A halogen bond (XB) occurs when there is evidence of a net attractive interaction between an electrophilic region associated with a halogen atom in a molecular entity and a nucleophilic region in another, or the same, molecular entity. This type of interaction can be decomposed in terms of electrostatic, orbital mixing charge-transfer (CT) and dispersion terms. Halogen bonding occurs in various biological systems and processes, so it can be utilized in drug design.
XB is also being exploited for a range of functional applications, e.g. crystal engineering, supramolecular chemistry, polymer sciences, liquid crystals, conductive materials and medicinal chemistry.
A halogen atom makes an electrostatic contact when its electropositive polar region (region most distant from the atom to which halogen is covalently bonded) approaches a negatively charged species. Only electrophilic halogen makes a halogen bond. Halogens participating in halogen bonding include: iodine (I), bromine (Br), chlorine (Cl), and sometimes fluorine (F). All four halogens are capable of acting as XB donors (as proven through theoretical and experimental data) and follow the general trend: F < Cl < Br < I, with iodine normally forming the strongest interactions.
Dihalogens (I2, Br2, etc.) tend to form strong halogen bonds with a variety of Lewis bases. In fact, while an electrostatic interpretation has been proposed, a simple orbital mixing model (electron donation from the LP (HOMO) on the XB acceptor to the antibonding σ*-MO (LUMO) on the dihalogen accounts for the nature of the interaction, and nicely explains spectroscopic features (FT-Raman spectroscopy) as well as the typical elongation on the dihalogen molecules occurring upon the adduct formation. The strength and effectiveness of chlorine and fluorine in XB formation depend on the nature of the XB donor. If the halogen is bonded to an electronegative (electron withdrawing) moiety, it is more likely to form stronger bonds. For example, iodoperfluoroalkanes are well-designed for XB crystal engineering. In addition, this is also why F2 can act as a strong XB donor, but fluorocarbons are weak XB donors because the alkyl group connected to the fluorine is not electronegative. In addition, the Lewis base (XB acceptor) tends to be electronegative as well and anions are better XB acceptors than neutral molecules.
Halogen bonds are strong, specific, and directional interactions that give rise to well-defined structures. Halogen bond energy range from 10–200 kJ/mol. The strength of XB allows it to compete with the strength of hydrogen bonds, which are a little bit weaker in energy. Another contributing factor to halogen bond strength comes from the short distance between the halogen (Lewis acid, XB donor) and Lewis base (XB acceptor). The attractive nature of halogen bonds result in the distance between the donor and acceptor to be shorter than the sum of Van der Waals radii. The XB interaction becomes stronger as the distance decreases between the halogen and Lewis base. Anions are usually better XB acceptors than neutral species: more dissociated an ion pair is, stronger is the XB formed with the anion.
Many covalently-bonded halogens have unexpected polar region of positive electrostatic potential along the extensions of the covalent bond, called σ-hole. In the free atom X, the valence configuration present an half-filled pz orbital. In a molecule RX, the R−X bond (which lies along the z axis) involve the half-filled pz orbital, probably with some level of sp-hybridization. Considering little hybridization, the bonds are being formed by electrons in nearly pure p orbital. The outer lobe of this orbital is deficient in electronic charge and it could develop a positive potential. The σ-hole is the electron-deficient outer lobe of a half-filled p bonding orbital.
Due to the higher electronegativity of fluorine, a disproportionally large share of bonding electrons is near the F atom. Also a greater degree of sp-hybridization along the RX (X=F) bond axis is observed. These effects contribute to neutralize the σ-hole in the fluorine atom, explaining the presence of fewer known example of halogen bonds involving fluorine.
Halogen bonds tend to form at 180° angles, which was shown in Odd Hassel’s studies with bromine and 1,4-dioxane in 1954. Halogen bonding is a highly directional interaction and short interactions are more directional then long ones. The consequence of the anisotropic distribution of electron density around the halogen atom is that the angle between the covalent and non-covalent bonds around the halogen atom in is approximately 180°. Notably this direction is the same as that of the axis of σ*-MO (LUMO) on X–Y, so that geometrical factors cannot help discriminating between the electrostatic and CT bonding models. Indeed, the σ-hole reflects the topology of low-lying virtual MO along with the charge distribution of the XB donor.
When n-electron donors such as ethers and amines function as XB acceptors the interaction preferentially develops along the axis of the donated lone pair of electrons. In halogen bonds involving pyridine derivatives, the XY covalent bond is approximately coplanar with the pyridine ring, and the two angles are about 120°. Carbonyl, this-, and selenocarbonyl groups, with a trigonal planar geometry around the Lewis donor atom, can act as acceptors for one or two XBs.
In both cases, D is the atom, group, or molecule that donates electrons to the electron poor species H-R or X-R. H is the hydrogen atom involved in hydrogen bonding, and X is the halogen atom involved in halogen bonding. Note the halogen bond donor (atom X) accepts electrons while the halogen bond acceptor (entity D) donates electrons.
A parallel relationship can easily be drawn between halogen bonding and hydrogen bonding. In both interactions, an electron donor/electron acceptor relationship exists. In hydrogen bonding, a hydrogen atom acts as the electron acceptor and forms a non-covalent interaction by accepting electron density from an electron rich site. In halogen bonding, a halogen atom is the electron acceptor. The difference between the two types of interaction is what species can act as the electron donor, D.
Simultaneously, the normal covalent bond between H or X and R weakens, so the electron density on H or X appears to be reduced. Electron density transfers results in a penetration of the van der Waals volumes.
Halogen bonding is a interaction more directional than hydrogen bonding and XBs could have greater bond energies than hydrogen bonds, though they are often comparable.
In 1814, Jean-Jacques Colin described the formation of a liquid (with a metallic lustre) when he mixed together dry gaseous ammonia and dry iodine. The precise composition of the resulting I2···NH3 complex was established fifty years later by Frederick Guthrie. In his experiment, he added I2 to aqueous ammonia. The true nature of the molecular interaction was perhaps first understood following Robert S. Mulliken's groundbreaking discoveries on charge-transfer interactions, and their detailed description by Odd Hassel.
In 1950s, Mulliken developed a detailed theory of electron donor-acceptor complexes, classifying them as being outer or inner complexes. Outer complexes were those in which the intermolecular interaction between the electron donor and acceptor were weak and had very little charge transfer. Inner complexes have extensive charge redistribution. Mulliken’s theory has been used to describe the mechanism by which XB formation occurs.
Around the same time period that Mulliken developed his theory, crystallographic studies performed by Hassel began to emerge and became a turning point in the comprehension of XB formation and its characteristics.
The first X-ray crystallography study from Hassel’s group came in 1954. In the experiment, his group was able to show the structure of bromine 1,4-dioxanate using X-ray diffraction techniques. The experiment revealed that a short intermolecular interaction was present between the oxygen atoms of dioxane and bromine atoms. The O−Br distance in the crystal was measured at 2.71 Å, which indicates a strong interaction between the bromine and oxygen atoms. In addition, the distance is smaller than the sum of the van der Waals radii of oxygen and bromine (3.35 Å). The angle between the O−Br and Br−Br bond is about 180°. This was the first evidence of the typical characteristics found in halogen bond formation and led Hassel to conclude that halogen atoms are directly linked to electron pair donor with a bond direction that coincides with the axes of the orbitals of the lone pairs in the electron pair donor molecule.
In 1969, Hassel was awarded the Nobel Prize in Chemistry for his outstanding discovery that halogens can act as electrophilic, electron acceptors, and self-assemble into highly directionally organised crystalline charge-transfer complexes in presence of electron donors. An early review about electron donor-acceptor was provided by Bent in 1968. The use of the term "halogen bond" was not implemented until 1978 by Dumas and coworkers, who investigated complexes of CCl4, CBr4, SiCl4, and SiBr4 with tetrahydrofuran, tetrahydropyran, pyridine, anisole, and di-n-butyl ether in organic solvents.
However, it was not until the mid-1990s, that the nature and applications of the halogen bond began to be intensively studied. Systematic and extensive work by Legon and coworkers, who studied a wide variety of halogen-bonded adducts formed in the gas phase via microwave spectroscopy, drew attention to the similarities between halogen-bonding and better-known hydrogen-bonding interactions.
In 2007, computational calculations by Politzer and Murray were particularly significant as they revealed that the high directionality of the halogen bond is the result of an anisotropic distribution of electron density around the halogen nucleus and paved the way to the definition of the “σ-hole”. The anisotropic charge distribution of the σ-hole was experimentally observed using Kelvin probe force microscopy.
Crystal engineering is a growing research area that bridges solid-state and supramolecular chemistry. This unique field is interdisciplinary and merges traditional disciplines such as crystallography, organic chemistry, and inorganic chemistry. The scope of crystal engineering is the utilization of the intermolecular interactions for crystallization and for the development of new substances with different desired physicochemical properties. Before the discovery of halogen bonding, hydrogen bonding, coordination chemistry and inter-ion interactions are used for the development of liquid-crystalline and solid-crystalline materials. Since the discovery of halogen bonding, new molecular assemblies exist. Furthermore, halogen bonding is employed for the organization of radical cationic salts, fabrication of molecular conductors. Due to its directionality and to the unique chemical nature of XBs, this intermolecular interaction serves as an additional tool for the development of crystal engineering.
The first reported use of halogen bonding in liquid crystal formation was by Nguyen and Bruce. In an effort to form liquid crystals, alkoxystilbazoles and pentafluoroiodobenzene were used. Previous studies by Metrangolo and Resnati demonstrated the utility of pentafluoroiodobenzene for solid-state structures. Various alkoxystilbazoles have been utilized for nonlinear optics and metallomesogens. Using another finding of Resnati (e.g. N−I complexes form strongly), the group engineered halogen-bonded complexes with iodopentafluorobenzene and 4-alkoxystilbazoles. X-ray crystallography revealed a N−I distance of 2.811(4) Å and the bonding angle to be 168.4°. Similar N−I distances were measured in solid powders. The N−I distance discovered is shorter than the sum of the van der Waals radii for nitrogen and iodine (3.53 Å). The single crystal structure of the molecules indicated that no quadrupolar interactions were present. The silsesquioxanes complexes in figure were found to be liquid-crystalline.
To test the notion of polarizability involvement in the strength of halogen bonding, bromopentafluorbenzene was used as a Lewis base. Consequently, verification of halogen bond complex formation wasn’t obtained. This finding provides more support for the dependence of halogen bonding on atomic polarizability. Utilizing similar donor-acceptor frameworks, the authors demonstrated that halogen bonding strength in the liquid crystalline state is comparable to the hydrogen-bonded mesogens.
Applications utilizing properties of conjugated polymers emerged from work done by Heeger, MacDiarmid, and Shirakawa with the discovery that polyacetylene is a conducting, albeit difficult to process material. Since then, work has been done to mimic this conjugated polymer’s backbone (e.g., poly(p-phenylenevinylene)). Conjugated polymers have many practical applications, and are used in devices such as photovoltaic cells, organic light-emitting diodes, field-effect transistors, and chemical sensors. Goroff et al. prepared ordered poly(diiododiacetylene) (PIDA) via prearrangement of monomer (2) with a halogen bond scaffolding. PIDA is an excellent precursor to other conjugated polymers, as Iodine can be easily transformed. For instance, C−I cleavage is possible electrochemical reduction.
Crystal structures of monomer (2) are disordered materials of varying composition and connectivity. Hosts (3-7) were investigated for their molecular packing, primarily by studying cocrystals of monomer (2) and respective host. Both (3) and (4) pre-organized monomer (2), but steric crowding around the iodines prevented successful topological polymerization of the monomer. Hosts (5-7) utilize hydrogen bonds and halogen bonds to hold monomer (2) at an optimal distance from each other to facilitate polymerization.
In fact, when host (7) was used, polymerization occurred spontaneously upon isolation of the cocrystals. Crystal structures show the polymer strands are all parallel to the hydrogen-bonding network, and the host nitriles are each halogen-bonded to iodine atoms. Half of the iodine atoms in (1) in the crystal are in close contact to the oxamide oxygen atoms. Oxygen atoms of host (7) are acting as both hydrogen and halogen bond acceptors. This weak halogen bond further stabilizes this cocrystal, as shown in the figure below.
Porous structures have a variety of uses. Many chemists and material scientists are working to improve metal-organic frameworks (MOFs) to store hydrogen to use in cars. These highly organized crystalline inclusion complexes have potential uses in catalysis and molecular separation devices. Molecular organization is often controlled via intermolecular forces such as hydrogen bonding. However, utilizing hydrogen bonding often limits the range of pore sizes available due to close packing.
Pigge et al., utilized halogen bonding interactions between amines, nitrogen heterocycles, carbonyl groups, and other organic halides, to construct their porous structures. This is significant because organic crystalline networks mediated by halogen bonds, an interaction significantly weaker than hydrogen bond, are rare.
Compounds 1 and 2 (shown in figure) forms porous crystal structures, obtained in a variety of solvents, such as dichloromethane, pyridine, and benzene. The authors note that the porous inclusion complexes appear to be mediated in part by unprecedented I-π interactions and by halogen bond between iodine and carbonyl groups. The crystal structure (shown in figure) come together in a triangular array and molecules of 2 are approximately symmetric. Additionally, all of the sets of halogen bonding interactions are not identical, and all of the intermolecular interactions between halogen and halogen bond acceptor slightly exceed the sum of the Van der Waals radius, signifying a slightly weaker halogen bond, which leads to more flexibility in the structure. The 2D layers stack parallel to each other to produce channels filled with solvent.
Solvent interactions are also noted in the formation of the hexagonal structures, especially in pyridine and chloroform. Initially, crystals that form these solutions form channeled structures. Over time, new needle-like solvate-free structures form are packed tighter together, and these needles are actually the thermodynamically favored crystal. The authors hope to use this information to better understand the complementary nature of hydrogen bonds and halogen bonds in order to design small molecules predict structures.
For some time, the significance of halogen bonding to biological macromolecular structure was overlooked. Based on single-crystal structures in the protein data bank (PDB) (July 2004 version), a study by Auffinger and others on single crystals structures with 3 Å resolution (or better) into the PDB revealed that over 100 halogen bonds were found in six halogenated-base nucleic acid structures and sixty-six protein-substrate complexes for halogen-oxygen interactions. Although not as frequent as halogen-oxygen interactions, halogen-nitrogen and halogen-sulfur contacts were identified as well. These scientific findings provide a unique basis for elucidating the role of halogen bonding in biological systems.
On the biomolecular level, halogen bonding is important for substrate specificity, binding and molecular folding. In the case of protein-ligand interactions, the most common charge-transfer bonds with polarizable halogens involve backbone carbonyls and/or hydroxyl and carboxylate groups of amino acid residues. Typically in DNA and protein-ligand complexes, the bond distance between Lewis base donor atoms (e.g. O, S, N) and Lewis acid (halogen) is shorter than the sum of their Van der Waals radius. Depending on the structural and chemical environment, halogen bonding interactions can be weak or strong. In the case of some protein-ligand complexes, halogen bonds are energetically and geometrically comparable to that of hydrogen bonding if the donor-acceptor directionality remains consistent. This intermolecular interaction has been shown to be stabilizing and a conformational determinant in protein-ligand and DNA structures.
For molecular recognition and binding, halogen bonding can be significant. An example of this assertion in drug design is the substrate specificity for the binding of inhibitor IDD 594 to human aldose reductase, as shown in the figure. E.I. Howard reported the best resolution for this monomeric enzyme. This biological macromolecule consists of 316 residues, and it reduces aldoses, corticosteroids, and aldehydes. D-sorbitol, a product of the enzymatic conversion of D-glucose, is thought to contribute to the downstream effects of the pathology of diabetes. Hence, inhibiting this enzyme has therapeutic merit and IDD 594 is a therapeutic candidate for the treatment of diabetic complications.
Aldehyde-based and carboxylate inhibitors are effective but toxic because the functional activity of aldehyde reductase is impaired. Carboxylate and aldehyde inhibitors were shown to hydrogen bond with Trp 111, Tyr 48, and His 110. The “specificity pocket,” created as a result of inhibitor binding, consists of Leu 300, Ala 299, Phe 122, Thr 113, and Trp 111. For inhibitors to be effective, the key residues of interaction were identified to be Thr 113 and Trp 111. IDD 594 was designed such that the halogen would provide selectivity and be potent. Upon binding, this compound induces a conformational change that causes halogen bonding to occur between the oxygen of the Thr and the bromine of the inhibitor. The bond distance was measured to be 2.973(4) Å. It is this O−Br halogen bond that contributes to the large potency of this inhibitor for human aldose reductase rather than aldehyde reductase.