Bifurcated (three-centered) halogen bonds — a single acceptor atom simultaneously engaged by two halogen-bond donors — were shown by Cinčić, Friščić and Jones (CrystEngComm, 2011), through new cocrystal structures and a Cambridge Structural Database survey, to be real but rare, largely charge-assisted, and not reliably predictable from molecular structure, with no design rules to favor bifurcation over the default two-center halogen bond. Later work resolved parts of this question: a survey of over 22,000 database halogen bonds quantified bifurcation rarity (under 2%) and near-symmetric geometry; quantum-chemical studies established negative cooperativity and electrostatic-potential and charge-density anisotropy as intrinsic reasons bifurcated arrangements are usually weaker than linear ones; and converging ortho-diiodo donors were shown to create a wide single-minimum potential that enables bifurcated binding at aromatic nitrogen acceptors — a local design heuristic confirmed by a dedicated database survey and targeted cocrystallization. The general core remains open: no published criteria are necessary and sufficient, or validated as generally predictive, for bifurcation versus two-center binding across acceptor chemistries and donor types. This problem asks for those criteria and their validation as quantitative, experimentally demonstrated design rules that would make bifurcated halogen-bond acceptors a deliberately designable supramolecular synthon.

Background

A halogen bond (XB) is an attractive noncovalent interaction between an electrophilic region on a covalently bound halogen atom (usually iodine or bromine) and a Lewis-basic acceptor such as N, O, S, an anion, or a π-system. The electrophilic region, called a σ-hole, is a patch of positive molecular electrostatic potential on the extension of the C–X bond axis, produced by the anisotropy of the halogen's electron density; its counterpart on the acceptor is a lone pair or region of negative electrostatic potential. Because the σ-hole is directional, two-center XBs (one donor halogen, one acceptor site) are among the most reliable tools of crystal engineering — the deliberate assembly of multi-component molecular crystals (cocrystals, salts) from recurring interaction motifs called supramolecular synthons; two-center XBs based on diiodotetrafluorobenzene donors now underpin a wide range of multicomponent assemblies and functional materials (reviewed by Ding et al., 2020).

A bifurcated (three-centered) halogen bond is the motif in which a single acceptor atom simultaneously binds two halogen-bond donors (donor–X···A···X–donor); this record concerns bifurcated acceptors. By analogy with multi-center hydrogen bonds, such motifs would be attractive design elements, but in 2011 Cinčić, Friščić and Jones showed experimentally and through a Cambridge Structural Database (CSD — the repository of essentially all published small-molecule crystal structures) survey that they are problematic as synthons. Their three new structures illustrated the full behavioral spectrum: a dithiane cocrystal (with the donor 1,4-diiodotetrafluorobenzene) in which each sulfur atom accepts two I···S bonds (3.38 and 3.40 Å) to form ladder-like chains; an acridine decamer (with the vicinal donor 1,2-diiodotetrafluorobenzene) whose central nitrogen accepts two I···N bonds (3.12 and 3.16 Å); and a triphenylphosphine oxide cocrystal in which the strongly basic oxygen did not bifurcate at all but formed one very short, near-linear I···O bond (2.75 Å) in wheel-and-axle assemblies. Their CSD survey (v5.31; distance cutoffs 3.5 Å for N/O/F, 3.8 Å for S/Cl, 3.9 Å for Br, 4.1 Å for I) found bifurcated acceptors to be rare — for example only 8 of 284 structures with short I···N contacts — and largely confined to salts with anionic, charge-assisted acceptors; among neutral cocrystals, sulfur and oxygen acceptors dominated. They concluded that three-centered XBs are real and acceptor-specific in geometry but rare and not reliably predictable, with no design rules to favor bifurcation over the default two-center bond; explanations based on acceptor hybridisation or adjacent π-systems remained conjectural.

Subsequent work sharpened the picture. A 2022 survey of more than 22,000 CSD halogen bonds (X = Cl, Br, I) by Michalczyk, Zierkiewicz and Scheiner found fewer than 2% bifurcated, with iodine dominating bifurcated arrangements; the two bonds of a bifurcated pair are typically near-symmetric (lengths within about 0.05 Å, angles within a few degrees), and quantum calculations show negative cooperativity — each bond is weakened and lengthened by the other. Model-compound studies traced the intrinsic rarity of bifurcation to the shape of the electrostatic potential, the anisotropy of the interacting orbitals, and the angular extent of the total electron density, with σ-hole–π bifurcated bonds also characterized theoretically; acceptor binding was further shown to reduce the electrostatic-potential maximum, and hence the bonding capability, of the remaining donor iodines on the same molecule (anticooperativity). On the exploitation side, converging ortho-diiodo donors were shown computationally and by a dedicated CSD survey to create a single wide potential minimum that enables bifurcated binding even at weakly basic aromatic (sp2) pyridine nitrogens — a local design heuristic — and deliberate bifurcated-XB assemblies such as supramolecular double helices, metal-based and charge-assisted systems, acceptors extended to heavier pnictogens (heavier group-15 elements — phosphorus, arsenic and antimony), and the first symmetrical bifurcated XB in a protein–ligand structure (DNA gyrase inhibitors) have been reported. What no study provides is the general, validated criterion: given a donor–acceptor pair and crystal context, when will the acceptor bifurcate rather than form two-center bonds?

Problem Statement

The open question, as yet unanswered: identify the specific electronic, steric, hybridisation and donor-geometry criteria that are necessary and sufficient for a halogen-bond acceptor site to adopt a bifurcated (three-centered) X···A···X arrangement rather than one or more conventional two-center halogen bonds — covering neutral and charge-assisted (ionic) acceptors across N, O, S and heavier acceptor chemistries, and organic iodine/bromine donors as well as inorganic donors such as I2I_2 and interhalogens — and validate these criteria as quantitative design rules. Validation has two required legs: (i) predictive, not merely descriptive, performance on systematic database data, tested against held-out crystal structures under uniform geometric criteria; and (ii) targeted experimental crystal engineering, in which cocrystals or salts designed according to the rules realize the predicted bifurcated motifs, as demonstrated by single-crystal X-ray diffraction. For leg (i), the operational definition of the target motif — the X···A distance cutoffs, the angular condition, and whether asymmetric bifurcated geometries count — must be stated explicitly before testing, fixed independently of the criterion under test, and applied uniformly; the predictive performance must then be reported under both the cutoffs of the 2011 survey (3.5 Å for N/O/F, 3.8 Å for S/Cl, 3.9 Å for Br, 4.1 Å for I) and the near-symmetry window of the 2022 survey (bond lengths within about 0.05 Å, angles within a few degrees), so that the verdict does not hinge on a single threshold choice. The overarching objective is to establish whether bifurcated halogen bonds, like two-center halogen bonds and multi-center hydrogen bonds, can serve as a deliberately designable supramolecular synthon, and if so under what explicit conditions. A resolution is either (a) a criterion meeting these requirements, validated by both legs, or (b) a demonstration that no criterion built from the descriptors named above (electronic, steric, hybridisation and donor-geometry descriptors) is generally necessary and sufficient. Branch (b) is satisfied either by an argument that rules out the whole descriptor class or by a matched set of structures that agree in those descriptors yet differ in bifurcation outcome; a survey of individually failing descriptors, or a single unmatched counterexample, does not resolve it. The list of acceptor chemistries and donor types above states the intended scope of the question, not an independent requirement that each class be resolved separately.

References

  1. Dominik Cinčić, Tomislav Friščić, William Jones (2011). Experimental and database studies of three-centered halogen bonds with bifurcated acceptors present in molecular crystals, cocrystals and salts. CrystEngComm 13, 3224. DOI: 10.1039/c0ce00699h. https://doi.org/10.1039/c0ce00699h
  2. Mariusz Michalczyk, Wiktor Zierkiewicz, Steve Scheiner (2022). Crystal Structure Survey and Theoretical Analysis of Bifurcated Halogen Bonds. Crystal Growth & Design 22, 6521-6530. DOI: 10.1021/acs.cgd.2c00726. https://doi.org/10.1021/acs.cgd.2c00726
  3. Nikola Bedeković, Antonio Magnabosco, Vladimir Stilinović, Dominik Cinčić (2026; published online December 2025). Cocrystals of 1,2-Diiodotetrafluorobenzene with Pyridine Derivatives: Pyridine Nitrogen as a Bifurcated Acceptor with Ortho-Diiodo Halogen Bond Donors. Crystal Growth & Design 26, 554-564. DOI: 10.1021/acs.cgd.5c01439. https://doi.org/10.1021/acs.cgd.5c01439
  4. Nikola Bedeković, Tomislav Piteša, Mihael Eraković, Vladimir Stilinović, Dominik Cinčić (2022). Anticooperativity of Multiple Halogen Bonds and Its Effect on Stoichiometry of Cocrystals of Perfluorinated Iodobenzenes. Crystal Growth & Design 22, 2644-2653. DOI: 10.1021/acs.cgd.2c00077. https://doi.org/10.1021/acs.cgd.2c00077
  5. Steve Scheiner (2021). Comparison of Bifurcated Halogen with Hydrogen Bonds. Molecules 26, 350. DOI: 10.3390/molecules26020350. https://doi.org/10.3390/molecules26020350
  6. Ekaterina Bartashevich, Elena Troitskaya, Ángel Martín Pendás, Vladimir Tsirelson (2015). Understanding the bifurcated halogen bonding N⋯Hal⋯N in bidentate diazaheterocyclic compounds. Computational and Theoretical Chemistry 1053, 229-237. DOI: 10.1016/j.comptc.2014.09.024. https://doi.org/10.1016/j.comptc.2014.09.024
  7. Martin Novák, Cina Foroutan-Nejad, Radek Marek (2015). Asymmetric bifurcated halogen bonds. Physical Chemistry Chemical Physics 17, 6440-6450. DOI: 10.1039/c4cp05532b. https://doi.org/10.1039/c4cp05532b
  8. Mehdi D. Esrafili, Esmail Vessally, Mohammad Solimannejad (2016). Symmetric bifurcated halogen bonds: substituent and cooperative effects. Molecular Physics 114, 3610-3619. DOI: 10.1080/00268976.2016.1253882. https://doi.org/10.1080/00268976.2016.1253882
  9. Ziyu Wang (2022). Bifurcated Halogen Bond-Driven Supramolecular Double Helices from 1,2-Dihalotetrafluorobenzene and 2,2′-Bi(1,8-naphthyridine). Crystals 12, 937. DOI: 10.3390/cryst12070937. https://doi.org/10.3390/cryst12070937
  10. Katarina Lisac, Filip Topić, Mihails Arhangelskis, Sara Cepić, Patrick A. Julien, Christopher W. Nickels, Andrew J. Morris, Tomislav Friščić, Dominik Cinčić (2019). Halogen-bonded cocrystallization with phosphorus, arsenic and antimony acceptors. Nature Communications 10, 61. DOI: 10.1038/s41467-018-07957-6. https://doi.org/10.1038/s41467-018-07957-6
  11. Anja Kolarič, Thomas Germe, Martina Hrast, Clare E. M. Stevenson, David M. Lawson, Nicolas P. Burton, Judit Vörös, Anthony Maxwell, Nikola Minovski, Marko Anderluh (2021). Potent DNA gyrase inhibitors bind asymmetrically to their target using symmetrical bifurcated halogen bonds. Nature Communications 12, 150. DOI: 10.1038/s41467-020-20405-8. https://doi.org/10.1038/s41467-020-20405-8
  12. Xue-Hua Ding, Yong-Zheng Chang, Chang-Jin Ou, Jin-Yi Lin, Ling-Hai Xie, Wei Huang (2020). Halogen bonding in the co-crystallization of potentially ditopic diiodotetrafluorobenzene: a powerful tool for constructing multicomponent supramolecular assemblies. National Science Review 7, 1906-1932. DOI: 10.1093/nsr/nwaa170. https://doi.org/10.1093/nsr/nwaa170
  13. Daniel P. Devore, Kevin L. Shuford (2024). Data and Molecular Fingerprint-Driven Machine Learning Approaches to Halogen Bonding. Journal of Chemical Information and Modeling 64, 8201-8214. DOI: 10.1021/acs.jcim.4c01427. https://doi.org/10.1021/acs.jcim.4c01427