Crystal Structure Survey and Theoretical Analysis of Bifurcated Halogen Bonds

The Halogen Bond to Ethers - Prototypic Molecules and Experimental Electron Density

Halogen bonds to dialkyl ether molecules have remained largely unexplored. We here address the synthesis and the structural chemistry of the first halogen-bonded noncyclic alkyl ethers, combining 1,4-diiodotetrafluorobenzene and the prototypic or commonly used ethers dimethyl ether, tetrahydrofuran, and methyl-tert-butyl ether as halogen acceptors. Two different structural motifs based on moderately strong halogen bonds were obtained: Discrete trimolecular aggregates are formed, and unexpected halogen-bonded supramolecular chain adducts feature oxygen-bifurcated halogen bonds with 1:1 donor:acceptor ratio. Both structure types may be selectively obtained even for the same ether by adjusting the stoichiometry in the crystallization experiments. The geometric features of the etheric oxygen center were found to be flexible, in contrast to the almost linear geometry about the halogen donor atom. A high-resolution X-ray diffraction experiment on the extended adduct of dimethyl ether allowed us to study the electronic details of the acceptor-bifurcated I···O···I halogen bonds. The electron density in the bond critical points and derived properties such as the Laplacian indicate essentially electrostatic interactions and explain the geometrical flexibility of ethers in halogen bonds. Our studies demonstrate the great versatility of ethers as halogen bond acceptors, that can occur in many geometrical arrangements and whose contribution to nature's structural designs should not be underestimated.

Does a halogen bond require positive potential on the acid and negative potential on the base?

It is usually expected that formation of a halogen bond (XB) requires that a region of positive electrostatic potential associated with a σ or π-hole on the Lewis acid will interact with the negative potential of the base, either a lone pair or π-bond region. Quantum calculations of model systems suggest this not to be necessary. The placement of electron-withdrawing substituents on the base can reverse the sign of the potential in its lone pair or π-bond region to positive, and this base can nonetheless engage in a XB with the positive σ-hole of a Lewis acid. The reverse scenario is also possible in certain circumstances, as a negatively charged σ-hole can form a XB with the negative lone pair region of a base. Despite these classical Coulombic repulsions, the overall electrostatic interaction is attractive in these XBs, albeit only weakly so. The strengths of these bonds are surprisingly insensitive to changes in the partner molecule. For example, even a wide range in the depth of the σ-hole of the approaching acid yields only a minimal change in the strength of the XB to a base with a positive potential.