Theoretical studies on pteridines. 2. Geometries,tautomer, ionization and reduction energies of substrates and inhibitors of dihydrofolate reductase

Ab initio SCF calculations with STO-3G and 3-21G basis sets are reported for approximately 85 pteridines of interest to the study of the reaction and inhibition mechanisms of dihydrofolate reductase. These include tautomeric, protonated, deprotonated, reduced and 6-substituted forms of the 2-amino-4-oxo- and 2,4-diamino-pteridines, many of which are not easily amenable to experimental investigation. Full geometry optimizations at the SCF/STO-3G level for 30 such pteridines have been performed. A step-wise computational protocol designed to identify the minimum level of theory necessary for reliable prediction of relative tautomer, reduction and ionization energies has been developed in an effort to minimize the cost of calculations for this reasonably large N-heterobicylic system. In general, SCF/STO-3G results were found to be inadequate while SCF/3-21G results obtained with STO-3G optimized geometries agreed with all available experimental evidence with the exception of the relative acidity of 6-methyl-7,8-dihydropterin. Correlation graphs relating experimental pK_a's to the calculated ionization energies are presented: these are of potential predictive value. An analysis is given of the importance of resonance substructures, such as the guanidinium and extended-guanidinium groups, in stabilizing some preferred tautomeric and ionized forms, and in explaining the observed geometry changes.     

Semiempirical NDDO calculations with STO-3G and 4-31G basis sets

Possible refinements of semiempirical methods include the use of larger basis sets and of correlated wave functions. These possibilities are investigated in semiempirical NDDO SCF calculations with the STO-3G and 4-31G basis sets, and in correlated calculations at the STO-3G level. The present approach is characterized by the analytical evaluation of all one-center terms and two-electron integrals, and the semiempirical adjustment of the remaining one-electron integrals and the nuclear repulsions. The NDDO SCF results tend to reproduce the correspondingab initio results more closely than experimental data, even if they are parametrized with respect to experiment. The explicit inclusion of electron correlation at the STO-3G level improves the calculated results only slightly.     

Theoretical study of hydrogen-bonded formaldehyde complexes

The hydrogen-bonded complexes involving formaldehyde and a series of proton donors of varying strengths, have been investigated at different levels of ab initio MO theory. The structures of the studied complexes were SCF optimized at the 6-31G basis set level. The binding energy was estimated employing basis set superposition correction, zero-point vibrations and MP2 correlation contribution at the different basis set: STO-3G; 6-31G; MP2/6-31G; 6-31G**; MP2/6-31G**; 6-311G(2d, 2p) and MP2/6-311G(2d, 2p). Linear relationships were found of the calculated binding energy with: the calculated shift in the carbonyl stretching frequency, the changes in carbonyl bond length and the optimum value of hydrogen-bond distance; furthermore the calculations confirm a parallel trend between the proton-donor ability and the strength of the hydrogen bond.     

Oxygen-induced next-nearest neighbour effects on the C1s-levels in polymer XPS-spectra

This letter reports the results of ab initio quantum chemical calculations on the C_1s core levels of model systems for a number of oxygen containing polymers. Conformational effects were studied. SCF calculations have been carried out with STO-3G and 4-31G basis sets, and Koopmans' theorem was applied to obtain the core-level binding energies. To evaluate the performance of the procedure SCF calculations were carried out on polyacrylic acid. The existence of oxygen-induced secondary substituent effects in the XPS-(ESCA-)spectra is discussed. A comparison with semi-empirical CNDO/2 results from Clark and Thomas has been made.     

Ab initio evaluation of oxidation numbers in some substituted benzenes

Giambiagi’s definition of oxidation number of an atom in a molecule has been applied successfully in theab initio SCF theory to calculate oxidation numbers using STO-3G and 4–31G basis sets for some substituted benzenes involving -F, -OH, -CH₃ and -NH₂ as substituents. The present study suggests that the oxidation numbers also seem to be indicative of their orientational behaviour like the net atomic charges.     

Optimized Structures and Relative Stabilities of the Carboranes from Ab Initio Calculations

Ab initio calculations at the STO-3G level were performed on almost all of the possible isomers for the entire series of closo-carboranes, C₂B_n-2H_n, 5 ? n ? 12. Geometry optimizations using the gradient method were also included in all calculations. We report here the relative energies obtained for the various isomers as well as the optimized structures. These calculations confirm our previous predictions of relative stabilities obtained from topological charge stabilization. Comparisons of our structures with those from experimental data provide us with a measure of reliability for bond distances obtained using ab initio SCF MO calculations at the STO-3G level. Results from the geometry optimization substantiated the experimentally known fluxional behavior of the 8 and 11 atom polyhedra.     

Geometry optimization and electronic structures of molecules H_3AXAH_3

The geometries of molecules H_3AXAH_3(X=O,S,Se and A=C,Si)have been optimizedusing STO-3G ab initio calculations and gradient method and the results are in good agreement withreported experimental values.From the STO-3G optimized geometries,we have also calculated theelectronic structures of these molecules using 4-31G and 6-31G basis sets to obtain the MO energies.atomic net charges and dipole moments.The ionization potentials calculated by 6-31G basis set are ingood agreement with experimental values.     

Theoretical Modeling of the Stereoisomerization of Tetracoordinated Be(II) Bis(chelate) Complexes

Quantum-chemical (semiempirical MNDO and ab initio RHF SCF with an STO-3G basis set) calculations were performed for stereoisomerization of tetracoordinated Be(II) bis(chelate) complexes with BeN₂O₂ coordination unit. The most probable pathway for their monomolecular isomerization was found to involve, in agreement with experimental data, cleavage of a Be–N coordination bond to form a tricoordinated Be intermediate (dissociation–recombination mechanism) rather than a competitive polytopal rearrangement. The planar structure of the intermediate detected upon thermo- and photoexcitation of the complexes was predicted from our calculations.     

A non-empirical LCAO MO SCF investigation of electronic relaxation accompanying core ionization in a series of saturated alkanes

Ab initio LCAO MO ΔSCF binding energies and relaxation energies have been computed at the STO-4.31G level for linear and branched alkanes (up to C₆) and are compared with experimental data. The structural dependence of binding energies and relaxation energies is discussed, and it is concluded that binding enegy shifts in these alkanes (compared with methane) arise mainly from relaxation energy changes.     

Electron donor-acceptor complexes: Evaluation of MNDO as a computational tool to probe intermolecular interactions

Donor-acceptor pairs form EDA complexes that exist as conformational isomers exhibiting different ground-state and photochemical properties. We have sought a rapid, general, and accurate quantum mechanical computational method to generate potential energy surfaces that are representative of the donor-acceptor intermolecular interactions at the self-consistent field (SCF) level. The semiempirical molecular orbital (MO) method MNDO has been compared to ab initio methods to assess its behavior with respect to energy, dipole moment and ionization potential shifts. MNDO correctly distinguishes between repulsive and bound EDA complex states at the SCF level and produces potential curves that are smooth and free of spurious minima or cusps. MNDO curves are systematically more repulsive than those for ab initio STO-3G calculations; calculated interaction energies exhibit a mean absolute deviation of 2.90 kcal/mol. MNDO appears to provide a reliable qualitative estimate of the nondispersion portion of the interaction energy. Limitations and errors arising from minimal basis sets, single determinants, and neglect of dispersion are discussed.     

The electronic structure of small lithium clusters

Ab initio molecular orbital calculations using the STO-6G and STO6-21G basis sets have been performed for the cluster series Li _n ⁺ , Li _n , and Li _n ^– (wheren=2–7). Thirty-two optimized structures are discussed and reported, many of which (especially for the anionic structures) have not yet been considered. The calculations suggest that for all three species the optimum geometries are planar. Of the two levels of theories that were investigated, STO-6G//STO-6G and STO6-21G//STO-6G, the latter hybrid theory was found to be less reliable. In particular, for the anionic structures these calculations should provide a platform from which more sophisticated, i.e., configuration interaction, geometry optimization can be performed.     

Tautomerism in alloxan

Ab initio STO-3G, 3-21G and 6-31G calculations have been used to investigate the energetics of the tautomerism in alloxan. The geometries of the tautomers have been fully optimized at STO-3G level. The results indicate that tautomerism in alloxan in the vapour is highly unlikely, the trioxo structure being by far the most stable structure. The population analysis of the alloxan anion gives evidence that the preferred protonation site is offered by the central oxygen atom, and rules out the opposite oxygen atom as a possible protonation site.     

Triaziridines. Part III. Triaziridine,azimine, and triazene: A SCF study of the energy and structure of N3H3-isomers

The portions of the N₃H₃ singlet potential energy surface corresponding to triaziridines ( 1 ), azimines ( 2 ) and triazenes ( 3 ) have been calculated by ab initio SCF using 3-21G, 6-31G, and 6-31G** basis sets. Minima and transition states were located by force gradient geometry optimization. The most important computation results are: (1) Triaziridines ( 1 ): The configuration at the 3 N-atoms is pyramidal. There are 2 stereoisomers, 1a and 1b . The c,t-isomer 1a has less energy than the c,c-isomer 1b . The 2 stereoisomerizations by N-inversion hve rather high activation energies. The N,N bonds in 1 are longer and weaker (STO-3G estimation) than in hydrazine. The N-homocycle 1 exhibits less ring strain than the C-homocycle cyclopropane or three-membered heterocycles. (2) Azimine ( 2 ): All 6 Atoms are in the same plane. There are 3 stereoisomers, 2a, 2b , and 2c . The order of ground state energies is (Z,Z) < (E,Z) ? (E,E). The 2 N,N bond lengths correspond to multiplicity 1½. The electronic structure of 2 corresponds to a 1,3-dipole with almost equal delocalization of the 4 π-electrons over all 3 N-atoms. The negative net charge at the central N-atom is much less than that at the terminal N-atoms. Azimines should behave as π-donors in complexation with transition metals (3) Triazene ( 3 ): All 6 atoms are in the same plane. There are 2 stereoisomers, 3a and 3b . The order of ground-state energies is (E) < (Z). The stereoisomerization proceeds as pure N-inversion. N-Inversion has a high energy barrier inversion at N(1) is faster than at N(2). One of the N,N bond lengths is typical for a double, the other for a single bond. The electronic structure of triazene 3 entails rather localized π- and p-electron pairs at N(1),N(2) and at N(3). Triazenes should behave as p-donors in complexation with transition metals. (4) -N₃H₃-Isomers: The order of ground-state energies is 3 < 2 < 1 . The energy differences between these constitutional isomers are much larger than between the stereoisomers of each. The [1,2]-H shifts for conversions of 2 to 3 and the [1,3]-H shift for tautomerization of 3 have relatively high activation energies; both shifts can be excluded as modes of thermal, unimolecular transformations.     

The ring-opening of an unsymmetrical tetrahedral intermediate - 2-hydroxy-1,3-oxathiolane

Ab initio calculations were performed on 2-hydroxy-1,3-oxathiolane: and the two products of its breakdown: Complete geometry optimizations were performed at minimal (STO-3G) and split-valence (3-21G) basis set levels. In addition, a single point calculation was performed at 6-31G* level withd orbitals added on sulfur only. The conformation of the oxathiolane intermediate and its stability relative to the breakdown products was investigated. The STO-3G basis set gave an envelope form while 3-21G gave the twist form of the five-membered ring as the most stable. For all three basis sets the ester product was more stable than thioester.     

Determination and characterization of a transition structure for water–formaldehyde addition

Quantum mechanical calculations of the geometric, energetic, electronic, and vibrational features of a transition structure for gas-phase water–formaldehyde addition (FW1?) are described, and a new transition-structure search algorithm is presented. Basis-set-dependent effects are assessed by comparisons of computed properties obtained from self-consistent field (SCF) molecular orbital (MO) calculations with STO-3G, 4-31G, and 6-31G^** basis sets in the absence of electron correlation. The results obtained suggest that STO-3G-level calculations may be sufficiently reliable for the prediction of the transition structure of FW1? and for the transition structures of related carbonyl addition reactions. Moreover, the calculated activation energy for formation of FW1? from water and formaldehyde (?44 kcal mol^?1) is very similar in all three basis sets. However, the energy of formaldehyde hydration predicted by STO-3G (? ?45 kcal mol^?1) is about three times larger than that predicted by the other two basis sets, with the activation energy for dihydroxymethane dehydration also being too large in STO-3G. Calculated force constants in all three basis sets are generally too large, leading to vibrational frequencies that are also too large. However, uniformly scaled force constants (in internal coordinates) give much better agreement with experimental frequencies, scaled 4-31G force constants being slightly superior to scaled STO-3G force constants.     

A conformational analysis of the six isomers of thiobispyridine

The conformational behaviour of the six isomers of thiobispyridine has been investigated using ab initio STO-3G*//rigid-roto, STO-3G*//STO-3G* and 6–31G**//STO-3G* molecular orbital models. The analysis reveals both the importance of optimising critical structure parameters and the basis set dependence of calculated rotational barrier heights. The most reliable model (6–31G**//STO-3G*) clearly indicates that the minimum energy conformers are not planar and that energy barriers between 30–100 kJ mol^?1 restrict inter-conversion to planar structures, thereby preventing conjugation between the p-electrons of the sulfur atom and the π system of both pyridine rings. From the calculated barrier heights, two mechanisms can be employed to explain conformer interconversion about the C? S bond: a disrotatory one-ring flip or a conrotatory two-ring flip mechanism. Where comparisons can be made (eg. 2,2′-thiobispyridine), dipole moment calculations are shown to be in good agreement with experiment. Finally, of the six isomers, appropriately substituted 2,2′, 2,3′- and 2,4′-thiobispyridines are most prone to a Smiles rearrangment.     

Comparative study of the molecular electrostatic potential obtained from different wavefunctions. Reliability of the semiempirical MNDO wavefunction

A systematic analysis of the molecular electrostatic potential (MEP) is presented. This study has been performed with a twofold purpose: first, to study the MEP dependence with regard to the quality of the basis set used to compute the ab initio SCF wavefunction and second, to develop and to assess a new strategy for computing isoelectrostatic potential maps using the semiempirical MNDO wavefunction. The only differences between this procedure and the ab initio SCF MEP computation lie in the freezing of the inner electrons and in the origin of the first-order density matrix. The statistical analysis of MEPs computed for a large number of molecules from MNDO wavefunction and ab initio SCF wavefunctions obtained using STO-3G, 4-31G, 6-31G, 4-31G*, 6-31G*, and 6-31G** basis sets points out the ability of any wavefunction to reproduce the general topological characteristics of the MEP surfaces. Nevertheless, split-valence basis sets including polarization functions are necessary to obtain accurate MEP minimum energy values. MNDO wavefunction tends to overestimate the MEP minima depth by a constant factor and shows an excellent ability to reflect the relative variation of MEP minima energies derived from a rather sophisticated (6-31G*) basis set, lacking of the shortcomings detected in the semiempirical CNDO approximation.     

Ab initio quantum mechanical calculations of energy,geometry, vibrational frequencies and IR intensities of tetraphosphorus tetrasulphide, α-P4S4(D2d), and vibrational analysis of As4S4 and As4Se4

Geometry, vibrational frequencies and IR intensities are calculated for α-P₄S₄ by scaled quantum mechanical calculations at the 6-31G*/SCF and STO-3G*/SCF levels. For both basis sets the frequencies are scaled with factors close to or equal to those found for P₄S₃, and based on these results a revised assignment is proposed. The α-P₄S₄ force field is transferred to the isostructural As₄S₄ and As₄Se₄ molecules and rescaled, and based on a good fit to experimental frequencies a new assignment is also proposed for these compounds.     

C80, C86, C88: Semiempirical and ab initio SCF calculations

Three complete sets of the isolated-pentagon-rule fullerene isomers are considered—₈₀ (7 species), C₈₆ (19 species), and C₈₈ (35 species). The interisomeric separation energetics is essentially similar at semiempirical (AM1, SAM1) and ab initio (HF/STO-3G, HF/3-21G, HF/4-31G) levels. For the C₈₀ set, the HF/4-31G computations explain why the lowest-energy isomer is not experimentally observed. The computations also agree with observations on C₈₆ and C₈₈. © 1997 John Wiley & Sons, Inc. Int J Quant Chem 63: 529–535, 1997