum simulation results using selected models are compared with those obtained through direct regression analysis (i.e., the standard simulation methodology (27)) for a group of keto steroid molecules. Another area of interest involves investigating alternative means for characterizing atomic structural environments in order to enhance the current model selection capabilities.…”

Prediction of carbon-13 nuclear magnetic resonance chemical shifts by artificial neural networks

“…McClelland and Rumelhart37 suggest using a momentum term, a, to increase the learning rate without leading to oscillation while moving along the error surface during training. The momentum term is incorporated into BP in the following manner: Aw¡j(n + 1) = i)bpflpj + aAWijin) (9) where n indicates the epoch or presentation number and a is the momentum. In all of the studies presented here the momentum term was held constant at 0.75.…”