Spectrum prediction

Several empirical approaches for NMR spectra prediction are based on the availability of large NMR spectral databases. By using special methods for encoding substructures that correspond to particular parts of the NMR spectrum, the correlation of substructures and partial spectra can be modeled. Substructures can be encoded by using the additive model greatly developed by Pretsch [11] and Clerc [12]. The authors represented skeleton structures and substituents by individual codes and calculation rules. A more general additive model was introduced... 

In this second empirical approach, which has also been used for C NMR spectra, predictions are based on tabulated chemical shifts for classes of structures, and corrected with additive contributions from neighboring functional groups or substructures. Several tables have been compiled for different types of protons. Increment rules can be found in nearly any textbook on NMR spectroscopy. 

Poroikov VV, Filimonov DA, Borodina YuV, Lagunin AA, Kos A. Robustness of biological activity spectra predicting by computer program PASS for non-congeneric sets of chemical compounds. J Chem Inf Comput Sci 2000 40 1349-55. 

Baumann, K., Clerc, J. T. Computer-assisted IR spectra prediction - linked similarity searches for structures and spectra. Anal. Chim. Acta 1997, 348, 327-343. 

W.R. Hruschka and K. Norris, Least squares curve fitting of near-infrared spectra predicts protein and moisture content in ground wheat, Appl. Spectrosc., 36, 261-265 (1982). 

It is often the case that the putative structure assigned to an RI is so unusual that vibrational and electronic spectra of analogous molecules are unavailable. How then can an experimentalist know whether an observed IR and/or UV-vis spectmm actually does belong to the putative structure assigned to the RI As exemphfied in the sections on matrix-isolated benzynes in Chapter 16 in this volume and also discussed in Chapter 17 in this volume, suitable electronic structure calculations can provide accurate predictions of the IR spectra and UV-vis excitation energies for RIs. The IR and UV-vis spectra, predicted for a particular structure, can then be compared with those obtained experimentally. 

The structure of IFN-x was also examined by CD [10]. Analysis of the IFN-x spectra predicts that the secondary structural elements derived from CD spectra indicate approximately 70% a-helix. The remainder of the molecule is either predicted to be random or a combination of (3 sheet and turn. Since it is known that algorithms that predict secondary structures from CD spectra are most accurate at identifying a helices, we are confident that IFN-x is mainly a helical. The CD spectra for the synthetic peptides of IFN-x were also obtained. The peptides IFN-x(l-37), IFN-x(62-92), IFN-x(l 19-150), and IFN-x(139-172) all show the presence of a helix, while IFN-x(34-64) and IFN-x(90-122) are mainly random. The presence of an a helix in the peptides supports the CD analysis of the intact protein and also roughly indicates the location of helical segments. 

The evaluation and interpretation of NMR parameters to establish molecular structures is usually a tedious task. An alternative way to elucidate a molecular structure is to directly compare its measured NMR spectrum - serving here as a fingerprint of the investigated molecule - with the corresponding spectra of known compounds. An expert system combining a comprehensive data base of NMR spectra with associated structures, NMR spectra prediction and structure generators not only facilitates this part of the NMR analysis but makes structure elucidation more reliable and efficient. 

Infrared and Raman spectra of stable carbocations have been obtained71,72 and are in complete agreement with their electron-deficient structures. Infrared spectra of alkyl cations and their deuteriated analogs correspond to the spectra predicted by calculations based on molecular models and force constants. Thus, these spectra can be used in the identification of stable carbocations. 

Hruschka, W.R. and Norris, K., Least Squares Curve Fitting of Near-Infrared Spectra Predicts Protein and Moisture Content in Ground Wheat Appl. Spectrosc. 1982, 36, 261-265. 

Fig.l. Level spectra predicted by various boson theories in the S0(6) limit of the Ginocchio model. 

As a result of the studies of the molecules in Figures 2.13 and 2.19, it is now clear that the VCD spectra predicted using Stephens equation for vibrational rotational strengths,... 

Now you have to be creative and invent a series of structures that appear to satisfy all the spectroscopic criteria you have so far, and test each structure critically against the spectra. Predict the chemical shifts do they really match Predict all the multiplet patterns do they match well Predict the NOEs and mass spectral fragmentation patterns and compare them with the experimental results. Discard any structures that clearly fail a test. Those that pass should be subjected to further, more searching tests. 

An LCAO (linear combination of atomic orbitals) local-density functional approach was used to calculate the band structures of a series of polymer chain conformations unsubstituted polysilane in the all-trans conformation and in a 411 helical conformation, and all-trans poly(dimethylsilane). Calculated absorption spectra predict a highly anisotropic absorption for the all-trans conformation of polysilane, with the threshold absorption peak arising strictly from polarizations parallel to the chain axis. The absorption spectrum for the helical conformation is much more isotropic. Results for the dimethyl-substituted polysilane chain suggest that the states immediately surrounding the Fermi level retain their silicon-backbone a character upon alkyl-group substitution, although the band gap decreases by I eV because of contributions from alkyl substituent states both below the valence band and above the conduction band to the frontier states. 

Baumann, K., Affolter, C, Pretsch, E. and Clerc, J.T. (1997a). Numerical Structure Representation and IR Spectra Prediction. Mikrochim.Acta, 14, 275-276. 

We illustrate the variation in predicted IR and VCD spectra resulting from variation in functional and basis set in Figs. 2-5, where spectra predicted for the small rigid chiral molecule propylene oxide (methyloxirane), 1, are shown. In Figs. 2 and 4, the basis set is varied with the functional held constant in Figs. 3 and 5, the basis set is fixed and the functional is varied. Over the range of basis sets employed one observes convergence to the complete basis set limit. The cc-pVTZ basis set provides results... 

Recently, small silicon clusters such as Si4, Si6 and Siv have been isolated and their structures determined by comparing Raman spectra predicted from ab initio quantum mechanical calculations with those observed in N2 matrices [1611]. The numbers of Raman-active fundamentals should be three, five, and five for the predicted structures of a planar Si4 rhombus (D2/ ), a distorted Si6 octahedron (D4/ ), and a pentagonal bipyramidal Siv (Ds/ ), respectively. Figure 2.61 shows the Raman spectra of these clusters as obtained by Honea et al. [1611]. It is seen that the observed spectra are in excellent agreement with the theoretical spectra, except for the mode of Si4 (which... 

Fig. 6.21 Mossbauer spectra of 10% Fe in NH4AI(S04)2.12D20 at three temperatures. The spectra predicted by the spin Hamiltonian for the three Kramers doublets are shown. [Ref. 119, Fig. 1]...

---