Development of high-level quantum chemical methods
One of the major goals of modern quantum chemistry is to develop highly accurate methods that provide balanced results over a wide range of molecular structures. The development of general, highly accurate methods is seriously hampered by the complexity of the models. In general, the more accurate results are provided by a quantum chemical method, the more complicated the method is. In practice, beyond a particular level, the implementation of the methods is not possible in the traditional way, that is, by writing the code “by hand”.
In recent years, we have developed automated programming techniques that resolve this issue. These tools are capable of automatically deriving and solving the equations of quantum chemical methods of arbitrary complexity. The procedure has been successfully applied to the implementation of the most accurate ab initio theory, the coupled-cluster (CC) approach. The automated techniques also enabled the implementation of energy derivatives, and thus the calculation of molecular properties (geometries, dipole and higher-order moments, vibrational frequencies and intensities, NMR chemical shifts, polarizabilities, excitation energies, transition moments, etc.) with high precision. Recently, approximate CC methods have been developed, which can provide better computational efficiency without sacrificing accuracy. Employing these models, high-accuracy calculations (e.g., accuracy of better than 0.5 kJ/mol for thermochemical properties) can now be performed for small and medium-sized molecules and radicals.
The new schemes have been applied to chemical problems, in particular, to the calculation of thermochemical properties of species relevant to combustion and atmospheric chemistry, establishing new reference data and often resulting in revision of the experimental values. The developed software is available to the scientific community (see http://www.mrcc.hu/) and is used by numerous research groups for highly accurate calculations.
Synthetic fluorescent sensors for recognition of biomolecules
In a joint project with the synthetic chemists of our university we develop new fluorescent molecular sensors, capable of recognizing essential constituents of living organisms, like amino acids, nucleotides (key compounds in intracellular energy transfer) and biologically and environmentally important anions and cations. Using such sensors in fluorescence imaging techniques, the distribution of a selected component, which is non-fluorescent in the absence of the molecular sensor, can be visualized. The photophysical properties of the dyes and their changes upon the binding of the above constituents are characterized on the basis of spectroscopic experiments and quantum chemical calculations.
Circular dichroism spectroscopy
In the past few years circular dichroism (CD) spectroscopy has become a generally applicable and efficient method for the elucidation of the absolute configuration and conformation of all types of chiral molecules. This is due to the rapid development of the quantum chemical methods applicable to the calculation of CD spectra.
We apply CD spectroscopy for studying the stereo structure of newly synthesized, chiral compounds. The spectra are interpreted with help of quantum chemical calculations. Recently, we have published detailed studies on the CD spectra and stereo structure of chalcone epoxides, phospholene oxides and cyclohexane based supramolecular receptors.
Theoretical calculation of vibrational spectra
The most reliable interpretation of the experimental vibrational (infrared and Raman) spectra is based upon quantum chemical calculations. Such calculations yield the optimized geometry of the investigated molecule, the harmonic vibrational force field, the frequencies and the infrared and Raman intensities of the vibrational normal modes. The fast computers of our days and the high level computational methods provide excellent results for medium size molecules.
Our calculations concern two topics:
- Adsorption of organic molecules on surfaces. The adsorption caused frequency shifts provide information on adsorbate-adsorbent interactions. As an initial step, we studied the vibrational spectra of several compound occurring frequently as adsorbates (5- and 6-membered N-aromatic parent compounds [00/3,03/3,04/5], O-heteroaromatics [04/2]). Our present work involves the computational modelling of adsorbent surfaces and the calculation of the vibrational spectra of the above molecules adsorbed on them. A special part in this field is the interpretation of the vibrational spectra of contact materials [06/1] that help in the better exploitation of ores during the flotation.
- Biological applications. We studied the vibrational spectra of several biologically active molecules for finding the spectra – structure correlations and the active parts of the molecules (orotic acid [00/2], phenothiazole derivatives [06/4], pinosylvin [02/2], resveratrol [07/3], gallic acid [02/5,07/2] phenols [02/2,07/2]). The investigation of calixarenes [02/1,07/7] helps in the modelling of ion channels of cell membranes. The study of supramolecular receptors, like calixarenes and b-cyclodextrine, will help the interpretation of the spectra of their complexes. Study of the vibrational spectrum – biological activity corralation on aflatoxins [06/2,07/8].
Investigation of hindered large amplitude motions (LAM) on high resolution IR spectra.
The spectrum analysis is based on the coupled vibration-rotation Hamiltonian which can take account of the Coriolis, Fermi, etc. perturbations. The investigated molecules have been methyl amine and formamide. In the case of methyl amine periodic K quantum number dependences were observed for the upper wagging levels which are perturbed by the second overtone levels of internal rotation. An extremely high resonance and some smaller resonances were found in the spectrum of formamide caused by Coriolis and Fermi couplings. The resolution of IR spectra was 0.001- 0.002 cm-1. MW and MWW data have been also used for the analysis.
The treatment of hindered LAM molecules requires elaboration of special theoretical models and programs. The goal is the determination of the potential surface from experimental spectrum data which requires the knowledge of the flexible geometry surface of the molecule. It is not possible to determine simultaneously both surfaces. The resolution of the problem is a combination of the high resolution spectra with the geometry surface calculated by quantum chemical methods. The procedure was applied to methyl amine. Unfortunately the potential surface determined by this way shows some dependence from the applied quantum chemical method and basis set.