Halogen bonding and the design of new materials: organic bromides, chlorides and perhaps even fluorides as donors

In some halides RX, the halogen X has a region of positive electrostatic potential on its outermost portion, centered around the extension of the R−X bond. The electrostatic attraction between this positive region and a lone pair of a Lewis base is termed halogen bonding. The existence and magnitudes of such positive potentials on some covalently bonded halogens, and the characteristic directionality of the interaction, can be explained in terms of the degree of sp hybridization and polarizability of X and the electronegativity of R. Halogen bonding increases in strength in the order Cl < Br < I; fluorine is frequently said to not form halogen bonds, although a notable result of the present study is computational evidence that it does have the capability of doing so, if R is sufficiently electron withdrawing. An increasingly important application of halogen bonding is in the design of new materials (e.g., crystal engineering). In this paper, we present the calculated energies of a series of halogen-bonding interactions that could be the basis for forming linear chains, of types X----X----X---- or X----Y----X----Y----. We focus upon chlorides and bromides, and nitrogen bases. The B3PW91/6-311G(3df,2p) and MP2/6-311++G(3df,2p) procedures were used. We show how the computed electrostatic potentials (B3PW91/6-31G**) can provide guidance in selecting appropriate halide/base pairs. Figure Computed electrostatic potential of CH₃CH₂Br on the molecular surface defined by the 0.001-au contour of the electronic density. The bromine is facing the reader, and has a small positive (green) region centered around the intersection of the C–Br axis with the surface     

Interactions in pseudorotoxanes based on crown ether-secondary ammonium motifs. A theoretical study

A theoretical analysis of the nature of the interactions in dibenzo[24]crown-8 (DB24C8)-n-dibutylammonium (DBM)—pseudorotaxane complex at the MP2 and DFT levels shows that the main contribution to the binding energy is the electrostatic interaction with moderate (20–25%) correlation stabilization. The total binding energy in the DB24C8-DBM complex represents a sum of the binding energies of two NH–O and one CH–O hydrogen bonds and the latter constitutes about 25% of the total interaction energy, giving the total binding energy of −41.2 kcal mol⁻¹ at the BHandHLYP/6-311++G** level. Deprotonation of the DB24C8-DBM complex reduces the binding energy by some 50 kcal mol⁻¹, giving metastable complexes DB24C8-DBA-1 or DB24C8-DBA-2, which will dissociate to give free crown ether and n-dibutylamine because of the strong exchange repulsion that prevails in neutral complexes. Figure Formation of DB24C8-DBM pseudorotoxane complex     

Blue shifts <Emphasis Type="Italic">vs</Emphasis> red shifts in σ-hole bonding

σ-Hole bonding is a noncovalent interaction between a region of positive electrostatic potential on the outer surface of a Group V, VI, or VII covalently-bonded atom (a σ-hole) and a region of negative potential on another molecule, e.g., a lone pair of a Lewis base. We have investigated computationally the occurrence of increased vibration frequencies (blue shifts) and bond shortening vs decreased frequencies (red shifts) and bond lengthening for the covalent bonds to the atoms having the σ-holes (the σ-hole donors). Both are possible, depending upon the properties of the donor and the acceptor. Our results are consistent with models that were developed earlier by Hermansson and by Qian and Krimm in relation to blue vs red shifting in hydrogen bond formation. These models invoke the derivatives of the permanent and the induced dipole moments of the donor molecule. Figure Computed electrostatic potential on the molecular surface of Cl-NO₂. Color ranges, in kcal mol⁻¹, are: red, greater than 25; yellow, between 10 and 25; green, between 0 and 10; blue, between −4 and 0; purple, more negative than −4. The chlorine is facing the viewer, to the right. Note the yellow region of positive potential on the outer side of the chlorine, along the extension of the N–Cl bond. The blue region shows the sides of the chlorine to have negative potentials. The calculations were at the B3PW91/6–31G(d,p) level.     

Theoretical study of [ XN5 ]− (X=O, S, Se, Te) systems

A series of [XN₅]⁻ (X=O, S, Se, Te) compounds has been examined with ab initio and Density Functional Theory (DFT) methods. The five-membered nitrogen ring series of structures are global minima and may exist or be characterized due to their significant dissociation barriers (29.7–32.7 kcal mol⁻¹). Nucleus-independent chemical shifts (NICS) criteria and the presence of (4n+2) π-electrons confirmed that the five-membered nitrogen ring in their structures exhibits characteristics of aromaticity. Thus, the strong stability of the five-membered nitrogen ring structures may be attributed partially to their aromaticity.      

Why are dimethyl sulfoxide and dimethyl sulfone such good solvents?

We have carried out B3PW91 and MP2-FC computational studies of dimethyl sulfoxide, (CH₃)₂SO, and dimethyl sulfone, (CH₃)₂SO₂. The objective was to establish quantitatively the basis for their high polarities and boiling points, and their strong solvent powers for a variety of solutes. Natural bond order analyses show that the sulfur–oxygen linkages are not double bonds, as widely believed, but rather are coordinate covalent single S⁺→O⁻ bonds. The calculated electrostatic potentials on the molecular surfaces reveal several strongly positive and negative sites (the former including σ-holes on the sulfurs) through which a variety of simultaneous intermolecular electrostatic interactions can occur. A series of examples is given. In terms of these features the striking properties of dimethyl sulfoxide and dimethyl sulfone, their large dipole moments and dielectric constants, their high boiling points and why they are such good solvents, can readily be understood. Figure Dimers of dimethyl sulfoxide (DMSO; left) and dimethyl sulfone (DMSO2; right) showing O S—O -hole bonding and C H—O hydrogen bonding. Sulfur atoms are yellow, oxygens are red, carbons are gray and hydrogens are white     

ONIOM DFT/PM3 calculation on the interaction between STI-571 and abelson tyrosine kinase

The ONIOM2 (B3LYP/6–31G (d, p): PM3) and B3LYP/6–31G (d, p) methods were applied to investigate the interaction between STI-571 and abelson tyrosine kinase binding site. The complex of N-[4-methyl-3-(4-pyridin-3-yl-pyrimidin-2-ylamino)- phenyl]-benzamide (part of STI-571) and related 16 amino acid residues were found at B3LYP/6–31G (d, p) level to have hydrogen bonds and π....π stacking interaction, their binding energy via HAF optimization was −20.4 kcal mol⁻¹. The results derived from this study agreed well with the reported observation. Figure Optimized structure of STI-571 and Thr315 in abelson tyrosine kinase based on ONIOM2 method     

Boron nitride cages from B12N12 to B36N36: square–hexagon alternants vs boron nitride tubes

The structures and stabilities of square–hexagon alternant boron nitrides (B _x N _x , x=12–36) vs their tube isomers containing octagons, decagons and dodecagons have been computed at the B3LYP density functional level of theory with the correlation-consistent cc-pVDZ basis set of Dunning. It is found that octagonal B₂₀N₂₀ and B₂₄N₂₄ tube structures are more stable than their square–hexagon alternants by 18.6 and 2.4 kcal mol⁻¹, respectively, while the square–hexagon alternants of other cages are more stable. Trends in stability as a function of cluster size are discussed.Figure The octagonal B₂₀N₂₀ and B₂₄N₂₄ tube structures are more stable than their square-hexagon alternant cagesDedicated to Professor Dr. Paul von Ragué Schleyer on the occasion of his 75th birthday     

Ab initio analysis of the Cope rearrangement of germacrane sesquiterpenoids

The energetics of the Cope rearrangement of 17 germacrane sesquiterpenoids to their respective elemane forms have been calculated using both density functional theory (B3LYP/6-31G*) and post Hartee-Fock (MP2/6-31G**) ab initio methods. The calculations are in qualitative agreement with experimentally observed Cope rearrangements, but the two methods give slightly different results. MP2 calculations generally show more favorable elemene energies compared to the respective germacrenes (by around 3–4 kcal mol⁻¹) and smaller activation energies (by 2–3 kcal mol⁻¹). Additionally, neither method is accurate enough to consistently reproduce the germacrene/elemene equilibrium. Apparently, the generally small energy differences between the two forms in these sesquiterpenoids cannot be adequately reproduced at these levels of calculation. Figure The Cope rearrangement of the germacrane sesquiterpenoid bacchascandon to the elemane shyobunone     

Aromatic C20F20 cage and its endohedral complexes X@C20F20 (X = H-, F-, Cl-, Br-, H, He)

The structure and stability of endohedral X@C₂₀F₂₀ complexes (X = H⁻, F⁻, Cl⁻, Br⁻, H, He) have been computed at the B3LYP level of theory. All complexes in I _h symmetry were found to be energy minimum structures. H⁻@C₂₀F₂₀ and F⁻@C₂₀F₂₀ complexes have negative inclusion energies, while other complexes have positive inclusion energies. Similarity between C₂₀F₂₀ and C₂₀H₂₀ has been found for X = H and He. On the basis of the computed nucleus independent chemical shift values at the cage center, both C₂₀F₂₀ and C₂₀F₂₀ are aromatic. Figure Endohedral X@C₂₀F₂₀ complexes     

Halogen bonding: the σ-hole

Halogen bonding refers to the non-covalent interactions of halogen atoms X in some molecules, RX, with negative sites on others. It can be explained by the presence of a region of positive electrostatic potential, the σ-hole, on the outermost portion of the halogen’s surface, centered on the R–X axis. We have carried out a natural bond order B3LYP analysis of the molecules CF₃X, with X = F, Cl, Br and I. It shows that the Cl, Br and I atoms in these molecules closely approximate the configuration, where the z-axis is along the R–X bond. The three unshared pairs of electrons produce a belt of negative electrostatic potential around the central part of X, leaving the outermost region positive, the σ-hole. This is not found in the case of fluorine, for which the combination of its high electronegativity plus significant sp-hybridization causes an influx of electronic charge that neutralizes the σ-hole. These factors become progressively less important in proceeding to Cl, Br and I, and their effects are also counteracted by the presence of electron-withdrawing substituents in the remainder of the molecule. Thus a σ-hole is observed for the Cl in CF₃Cl, but not in CH₃Cl. Figure Schematic representation of the atomic charge generation. The molecular electrostatic potential (MEP) is calculated using the AM1* Hamiltonian. The semiempirical MEP is then scaled to DFT or ab initio level and atomic charges are generated from it by the restrained electrostatic potential (RESP) fit method.     

Ab initio and DFT study on the electrophilic addition reaction of bromine to tetracyclo[5.3.0.02,6.03,10]deca–4,8–diene

The electronic and geometric structures of tetracyclo[5.3.0.0^2,6.0^3,10]deca-4,8-diene (hypostrophene) have been investigated by ab initio and DFT/B3LYP methods using the 6-31G* and 6-311G* basis sets. The double bonds of hypostrophene are endo-pyramidalized. The cationic intermediates and products formed in the addition reaction have been investigated using the HF/6-311G*, HF/6-311G**, and B3LYP/6-311G* methods. The bridged bromonium cation was more stable than the U-type cation. Considering that the bridged cation does not isomerize to the less stable U-type cation, it is not possible for the U-type product to be obtained in the reaction. The bridged bromonium cation transformed into the more stable N-type cation and the N-type product was obtained via this cation. The thermodynamic stability of the exo, exo and exo, endo isomers of the N-type dibromide molecule were almost identical. The N-type product was 16.6 kcal mol⁻¹ more stable than the U-type product. Figure General energy diagram of the hypostrophene–bromine (HS–Br₂) system (kcal mol⁻¹) (MP2/6-311G*//HF/6-311G*)     

Synchronous fluorescence determination and molecular modeling of 5-Aminosalicylic acid (5-ASA) interacted with human serum albumin

In this paper, we proposed a new method for the determination of either human serum albumin (HSA) or 5-Aminosalicylic acid (5-ASA) by synchronous fluorescence spectra and examined the interaction between them using the molecular modeling method under simulative physiological conditions. The optimum conditions of synchronous fluorometric determination of HSA were investigated and the method was successfully applied to the determination of 5-ASA added to serum, urine, and saliva samples. The linear range of the determination of HSA and 5-ASA were 1.60 – 414 μg mL⁻¹ and 0.76 –22.95 μg mL⁻¹, the detection limits were 0.552 μg mL⁻¹ and 0.38 μg mL⁻¹, respectively. In addition, the effect of various common ions on the determination of HSA with 5-ASA was also discussed at room temperature. Figure The salicylic acid moiety is located within the binding pocket. The ring of 5-ASA was inserted in the hydrophobic cavity of site I, and it is important to note that the residue ARG-218 and the trptophan residue of HSA (Trp214) are in close proximity to the ring of 5-ASA suggesting the existence of hydrophobic interaction between them.     

G2, G3, and complete basis set calculations of the thermodynamic properties of cis- and trans-triazene

Following our recent study on triazane, we present a follow-up study on the thermodynamic properties of triazane’s unsaturated analog, triazene. We predict optimized structural parameters, vibrational frequencies, enthalpies of formation, enthalpies of combustion, specific enthalpies of combustion, and proton affinities. Our results indicate that the cis form of triazene has a specific enthalpy of combustion of −15.2 kJ g⁻¹ and the trans form has a specific enthalpy of combustion of −14.7 kJ g⁻¹. Figure Structures of cis- and trans-triazane, N₃H₃     

σ-hole bonding between like atoms; a fallacy of atomic charges

Covalently bonded atoms, at least in Groups V–VII, may have regions of both positive and negative electrostatic potentials on their surfaces. The positive regions tend to be along the extensions of the bonds to these atoms; the origin of this can be explained in terms of the σ-hole concept. It is thus possible for such an atom in one molecule to interact electrostatically with its counterpart in a second, identical molecule, forming a highly directional noncovalent bond. Several examples are presented and discussed. Such “like-like” interactions could not be understood in terms of atomic charges assigned by any of the usual procedures, which view a bonded atom as being entirely positive or negative. Figure Calculated electrostatic potential on the surface of SCl₂. The sulfur is in the foreground, the chlorines are at the back. Color ranges (kcal mol⁻¹): purple negative, blue between 0 and 8, green between 8 and 15, yellow between 15 and 20, red more positive than 20. Note that the sulfur has regions of both positive (red) and negative (purple) electrostatic potential     

Molecular surface electrostatic potentials and anesthetic activity

General anesthetics apparently act through weak, noncovalent and reversible interactions with certain sites in appropriate brain proteins. As a means of gaining insight into the factors underlying anesthetic potency, we have analyzed the computed electrostatic potentials V _S(r) on the surfaces of 20 molecules with activities that vary between zero and high. Our results are fully consistent with, and help to interpret, what has been observed experimentally. We find that an intermediate level of internal charge separation is required; this is measured by Π, the average absolute deviation of V _S(r), and the approximate window is 7 < Π < 13 kcal mol⁻¹. This fits in well with the fact that anesthetics need to be lipid soluble, but also to have some degree of hydrophilicity. We further show that polyhalogenated alkanes and ethers, which include the most powerful known anesthetics, have strong positive potentials, V _S,max, associated with their hydrogens, chlorines and bromines (but not fluorines). These positive sites may impede the functioning of key brain proteins, for example by disrupting their normal hydrogen-bond patterns. It has indeed been recognized for some time that the most active polyhalogenated alkanes and ethers contain hydrogens usually in combination with chlorines and/or bromines. Figure The computed HF/6-31G* electrostatic potential, in kcal mol⁻¹, on the 0.001 electrons/bohr³ surface of halothane, CF₃CHBrCl. The color ranges are: red, more positive than 25; yellow, between 15 and 25; green between 0 and 15; blue, between −10 and 0. The strongly positive (red) potential is due to the hydrogen; the yellow and green positive regions at the right are on the bromine surface Proceedings of “Modeling Interactions in Biomolecules II”, Prague, September 5th–9th, 2005.     

Theoretical 49Ti NMR chemical shifts

⁴⁹Ti chemical shifts for a total of 20 titanium complexes are reported, and several levels of theory are evaluated in order to identify a reliable approach for the calculation of titanium NMR data. The popular B3LYP/6–31G(d)//B3LYP/6–31G(d) proves to give very good agreement with experimental data over a range from 1,400 to −1,300 ppm. The MP2/6–31G(d)//MP2/6–31G(d) level computes even smaller average deviations but fails for TiI₄. This behavior together with its huge demand for computational resources requires careful handling of this theoretical level. In addition, NMR data for five titanium fulvene (or related) complexes are given. Dedicated to Professor Dr. Paul von Ragué Schleyer on the occasion of his 75th birthday     

Probing the nature of hydrogen bonds in DNA base pairs

Energy decomposition analyses based on the block-localized wave-function (BLW-ED) method are conducted to explore the nature of the hydrogen bonds in DNA base pairs in terms of deformation, Heitler–London, polarization, electron-transfer and dispersion-energy terms, where the Heitler–London energy term is composed of electrostatic and Pauli-exchange interactions. A modest electron-transfer effect is found in the Watson–Crick adenine–thymine (AT), guanine–cytosine (GC) and Hoogsteen adenine-thymine (H-AT) pairs, confirming the weak covalence in the hydrogen bonds. The electrostatic attraction and polarization effects account for most of the binding energies, particularly in the GC pair. Both theoretical and experimental data show that the GC pair has a binding energy (−25.4 kcal mol⁻¹ at the MP2/6-31G** level) twice that of the AT (−12.4 kcal mol⁻¹) and H-AT (−12.8 kcal mol⁻¹) pairs, compared with three conventional N-H···O(N) hydrogen bonds in the GC pair and two in the AT or H-AT pair. Although the remarkably strong binding between the guanine and cytosine bases benefits from the opposite orientations of the dipole moments in these two bases assisted by the π-electron delocalization from the amine groups to the carbonyl groups, model calculations demonstrate that π-resonance has very limited influence on the covalence of the hydrogen bonds. Thus, the often adopted terminology “resonance-assisted hydrogen bonding (RHAB)” may be replaced with “resonance-assisted binding” which highlights the electrostatic rather than electron-transfer nature of the enhanced stabilization, as hydrogen bonds are usually regarded as weak covalent bonds. Figure Electron density difference (EDD) maps for the GC pair: a shows the polarization effect (isodensity 1.2×10⁻³ a.u.); b shows the charge transfer effect (isodensity 2×10⁻⁴ a.u.) Dedicated to Professor Paul von Ragué Schleyer on the occasion of his 75th birthday     

Structures of inclusion complexes of halogenbenzoic acids and α-cyclodextrin based on AM1 calculations

Semi-empirical quantum mechanics calculations using AM1 (Austin Method 1) were carried out for various host-guest combinations of α-cyclodextrin and mono-halogen benzoic acids. The energetically favorable inclusion structures were identified. The AM1 results show that α-cyclodextrin complexes with mono-halogen benzoic acid acids (where the halogen is chlorine, bromide, iodine) as guest compounds are more stable in the “head first” position than in the “tail-first” position for meta and para isomers while ortho mono-halogen benzoic acids complexes with α-cyclodextrin are more stable in “tail-first” position. The calculated structures were found to be in good agreement with those obtained from crystalographic databases.