The perfluoroaryl⋯aryl interaction, the most important subset of π-hole⋯π bonding, refers to the attractive stacking interaction between a perfluoroaryl group and an aryl group. In contrast to the aryl⋯aryl interaction with the same size, the much stronger perfluoroaryl⋯aryl interaction has its own characteristics and applications. A brief history of the development of the perfluoroaryl⋯aryl interaction was given first in this review, followed by an overview of the state-of-the-art of the nature of the perfluoroaryl⋯aryl interaction. Much attention was paid to the application of the perfluoroaryl⋯aryl interaction both in the traditional research fields such as crystal engineering and organic luminescent materials and in the hot research fields such as photovoltaics materials and biological engineering. It is believed that this timely and comprehensive review provides a foundation and guide for the future development and application of the perfluoroaryl⋯aryl interaction.

The element fluorine (F) is abundant in inorganic substances. However, there are only twelve F-containing organic compounds found in nature.1 The introduction of F atoms in small organic molecules or macromolecules can alter considerably their physicochemical properties.2 Hence, fluorination has become an efficient strategy to develop and/or utilize new types of noncovalent interactions. The rediscovery of the halogen bond provides a typical example of successful utilization of the fluorination strategy: most of the halogen bond donors are fluorinated organic compounds.3 The perfluoroaryl⋯aryl interaction usually refers to the noncovalent stacking interaction between the perfluoroaryl and aryl rings. This is due to that the other possible noncovalent interactions between the perfluoroaryl and aryl rings have their own specific names such as the C−H⋯F, C−F⋯π (perfluoroaryl ring) and F⋯F interactions, and these noncovalent interactions are always quite weak.4 If the perfluoroaryl⋯aryl interaction occurs between the perfluoroarene and arene molecules, it can also be called the perfluoroarene⋯arene interaction. The terms “aryl⋯perfluoroaryl interaction” and “arene⋯perfluoroarene interaction” were also used in some references. To avoid confusion, in this review, we will uniformly use the term “perfluoroaryl⋯aryl interaction.” Compared to other well-known noncovalent interactions such as hydrogen bond, halogen bond, and arene⋯arene stacking interaction, the perfluoroaryl⋯aryl interaction does not exist in nature, and it develops along with the artificial synthesis of perfluorinated organic compounds.1–5 There are a large number of review articles on the hydrogen bond, halogen bond, or arene⋯arene stacking interaction, but no topic review article on the perfluoroaryl⋯aryl interaction has been published to date. With recent rapid progress and wide application of the perfluoroaryl⋯aryl interaction in many fields, it is necessary and timely to present a comprehensive review article on its past, present, and future.

In 1960, Patrick and Prosser reported the 1:1 molecular complex between hexafluorobenzene (HFB) and benzene in the top journal Nature.6 This Nature article is less than one page in length, but has opened up a new era for the application and development of the perfluoroaryl⋯aryl interaction. The melting points of hexafluorobenzene and benzene are 5.0 and 5.4 °C, respectively, while the 1:1 mixture of the two substances has a higher melting point of 23.7 °C. Evidently, the much higher melting point of the 1:1 mixture can be attributed to the much stronger intermolecular force between hexafluorobenzene and benzene. The structures at different phases of the 1:1 solid adduct of hexafluorobenzene with benzene/deuterobenzene have been determined by the x-ray single-crystal diffraction techniques and powder diffraction methods.7,8 The hexafluorobenzene and benzene/deuterobenzene molecules in these structures are arranged alternately and stacked in infinite columns (Fig. 1). The columnar nature of these structures indicates the dominant role of the perfluoroaryl⋯aryl interaction between hexafluorobenzene and benzene/deuterobenzene in the formation of the solid adduct of the two substances.

FIG. 1.

The spacefilling structure of the 1:1 solid adduct of hexafluorobenzene with benzene viewed along the b axis in the low-temperature phase. Color code: carbon, gray; hydrogen, white; fluorine, yellow green.

FIG. 1.

The spacefilling structure of the 1:1 solid adduct of hexafluorobenzene with benzene viewed along the b axis in the low-temperature phase. Color code: carbon, gray; hydrogen, white; fluorine, yellow green.

Close modal

Patrick and Prosser had thought that the complex between hexafluorobenzene and benzene might be the charge-transfer-type complex. However, their measurements based on absorption spectra did not support such a view.6 The unexpected experimental results may be caused by two reasons: (1) the sensitivity of absorption spectra is not high enough to detect the complex in which the charge transfer is not very strong; (2) no charge transfer occurs upon the formation of the complex, at least at the ground state. About twenty years later, the molecular quadrupole moments of hexafluorobenzene and benzene in both the liquid and gas phases were determined by using the electric field-gradient birefringence method.9,10 The quadrupole moments of hexafluorobenzene and benzene in dilute solutions in CCl4 are (31.7 ± 1.7)×10−40 and –(33.3 ± 2.1)×10−40 C m2, respectively,9 and in the gas phase are (31.7 ± 1.7)×10−40 and –(29.0 ± 1.7)×10−40 C m2, respectively.10 The gas-phase values are in good agreement with the liquid-phase values. The experimental determinations of the signs and magnitudes of the quadrupole moments of hexafluorobenzene and benzene have laid a solid foundation for the study of the quadrupole–quadrupole interaction between the two molecules. It must be pointed out here that the quadrupole–quadrupole interaction alone cannot accurately describe the perfluoroaryl⋯aryl interaction between hexafluorobenzene and benzene. Vinter et al. reported that the perfluoroaryl⋯aryl interaction can be expressed as a sum of atom-based multipole and van der Waals (vdW) interactions.11 They calculated the interaction energy of the complex between hexafluorobenzene and benzene by using the force field including the extended electron distribution and found that the calculated value of −5.6 kcal/mol was in good agreement with the estimated experimental value of about −7.0 kcal/mol. The binding strength of the perfluoroaryl⋯aryl interaction between hexafluorobenzene and benzene is slightly larger than that of the well-known O−H⋯O hydrogen bond in the water dimer.12 

Considering the importance of the quadrupole–quadrupole interaction for the study of the complex between hexafluorobenzene and benzene, it is significant to know the size of this energy term. In a symmetrically face-to-face complex between hexafluorobenzene and benzene, the interaction energy of two quadrupoles separated by a distance r can be calculated as follows:13,14
U ( r ) = 9 Q 1 Q 2 / ( 16 π ε 0 r 5 ) ,
(1)
where Q1 and Q2 are the molecular quadrupole moments of hexafluorobenzene and benzene, respectively. The experimental values for Q1, Q2, and r are 31.7 × 10−40 C m2, −29.0 × 10−40 C m2, and 3.5 Å, respectively.8,10 So, the quadrupole–quadrupole interaction energy of the complex between hexafluorobenzene and benzene is obtained to be −5.1 kcal/mol. The value of this energy term is very close to the value of the total interaction energy, which is −5.6 kcal/mol.11 In previous studies, it was generally accepted that the quadrupole–quadrupole interaction determined the structures of the cocrystals between hexafluorobenzene and benzene.14 

Williams analyzed in detail the structural phase transitions and molecular packing modes of the solid adduct of hexafluorobenzene with benzene based on the intermolecular quadrupole–quadrupole interactions.15 The molecular packing modes in the solid adduct of hexafluorobenzene with benzene were found to be different from those in the pure solid hexafluorobenzene or in the pure solid benzene (Fig. 2). As shown in Fig. 2, the hexafluorobenzene and benzene molecules contain many “separated charges.”15 Two like charges will repel each other, and two opposite charges will attract each other. Both the repulsion between two like charges and attraction between two opposite charges inevitably exist in the molecular solid. Hence, if only the electrostatic forces are considered in the intermolecular interactions, the molecular packing in the solid is the consequence of the balance between maximizing the electrostatic attraction and minimizing the electrostatic repulsion. In the pure solid benzene or pure solid hexafluorobenzene, the two lowest energy configurations for the dimers are the T-shaped and parallel-slipped configurations, and in the solid adduct of hexafluorobenzene with benzene, the molecules hexafluorobenzene and benzene stack alternatively to form a series of infinite columns.15 It seems that the molecular packing modes in these solids can be perfectly explained by employing the electrostatic interactions of quadruple moments as shown in Fig. 2. However, the intercolumnar interactions in the solid adduct of hexafluorobenzene with benzene are much more complex and cannot be easily predicted by the electrostatic molecular quadrupole–quadrupole interactions only.

FIG. 2.

A schematic representation of the quadrupole–quadrupole interactions in the pure solid benzene (a) and solid adduct of hexafluorobenzene with benzene (b). Reproduced with permission from J. H. Williams, Acc. Chem. Res. 26, 593–598 (1993). Copyright 1993 American Chemical Society.15 

FIG. 2.

A schematic representation of the quadrupole–quadrupole interactions in the pure solid benzene (a) and solid adduct of hexafluorobenzene with benzene (b). Reproduced with permission from J. H. Williams, Acc. Chem. Res. 26, 593–598 (1993). Copyright 1993 American Chemical Society.15 

Close modal

In the crystal structure of the solid adduct of hexafluorobenzene with benzene in its lowest temperature phase IV viewed along the c crystallographic axis, the centroid distance between neighboring hexafluorobenzene and benzene molecules is about 3.77 Å.8 Using this centroid distance, the induction interaction energy between neighboring hexafluorobenzene and benzene was calculated to be a few milli-electron Volt and the electrostatic interaction energy was calculated to be 150 meV.16 Therefore, it is reasonable to neglect the contribution of the interaction energy in analyzing the structure–property relationship of the solid adduct of hexafluorobenzene with benzene. It was also found that the dispersion interaction energy between hexafluorobenzene and benzene is in magnitude comparable to the quadrupole–quadrupole electrostatic interaction energy. However, the role of the dispersion interaction energy was neglected by Williams et al. in rationalizing the crystal structure of the solid adduct of hexafluorobenzene with benzene.

Although Patrick and Prosser did not detect the contribution of charge transfer in the liquid mixtures of hexafluorobenzene and benzene using the absorption spectroscopy method in their seminal Nature article,6 the role of charge–transfer interaction in the complex between hexafluorobenzene and benzene and further in the other complexes bound by the perfluoroaryl⋯aryl interactions has still been the focus of many research studies. Klemperer et al. carried out a detailed molecular beam study on the dipole moment of the complex between hexafluorobenzene and benzene in the gas phase and reported a value of 0.44 ± 0.04 D.17 This means that considerable amounts of electron densities are transferred from benzene to hexafluorobenzene upon the complex formation in the gas phase.15 Evidently, such an appreciable charge transfer will lead to large changes of the ring-breathing modes of both the hexafluorobenzene and benzene. However, in the solid phase, the expected large frequency shifts were not observed in the vibrational spectra of the complex between hexafluorobenzene and benzene.16,18

Similarly, the study of Smith and Massey showed that perfluorotriphenylene (PF-Tp) and triphenylene (Tp) can also form the complex bound by the perfluoroaryl⋯aryl interaction.19 The direct experimental evidence is that the melting point of the 1:1 mixture of PF-Tp and Tp is 250–252 °C, whereas the melting points of pure PF-Tp and pure Tp are only 109 and 199 °C, respectively. The charge transfer was still not detected upon the formation of the complex between PF-Tp and Tp. It was supposed that the charge transfer bands might be observed if the electron-donating substituents were introduced to Tp. Grubbs et al. synthesized a chiral triphenylene derivative (OR-Tp) and studied the perfluoroaryl⋯aryl interaction between PF-Tp and OR-Tp (Fig. 3).20 Comparing the UV/Vis absorption spectra of PF-Tp (blue), OR-Tp (green), and the 1:1 mixture PF-Tp/OR-Tp in Fig. 3, the focus naturally falls on the low, broad absorption band at more than 360 nm in the UV/Vis spectra of PF-Tp/OR-Tp. The emergence of this new absorption band suggests that the complex PF-Tp/OR-Tp is likely to exhibit some characteristics of charge transfer. Such a new absorption band was not experimentally detected in other complexes bound by the perfluoroaryl⋯aryl interactions.

FIG. 3.

The molecular structures of PF-Tp and OR-Tp, and the UV/Vis absorption spectra of PF-Tp (blue), OR-Tp (green), and 1:1 mixture PF-Tp/OR-Tp of the two compounds (red). Reproduced with permission from Weck et al., Angew. Chem., Int. Ed. 38, 2741–2745 (1999). Copyright 1999 Wiley-VCH.20 

FIG. 3.

The molecular structures of PF-Tp and OR-Tp, and the UV/Vis absorption spectra of PF-Tp (blue), OR-Tp (green), and 1:1 mixture PF-Tp/OR-Tp of the two compounds (red). Reproduced with permission from Weck et al., Angew. Chem., Int. Ed. 38, 2741–2745 (1999). Copyright 1999 Wiley-VCH.20 

Close modal

Without any doubt, above-mentioned classic studies have laid a solid foundation for further development of the perfluoroaryl⋯aryl interaction. Most of the findings and conclusions of these studies are still frequently cited today in the scientific literature.

In order to rationalize the origin and nature of the noncovalent interactions, the concepts of σ-hole and π-hole were introduced sequentially from 2007 to 2010 and have now been generally accepted by the scientific community.21–26 The σ-hole and σ-hole bonding have been used to successfully explain the origin and nature of the halogen bond, chalcogen bond, tetrel bond, etc.24 Here, the π-hole and π-hole bonding will be employed to rationalize the origin and nature of the perfluoroaryl⋯aryl interaction.

1. The π-hole

The σ-hole was defined as “a region of low electron density along the extension of a σ bond.”21–25 As a counterpart of the σ-hole, the π-hole was referred to “a region of low electron density that is perpendicular to a portion of a molecular framework.”22,26 Furthermore, a σ-hole with positive electrostatic potentials was called the positive σ-hole and a π-hole with positive electrostatic potentials was called the positive π-hole, and vice versa.21–26  Figure 4 shows the electrostatic potential maps of two model molecules Cl2 and F2C=O. Unless otherwise stated, the electron densities and electrostatic potentials of selected model molecules in this review article were calculated at the reliable MP2/aug-cc-pVTZ theory level with the Gaussian 16 code.27 As can be seen in Fig. 4, the geometric difference between the σ-hole and π-hole is quite clear.

FIG. 4.

The molecular electrostatic potentials (kcal/mol) on the 0.001 a.u. molecular surfaces of Cl2 and F2C=O.

FIG. 4.

The molecular electrostatic potentials (kcal/mol) on the 0.001 a.u. molecular surfaces of Cl2 and F2C=O.

Close modal
Although the π-hole was defined based on the electron density, the molecular electrostatic potential map was always employed to confirm the existence of a π-hole. The molecular electrostatic potential, V(r), is an observable that can be determined by x-ray diffraction.28 A considerable amount of research has demonstrated that V(r) can serve as an important tool for predicting and interpreting the noncovalent interactions.21–26, V(r) at a point r is defined by the following equation in atomic unit:
V ( r ) = A Z A R A r ρ ( r ) d r r r ,
(2)
in which ZA located at RA is the charge on nucleus A and ρ(r′) represents the electron density distribution.29 The value of V(r) can either be positive or negative and is a net result of the interplay between the nuclear contribution and the electronic contribution. Hence, it must be careful to connect the V(r) value with only the local electron density. The positive V(r) value does not necessarily correspond to the electron-poor region, and the negative V(r) value does not necessarily correspond to the electron-rich region.30–32 By the same reason, the positive π-hole often does correspond to the electron-poor region, but not necessarily so.

There are many molecules containing the π-holes. Figure 5 summarizes some common π-hole molecules, and the geometries, energies, and nature of the π-holes in these molecules have been explored by many studies.22,23,25,26,33–35 The early examples of the π-holes were associated with the specific atoms. As shown in the first two rows of Fig. 5, the π-holes are above and below the specific atoms. Later the π-hole family has greatly expanded and at present contains more numbers, which are associated with two or more atoms. The last three rows of Fig. 5 show some common aromatic molecules having the multiple-atom π-holes. Evidently, the π-holes in these aromatic molecules are above and below the aromatic ring planes. It is reasonable that some molecules may have both the single-atom π-holes and multiple-atom π-holes. The 1,4-dinitrobenzene is such a molecule that has four π-holes above and below each NO2 nitrogen and two π-holes above and below the center of the benzene ring.26 The perfluoroarene molecules as well as their derivatives certainly have the π-holes. Figure 6 shows the molecular electrostatic potential maps of C6H3F3, C6H2F4, and C6F6 along with the positions of locally most positive electrostatic potentials. The locally most positive or most negative electrostatic potentials (VS,max or VS,min) on the molecular surfaces were always employed to locate the sites of noncovalent interactions.24 The molecules C6H3F3, C6H2F4, and C6F6 have four, three, and seven π-hole VS,max, respectively, above and below the benzene rings (Fig. 6). These π-hole VS,max are located in the craters on the molecular surfaces. Many molecules have more than one π-hole VS,max. All the π-hole VS,max in each molecule should be considered when investigating the π-hole bonding associated with this molecule.

FIG. 5.

The common π-hole molecules (Ch = chalcogen, EWG = electron-withdrawing group). The atoms shown in red are those with π-holes.

FIG. 5.

The common π-hole molecules (Ch = chalcogen, EWG = electron-withdrawing group). The atoms shown in red are those with π-holes.

Close modal
FIG. 6.

Top views of the molecular electrostatic potentials (kcal/mol) on the 0.001 a.u. molecular surfaces of C6H3F3 (a), C6H2F4 (b), and C6F6 (c). The small red circles represent the positions of locally most positive electrostatic potentials.

FIG. 6.

Top views of the molecular electrostatic potentials (kcal/mol) on the 0.001 a.u. molecular surfaces of C6H3F3 (a), C6H2F4 (b), and C6F6 (c). The small red circles represent the positions of locally most positive electrostatic potentials.

Close modal

The origin and nature of the π-hole are much more complicated than those of the σ-hole. This is because the σ-hole is often but not always on the outermost portion of the surface of a molecular entity, whereas the π-hole is always on the middle portion of the surface of a molecular entity and always associated with more than one atom in this molecular entity. Politzer et al. systematically explored the origin and nature of the π-hole.24,26,36,37 They concluded that much stronger electron-withdrawing substituents always lead to much more positive single-atom π-holes, and much more polarizable atoms always have much more positive single-atom π-holes.26 For some typical aromatic π-hole molecules such as hexafluorobenzene and 1,3,5-triazine, it had been thought that the positive π-holes on these molecules originate from the strong electron-withdrawing abilities of the fluorine atoms and ring nitrogen atoms.38–40 However, such a widespread assumption was challenged by Wheeler and co-workers.30,41,42 They stated with clear proofs that it is the through-space effect, rather than the decrease in the π-electron density, that leads to the formation of the positive π-hole on the molecular surface of hexafluorobenzene or 1,3,5-triazine. Figure 7 shows the comparisons of the electrostatic potentials and electron densities between benzene and 1,3,5-triazine. The electrostatic potentials along the Z axis starting from the center of mass of benzene are negative at the distances larger than 1.2 Å, whereas the electrostatic potentials of 1,3,5-triazine are all positive along the Z axis. This results in that some previous studies described benzene as π-electron-rich and 1,3,5-triazine as π-electron-deficient. In fact, comparing the electron densities of benzene with the electron densities of 1,3,5-triazine in Fig. 7, it can be found that the changes of the electron densities from benzene to 1,3,5-triazine are not very large at the distances larger than 1.2 Å, and the values of the electron densities of 1,3,5-triazine are even larger than the corresponding ones of benzene at the distances shorter than 1.2 Å. Obviously, the introduction of the three ring nitrogen atoms does not lead to the decrease in the π-electron densities. Using a series of electrostatic potential difference maps of 1,3,5-triazine relative to benzene, Wheeler and Bloom clearly demonstrated that it is the position of nuclear charges that leads to the positive electrostatic potentials above and below the ring plane of 1,3,5-triazine.42 The case for the hexafluorobenzene is almost the same,30 that is to say, the change of the electron density is a relatively minor cause of the positive electrostatic potentials on the molecular surface of hexafluorobenzene. This raises the question whether or not the regions above and below the centers of the rings of hexafluorobenzene and 1,3,5-triazine correspond to the π-holes. Politzer and co-workers answered this question and showed that these regions are indeed the regions of lower electron density and hence correspond to the π-holes.26 

FIG. 7.

The electrostatic potentials and electron densities along the Z axes starting from the centers of mass of benzene and 1,3,5-triazine.

FIG. 7.

The electrostatic potentials and electron densities along the Z axes starting from the centers of mass of benzene and 1,3,5-triazine.

Close modal

A π-hole with negative electrostatic potentials was called the negative π-hole. The studies of the negative π-holes are relatively rare, in comparison with those of the positive π-holes. Scheiner and co-workers reported a series of anions, which have the negative π-holes (Fig. 8).43–46 These anions include MX3 (M=Zn, Cd, Hg; X=Cl, Br, I), KrF5, XeF5, and XeCl5. There is another study, which stated that the halogenated allyl anions also have the negative π-holes.47 In fact, the regions above and below the allyl planes of these anions have much higher electron densities, and they are not of the π-hole.

FIG. 8.

Top (a) and side (b) views of the molecular electrostatic potentials (kcal/mol) on the 0.001 a.u. molecular surface of ZnBr3.

FIG. 8.

Top (a) and side (b) views of the molecular electrostatic potentials (kcal/mol) on the 0.001 a.u. molecular surface of ZnBr3.

Close modal

It can be seen from the above discussions that the π-hole does not necessarily correspond to a π-electron-deficient region, that is to say, the π-hole does not necessarily correspond to a flattening of the electron density in the π molecular orbital (MO) region. Therefore, it may not be appropriate to continue using the following terms in replace of the π-hole: π-acid, π-acidic ring, electron-deficient arene, π-deficient triazine ring, π-electron-poor area, etc.48–53 

2. The π-hole bonding

The π-hole bonding refers to the noncovalent interaction involving a π-hole. At present, the International Union of Pure and Applied Chemistry (IUPAC) has not recommended the definition of the π-hole bonding. The halogen bond and chalcogen bond have been defined by IUPAC in one continuous line or series.54,55 Hence, it should be reasonable to define the π-hole bonding as “the net attractive interaction between an electrophilic region associated with a π-hole in a molecular entity and a nucleophilic region in another, or the same, molecular entity.”

The well-known π-hole bonding examples are the so-called π-hole–anion interactions. Here, the term “π-hole–anion interaction” not the popular “anion–π interaction” is used because the anion–π interactions in which the π molecules do not have the π-holes are not the focus of this review article. There are many review articles discussed the π-hole–anion interactions directly or indirectly. The review article by Giese et al. in 2015 has focused on the π-hole–anion interactions involving fluoroarenes.56 Two recent review articles are those of Bianchi et al.57 and Wang et al.58 The π-hole is a region not a point. Hence, the binding sites of anions with π-hole molecules may not be limited to the centers of the π-hole rings. Figure 9 demonstrates the number of crystal structures for each binding motif between halide anion and pentafluorophenyl unit.59 The η3-type binding motif not the η6-type binding motif has the largest number of hits. Many factors lead to such a result. One of the most important factors is the effect of the other noncovalent interaction.58,60 This is quite different from the case in the gas phase, in which the η6-type binding structure is the most stable structure.61 In the gas phase, the large molecular clusters are hard to be formed, so the effect of the other noncovalent interaction can be neglected. The physical nature of the π-hole–anion interactions is quite clear now. For the π-hole molecules with large and positive permanent quadrupole moments, the binding strengths of the π-hole–anion interactions involving these molecules are determined mainly by the electrostatic energies, whereas for the π-hole molecules with negligible permanent quadrupole moments, the binding strengths of the π-hole–anion interactions involving these molecules are determined mainly by the polarization energies.62 Furthermore, Kim and co-workers carried out rigorous energy component analyses for a series of π-hole–anion interactions.63 They found that both the electrostatic and induction energies play dominant roles for the stabilities of the π-hole–anion interactions, and the contribution of the dispersion energy becomes much larger in the complex involving a much larger organic anion. In contrast to the conventional model for the π-hole–anion interaction in which the polarization of the aryl π system induced by the substituents plays a dominant role, Wheeler and Houk proposed a new additive model for the π-hole–anion interaction (Fig. 10). In this model, the π-hole–anion interactions originate mainly from the direct interactions between the anions and the local dipoles induced by the substituents, that is to say, the π-hole–anion interactions do not depend on the polarization of the aryl π system induced by the substituents. A similar explanation has been provided for the noncovalent interactions between π-hole azines and anions.44 

FIG. 9.

Structural analysis of the binding motifs between halide anions and pentafluorophenyl unit. The hapticity (η) refers to the number of the binding contacts between a halide anion and carbon atoms of the pentafluorophenyl unit. Reproduced with permission from Giese et al., Chem. Sci. 6, 354–359 (2015). Copyright 2015 Royal Society of Chemistry and Author(s), licensed under a Creative Commons Attribution License (CC BY).59 

FIG. 9.

Structural analysis of the binding motifs between halide anions and pentafluorophenyl unit. The hapticity (η) refers to the number of the binding contacts between a halide anion and carbon atoms of the pentafluorophenyl unit. Reproduced with permission from Giese et al., Chem. Sci. 6, 354–359 (2015). Copyright 2015 Royal Society of Chemistry and Author(s), licensed under a Creative Commons Attribution License (CC BY).59 

Close modal
FIG. 10.

Electrostatic model of through-space substituent effects in C6H6-nXn–Cl (X represents a substituent) complexes. Reproduced with permission from S. E. Wheeler and K. N. Houk, J. Phys. Chem. A 114, 8658–8664 (2010). Copyright 2010 American Chemical Society.41 

FIG. 10.

Electrostatic model of through-space substituent effects in C6H6-nXn–Cl (X represents a substituent) complexes. Reproduced with permission from S. E. Wheeler and K. N. Houk, J. Phys. Chem. A 114, 8658–8664 (2010). Copyright 2010 American Chemical Society.41 

Close modal

In the π-hole–anion interaction discussed above, the π-hole clearly refers to the positive π-hole not the negative π-hole. The noncovalent interactions between negative π-holes and anions do exist, but are rare. Scheiner and co-workers predicted computationally that the anions such as MX3 (M = Zn, Cd, Hg; X = Cl, Br, I) and AeY5 (Ae = Kr, Xe; Y = F, Cl) can have the negative π-holes, and can also bind the anions such as F, Cl, and CN.43–46 The negative π-holes on MX3and AeY5 are located at the regions that are perpendicular to the central portions of these planar anions (see Fig. 8 as an example). Figure 11 shows the structures of three representative complexes KrF5–NC, XeCl5–Cl, and XeF5–F which are bound by the π-hole–anion interactions involving the negative π-holes. Obviously, there are strong electrostatic repulsion interactions between the two anions in these complexes, which will lead to the instabilities of these anion–anion complexes in the gas phase. However, these anion–anion complexes may be stable in polar media. In polar aqueous solution, the binding energies of MX3–CN are in the range of 11–18 kcal/mol; the binding energies of stacked MX3–MX3 are less than 10 kcal/mol; the binding energies of KrF5–NC, XeCl5–Cl, and XeF5–F are no more than 2 kcal/mol.43–46 The authors of these theoretical articles believed that these complexes bound by the noncovalent interactions between negative π-holes and anions could be observed experimentally. Unfortunately, there are still no such experimental reports until now.

FIG. 11.

The structures of the complexes KrF5–NC (a), XeCl5–Cl (b), and XeF5–F (c) bound by the π-hole–anion interactions. Reproduced with permission from Grabarz et al., Molecules 26, 2116 (2021). Copyright 2021 Author(s), licensed under a Creative Commons Attribution License (CC BY).46 

FIG. 11.

The structures of the complexes KrF5–NC (a), XeCl5–Cl (b), and XeF5–F (c) bound by the π-hole–anion interactions. Reproduced with permission from Grabarz et al., Molecules 26, 2116 (2021). Copyright 2021 Author(s), licensed under a Creative Commons Attribution License (CC BY).46 

Close modal

If the anions in the complexes bound by the π-hole–anion interactions are replaced by the neutral molecules with the lone-pair electrons (LP) or π-electrons, the terms “π-hole–LP interactions” and “π-hole⋯π interactions” have always been used to describe these noncovalent interactions, respectively. Certainly, both the π-hole–LP interactions and π-hole⋯π interactions are also of the π-hole bonding. The π-hole–LP interactions were more frequently found in biological systems.64–67 Strictly speaking, the π-hole–anion interactions are the subset of the π-hole–LP interactions.66,67 The perfluoroaryl⋯aryl interactions are of the π-hole⋯π interactions. As will be demonstrated in this review article, the perfluoroaryl⋯aryl interactions are the most important subset of the π-hole⋯π interactions. We will discuss in detail the origin and nature of the perfluoroaryl⋯aryl interactions in Sec. II.

A perfluoro-compound or perfluorinated compound always refers to the organofluorine compound lacking C−H bonds. Hence, all of the perfluoroarenes or perfluoroaryl groups addressed in this review article do not contain the C−H bonds. The general aspects of the structures, energies, nature, and applications of perfluoroaryl⋯aryl interactions are presented in this review article. It is important to emphasize that we have not deliberately overlooked studies on perfluoroaryl⋯aryl interactions in the gas and liquid phases. Compared to the extensive research on perfluoroaryl⋯aryl interactions in the solid phase, studies on perfluoroaryl⋯aryl interactions in the gas and liquid phases are indeed limited.68,69 The perfluoroaryl⋯aryl interactions have been successfully applied in many different fields, ranging from chemistry and materials science to biology and medicine. Considering that the research on perfluoroaryl⋯aryl interactions in some application fields is not yet very systematic, in this review article, only the following application fields will be discussed: crystal engineering, liquid crystal materials, organic luminescent materials, perovskite and organic photovoltaics, photocatalysis, and biological engineering.

As mentioned above, besides the perfluoroarenes or perfluoroaryl groups, there are many other organic molecules containing the π-holes. Hence, the perfluoroaryl⋯aryl interactions are the subset of the π-hole⋯π interactions, which are again the subset of the π-hole bonding. The general theories of the π-hole⋯π interactions and even the π-hole bonding can be applied to the perfluoroaryl⋯aryl interactions. However, the perfluoroaryl⋯aryl interactions have their own characteristic nature, which deserves to be reviewed separately due to their importance in many fields.

It has been generally accepted that decomposition of the total interaction energy into physically or chemically meaningful energy components, which is known as the energy decomposition analysis (EDA), is the most efficient method to uncover the nature of the noncovalent interactions.70,71 The EDA methodology has been continuously developed and improved since the early works of Morokuma, Szalewicz and their co-workers.72,73 Until now, there are more than fifteen EDA schemes that have been proposed and applied.74 The variational and perturbation-based EDAs are currently popular, although there are also other types of EDA schemes such as the “Chemical Hamiltonian” method developed by Mayer and the “Experimental Quantum Chemistry” method developed by Rahm and Hoffmann.70,71,75,76 The symmetry-adapted perturbation theory (SAPT) developed by Jeziorski et al. is a perturbation-based and well-established EDA scheme that is usually referred to as the standard EDA model for the study of the noncovalent interactions.77,78 SAPT defines the total interaction energy ( E int SAPT) in terms of four physically meaningful energy components: electrostatics (Eelst), short-range exchange-repulsion (Eexch), induction (Eind), and dispersion (Edisp)
E int SAPT = E elst + E exch + E ind + E disp .
(3)

The SAPT scheme has some advantages over the other EDAs. First, SAPT calculates the total interaction energy directly via perturbation theory and therefore is free of basis set superposition error (BSSE). More importantly, SAPT avoids arbitrary definitions of the energy components, whereas most of the other EDAs have such inherent problems. The SAPT scheme has also some limitations that need to be overcome. One of the limitations is that the high-order SAPT calculations are far too expensive in both computational time and computational resources to be used for large noncovalent systems. Sherrill and co-workers have proved that the low-order SAPT method (SAPT0) in conjugation with a truncated Dunning aug-cc-pVDZ basis set (jun-cc-pVDZ) can provide reasonably accurate interaction energies.79–81 The low-cost SAPT0/jun-cc-pVDZ calculations with further application of the density fitting (DF) approximations make it possible to deal with large noncovalent systems with hundreds of atoms.82 Comparing with the accuracy and cost of the SAPT0/jun-cc-pVDZ calculation, Wang et al. found that the spin component scaled (SCS) SAPT0 scheme with standard Dunning basis set aug-cc-pVDZ (aVDZ) performs much better for the study of the nature of the π–π stacking interaction.83,84 Therefore, most of the EDA results in this section were obtained from the SCS-SAPT0/aVDZ calculations. Here, a more comprehensive description of the SAPT scheme is obviously unnecessary. More details for the SAPT methodology and development can be found in the excellent review articles by Szalewicz, Jansen, and Patkowski.77,78,85,86

The complex between hexafluorobenzene and benzene (C6F6–C6H6) is the simplest model for the study of the nature of the perfluoroaryl⋯aryl interaction. As in the case of the benzene dimer,87 the sandwich, T-shaped, and parallel-displaced configurations were always selected as prototypes in the gas-phase calculations of the complex C6F6–C6H6.88 In this review article, we only consider the sandwich and parallel-displaced configurations of C6F6–C6H6 because the T-shaped configurations are not related to the perfluoroaryl⋯aryl interactions. Figure 12 shows the S-1, S-2, and PD configurations of C6F6–C6H6 optimized at the PBE0-D3/def2-TZVPP theory level.89 PBE0-D3 is a dispersion-corrected hybrid functional, which performs excellently for the π-stacked complexes.89 The interaction energies of the three configurations were calculated to be −5.56, −5.55, and −5.95 kcal/mol, respectively, at the coupled-cluster level of theory with singles, doubles, and perturbative triples [CCSD(T)] in the complete basis set (CBS) limit. The interaction energy of the S-1 configuration is almost the same as that of the S-2 configuration, which indicates that the C6F6 or C6H6 molecule can rotate freely around the C6 axis of the complex. The PD configuration has the most negative interaction energy of −5.95 kcal/mol. The interaction energy differences of the three configurations are less than or equal to 0.4 kcal/mol. This means that the potential energy surfaces are very flat along the sliding coordinates. Tsuzuki and co-workers investigated the noncovalent interactions between C6F6 and C6H6 by using the quantum chemical ab initio calculations.88 The estimated CCSD(T)/CBS interaction energies for the S-1 and PD configurations of C6F6–C6H6 were found to be −5.07 and −5.38 kcal/mol, respectively, which are close to the corresponding values discussed above. The nature of the perfluoroaryl⋯aryl interactions in C6F6–C6H6 was further analyzed via the EDA calculations. In the EDA scheme employed by Tsuzuki and co-workers, the electrostatic and induction energies were calculated with the method of distributed multipole analysis;90 the dispersion energy was calculated simply as the difference between the CCSD(T)/CBS interaction energy and the corresponding Hartree–Fock (HF) interaction energy; the exchange-repulsion energy was calculated as the difference between the HF interaction energy and the sum of the electrostatic and induction energies. Their EDA results showed that the dominant attractive component of the perfluoroaryl⋯aryl interaction in C6F6–C6H6 is the dispersion term, and the directionality of the perfluoroaryl⋯aryl interaction is determined by both the dispersion and electrostatic terms. Wang et al. carried out energy decomposition analyses for the S-1, S-2, and PD configurations of C6F6–C6H6 at the SCS-SAPT0/aVDZ level of theory.89 As shown in Table I, the attractive dispersion interaction plays a dominant role for the stabilization of each of the three configurations of C6F6–C6H6, and there is also a substantial electrostatic contribution (37%) to the total intermolecular attractive interaction. In comparison, the induction energy only accounts for about 5% of the total attractive interaction energy. These results are in agreement with previous EDA calculations on the complex C6F6–C6H6.88,91–93 The EDA results for the complexes C6Cl6–C6H6 and C6Br6–C6H6 are listed in Table I for the purpose of comparison. From C6F6–C6H6 to C6Cl6–C6H6 and C6Br6–C6H6, the proportion of electrostatic component in total attractive interaction decreases, while the proportion of dispersion interaction in total attractive interaction increases. This is reasonable due to the smaller size and stronger electron-withdrawing ability of the F atom as compared to the Cl and Br atoms. Such a comparison throws light on the main characteristics of the perfluoroaryl⋯aryl interactions.

FIG. 12.

Top (left) and side (right) views of the sandwich-eclipsed (S-1), sandwich-staggered (S-2), and parallel-displaced (PD) configurations of the complex C6F6–C6H6. The blue numbers are the interaction energies in kcal/mol. Color code: carbon, gray; hydrogen, white; fluorine, light blue.

FIG. 12.

Top (left) and side (right) views of the sandwich-eclipsed (S-1), sandwich-staggered (S-2), and parallel-displaced (PD) configurations of the complex C6F6–C6H6. The blue numbers are the interaction energies in kcal/mol. Color code: carbon, gray; hydrogen, white; fluorine, light blue.

Close modal
TABLE I.

The EDA results at the SCS-SAPT0/aVDZ level for the three configurations of C6F6–C6H6. The EDA results of C6Cl6–C6H6 and C6Br6–C6H6 are listed only for the purpose of comparison. The unit for all energy terms is kcal/mol.a

Complex Configuration Eelst Eexch Eind Edisp Eelst%b Eind%b Edisp%b
C6F6–C6H6  S-1  −4.48  6.55  −0.61  −6.94  37  58 
S-2  −4.46  6.52  −0.60  −6.94  37  58 
PD  −4.99  7.68  −0.87  −7.64  37  57 
C6Cl6–C6H6  S-1  −5.38  9.44  −0.73  −11.23  31  65 
S-2  −5.51  9.60  −0.70  −11.50  31  65 
PD  −5.67  10.24  −0.92  −11.70  31  64 
C6Br6–C6H6  S-1  −5.63  10.22  −0.79  −12.25  30  66 
S-2  −5.80  10.41  −0.75  −12.64  30  66 
PD  −5.93  11.00  −0.98  −12.70  30  65 
Complex Configuration Eelst Eexch Eind Edisp Eelst%b Eind%b Edisp%b
C6F6–C6H6  S-1  −4.48  6.55  −0.61  −6.94  37  58 
S-2  −4.46  6.52  −0.60  −6.94  37  58 
PD  −4.99  7.68  −0.87  −7.64  37  57 
C6Cl6–C6H6  S-1  −5.38  9.44  −0.73  −11.23  31  65 
S-2  −5.51  9.60  −0.70  −11.50  31  65 
PD  −5.67  10.24  −0.92  −11.70  31  64 
C6Br6–C6H6  S-1  −5.63  10.22  −0.79  −12.25  30  66 
S-2  −5.80  10.41  −0.75  −12.64  30  66 
PD  −5.93  11.00  −0.98  −12.70  30  65 
a

All data are taken from Ref. 89.

b

Contribution to the total attractive interactions.

In a seminal work in 1990, Hunter and Sanders proposed a simple model based on the quadrupole moments of aromatic molecules to explain the geometrical preferences of the ππ interactions.94 By using this model, the configuration preferences of the benzene dimer were successfully explained. The T-shaped and parallel-displaced configurations of the benzene dimer are more stable than the sandwich configuration of the benzene dimer because the quadrupole–quadrupole interaction is attractive in the T-shaped or parallel-displaced configuration but repulsive in the sandwich configuration. The quadrupole moment of monomer C6F6 has the opposite sign to that of monomer C6H6, and therefore, the quadrupole–quadrupole electrostatic interactions are attractive in the cofacial configurations of the complex C6F6–C6H6. According to the Hunter–Sanders model, the sandwich configuration of C6F6–C6H6 should be more stable than the parallel-displaced configuration. However, such a deduction is not supported by above CCSD(T)/CBS interaction energies for the S-1, S-2, and PD configurations of C6F6–C6H6. Herbert and co-workers resolved this contradiction by employing an “extended” SAPT scheme incorporated with the many-body dispersion effect (XSAPT + MBD).95–97 In order to rationalize the configuration preferences of C6F6–C6H6, they used a “van der Waals (vdW) model” to replace the quadrupolar electrostatics model of Hunter and Sanders.94 In this vdW model, the vdW interaction potential was defined as the sum of Pauli repulsion and dispersion and found to be the driving force toward the parallel-displaced configuration of C6F6–C6H6. Figure 13 illustrates the potential energy curves along the sliding coordinates for C6F6–C6H6. The total XSAPT + MBD interaction energy is the sum of vdW and electrostatics + induction components. At the XSAPT + MBD theory level, the parallel-displaced configuration is a local minimum and the sandwich configuration is a saddle point, which is consistent with the CCSD(T)/CBS results. The electrostatics + induction potential curve incorrectly shows a clear preference for the sandwich configuration of C6F6–C6H6, as does the Hunter–Sanders model. On the contrary, the vdW model correctly predicts the saddle point corresponding to the sandwich configuration and explains the preference for the parallel-displaced configuration of C6F6–C6H6. Fink et al. also emphasized the important role of the exchange-repulsion component in determining the structures of aggregates including the complex C6F6–C6H6.98 Gung and Amicangelo investigated computationally the substituent effects on the perfluoroaryl⋯aryl interactions in C6F6–C6H5X (X = CN, F, H, CH3, NH2, N(CH3)2).99 They pointed out that the attractive “charge-transfer” interaction contributes significantly to the unusually strong binding energy of the complex formed by C6F6 with N,N-dimethylaniline. Although they did not perform the energy decomposition analysis for this complex, the experimental results supported their hypothesis.100,101 In the SAPT formalism, the charge-transfer energy is covered by the induction term. Therefore, the study of Gung and Amicangelo indicates that the induction energy can also be an indispensable factor in the study of the perfluoroaryl⋯aryl interaction.

FIG. 13.

The potential energy curves along the sliding coordinates. Atom colors: carbon, gray; hydrogen, white; fluorine, sky blue. Reproduced with permission from K. Carter-Fenk and J. M. Herbert, Chem. Sci. 11, 6758–6765 (2020). Copyright 2020 Royal Society of Chemistry and Author(s), licensed under a Creative Commons Attribution License (CC BY).95 

FIG. 13.

The potential energy curves along the sliding coordinates. Atom colors: carbon, gray; hydrogen, white; fluorine, sky blue. Reproduced with permission from K. Carter-Fenk and J. M. Herbert, Chem. Sci. 11, 6758–6765 (2020). Copyright 2020 Royal Society of Chemistry and Author(s), licensed under a Creative Commons Attribution License (CC BY).95 

Close modal

Jin et al. synthesized a series of luminescent cocrystals formed by haloperfluorobenzenes with pyrene and fluoranthene, respectively, and investigated their structures and properties.102,103 The single-crystal x-ray diffraction analyses revealed that the perfluoroaryl⋯aryl interactions occur in all these cocrystals. Figure 14 shows the perfluoroaryl⋯aryl interactions in the phosphorescent cocrystals formed by pyrene (Pyr) with hexafluorobenzene (HFB), chloroperfluorobenzene (Cl-5FB), bromoperfluorobenzene (Br-5FB), 1,2-dibromotetrafluorobenzene (1,2-DBrTFB), 1,3-dibromotetrafluorobenzene (1,3-DBrTFB), and 1,4-dibromotetrafluorobenzene (1,4-DBrTFB), respectively. Table II summarizes the EDA results for the six complexes occurred in their corresponding cocrystal structures. The perfluoroaryl⋯aryl interactions in the six complexes are determined mainly by the dispersion and electrostatic terms, as is the case in the complex C6F6–C6H6. The induction energy accounts for only 6%–7% of the total attractive interaction energy. For the six complexes, with the increase in the surface area of haloperfluorobenzene, the dispersion contribution becomes larger and correspondingly the contribution of the electrostatic term becomes smaller. It is interesting to know what happens to each energy component of the perfluoroaryl⋯aryl interaction when the surface area of aryl/arene becomes larger and larger. Using coronene 43 (C110H26) as a model of graphene, Wang and co-workers studied the structure, energy, and nature of the noncovalent interaction between C6F6 and graphene.83,104 The EDA results for the PD configuration of the complex C6F6–C110H26 are listed in Table III, together with the EDA results for the PD configurations of C6F6–C6H6 and C6F6–C16H10 (Pyrene). For the three complexes, with the increase in the surface area of arene, the dispersion contribution also becomes larger, and correspondingly, the contribution of the electrostatic term also becomes smaller. The induction energy contributes only 5%–7% to the total attractive interaction energy. Although the electrostatic contribution becomes smaller, it is still far greater than the induction contribution.

FIG. 14.

The perfluoroaryl⋯aryl interactions in the cocrystal structures of (a) Pyr·HFB; (b) Pyr·Cl-5FB; (c) Pyr·Br-5FB; (d) Pyr·1,2-DBrTFB; (e) Pyr·1,3-DBrTFB; and (f) Pyr·1,4-DBrTFB, respectively. Reproduced with permission from Pang et al., Cryst. Growth Des. 15, 4938–4945 (2015). Copyright 2015 American Chemical Society.102 

FIG. 14.

The perfluoroaryl⋯aryl interactions in the cocrystal structures of (a) Pyr·HFB; (b) Pyr·Cl-5FB; (c) Pyr·Br-5FB; (d) Pyr·1,2-DBrTFB; (e) Pyr·1,3-DBrTFB; and (f) Pyr·1,4-DBrTFB, respectively. Reproduced with permission from Pang et al., Cryst. Growth Des. 15, 4938–4945 (2015). Copyright 2015 American Chemical Society.102 

Close modal
TABLE II.

The EDA results at the SCS-SAPT0/aVDZ level for the six complexes considered. The unit for all energy terms is kcal/mol.a

Complex Eelst Eexch Eind Edisp Eelst%b Eind%b Edisp%b
PyrHFB  −8.48  16.12  −1.80  −17.02  31  62 
PyrCl-5FB  −8.45  16.54  −1.81  −18.07  30  64 
PyrBr-5FB  −8.25  16.16  −1.74  −18.13  29  65 
Pyr1,2-DBrTFB  −8.77  17.63  −1.87  −19.69  29  65 
Pyr1,3-DBrTFB  −8.50  17.00  −1.80  −19.54  28  66 
Pyr1,4-DBrTFB  −8.43  17.40  −1.80  −19.89  28  66 
Complex Eelst Eexch Eind Edisp Eelst%b Eind%b Edisp%b
PyrHFB  −8.48  16.12  −1.80  −17.02  31  62 
PyrCl-5FB  −8.45  16.54  −1.81  −18.07  30  64 
PyrBr-5FB  −8.25  16.16  −1.74  −18.13  29  65 
Pyr1,2-DBrTFB  −8.77  17.63  −1.87  −19.69  29  65 
Pyr1,3-DBrTFB  −8.50  17.00  −1.80  −19.54  28  66 
Pyr1,4-DBrTFB  −8.43  17.40  −1.80  −19.89  28  66 
a

Data from Ref. 102.

b

Contribution to the total attractive interactions.

TABLE III.

The EDA results at the SCS-SAPT0/aVDZ level for the three complexes containing C6F6. The unit for all energy terms is kcal/mol.a

Complex Eelst Eexch Eind Edisp Eelst%b Eind%b Edisp%b
C6F6–C6H6  −4.99  7.68  −0.87  −7.64  37  57 
C6F6–C16H10  −8.48  16.12  −1.80  −17.02  31  62 
C6F6–C110H26  −5.78  15.18  −1.59  −23.03  19  76 
Complex Eelst Eexch Eind Edisp Eelst%b Eind%b Edisp%b
C6F6–C6H6  −4.99  7.68  −0.87  −7.64  37  57 
C6F6–C16H10  −8.48  16.12  −1.80  −17.02  31  62 
C6F6–C110H26  −5.78  15.18  −1.59  −23.03  19  76 
a

Data from Refs. 83, 89, and 102.

b

Contribution to the total attractive interactions.

In summary, the SAPT analysis provides physical insights into the nature of the perfluoroaryl⋯aryl interactions. All of the four energy components—electrostatics, exchange-repulsion, induction, and dispersion—should be considered in the study of the perfluoroaryl⋯aryl interactions. In most cases, the dispersion energies play dominant roles for the structures and stabilization of the perfluoroaryl⋯aryl interactions; the electrostatic energies play secondary roles and the induction energies play minor roles. Some studies showed that the induction energies may play key roles for the structures and stabilization of the so-called charge-transfer complexes bound by the perfluoroaryl⋯aryl interactions. Unfortunately, reliable EDA results for these complexes are rather limited.

In recent years, Politzer, Murray, and Clark reemphasized the distinction between mathematical model and physical reality in the study of the noncovalent interactions.36,105–108 In fact, there has been some dispute as to whether or not the mathematical model or the physical reality is the more fundamental in the chemical research.109,110 Bader stated that “those who use some form of energy partitioning analysis (EPA) in studies of chemical bonding arrive at conclusions that are difficult to criticize because they lie beyond the boundaries of physics.”109 Frenking et al. rejected Bader's declaration and wrote that “chemical research begins where the physics of Richard Bader ends.”110 We believe that such a dispute will never end. Without any doubt, the coexistence of various types of theoretical methods either based on the mathematical models or based on the physical observables will be more capable of promoting the development of chemical research.

Agreeing with Bader's point of view, Politzer, Murray, and Clark also stated that the energy components in EDA procedures are intrinsically inseparable and the arbitrary and artificial EDA procedures cannot give physical predictions.105,108 According to the Hellmann–Feynman theorem,111,112 the forces on the nuclei in a molecule or complex are purely electrostatic forces. Therefore, Politzer, Murray, and Clark proposed to use the Coulombic interactions to describe the noncovalent interactions.36,105–108 Their main points of view are as follows: (1) The Coulombic nature of the noncovalent interactions is supported by the Hellmann–Feynman theorem. (2) The σ-hole bonding and π-hole bonding can be properly described by the Coulombic interactions, which encompass both electrostatics and polarization. (3) Charge transfer is equivalent to polarization. (4) Dispersion is simply part of polarization. (5) Pauli or exchange repulsion is part of the mathematical model and should not be included in analyzing Coulombic interactions. Some studies questioned the electrostatic interpretations of noncovalent interactions.113–115 Clark et al. pointed out that polarization is an inherent part of electrostatic interaction and must be considered in a correct electrostatic treatment of noncovalent interaction.36,116 Politzer et al. plotted the molecular electrostatic potential maps for a series of hydrocarbons and perfluorocarbons.117 Using the molecular electrostatic potentials, they successfully explained why perfluorobenzene or perfluoronaphthalene can form infinite stacks alternating with aromatic hydrocarbons.117–119 However, it was also found by them that the electrostatic potentials alone cannot well explain why linear perfluoroalkanes do not mix easily with linear alkanes and the polarizabilities of linear perfluoroalkanes and alkanes must be considered together.117 

In several studies, Politzer, Murray, and Clark tried to calculate the interaction energies of σ-hole and π-hole complexes according to the regression equation based on electrostatics and polarization.26,32,120,121 For the π-hole complexes in which HCN and NH3 act as the electron donors, they found that the following equation could be used to calculated the interaction energies:
Δ E ( int ) = c 1 [ V S , max ] + c 2 [ ε ( R ) ] 2 + c 3 [ α ] + c 4 ,
(4)
in which VS,max is the most positive surface electrostatic potential of the π-hole, ɛ(R) is the electric field that the π-hole molecule exerts on the negative site, and α is the average polarizability of HCN or NH3. Figure 15 shows the correlation between the ΔE(int) predicted with Eq. (4) and the ΔE(int) calculated with the conventional supermolecule method. The squared correlation coefficient R2 is 0.952, which means that the correlation is pretty good and further indicates that the interaction energies of the π-hole complexes can be reasonably expressed by Eq. (4). Later, Politzer and Murray expanded the database of 21 π-hole complexes to a larger database of 33 π-hole complexes.32 Accordingly, they modified their regression model and also introduced the most negative surface electrostatic potential of the negative site (VS,min). Finally, it was found that the following equation can used to predict the interaction energies of the π-hole complexes:
Δ E ( int ) = c 1 [ ε ( R ) ] 2 + c 2 [ α ] + c 3 [ V S , min ] + c 4 .
(5)
For these 33 π-hole complexes, the correlation between the ΔE(int) predicted with Eq. (5) and the ΔE(int) calculated with the quantum chemical method has an R2 of 0.952. If the complex F2Si=O–NCH is excluded from the database, the value of R2 increases to 0.974. Evidently, the Eq. (5) performs very well for predicting the binding strengths of the π-hole complexes. The variables in the two regression equations Eqs. (4) and (5) are the components of either electrostatics or polarization. The good performance of both Eqs. (4) and (5) reflects the Coulombic nature of the π-hole bonding.
FIG. 15.

Correlation of predicted ΔE(int) with computed ΔE(int). ΔE(int) is in kcal/mol. Reproduced with permission from Politzer et al., J. Phys. Chem. A 123, 10123–10130 (2019). Copyright 2019 American Chemical Society.120 

FIG. 15.

Correlation of predicted ΔE(int) with computed ΔE(int). ΔE(int) is in kcal/mol. Reproduced with permission from Politzer et al., J. Phys. Chem. A 123, 10123–10130 (2019). Copyright 2019 American Chemical Society.120 

Close modal

The perfluoroaryl⋯aryl interactions were not included in the database of 33 π-hole complexes built by Politzer and Murray. However, in principle, the Eq. (4) or Eq. (5) can also be applied to predict the interaction energies of the perfluoroaryl⋯aryl interactions because the perfluoroaryl⋯aryl interactions are of the type of π-hole bonding. It needs to be specifically pointed out that, for the calculations of the interaction energies of the π-hole complexes with Eq. (4) or Eq. (5), the greatest difficulty lies in how to obtain the accurate values of VS,max, VS,min, ɛ(R), and α.

The study on the nature of the perfluoroaryl⋯aryl interaction can be simplified as the study on the effects of F-substituents in the aryl⋯aryl interaction. Wheeler et al. proposed a local, direct interaction model to explain the substituent effects in the π-stacking interactions.122–126 In this new model, the substituent effect on the π-stacking interaction mainly depends on the direct electrostatic interaction between the substituent and the proximal vertex of the unsubstituted ring (Fig. 16).122 In a subsequent study, Wheeler and co-workers reported the broad transferability of substituent effects among diverse π-stacked dimers and emphasized the importance of the electric fields of the unsubstituted rings for explaining these transferable phenomena.123 The local, direct interaction model was correspondingly modified to the one, which is based on the electrostatic interaction between the local dipole moment associated with the substituent and the electric field of the unsubstituted ring.124 According to the conventional view, the electron-withdrawing substituents will decrease the electron density above and below an aryl ring and strengthen the aryl⋯aryl stacking interaction, while the electron-donating substituents will increase the electron density above and below an aryl ring and weaken the aryl⋯aryl stacking interaction. However, the local, direct interaction model of Wheeler et al. is independent of the aryl π-electron density and focuses only on the local environment of the substituent. The dimers formed by substituted benzenes (C6H5X) with 1,2,3-trifluorobenzene (C6H3F3) and C6F6, respectively, provide a good example of using the local, direct interaction model to explain the substituent effects in the π-stacking interactions. Figure 17 shows the sandwich structures of the dimers C6H5X–C6H3F3 and C6H5X–C6F6 and the correlation between E int rel(C6H5X–C6H3F3) and E int rel(C6H5X–C6F6). The E int rel(C6H5X–C6H3F3) is the C6H5X–C6H3F3 interaction energy relative to the C6H6–C6H3F3 interaction energy, and the E int rel(C6H5X–C6F6) is the C6H5X–C6F6 interaction energy relative to the C6H6–C6F6 interaction energy. The correlation coefficient (r) in Fig. 17 is 0.97, which shows a strong linear correlation between the relative interaction energies of C6H5X–C6H3F3 and C6H5X–C6F6. The strong linear correlation between the relative interaction energies means the strong linear correlation between the substituent effects. That is, changing the unsubstituted ring from C6H3F3 to C6F6 does not alter the substituent effects. This phenomenon cannot be explained by the molecular quadrupole moments or π-resonance effects. It can be clearly seen from Fig. 17 that the local environments of the substituents in C6H5X–C6H3F3 and C6H5X–C6F6 are the same. Hence, the local, direct interaction model explains well the strong linear correlation between the substituent effects in C6H5X–C6H3F3 and C6H5X–C6F6. Wheeler also found that the interaction energy of the 1,3,5-trifluorobenzene dimer is –5.9 kcal/mol, which is very close to the C6H6–C6F6 interaction energy with a value of –6.3 kcal/mol.125 Such a calculated result is in agreement with the prediction of the local, direct interaction model.

FIG. 16.

The local, direct interaction model of Wheeler. Reproduced with permission from S. E. Wheeler, J. Am. Chem. Soc. 133, 10262–10274 (2011). Copyright 2011 American Chemical Society and Author(s), licensed under an ACS AuthorChoice License.122 

FIG. 16.

The local, direct interaction model of Wheeler. Reproduced with permission from S. E. Wheeler, J. Am. Chem. Soc. 133, 10262–10274 (2011). Copyright 2011 American Chemical Society and Author(s), licensed under an ACS AuthorChoice License.122 

Close modal
FIG. 17.

Correlation between E int rel(C6H5X–C6H3F3) and E int rel(C6H5X–C6F6). X denotes the substituent. Reproduced with permission from S. E. Wheeler, J. Am. Chem. Soc. 133, 10262–10274 (2011). Copyright 2011 American Chemical Society and Author(s), licensed under an ACS AuthorChoice License.122 

FIG. 17.

Correlation between E int rel(C6H5X–C6H3F3) and E int rel(C6H5X–C6F6). X denotes the substituent. Reproduced with permission from S. E. Wheeler, J. Am. Chem. Soc. 133, 10262–10274 (2011). Copyright 2011 American Chemical Society and Author(s), licensed under an ACS AuthorChoice License.122 

Close modal

The success of the local, direct interaction model in explaining the substituent effects on the π-stacking interactions provides new insight into the nature of the perfluoroaryl⋯aryl interaction. Analyses and explanations of the perfluoroaryl⋯aryl interactions become very simple if the local, direct interaction model is applied. We just need to consider the electrostatic interaction between the local dipole associated with the C–F bond of perfluoroarene and the local dipole associated with the C–H bond of arene because such a local electrostatic interaction plays a dominant role for determining the structure and stabilization of the whole perfluoroaryl⋯aryl interaction. This makes it possible to study the perfluoroaryl⋯aryl interactions between large systems. By contrast, the popular quadrupole-based models and π-polarization-based models fail to rationalize the nature of the perfluoroaryl⋯aryl interaction.

The molecular electrostatic potential maps are extensively applied in the study of the noncovalent interactions. Before discussing the theoretical basis of the local, direct interaction model, it would be helpful to clarify once again the substituent effects on the molecular electrostatic potentials above or below the ring planes of arenes. Here, we only focus on the effects of the F substituents. According to the definition of the molecular electrostatic potential [see Eq. (2)], both the nuclear charge distribution and the electron density distribution should be considered when analyzing and interpreting the molecular electrostatic potentials. As mentioned earlier in the Introduction, the through-space effects of the F substituents not the π-electron densities of the aromatic rings play dominant roles for the formations of the positive electrostatic potentials on the molecular surfaces of perfluoroarenes. Wheeler and Houk employed an additive electrostatic potential model to assess the contributions of π-resonance and inductive/field effects on the electrostatic potentials of perfluoroarenes.30, Figure 18 demonstrates the real and additive electrostatic potential maps for C6F6 and 1,3,5-tris(perfluorophenethynyl)benzene. The additive electrostatic potentials of C6F6 and 1,3,5-tris(perfluorophenethynyl)benzene were calculated by simply adding the electrostatic potentials of hydrogen capped-substituents to the electrostatic potentials of C6H6 and 1,3,5-trisphenethynylbenzene, respectively. In Fig. 18, the additive electrostatic potential maps are very similar to the real electrostatic potential maps, although the σ- and π-electron distributions of the substituted and unsubstituted rings have no any changes upon the calculations of the additive electrostatic potentials. This indicates that, upon F substitutions, the changes of the electrostatic potentials above and below the aromatic rings do not arise from the changes of the π-electron densities, but are the result of the through-space effects of the local dipoles associated with the F substituents. These calculated results reveal the local nature of the substituent effects in the π-stacking interactions and lay a solid foundation for the local, direct interaction model of Wheeler and co-workers. On the other hand, some EDA studies on the substituted benzene dimers also support the electrostatic basis of the local, direct interaction model.127–129 It has been accepted that the dispersion energy is the dominant attractive component in the π-stacking interactions. However, the important impacts of the dispersion interactions on different substituted benzene dimers are almost canceled by the accompanying changes of the exchange-repulsion and induction interactions. This leads to that the substituent effects on the π-stacking interactions are actually governed by the electrostatic interactions, which is consistent with the theoretical basis of the local, direct interaction model. Sherrill and Parrish evaluated the reliability of the local, direct interaction model in situ by using the functional-group partition of symmetry-adapted perturbation theory.130 Their calculated results provided direct evidence for the correctness of the local, direct interaction model.

FIG. 18.

The real (left) and additive (right) electrostatic potential maps for C6F6 and 1,3,5-tris(perfluorophenethynyl)benzene. Reproduced with permission from S. E. Wheeler and K. N. Houk, J. Chem. Theory Comput. 5, 2301–2312 (2009). Copyright 2009 American Chemical Society.30 

FIG. 18.

The real (left) and additive (right) electrostatic potential maps for C6F6 and 1,3,5-tris(perfluorophenethynyl)benzene. Reproduced with permission from S. E. Wheeler and K. N. Houk, J. Chem. Theory Comput. 5, 2301–2312 (2009). Copyright 2009 American Chemical Society.30 

Close modal

Hulliger et al. reviewed the role of the F atom in crystal engineering about twenty years ago.4 They categorized the F interactions into the following four types: C−H⋯F, C−F⋯π(perfluoroaryl ring), F⋯F, and perfluoroaryl⋯aryl interactions. In crystal engineering, only the perfluoroaryl⋯aryl interaction is increasingly showing its importance because the C−H⋯F, C−F⋯π(perfluoroaryl ring), and F⋯F interactions are relatively very weak.131–153,155–165 Especially, the crystals assembled either primarily or secondarily by the perfluoroaryl⋯aryl interactions have certain advantages in the design of functional materials with attractive properties and applications due to the planar structures of the perfluoroaryl and aryl rings.

From 1971 to 1988, Dahl studied and reported a series of crystal structures of the complexes formed by hexafluorobenzene with aromatic compounds.131,132 Later, Marder and his collaborators synthesized and characterized more cocrystals of perfluoroarenes with arenes or their derivatives.118,119,133–144 Marder and his collaborators' systematic research significantly enriched and expanded the application of perfluoroaryl⋯aryl interactions in crystal engineering.

The perfluoroaryl⋯aryl interactions combined with the hydrogen bonds are often utilized to design and grow crystals. Note that the hydrogen bonds are undoubtedly the most common noncovalent interactions in the field of crystal engineering.145 The interplay between perfluoroaryl⋯aryl interaction and hydrogen bond in controlling the cocrystal structures has been investigated by several studies.146–151 As a π-hole molecule, pentafluorophenol (pfp) displays the ability to control the microscopic stacking pattern of crystals through the perfluoroaryl⋯aryl interactions, just as the perfluorobenzene. In addition, the hydrogen bonds between the hydroxyl groups at the molecular edge of pfp may also be another factor regulating the crystal microstructure. The powder neutron diffraction analysis at 4 K reveals dense packing of pfp and phenol (p) in 1:1 ratio in the crystal state (Fig. 19).146 The pfp and p molecules are stacked alternately to form a one-dimensional columnar structure through the perfluoroaryl⋯aryl interactions. The columnar structures are connected together through the O−H⋯O hydrogen bonds between the OH groups of pfp and p to further construct the cocrystal. In the crystal structure of the cocrystal between pfp and 1,3,5-tris-(phenylethynyl)benzene (TPEB), three pfp molecules construct a trimer (pfp3) through three intermolecular O−H⋯O hydrogen bonds and three weak F⋯F interactions, and each TPEB molecule is surrounded by three pfp3 trimers (Fig. 20).147 TPEB and pfp3 trimers are connected together through a large number of weak C–H⋯F hydrogen bonds, forming a two-dimensional network. Then, the three-dimensional structure is constructed by the perfluoroaryl⋯aryl interaction of each pfp ring with the benzene ring and acetylene group of TPEB between the adjacent layers. In this work, pfp displays the flexibility in crystal engineering and the ability to induce the corresponding triaryl molecules to stabilize in planar configuration.

FIG. 19.

Molecular packing modes in the cocrystal pfp•p viewed along the a-axis. Color scheme: white, H; gray, C; red, O; light-green, F. The H atoms of the hydroxyl groups are disordered.

FIG. 19.

Molecular packing modes in the cocrystal pfp•p viewed along the a-axis. Color scheme: white, H; gray, C; red, O; light-green, F. The H atoms of the hydroxyl groups are disordered.

Close modal
FIG. 20.

The noncovalent interactions (blue dashed lines) in one layer of the cocrystal pfp•TPEB (a). The perfluoroaryl⋯aryl interactions between three adjacent layers of the cocrystal pfp•TPEB (b). Color scheme: white, H; gray, C; red, O; and light-green, F.

FIG. 20.

The noncovalent interactions (blue dashed lines) in one layer of the cocrystal pfp•TPEB (a). The perfluoroaryl⋯aryl interactions between three adjacent layers of the cocrystal pfp•TPEB (b). Color scheme: white, H; gray, C; red, O; and light-green, F.

Close modal

The perfluoroaryl⋯aryl interaction also shows its value in the construction of supramolecular host-guest materials. In 2012, Mader and co-workers reported the interesting crystal structure of the 2:1 cocrystal of hexafluorobenzene with pyrene-2,7-bis(4,4,5,5-tetramethyl-[1,3,2]dioxaborolane).143 In this crystal structure, the host molecules (pyrene-2,7-bis(4,4,5,5-tetramethyl-[1,3,2]dioxaborolane)) construct channel structures in which the guest molecules (hexafluorobenzene) are confined by perfluoroaryl⋯aryl interactions. Figure 21 demonstrates that the phenylboronic acid catechol ester (be) and trans-pentafluorostilbazole (pf-sbz) are selected as host molecules to construct electron-deficient channels via B→N bond. This very first example of fluorinated boron host channel enables to confine π-electron-rich molecules as guests, such as those commonly used benzene, toluene, and o-xylene in the petrochemical industry.152 Taking toluene as an example, the single-crystal x-ray diffraction analysis revealed that the toluene molecule could be captured in the host channels via face-to-face perfluoroaryl⋯aryl interactions between fluorinated rings and toluene rings, with a centroid–centroid distance of 3.790 Å. This study indicates the possibility of using the perfluoroaryl⋯aryl interactions to construct crystalline materials for storage and separation of important aromatic petrochemicals.

FIG. 21.

(a) Self-assembly of be-pf-sbz⊃guest. (b) The guest molecules. (c) The channel filled with the guest molecules. Reproduced with permission from Campillo-Alvarado et al., Front. Chem. 7, 695 (2019). Copyright 2019 Author(s), licensed under a Creative Commons Attribution License (CC BY).152 

FIG. 21.

(a) Self-assembly of be-pf-sbz⊃guest. (b) The guest molecules. (c) The channel filled with the guest molecules. Reproduced with permission from Campillo-Alvarado et al., Front. Chem. 7, 695 (2019). Copyright 2019 Author(s), licensed under a Creative Commons Attribution License (CC BY).152 

Close modal

Wang et al. proposed a superstacking self-assembly approach, which can be used to precisely design and finely synthesize the organic topological heterostructures for future integrated optoelectronics.153 In this study, in addition to the crystal of pure benzo[ghi]perylene (BGP), the authors also synthesized three cocrystals formed by BGP with 1,2,4,5-tetracyanobenzene (TCNB), 2,3,5,6-tetrafluoroterephthalonitrile (TFP), and octafluoronaphthalene (OFN), respectively. Figure 22 schematically illustrates the molecular-puzzle approach based on noncovalent interactions and the crystal-puzzle approach based on lattice matching. The calculated interaction energies of the dimers BGP–BGP, BGP–TCNB, BGP–TFP, and BGP–OFN are −1.33, −3.14, −2.84, and −2.15 kcal/mol, respectively (Fig. 22). These values are clearly unreliable because the authors used incorrect calculation methods. As a comparison the interaction energies of the T-shaped and parallel-displaced configurations of the much smaller benzene dimer are about −3.00 kcal/mol,154 that is to say, the binding strengths of the noncovalent interactions in Fig. 22 were greatly underestimated. However, these incorrect values of interaction energies do not affect the accuracy of the experimental results. The crystal structures of the cocrystals BGP–TFP and BGP–OFN indicate that the BGP-TFP and BGP-OFN microrods are formed mainly by the perfluoroaryl⋯aryl interactions. At the reliable PBE0-D3/def2-TZVPP theory level,89 the interaction energies of the dimers BGP–TFP and BGP–OFN taken from the crystal structures are −14.67 and −16.01 kcal/mol, respectively, which shows that the perfluoroaryl⋯aryl interactions in the two cocrystals are quite strong. In subsequent series of studies, the perfluoroaryl⋯aryl interactions involving TFP and OFN were frequently employed by Wang and co-workers to construct the organic heterostructures for different applications.155–157 The excellent work of Wang et al. greatly extends the application range of the perfluoroaryl⋯aryl interaction.

FIG. 22.

The schematic illustrations of the molecular-puzzle approach based on noncovalent interactions and the crystal-puzzle approach based on lattice matching (a). The fluorescence (FL) microscopy images of benzo[ghi]perylene (BGP) microplates (b), BGP-1,2,4,5-tetracyanobenzene (TCNB) microrods (c), BGP-2,3,5,6-tetrafluoroterephthalonitrile (TFP) microrods (d), and BGP-octafluoronaphthalene (OFN) microrods (e) under the UV excitation of (330–380 nm). Reproduced with permission from Zhuo et al., CCS Chem. 2, 413–424 (2020). Copyright 2020 Chinese Chemical Society. This figure has been published in CCS Chemistry 2020; “Super-Stacking Self-Assembly of Organic Topological Heterostructures” is available online at https://doi.org/10.31635/ccschem.020.202000171.153 

FIG. 22.

The schematic illustrations of the molecular-puzzle approach based on noncovalent interactions and the crystal-puzzle approach based on lattice matching (a). The fluorescence (FL) microscopy images of benzo[ghi]perylene (BGP) microplates (b), BGP-1,2,4,5-tetracyanobenzene (TCNB) microrods (c), BGP-2,3,5,6-tetrafluoroterephthalonitrile (TFP) microrods (d), and BGP-octafluoronaphthalene (OFN) microrods (e) under the UV excitation of (330–380 nm). Reproduced with permission from Zhuo et al., CCS Chem. 2, 413–424 (2020). Copyright 2020 Chinese Chemical Society. This figure has been published in CCS Chemistry 2020; “Super-Stacking Self-Assembly of Organic Topological Heterostructures” is available online at https://doi.org/10.31635/ccschem.020.202000171.153 

Close modal

Liquid crystals have properties of both liquids and crystals, and they are particularly useful in modern life and industry, providing display functions for many electronic products and instruments. In 1999, Marder et al. reported that the liquid crystal phase could be greatly stabilized by the perfluoroaryl⋯aryl interactions.166  Figure 23 shows the complementary perfluoroaryl⋯aryl interactions between 1,4-bis(phenylethynyl)-tetrafluorobenzene and 1,4-bis(pentafluorophenylethynyl)-benzene for the construction of a two-component liquid crystal phase. This is probably the most classic liquid crystal material assembled by the strong face-to-face perfluoroaryl⋯aryl interactions combining with the weak C–H⋯F–C in-plane hydrogen bonds. The 1:1 complex exhibits nematic liquid-crystalline phase behavior that neither of its two pure components possesses. An important character of this two-component system is that the liquid-crystalline phase grows along the one-dimensional direction based on the perfluoroaryl⋯aryl interactions. The 1:1 arene⋯perfluoroarene complexes are more common, while 2:1 arene⋯perfluoroarene complexes are very rare. Marder and co-workers reported the unusual structure and liquid-crystalline phase behavior of a 2:1 complex between 1,4-bis(phenylethynyl)-benzene and 1,4-bis(pentafluorophenylethynyl)tetrafluorobenzene.139 The phase behavior of this 2:1 complex is very complex, which indicates that the energies of alternative arrangements are close.

FIG. 23.

The complementary perfluoroaryl⋯aryl interactions between 1,4-bis(phenylethynyl)-tetrafluorobenzene and 1,4-bis(pentafluorophenylethynyl)-benzene for the construction of the liquid crystal phase.

FIG. 23.

The complementary perfluoroaryl⋯aryl interactions between 1,4-bis(phenylethynyl)-tetrafluorobenzene and 1,4-bis(pentafluorophenylethynyl)-benzene for the construction of the liquid crystal phase.

Close modal

Another classic example of the liquid crystalline formed by the perfluoroaryl⋯aryl interaction is the 1:1 mixture of PF-Tp and OR-Tp.20 The chemical structures of PF-Tp and OR-Tp have been shown in Fig. 3. When the 1:1 mixture of PF-Tp and OR-Tp was cooled to 90.8 °C from the isotropic melt, the discotic hexagonal mesophases were observed (Fig. 24). The results of powder x-ray diffraction spectroscopy demonstrated the highly ordered columnar alignments of the mesogens, which could be attributed to the strong perfluoroaryl⋯aryl interactions between PF-Tp and OR-Tp. At the same time, the temperature range of the liquid crystalline mesophase was found to be largely enlarged by adding PF-Tp to OR-Tp. Their results clearly show that the perfluoroaryl⋯aryl interaction plays a crucial role in both stabilizing the liquid crystalline mesophase and enlarging the temperature range of the mesogenic phase.

FIG. 24.

Polarized optical micrograph of the fan-shaped texture at the temperature of 90.8 °C upon cooling of a 1:1 mixture of PF-Tp and OR-Tp from the isotropic state. Reproduced with permission from Weck et al., Angew. Chem., Int. Ed. 38, 2741–2745 (1999). Copyright 1999 John Wiley and Sons Ltd.20 

FIG. 24.

Polarized optical micrograph of the fan-shaped texture at the temperature of 90.8 °C upon cooling of a 1:1 mixture of PF-Tp and OR-Tp from the isotropic state. Reproduced with permission from Weck et al., Angew. Chem., Int. Ed. 38, 2741–2745 (1999). Copyright 1999 John Wiley and Sons Ltd.20 

Close modal

Kishikawa et al. further expanded the liquid crystal materials based on the perfluoroaryl⋯aryl interactions.167–169 They demonstrated that some non-disk-shaped molecules which contain both the pentafluorophenyl group and the nonfluorinated phenyl group could self-organize into ordered columnar superstructures in the mesophases via the strong face-to-face perfluoroaryl⋯aryl interactions in the core parts of the columns (Fig. 25).167 The compounds 16 in Fig. 25 all showed the liquid crystal phases. As a contrast, the nonfluorinated benzoate such as 3,4,5-tris(decyloxy)benzyl benzoate was found not to show any liquid crystal phase. A continuous work of Kishikawa's group further showed that the type of columnar liquid crystal phases and the macroscopic morphologies of the liquid crystal domains could be effectively changed according to the length and balance of the alkyl chains in 3,4,5-trialkoxybenzyl pentafluorobenzoates (Fig. 26).169 Although the intercolumnar interactions between alkyl chains of neighboring columns are important for the stabilization of the columnar liquid crystal phases, the perfluoroaryl⋯aryl interactions are still crucial for generating the columnar liquid crystal phases. Twieg et al. synthesized a series of cyanobiphenyl and cyano-p-terphenyl derivatives with a variety of multifluorophenyloxy termini and studied the effect of the multifluorophenyloxy termini on the liquid crystalline behavior.170 Taking 1-(4-cyanobiphenyl-4'-yloxy)-6-(phenyloxy)hexane (PHC6OCB) and 1-(4-cyanobiphenyl-4'-yloxy)-6-(pentafluorophenyloxy)hexane (PFC6OCB) as examples, their research found that PHC6OCB does not exhibit any mesogenic properties, whereas PFC6OCB shows a nematic phase. They attributed the liquid crystalline behavior of PFC6OCB to the perfluoroaryl⋯aryl interactions between PFC6OCB molecules, as explained by Kishikawa.168 

FIG. 25.

Above: The columnar superstructure constructed by the perfluoroaryl⋯aryl interactions (green disk: perfluorinated benzene ring; orange disk: nonfluorinated benzene ring; gray block: trialkoxy groups). Below: The ribbonlike textures (left) and focal conic textures (right) of compound 2. Reproduced with permission from Kishikawa et al., Angew. Chem., Int. Ed. 46, 764–768 (2007). Copyright 2007 John Wiley and Sons Ltd.167 

FIG. 25.

Above: The columnar superstructure constructed by the perfluoroaryl⋯aryl interactions (green disk: perfluorinated benzene ring; orange disk: nonfluorinated benzene ring; gray block: trialkoxy groups). Below: The ribbonlike textures (left) and focal conic textures (right) of compound 2. Reproduced with permission from Kishikawa et al., Angew. Chem., Int. Ed. 46, 764–768 (2007). Copyright 2007 John Wiley and Sons Ltd.167 

Close modal
FIG. 26.

Above: Self-organization of trialkoxybenzyl pentafluorobenzoates into the rack-gear structure (green disk: perfluorinated benzene ring; yellow disk: nonfluorinated benzene ring; gray block: trialkoxy groups). Below: The fan-shaped and/or curved tape textures and linear tape textures regulated by the lengths of the alkyl chains. Reproduced with permission from Kishikawa et al., Liq. Cryst. 50, 319–330 (2023). Copyright 2023 Taylor and Francis Ltd.169 

FIG. 26.

Above: Self-organization of trialkoxybenzyl pentafluorobenzoates into the rack-gear structure (green disk: perfluorinated benzene ring; yellow disk: nonfluorinated benzene ring; gray block: trialkoxy groups). Below: The fan-shaped and/or curved tape textures and linear tape textures regulated by the lengths of the alkyl chains. Reproduced with permission from Kishikawa et al., Liq. Cryst. 50, 319–330 (2023). Copyright 2023 Taylor and Francis Ltd.169 

Close modal

Maly et al. synthesized two novel polycatenar dibenzopentacenequinones, 2,3,6,7-tetrakis(decyloxy)dibenzo[a,c]pentacene-10,17-dione (TDDBPQ) and its fluorinated analog 12,13,14,15-tetrafluoro-2,3,6,7-tetrakis(decyloxy)dibenzo[a,c]pentacene-10,17-dione (TF-TDDBPQ), both of them showed columnar liquid crystal phases.171 The temperature range of the mesophase of TDDBPQ is 148–177 °C, while the temperature range of the mesophase of TF-TDDBPQ expands to 121–336 °C. Compared with nonfluorinated TDDBPQ, the significantly improved stabilization of the columnar mesophase of TF-TDDBPQ was attributed to the much stronger perfluoroaryl⋯aryl interactions between TF-TDDBPQ molecules (Fig. 27).

FIG. 27.

The columnar mesophase (left) of TF-TDDBPQ stabilized by the perfluoroaryl⋯aryl interactions (right). Reproduced with permission from Yardley et al., Org. Lett. 21, 10102 − 10105 (2019). Copyright 2019 American Chemical Society.171 

FIG. 27.

The columnar mesophase (left) of TF-TDDBPQ stabilized by the perfluoroaryl⋯aryl interactions (right). Reproduced with permission from Yardley et al., Org. Lett. 21, 10102 − 10105 (2019). Copyright 2019 American Chemical Society.171 

Close modal

The partial fluorination of polycyclic aromatic compounds has become a commonly used approach in the induction and stabilization of the columnar mesophases.172,173 Donnio and co-workers fully investigated the liquid crystalline behavior and optical properties of some partially fluorinated polycyclic aromatic compounds (Fig. 28).172 As shown in Fig. 28, TPn, 1F-TPn, 1F-TPn, and BTPn do not exhibit any mesophases, while all of the other compounds show the mesophases. The polar “Janus” mesogen 4F-TP3 has the highest clearing temperature and broadest temperature range of the columnar hexagonal mesophase among the 4F-TPn series, and the other Janus mesogen 6F-BTP6 has the highest clearing temperature and broadest temperature range of the columnar hexagonal mesophase among the 6F-BTPn series. From the 4F-TPn series to the 6F-BTPn series, the mesomorphous ranges are greatly expanded and the mesophase stabilities are significantly enhanced. These results can be explained by the complementary perfluoroaryl⋯aryl interactions between these Janus molecules. The perfluoroaryl⋯aryl interactions between these Janus molecules are much stronger than the aryl⋯aryl interactions between their nonfluorinated homologs, and the perfluoroaryl⋯aryl interactions between 6F-BTPn molecules are much stronger than the corresponding ones between 4F-TPn molecules. Such a conclusion was also supported by the study on the optical properties of these compounds.

FIG. 28.

Mesophase diagrams of the compounds shown on top (R: CnH2n+1; Cr: Crystalline phase; Colhex: Columnar hexagonal mesophase; [Colhex]: Columnar hexagonal mesophase upon cooling; Colobl: Oblique columnar mesophase). Reproduced with permission from Zhou et al., Chem. - Eur. J. 29, e202301829 (2023). Copyright 2023 Wiley-VCH GmbH.172 

FIG. 28.

Mesophase diagrams of the compounds shown on top (R: CnH2n+1; Cr: Crystalline phase; Colhex: Columnar hexagonal mesophase; [Colhex]: Columnar hexagonal mesophase upon cooling; Colobl: Oblique columnar mesophase). Reproduced with permission from Zhou et al., Chem. - Eur. J. 29, e202301829 (2023). Copyright 2023 Wiley-VCH GmbH.172 

Close modal

From the important studies mentioned above, we can see that the perfluoroaryl⋯aryl interactions between different molecules or between identical Janus molecules play crucial roles in inducing or stabilizing the mesophases of liquid crystals. This methodology should be very useful for the design and synthesis of new-generation liquid crystal materials.

Many review articles on the organic luminescent materials have been published in recent years, but only a few of them concerned about the important role of the perfluoroaryl⋯aryl interaction in the field of organic luminescent materials.174,175 The review articles of Refs. 174 and 175 both mentioned the important works by Sun and co-workers in applying the perfluoroaryl⋯aryl interaction to organic luminescent cocrystals.176,177 In one study utilizing the competition between perfluoroaryl⋯aryl and charge-transfer interactions, Sun et al. synthesized the TCNB-doped cocrystals pyrene–OFN and pyrene–TFP, which show tunable emission colors.176 Such a simple method can be extended to the design and synthesis of novel organic light-harvesting systems. In the other study, Sun and co-workers constructed the two-dimensional nanorod meshes from the organic luminescent cocrystals formed by benzoperylene with 1,3-dicyanotetrafluorobenzene, 1,4-dicyanotetrafluorobenzene, and OFN, respectively, by using the benzoperylene microsheets (MSs) as templates.177 The crystal transformation was driven by the perfluoroaryl⋯aryl interaction. The nanorod meshes with different luminescence have potential applications in optoelectronic devices.

The perfluoroaryl⋯aryl interaction can be utilized to enhance the photoluminescence quantum yield (PLQY) of some aromatic fluorophores. As mentioned above, as far back as 2001, Marder and co-workers reported the crystal structure of 1:1 complexes of OFN with the arenes anthracene, phenanthrene, pyrene, and triphenylene.119 However, they did not investigate the luminescent properties of these complexes. Recent studies have shown that the solid PLQYs of polycyclic aromatic hydrocarbons such as anthracene, pyrene, perylene (Per), and coronene (Cor) could be greatly improved as the OFN molecules were intercalated between these aromatic fluorophores via the perfluoroaryl⋯aryl interactions.178,179 As can be seen in Fig. 29, the introduction of OFN leads to much brighter luminescence of the cocrystals Per/OFN and Cor/OFN under a 365 nm UV lamp. The intercalation of the OFN molecules into the packed structures of polycyclic aromatic hydrocarbons reduces the aggregation caused quenching (ACQ) effects of these aromatic fluorophores and therefore improves their PLQYs. These results clearly show that traditional ACQ molecules can be used to fabricate solid emissive materials with high-emission efficiency by using the perfluoroaryl⋯aryl interactions. This will greatly expand the family of organic emitting materials. In a very recent article, Yang and co-workers reported similar experimental phenomena; furthermore, they discovered that the chiral crystals self-assembled by the perfluoroaryl⋯aryl interactions and C–H⋯O hydrogen bonds could exhibit strong mechanoluminescence emissions.180 

FIG. 29.

The formation of the cocrystals Per/OFN (a) and Cor/OFN (b), and the photographs of powders as well as aqueous dispersions for Per microsheet (MS), Per/OFN microwire (MW), OFN crystal, Cor microwire (MW), and Cor/OFN microwire (MW) under daylight (c) and a 365 nm UV lamp (d). Reproduced with permission from Huang et al., Nat. Commun. 10, 169 (2019). Copyright 2019 Author(s), licensed under a Creative Commons Attribution 4.0 International License.179 

FIG. 29.

The formation of the cocrystals Per/OFN (a) and Cor/OFN (b), and the photographs of powders as well as aqueous dispersions for Per microsheet (MS), Per/OFN microwire (MW), OFN crystal, Cor microwire (MW), and Cor/OFN microwire (MW) under daylight (c) and a 365 nm UV lamp (d). Reproduced with permission from Huang et al., Nat. Commun. 10, 169 (2019). Copyright 2019 Author(s), licensed under a Creative Commons Attribution 4.0 International License.179 

Close modal

The circularly polarized luminescence (CPL)-active materials have broad application prospects in areas such as biosensing, 3D optical displays, information encryption and storage, and optoelectronic devices due to their unique chiro-optical properties. Exploring and developing new materials with circularly polarized luminescent properties has become one of the current research hotspots. Similarly, OFN molecules are often employed to enhance the performances of CPL-active materials through the perfluoroaryl⋯aryl interactions. Duan et al. first applied the perfluoroaryl⋯aryl interactions to the chiral emissive polycyclic aromatic hydrocarbons (Fig. 30).181 The molecular structures of R/S-Py and R/S-An are shown in Fig. 30. Amorphous R/S-Py and R/S-An exhibited low photoluminescence quantum yield (ΦPL) and small luminescence dissymmetry factor (glum). The chiral R/S-Py and R/S-An can assemble with achiral OFN via strong perfluoroaryl⋯aryl interactions. The formed supramolecular co-assemblies possess both higher ΦPL and larger glum. The perfluoroaryl⋯aryl interaction demonstrated its powerful capabilities to amplify the circular polarization of CPL and to improve the photoluminescence quantum yield.

FIG. 30.

The circular polarization and luminescence efficiency boosted by the perfluoroaryl⋯aryl interactions. Reproduced with permission from Zhang et al., Angew. Chem., Int. Ed. 60, 4575–4580 (2021). Copyright 2021 John Wiley and Sons Ltd.181 

FIG. 30.

The circular polarization and luminescence efficiency boosted by the perfluoroaryl⋯aryl interactions. Reproduced with permission from Zhang et al., Angew. Chem., Int. Ed. 60, 4575–4580 (2021). Copyright 2021 John Wiley and Sons Ltd.181 

Close modal

The synergistic effects of perfluoroaryl⋯aryl interactions and other noncovalent interactions such as hydrogen bonds and electrostatic interactions have been utilized to precisely synthesize the CPL-active materials across multiple scales and/or to rationally regulate the performances of CPL-active materials. The CPL amplification and/or CPL inversion promoted by the perfluoroaryl⋯aryl interactions jointly with other noncovalent interactions have been reported by several studies.182–184 Xing et al. constructed a series of functional chiroptical materials by employing orthogonal perfluoroaryl⋯aryl interactions and hydrogen bonds (Fig. 31).183,184 In addition to the nanostructure transformation, the chiroptical inversion phenomenon was also observed upon the formation of these supramolecular chiral materials. These studies show again that co-assembly with OFN to form the perfluoroaryl⋯aryl interaction is a simple but efficient approach to regulate the structures and properties of chiroptical materials.

FIG. 31.

(a) Molecular structures of PM, PT, Mm, and OFN, and the perfluoroaryl⋯aryl interaction (AP) formed by OFN with pyrene ring as well as the cyclic double hydrogen bonds (H-bonds) formed by Mm with carboxyl group. (b) Schematic representation of the formation of hierarchical co-assemblies via AP and H-bonds. Reproduced with permission from Liang et al., ACS Appl. Mater. Interfaces 13, 29170–29178 (2021). Copyright 2021 American Chemical Society.183 

FIG. 31.

(a) Molecular structures of PM, PT, Mm, and OFN, and the perfluoroaryl⋯aryl interaction (AP) formed by OFN with pyrene ring as well as the cyclic double hydrogen bonds (H-bonds) formed by Mm with carboxyl group. (b) Schematic representation of the formation of hierarchical co-assemblies via AP and H-bonds. Reproduced with permission from Liang et al., ACS Appl. Mater. Interfaces 13, 29170–29178 (2021). Copyright 2021 American Chemical Society.183 

Close modal

In addition to the OFN molecule, other molecules containing the perfluoroaryl groups can also regulate the luminescent performance of organic materials via the perfluoroaryl⋯aryl interactions. Recently, Liu and co-workers reported that the fluorescence (FL) and CPL activities of nonemissive helical cage rotors could be switched on step by step upon mounting the inner and outer tris(pentafluorophenyl)borane (TFPB) brakes (Fig. 32).185 One helical cage was constructed by three chiral pyromellitic diimide molecules and two tris(4-formylphenyl)amine molecules. The phenyl rings in the helical cage can rotate freely, which leads to that the helical cage cannot emit light efficiently either in solution or in the solid states. The mounting of the inner and outer TFPB brakes to the helical cage rotors hinders the free rotations of the phenyl rings through the perfluoroaryl⋯aryl interactions and hydrogen bonds, and further switches on the FL and CPL activities of helical cages in a step-by-step manner. This work demonstrates a simple but effective strategy to regulate the FL and CPL activities of cage rotor systems by using the perfluoroaryl⋯aryl interactions.

FIG. 32.

The FL and CPL activities of nonemissive helical cage rotors are switched on step by step upon mounting the inner and outer TFPB brakes. Reproduced with permission from Shang et al., J. Am. Chem. Soc. 145, 27639–27649 (2023). Copyright 2023 American Chemical Society.185 

FIG. 32.

The FL and CPL activities of nonemissive helical cage rotors are switched on step by step upon mounting the inner and outer TFPB brakes. Reproduced with permission from Shang et al., J. Am. Chem. Soc. 145, 27639–27649 (2023). Copyright 2023 American Chemical Society.185 

Close modal

The perfluoroaryl⋯aryl interaction can also play a critical role in the design of room temperature phosphorescence (RTP) materials.33,186 As shown in Fig. 33, the Ar-Ar and ArF-ArF molecules are packed via edge-to-face contacts, respectively, while the Ar-ArF molecules tend to form the face-to-face arrangements via the perfluoroaryl⋯aryl interactions [Figs. 33(d)–33(f)]. Crystal structures determine the properties of materials. Correspondingly, the Ar-Ar and ArF-ArF crystals have no RTP phenomenon, while the Ar-ArF crystals exhibit green RTP phenomenon after turning off the 365 nm UV lamp [Figs. 33(g)–33(i)]. The RTP performance of Ar-ArF was further optimized by means of halogen substitution. The synthetic Ar-ArF derivatives in Fig. 33(j) exhibit RTP emission in both solution and solid states, and PFP-F in its powder form achieves the longest RTP lifetime of 282 ms. In the crystal structures, all these Ar-ArF derivatives are packed in face-to-face modes due to the perfluoroaryl⋯aryl interactions. Theoretical calculations reveal that the face-to-face packing arrangement is a key factor in realizing the RTP emission because such an arrangement increases the spin-orbital coupling value, which is a key factor to the intersystem crossing process. This study shows that the formation of the perfluoroaryl⋯aryl interactions is a very efficient method for achieving the persistent RTP.

FIG. 33.

Molecular packing arrangements of benzene (Ar), hexafluorobenzene (ArF), and benzene/hexafluorobenzene (Ar/ArF), diphenyl (Ar-Ar), decafluorobiphenyl (ArF-ArF), and 2,3,4,5,6-pentafluoro-1,1′-biphenyl (Ar-ArF) (a)–(f); the phosphorescence spectra of Ar-Ar, ArF-ArF, and Ar-ArF, and corresponding photographs taken under and after the 365 nm UV irradiation at room temperature (g)–(i); the photographs of PFP-F, PFP-Cl, PFP-Br, and PFP-I taken under and after the 365 nm UV irradiation at room temperature (j). Reproduced with permission from Zhang et al., Sci. China: Chem. 65, 918–925 (2022). Copyright 2022 Science China Press.186 

FIG. 33.

Molecular packing arrangements of benzene (Ar), hexafluorobenzene (ArF), and benzene/hexafluorobenzene (Ar/ArF), diphenyl (Ar-Ar), decafluorobiphenyl (ArF-ArF), and 2,3,4,5,6-pentafluoro-1,1′-biphenyl (Ar-ArF) (a)–(f); the phosphorescence spectra of Ar-Ar, ArF-ArF, and Ar-ArF, and corresponding photographs taken under and after the 365 nm UV irradiation at room temperature (g)–(i); the photographs of PFP-F, PFP-Cl, PFP-Br, and PFP-I taken under and after the 365 nm UV irradiation at room temperature (j). Reproduced with permission from Zhang et al., Sci. China: Chem. 65, 918–925 (2022). Copyright 2022 Science China Press.186 

Close modal

Generally, a perovskite is any compound with the chemical formula ABX3, where A and B are two positively charged ions and X is a negatively charged ion. The perovskite discovered about 200 years ago will probably be the subverter of the next-generation solar cell materials. The layered organic–inorganic perovskites are a class of optoelectronic materials consisting of alternating inorganic perovskite slabs and organic spacer layers. Without changing the inorganic components, the device performance and stability of hybrid perovskites solar cells can be boosted only by changing the composition of the organic layer. Perfluoroarenes and further the perfluoroaryl⋯aryl interactions have been applied successfully to enhance the device performance and the stability of hybrid perovskite photovoltaics.187 

In 2002, the first two examples of intercalated organic–inorganic perovskites stabilized by perfluoroaryl⋯aryl interactions were reported by Mitzi et al.188 The two intercalated hybrid perovskites are (F5-PEA)2SnI4•(C6H6) and (PEA)2SnI4•(C6F6), in which F5-PEA represents the organic cation 2,3,4,5,6-pentafluorophenethylammonium and PEA denotes the organic cation 2-phenylethanaminium. (F5-PEA)2SnI4 and (PEA)2SnI4 are two parent perovskites, and C6H6 and C6F6 are two intercalated molecules. As shown in Fig. 34, in the crystal structure of C6H6-intercalated (F5-PEA)2SnI4, the offset face-to-face perfluoroaryl⋯aryl interactions are formed between each intercalated C6H6 molecule and its two nearest-neighbor C6F5 groups; in the isostructural crystal (PEA)2SnI4•(C6F6), the offset face-to-face perfluoroaryl⋯aryl interactions are formed between each intercalated C6F6 molecule and its two nearest-neighbor C6H5 groups. The intercalation experiments carried out in this study highlighted the key roles of the offset face-to-face perfluoroaryl⋯aryl interactions in stabilizing the organic–inorganic hybrid perovskites. The application of the perfluoroaryl⋯aryl interactions in the field of organic–inorganic hybrid perovskites provides a new direction for tailoring this exciting class of optoelectronic materials.

FIG. 34.

The perfluoroaryl⋯aryl interactions in the crystal structures of (F5-PEA)2SnI4•(C6H6) (left) and (PEA)2SnI4•(C6F6) (right). The C, N, and F atoms are shown in white, blue, and green. For clarity the H atoms are omitted. Reproduced with permission from Mitzi et al., Inorg. Chem. 41, 2134 − 2145 (2002). Copyright 2002 American Chemical Society.188 

FIG. 34.

The perfluoroaryl⋯aryl interactions in the crystal structures of (F5-PEA)2SnI4•(C6H6) (left) and (PEA)2SnI4•(C6F6) (right). The C, N, and F atoms are shown in white, blue, and green. For clarity the H atoms are omitted. Reproduced with permission from Mitzi et al., Inorg. Chem. 41, 2134 − 2145 (2002). Copyright 2002 American Chemical Society.188 

Close modal

As the expansion of above study, subsequent research revealed that both the device stability and photovoltaic efficiency of 2D hybrid perovskites-based solar cells could be significantly boosted by the perfluoroaryl⋯aryl interactions.189,190 Figure 35 shows the chemical structures of F5-PEA and PEA, and the local single-crystal structure of 2D organic–inorganic hybrid perovskite [(PEA)0.5(F5-PEA)0.5]2PbI4.190 It is clear in this hybrid perovskite that the offset face-to-face perfluoroaryl⋯aryl interaction is formed between the pentafluorophenyl ring of F5-PEA and the phenyl ring of PEA. Among the 2D hybrid perovskites-based solar cells with different ratios of F5-PEA/PEA, the 1:1 F5-PEA:PEA-based solar cell has the best stability and highest photovoltaic efficiency of about 11%, which indicates the pivotal role of the perfluoroaryl⋯aryl interactions. This study provides an example of how the perfluoroaryl⋯aryl interactions can tune the device stability and performance of organic–inorganic hybrid perovskites-based solar cells.

FIG. 35.

The chemical structures of F5-PEA and PEA, and the local single-crystal structure of 2D organic–inorganic hybrid perovskite [(PEA)0.5(F5-PEA)0.5]2PbI4. Reproduced with permission from Hu et al., ACS Mater. Lett. 1, 171 − 176 (2019). Copyright 2019 American Chemical Society.190 

FIG. 35.

The chemical structures of F5-PEA and PEA, and the local single-crystal structure of 2D organic–inorganic hybrid perovskite [(PEA)0.5(F5-PEA)0.5]2PbI4. Reproduced with permission from Hu et al., ACS Mater. Lett. 1, 171 − 176 (2019). Copyright 2019 American Chemical Society.190 

Close modal

Milić and co-workers explored the nanoscale phase segregation in perfluoroarene⋯arene π-systems for hybrid perovskites by using the nuclear magnetic resonance (NMR) crystallography as well as other experimental methods.191,192 They selected F5-PEA and PEA as monofunctional organic spacers for the Ruddlesden–Popper layered perovskites, and selected 2,3,5,6-tetrafluoro-1,4-phenylenedimethanammonium (F4-PDMA) and 1,4-phenylenedimethanammonium (PDMA) as bifunctional organic spacers for the Dion–Jacobson layered perovskites (Fig. 36). The results showed that both the 1:1 spacer mixture of F5-PEA and PEA and the 1:1 spacer mixture of F4-PDMA and PDMA lead to the nanoscale phase segregation. The assembly of the organic spacer layer is driven mainly by the perfluoroaryl⋯aryl interaction between F5-PEA and PEA or between F4-PDMA and PDMA. Furthermore, it was found that the solar cells based on the mixed-spacer Ruddlesden–Popper layered perovskites have much higher power conversion efficiencies and the operational stability than the ones based on the pure Ruddlesden–Popper layered perovskites. Similarly, the perfluoroaryl⋯aryl interaction between OFN and PEA has been utilized to adjust the phase distribution of quasi-2D perovskite light-emitting diodes.193 

FIG. 36.

Schematic representation of the perfluoroaryl⋯aryl interactions between spacer cations and the structures of the Ruddlesden–Popper layered perovskites and Dion–Jacobson layered perovskites. The cyan rods denote the corresponding spacer cations shown in the left sides. Reproduced with permission from Almalki et al., Nanoscale 14, 6771–6776 (2022). Copyright 2022 The Royal Society of Chemistry and Author(s), licensed under a Creative Commons Attribution-Non Commercial 3.0 Unported License (CC BY-NC).192 

FIG. 36.

Schematic representation of the perfluoroaryl⋯aryl interactions between spacer cations and the structures of the Ruddlesden–Popper layered perovskites and Dion–Jacobson layered perovskites. The cyan rods denote the corresponding spacer cations shown in the left sides. Reproduced with permission from Almalki et al., Nanoscale 14, 6771–6776 (2022). Copyright 2022 The Royal Society of Chemistry and Author(s), licensed under a Creative Commons Attribution-Non Commercial 3.0 Unported License (CC BY-NC).192 

Close modal

Like the perovskite photovoltaics, the organic photovoltaics are also of the next-generation solar cell materials. Polyacenes and their derivatives have widespread interest in organic photovoltaics. Anthony et al. studied the structures and electronic properties of 6,13-bis(triisopropylsilylethynyl)pentacene (TIPSP), 1,2,3,4-tetrafluoro-6,13-bis(triisopropylsilylethynyl)pentacene (F4-TIPSP), and 1,2,3,4,8,9,10,11-octafluoro-6,13-bis(triisopropylsilylethynyl)pentacene (F8-TIPSP) in their solid states (Fig. 37).194 The crystallographic results show that the average interplanar distance decreases on the order of TIPSP > F4-TIPSP > F8-TIPSP. The smaller interplanar distances of F4-TIPSP and F8-TIPSP compared to nonfluorinated TIPSP are caused by the perfluoroaryl⋯aryl interactions in F4-TIPSP and F8-TIPSP. The hole mobility is inversely proportional to the average interplanar distance between the pentacene planes in the solid state. Correspondingly, F8-TIPSP shows the highest hole mobility, while nonfluorinated TIPSP exhibits the lowest hole mobility. Different from the strategy of utilizing the perfluoroaryl⋯aryl interactions between the polyacene cores, Matsuo and co-workers utilized the perfluoroaryl⋯aryl interactions between the substituents of polyacenes to improve the charge-carrier mobility.195 To this end, they synthesized the target compound 5-pentafluorophenyl-11-phenyltetracene (F5-PPT) and the reference compound 5,11-diphenyltetracene (PPT) (Fig. 38). The crystal structures of F5-PPT and PPT jointly confirm that it is the perfluoroaryl⋯aryl interaction between the perfluorophenyl and phenyl substituents that leads to the smaller interplanar distance between the tetracene cores. The smaller interplanar distance means the higher charge-carrier mobility. As demonstrated in Fig. 38, the maximum photoconductivities of F5-PPT microcrystals are ten times that of PPT microcrystals, and the average hole mobility of F5-PPT is over a hundred times that of PPT. In the same manner, the perfluoroaryl⋯aryl interaction between the substituent and the polyacene core can also been utilized to control the packing of polyacene molecules in the solid state and further to tune their electronic properties. It has been reported that the compound 2,6-bis(pentafluorophenyl)anthracene exhibited excellent n-type semiconducting properties.196 Its crystal structure shows that the perfluoroaryl⋯aryl interactions occur between the pentafluorophenyl groups and the anthracene cores.197 

FIG. 37.

The chemical structures and hole mobilities of TIPSP, F4-TIPSP, and F8-TIPSP (a), and the perfluoroaryl⋯aryl interactions in the crystal structure of F8-TIPSP (b). Color code in the spacefill representation: white, H; gray, C; light-green, F; slightly desaturated yellow, Si.

FIG. 37.

The chemical structures and hole mobilities of TIPSP, F4-TIPSP, and F8-TIPSP (a), and the perfluoroaryl⋯aryl interactions in the crystal structure of F8-TIPSP (b). Color code in the spacefill representation: white, H; gray, C; light-green, F; slightly desaturated yellow, Si.

Close modal
FIG. 38.

The chemical structures, maximum photoconductivities (red numbers), and hole mobilities (blue numbers) of F5-PPT and PPT (a), and the perfluoroaryl⋯aryl interactions in the crystal structure of F5-PPT (b). Color code in the spacefill representation: white, H; gray, C; light-green, F.

FIG. 38.

The chemical structures, maximum photoconductivities (red numbers), and hole mobilities (blue numbers) of F5-PPT and PPT (a), and the perfluoroaryl⋯aryl interactions in the crystal structure of F5-PPT (b). Color code in the spacefill representation: white, H; gray, C; light-green, F.

Close modal

Multifluorination is an efficient strategy to design the polymer donors for organic solar cells. Wu and co-workers reported that the organic polymer solar cells containing 4,5,6,7-tetrafluoronaphtho[2,1-b:3,4-b']dithiophene (FNT) units have high photo-conversion efficiency of over 18%.198 In the crystalline state, the planar FNT molecules are stacked head-to-tail via the parallel-displaced perfluoroaryl⋯aryl interactions (Fig. 39). The short C–C interatomic distances of 3.346 Å indicate the strong perfluoroaryl⋯aryl interactions between FNT molecules. The density functional theory calculations further demonstrated good long-range planarity of the conjugated backbone of the FNT-based polymer.198 Therefore, it can be inferred that the strong perfluoroaryl⋯aryl interactions may also exist between the FNT-based polymer molecules. The formation of the perfluoroaryl⋯aryl interaction facilitates the hole transport and makes the FNT-based polymer a very promising donor material for organic polymer solar cells.

FIG. 39.

The chemical structure of FNT and the perfluoroaryl⋯aryl interactions in the crystal structure of FNT.

FIG. 39.

The chemical structure of FNT and the perfluoroaryl⋯aryl interactions in the crystal structure of FNT.

Close modal

Many studies have demonstrated that the formation of perfluoroaryl⋯aryl interaction can facilitate the occurrence of the photochemically induced [2 + 2] cycloaddition reaction in the solid state (Fig. 40).144,199–204 Photolysis of the cocrystal between trans-stilbene and trans-decafluorostilbene and photolysis of the homocrystal of trans-pentafluorostilbene are two of the most classic examples of [2 + 2] photocycloaddition reactions directed by the perfluoroaryl⋯aryl interactions.199 Both [2 + 2] photocycloaddition reactions occur at room temperature, with reaction yields exceeding 98%. As shown in Fig. 40, the formation of strong perfluoroaryl⋯aryl interactions leads to a more favorable arrangement of the C = C double bonds for the occurrence of [2 + 2] photocycloaddition reactions. In addition to the photodimerization reaction, the photopolymerization reaction occurs in the same manner. Sonoda and co-workers investigated the solid-state photoreactivities of (E,E,E)-1,6-diphenyl-1,3,5-hexatriene (DPH) and a series of ring-fluorinated (E,E,E)-DPHs.201 As expected, the crystals directed by the strong perfluoroaryl⋯aryl interactions possessed the highest photoreactive activities.

FIG. 40.

The [2 + 2] photodimerization (left) and photopolymerization (right) reactions in the solid state. The blue dash lines represent the perfluoroaryl⋯aryl interactions. Reproduced with permission from Coates et al., J. Am. Chem. Soc. 120, 3641 − 3649 (1998). Copyright 1998 American Chemical Society.199 

FIG. 40.

The [2 + 2] photodimerization (left) and photopolymerization (right) reactions in the solid state. The blue dash lines represent the perfluoroaryl⋯aryl interactions. Reproduced with permission from Coates et al., J. Am. Chem. Soc. 120, 3641 − 3649 (1998). Copyright 1998 American Chemical Society.199 

Close modal

The perfluoroaryl⋯aryl interactions between the photocatalysts and the substrates can also promote the photocatalytic reactions.205,206 In 2016, Zhang et al. reported that the perfluoroaryl⋯aryl interactions between pyrene-based photocatalysts (Py) and polyfluoroarenes (FA) facilitated the photocatalytic hydrodefluorination reactions of polyfluoroarenes.205  Figure 41 illustrates the proposed photocatalytic cycle for the hydrodefluorination reactions of polyfluoroarenes using pyrene-based photocatalysts. All the complexes in Fig. 41 are bound by the perfluoroaryl⋯aryl interactions. An inner-sphere electron transfer from Py to FA occurs in the anionic radical complex [Py:FA]•−. The inner-sphere process based on the perfluoroaryl⋯aryl interaction results in the formation of FA•− and promotes the occurrence of the hydrodefluorination reaction. The perfluoroaryl⋯aryl interaction associated with the catalytic activity of Py can be tuned by changing the substituent R of Py. Such a strategy can be used to enhance the catalytic activity of photocatalysts. Tang et al. described the other example of photoredox catalysis in which the perfluoroaryl⋯aryl interaction was utilized to construct the photocatalyst (Fig. 42).206 Tris(pentafluorophenyl)borane, B(C6F5)3, is a π-hole molecule, and its hydrate B(C6F5)3•H2O also serves as a very good hydrogen bond donor. The complex between B(C6F5)3•H2O and amide 2-phenyl-3,4-dihydroisoquinolin-1(2H)-one is formed through the perfluoroaryl⋯aryl interactions and O−H⋯O hydrogen bonds. Different from the monomers, the complex has the photoactive properties. When the complex B(C6F5)3•H2O/amide was used as the photocatalyst, a series of α-aminoamides could be synthesized under mild reaction conditions. On the other hand, it was noted that the reaction yields for N-aryl tetrahydroisoquinolines were generally good, while the reaction yields for N-alkyl tetrahydroisoquinolines were very low. This provided further evidence for the key role of the perfluoroaryl⋯aryl interaction in the photocatalytic syntheses of α-aminoamides.

FIG. 41.

Proposed photocatalytic cycle for the hydrodefluorination reactions of polyfluoroarenes using pyrene-based photocatalysts. The blue dashed lines represent the perfluoroaryl⋯aryl interactions. Reproduced with permission from Lu et al., J. Am. Chem. Soc. 138, 15805–15808 (2016). Copyright 2016 American Chemical Society.205 

FIG. 41.

Proposed photocatalytic cycle for the hydrodefluorination reactions of polyfluoroarenes using pyrene-based photocatalysts. The blue dashed lines represent the perfluoroaryl⋯aryl interactions. Reproduced with permission from Lu et al., J. Am. Chem. Soc. 138, 15805–15808 (2016). Copyright 2016 American Chemical Society.205 

Close modal
FIG. 42.

Proposed photocatalytic cycle for the synthesis of α-aminoamide. The red dot lines highlight the perfluoroaryl⋯aryl interactions. Reproduced with permission from Wang et al., iScience 26, 106528 (2023). Copyright 2023 Author(s), licensed under the CC BY-NC-ND license.206 

FIG. 42.

Proposed photocatalytic cycle for the synthesis of α-aminoamide. The red dot lines highlight the perfluoroaryl⋯aryl interactions. Reproduced with permission from Wang et al., iScience 26, 106528 (2023). Copyright 2023 Author(s), licensed under the CC BY-NC-ND license.206 

Close modal

There are only two naturally occurring fluorine-containing organic compounds, which are related to the nucleosides and amino acids: 4'-fluoro-5'-O-sulphamoyladenosine and 4-fluorothreonine.1 Chemical or biological synthesis must be carried out in order to introduce the F atoms into the biological systems. Hence, the perfluoroaryl⋯aryl interactions are always connected with the terms such as “protein engineering,” “genetic engineering,” and “pharmaceutical engineering.” Until now, the effects of the perfluoroaryl⋯aryl interactions on the structures and functions of the modified proteins or nucleic acids have been widely investigated.

Over the last few decades, many significant attempts have been made to expand the genetic alphabet by incorporating alternative base pairs into deoxyribonucleic acid (DNA) and ribonucleic acid (RNA).207–211 In addition to the hydrogen bonding,207 the perfluoroaryl⋯aryl interactions could also be utilized to design new base pairs.208–211 Mathis and Hunziker used phenyl-β-D-deoxyriboside (P, see Fig. 43) and pentafluorophenyl-β-D-deoxyriboside (F,5 see Fig. 43) as base replacements to investigate the effects of the number and position of the artificial base pair F5-P on the thermodynamic stability of a series of 10-mer DNA duplexes.208 Note that the artificial base pair F5-P is bound by the perfluoroaryl⋯aryl interaction. According to the melting temperature Tm determined by the UV melting experiments, it was found that the duplex stability is largely decreased when a single base-pair F5-P is incorporated into a 10-mer DNA duplex. Two base-pair replacements also lead to a decrease in the duplex stability compared to the unmodified 10-mer DNA duplex, whereas four consecutive base-pair replacements lead to an increase in the thermal stability of the 10-mer DNA duplex. Leumann et al. studied the perfluoroaryl⋯aryl interactions in a series of biphenyl–DNA duplexes.209  Figure 43 also shows the structures of the nucleoside analogues biphenyl-C-nucleoside (bph) and pentafluorobiphenyl-C-nucleoside (5Fbph). Similar to the cases for the artificial base pair F5-P, the Tm data show that the replacement of a single artificial base pair 5Fbph-bph destabilizes the DNA duplex, whereas three and four consecutive base-pair replacements stabilize the DNA duplexes. These results highlight the importance of the perfluoroaryl⋯aryl interactions in the novel design of artificial base pairs. In a recent article, Höbartner and co-workers implemented the perfluorotolane–tolane (THH) interactions as base-pair replacements in the DNA duplexes.210 The structures of tolane (THH) ether and perfluorotolane (TFF) ether are shown in Fig. 44. Evidently, the artificial base-pair TFF-THH is larger than the artificial base pairs F5-P and 5Fbph-bph. The THH/TFF moieties are connected to the acyclic 1,2-propanediol phosphodiester backbones to form a glycol nucleic acid (GNA), and THH/TFF moieties are connected to the acyclic 1,3-butanediol phosphodiester backbones to form a butyl nucleic acid (BuNA) (see Fig. 44). The GNA linker is shorter than the BuNA linker, which leads to obvious difference between the alternative base-pair TFF-THH attached to the GNA backbones and the alternative base-pair TFF-THH attached to the BuNA backbones in stabilizing the DNA duplex shown in Fig. 44. The thermodynamic analyses show that the alternative base-pair TFF-THH stabilizes the DNA duplex when it is attached to the BuNA backbones, but destabilizes the DNA duplex when it is attached to the GNA backbones. This study further demonstrates the potential of using the perfluoroaryl⋯aryl interactions to expand the genetic alphabet.

FIG. 43.

Structures of the nucleoside analogues P, F5, bph, and 5Fbph.

FIG. 43.

Structures of the nucleoside analogues P, F5, bph, and 5Fbph.

Close modal
FIG. 44.

Structures of THH ether, perfluorotolane (TFF) ether, glycol nucleic acid (GNA), and butyl nucleic acid (BuNA). X and Y in the deoxyribonucleic acid (DNA) duplex denote the modified positions.

FIG. 44.

Structures of THH ether, perfluorotolane (TFF) ether, glycol nucleic acid (GNA), and butyl nucleic acid (BuNA). X and Y in the deoxyribonucleic acid (DNA) duplex denote the modified positions.

Close modal

The roles of perfluoroaryl⋯aryl interactions in stabilizing peptides and proteins have been investigated by many studies. Waters et al. investigated quantitatively the effect of the phenylalanine–phenylalanine (F-F) and pentafluorophenylalanine–phenylalanine (f5F-F) interactions on the stabilization of the α-helices.211 Their results showed that the I, I + 4 F-F and f5F-F interactions at the internal positions of the peptides contribute an equivalent −0.27 kcal/mol to the stability of the α-helices, and positioning the I, I + 4 F-F and f5F-F interactions at the C-terminus leads to a much larger stabilization energy (ΔGF-F = −0.8 kcal/mol; ΔGF-f5F = −0.55 kcal/mol). It is unexpected that the f5F-F interactions are weaker than the F-F interactions. This may be because the geometries of the f5F-F interactions are not the optimal face-to-face stacking configurations in the peptide helices.

The villin headpiece subdomain VHP35 is a very good model protein for the study of the role of perfluoroaryl⋯aryl interaction in stabilizing protein structure.212–215 The hydrophobic core of VHP35 consists of three Phe side chains bound by the edge-to-face ππ stacking interactions (see Fig. 45). Gellman and co-workers studied the effects of phenylalanine → pentafluorophenylalanine mutations (Phe → f5-Phe) on the VHP35 stability.212 They found that perfluorination of the Phe-10 side chain, that is to say, the Phe10 → f5-Phe mutation, leads to the stabilization of folded structure of VHP35. However, the Phe6 → f5-Phe, Phe17 → f5-Phe, Phe6,10 → f5-Phe, Phe6,17 → f5-Phe, Phe10,17 → f5-Phe, and Phe6,10,17 → f5-Phe mutations all destabilize the folded structures of VHP35. The subsequent studies from Gao's group showed the complexity involved in analyzing the noncovalent interactions in the hydrophobic core of VHP35.213,215 By inspecting the electrostatic potential maps of Phe-6 and fully fluorinated Phe-10,214 it can be understood that the noncovalent interaction between Phe-6 and f5-Phe10 is not of the edge-to-face ππ stacking interaction. Note that there is an electrostatic repulsion between the F atom and the benzene ring. The noncovalent interaction between Phe-6 and f5-Phe10 is attractive although the two ring planes are not fully stacked. Therefore, such a nonoptimal interaction is still of the perfluoroaryl⋯aryl interaction. The perfectly face-to-face perfluoroaryl⋯aryl interactions have been found in the structure of fluorinated α2D dimer.215–218 As shown in Fig. 46, there are two pairs of identical face-to-face ππ stacking interactions (Phe10Phe29, Phe10Phe29) in the folded structure of α2D dimer. If one of the two Phe residues (for example, Phe29) is mutated by f5-Phe29, there will be two pairs of identical face-to-face perfluoroaryl⋯aryl interactions (Phe10f5-Phe29, Phe10f5-Phe29) in the fluorinated α2D dimer. The single-mutant (Phe → f5-Phe) α2D dimer has a melting temperature of 62.5 °C, nearly 34 °C higher than the wild-type α2D dimer.218 Additionally, the calculated thermodynamic data show that the single-mutant α2D dimer is more stable than the wild-type α2D dimer by about 5.0 kcal/mol.218 The higher thermal stability of the fluorinated α2D dimer was attributed mainly to the attractive perfluoroaryl⋯aryl interactions. On the other hand, Gao and co-workers also found that the perfluoroaryl⋯aryl interactions can direct specific protein–protein interactions: upon mixing the homodimers of the wild-type α2D with the homodimers of the double-mutant (Phe10 → f5-Phe10, Phe29 → f5-Phe29) α2D, the specific heterodimers between the wild-type α2D and the double-mutant α2D are rapidly formed.216 Collectively, these results clearly demonstrate the significant role of the perfluoroaryl⋯aryl interaction in determining the protein stability and protein–protein interaction.

FIG. 45.

The structure of VHP35 (PDB: 1YRF). The residues Phe-6, Phe-10, and Phe-17 are highlighted with the solid licorice representation.

FIG. 45.

The structure of VHP35 (PDB: 1YRF). The residues Phe-6, Phe-10, and Phe-17 are highlighted with the solid licorice representation.

Close modal
FIG. 46.

The structure of the α2D dimer (PDB: 1PQ6). The residues Phe-10 and Phe-29 are highlighted with the solid licorice representation.

FIG. 46.

The structure of the α2D dimer (PDB: 1PQ6). The residues Phe-10 and Phe-29 are highlighted with the solid licorice representation.

Close modal

Perfluoroarenes and their derivatives have emerged as very useful tools for the synthesis, labeling, modification, and conjugation of peptides and proteins.219–226, Figure 47 demonstrates some of the perfluorophenyl derivatives commonly used for the modification and conjugation of peptides and proteins. Antibody–drug conjugates are a rapidly growing class of anticancer agents, and each of them is composed of an antibody covalently bound to a cytotoxic payload via a chemical linker.224 Perfluoroarenes and their derivatives are potential chemical linkers for antibody–drug conjugates.219–230 Theoretically, the perfluoroaryl⋯aryl interactions can be often found in these engineered peptides or proteins containing the perfluoroaryl groups. Unfortunately, the roles of the perfluoroaryl⋯aryl interactions in these systems are seldom studied. Pentelute et al. proposed and developed the π-clamp chemistry for site-selectively modifying peptides and proteins.227–230 The π-clamp is a Phe-Cys-Pro-Phe (FCPF) tetrapeptide sequence, which can be used to site-selectively conjugate its cysteine thiol to a perfluoroaryl compound. As the name “π-clamp” states, the phenyl rings of the two Phe residues play a key role for the selective and efficient cysteine conjugation. Employing molecular dynamics simulations jointly with the density functional theory calculations, Pentelute and co-workers explored the reaction mechanism of the π-clamp-mediated conjugation.227 The other tetrapeptide sequence Gly-Cys-Pro-Gly (GCPG) was used for comparison. Figure 48 shows the calculated free-energy barriers for the nucleophilic aromatic substitution reactions of π-clamp and GCPG. In contrast to the perfluoroarylation reaction at the GCPG, the perfluoroarylation reaction at the π-clamp has a lower free-energy barrier of about 3 kcal/mol and has a much lower product free energy of about 7 kcal/mol. This indicates that the perfluoroarylation reaction at the π-clamp is both kinetically and thermodynamically more favorable than the perfluoroarylation reaction at the GCPG, which is in agreement with the experimental results. The stabilized perfluoroaryl⋯aryl interaction between perfluoroaryl group and Phe4 can be clearly seen in the structure of the product IV in Fig. 48. In the other article that follows in this series, Pentelute et al. carried out a more in-depth investigation on the structure and mechanism of the π-clamp-mediated cysteine perfluoroarylation.229 They built two structural models for the product of the perfluoroarylation reaction at the π-clamp based on the solid-state NMR techniques. There exists an obvious perfluoroaryl⋯aryl interaction between perfluoroaryl group and Phe1 in each of the two structural models. Employing a pyrenylalanine π-clamp mutant, Pentelute and co-workers also investigated the perfluoroaryl⋯aryl interaction between perfluoroaryl probe and pyrenyl side chain in solution. The perfluoroaryl⋯aryl interaction in the cocrystal between hexafluorobenzene and pyrene was reported as early as 2002.118 The perfluoroaryl–pyrenyl interaction is much stronger than the perfluoroaryl–phenyl interaction, due to the larger size of the pyrenyl group compared to the phenyl group. Therefore, the perfluoroaryl–pyrenyl interaction is relatively easier to be detected experimentally in solution compared to the perfluoroaryl–phenyl interaction. In this study, the existence of the perfluoroaryl–pyrenyl interaction in solution was proved by both the 19F solution NMR titration experiments and isothermal titration calorimetry measurements. Along this line, Pentelute et al. further optimized the π-clamp and designed two π-clamp mutants XCPX and XCX'X, in which X is 3-pyrenyl-L-Ala and X' is L-α-methylproline. It was found that the rate constant for the perfluoroarylation reaction at the XCPX is about 43-fold higher than that at the π-clamp, and the rate constant for the perfluoroarylation reaction at the XCX'X is about 85-fold higher than that at the π-clamp. These results further strengthen the importance of the perfluoroaryl⋯aryl interaction in the π-clamp chemistry. Liu and Huang highlighted the work by Pentelute and co-workers in 2016.228 The synthesis of the antibody–drug conjugate was illustrated in Fig. 49. Evidently, only the cysteine thiol in the π-clamp can be site-specifically modified by a perfluoroaryl compound. The antibody–drug conjugate was stabilized by the perfluoroaryl⋯aryl interaction. They pointed out that the discovery of the π-clamp provides a valuable method for site-selective modification of peptides and proteins. The π-clamp chemistry has been applied to the engineering of a series of Cas proteins, such as Cas9, dCas9, Cas12, and Cas13.230 Cheng et al. modified the Cas proteins with the π-clamp FCPF sequence and found that the FCPF-engineered Cas proteins can also be selectively recognized by the perfluoroaryl compounds.230 The engineered Cas proteins can be labeled or degraded by using the perfluoroaryl compounds as linkers. Although the role of the perfluoroaryl⋯aryl interaction was not studied in this work, it is believed that the perfluoroaryl⋯aryl interaction at least promotes the reaction of the site-selective modification of the π-clamp, which is inserted into the Cas protein, as has been revealed by the work of Pentelute et al. As such, this work provides a strong incentive to further study the role of the perfluoroaryl⋯aryl interaction in the bioconjugation field.

FIG. 47.

The perfluorophenyl derivatives commonly used for the modification and conjugation of peptides and proteins.

FIG. 47.

The perfluorophenyl derivatives commonly used for the modification and conjugation of peptides and proteins.

Close modal
FIG. 48.

The calculated free-energy barriers for the nucleophilic aromatic substitution reactions of π-clamp (red) and GCPG (gray). Reproduced with permission from Zhang et al., Nat. Chem. 8, 120–128 (2016). Copyright 2016 Springer Nature.227 

FIG. 48.

The calculated free-energy barriers for the nucleophilic aromatic substitution reactions of π-clamp (red) and GCPG (gray). Reproduced with permission from Zhang et al., Nat. Chem. 8, 120–128 (2016). Copyright 2016 Springer Nature.227 

Close modal
FIG. 49.

The synthesis of the antibody–drug conjugate stabilized by the perfluoroaryl⋯aryl interaction. Reproduced with permission from Y. Huang and L. Liu, Nat. Chem. 8, 101–102 (2016). Copyright 2016 Springer Nature.228 

FIG. 49.

The synthesis of the antibody–drug conjugate stabilized by the perfluoroaryl⋯aryl interaction. Reproduced with permission from Y. Huang and L. Liu, Nat. Chem. 8, 101–102 (2016). Copyright 2016 Springer Nature.228 

Close modal

After more than 60 years of research and development, it is appropriate to review the state-of-the-art of the study of the perfluoroaryl⋯aryl interaction with a brief look into the future. From above discussions, it can be clearly seen that the perfluoroaryl⋯aryl interaction has been applied to many different fields of science and engineering with its unique advantages, which indicates that it is indeed the most important subset of the π-hole⋯π bonding. In a recent article, Jelínek and co-workers reported a direct experimental observation of the π-hole in the molecule 9,10-dichlorooctafluoroanthracene by using the Kelvin probe force microscopy.231 This study confirms the theoretical prediction of the existence of the π-hole and demonstrates the great potential of the scanning probe microscopy for the study of the π-hole and π-hole bonding. The structures and binding energies of some noncovalent interactions have been recently investigated by using the scanning tunneling microscopy-based break junction technique.232–234 Expectedly, the structures and binding energies of the perfluoroaryl⋯aryl interactions can also be determined using such a technique.

According to the SAPT analyses, the perfluoroaryl⋯aryl interactions in most complexes are determined mainly by the dispersion and electrostatics components, and the dispersion term always plays a dominant role. For some charge-transfer complexes between perfluoroarenes and N-alkyl arylamines, it was supposed that the induction energies would play more important roles in stabilizing these complexes.99–101 In addition to the complexes between perfluoroarenes and N-alkyl arylamines, there are many other charge-transfer complexes, which are bound by the perfluoroaryl⋯aryl interactions, and many of them have been applied to organic semiconductors and conductors.235,236 Unfortunately, systematic studies on these so-called charge-transfer complexes are relatively scarce and the key roles of the induction energies in these complexes have not been rigorously proven. For a long time, the term charge transfer has been full of controversy in the field of noncovalent interaction. Systematically theoretical and experimental studies on these charge-transfer complexes are desired in the future.

The noncovalent interactions play crucial roles in separation science and molecular recognition.237 After the use of halogen bonds in the solid-phase extraction (SPE) and possible enantioselective recognition,238,239 the perfluoroaryl⋯aryl interactions were also applied to the design of chromatography stationary phase materials.240,241 Yan et al. synthesized perfluorophenyl-bonded silica sorbent for the SPE of sixteen polycyclic aromatic hydrocarbons in water.240 They found that, in contrast to the traditional octadecyl silica sorbent, the perfluorophenyl-bonded silica sorbent had much higher adsorbabilities for the polycyclic aromatic hydrocarbons. The combination of SPE with high performance liquid chromatography-fluorescence/ultraviolet detection method showed a great success in the analysis of sixteen polycyclic aromatic hydrocarbons in water samples. In addition, compared with the phenyl-bonded silica sorbent, the perfluorobenzene-bonded silica sorbent has obvious advantages in SPE because of the much stronger perfluoroaryl⋯aryl interactions between perfluorobenzene and polycyclic aromatic hydrocarbons. In the other study, Yan and co-workers expanded the scope of application of this method from the water samples to the soil samples.241 The two studies of Yan et al. highlight the remarkable application potential of the perfluoroaryl⋯aryl interaction in the field of analytical chemistry.

The high-pressure chemistry has been a rapidly developing field in recent years. The pressure-induced phase transition and polymerization of perfluoroarene⋯arene cocrystals have been investigated by some researchers.242–248 Marder and co-workers for the first time conducted a detailed study of the pressure evolution of the perfluoroarene⋯arene cocrystals using a combination of experimental and theoretical methods.246 They found that the high-pressure behavior of large perfluoroarene⋯arene cocrystals may be very different from that of small perfluoroarene⋯arene cocrystals.242,243,246 This finding suggests that there is still much to explore in the high-pressure chemistry of the perfluoroaryl⋯aryl interactions.

Nowadays, biomedicine holds a significant and influential position, with its involvement spanning across a wide array of fields, showcasing strong practicality. The great application potential of the perfluoroaryl⋯aryl interaction in the study of the antibody–drug conjugate has been demonstrated in Sec. III. A recent interesting study suggests that the perfluoroaryl⋯aryl interactions can also be utilized to fabricate two-dimensional peptide materials for proliferation and differentiation of muscle cells, which indicates the tremendous potential applications of the perfluoroaryl⋯aryl interactions in the development of next-generation biomaterials.249 The magnetic resonance imaging (MRI) is another promising research area, in which the perfluoroaryl⋯aryl interaction may be involved.250–252 The 19F MRI is gaining widespread interest due to its additional advantages over the conventional 1H MRI. The metal-free fluoropolymers are widely used as the 19F MRI contrast agents, and some of them contain the perfluoroaryl groups.250–252 The 19F MRI contrast agents can be conjugated with targeting agents to achieve in vivo site-specific imaging. Many targeting agents contain the aryl rings. Therefore, it is possible that the 19F MRI contrast agents can bind the targeting agents by the perfluoroaryl⋯aryl interactions. There are currently no reported research findings on this topic. This should be another highly promising and worthy area for in-depth study.

This work was supported by the Natural Science Foundation of Henan Province (No. 232300421147) and the Natural Science Foundation of China (No. 21974011).

The authors have no conflicts to disclose.

Weizhou Wang: Conceptualization (lead); Funding acquisition (lead); Project administration (lead); Writing – original draft (lead); Writing – review & editing (lead). Wen Xin Wu: Conceptualization (supporting); Writing – original draft (supporting); Writing – review & editing (supporting). Yu Zhang: Conceptualization (supporting); Writing – original draft (supporting); Writing – review & editing (supporting). Wei Jun Jin: Conceptualization (lead); Funding acquisition (lead); Project administration (lead); Writing – original draft (lead); Writing – review & editing (lead).

The data that support the findings of this study are available from the corresponding authors upon reasonable request.

1.
D.
O'Hagan
and
D. B.
Harper
, “
Fluorine-containing natural products
,”
J. Fluorine Chem.
100
,
127
133
(
1999
).
2.
I.
Ojima
,
Fluorine in Medicinal Chemistry and Chemical Biology
(
John Wiley & Sons
,
Chichester
,
2009
), pp.
3
523
.
3.
G.
Cavallo
,
P.
Metrangolo
,
R.
Milani
,
T.
Pilati
,
A.
Priimagi
,
G.
Resnati
, and
G.
Terraneo
, “
The halogen bond
,”
Chem. Rev.
116
,
2478
2601
(
2016
).
4.
K.
Reichenbächer
,
H. I.
Süss
, and
J.
Hulliger
, “
Fluorine in crystal engineering–‘the little atom that could
,’ ”
Chem. Soc. Rev.
34
,
22
30
(
2005
).
5.
L. M.
Salonen
,
M.
Ellermann
, and
F.
Diederich
, “
Aromatic rings in chemical and biological recognition: Energetics and structures
,”
Angew. Chem., Int. Ed.
50
,
4808
4842
(
2011
).
6.
C. R.
Patrick
and
G. S.
Prosser
, “
A molecular complex of benzene and hexafluorobenzene
,”
Nature
187
,
1021
1021
(
1960
).
7.
J. S. W.
Overell
and
G. S.
Pawley
, “
An x-ray single-crystal study of the molecular system C6F6·C6D6
,”
Acta Crystallogr., Sect. B
38
,
1966
1972
(
1982
).
8.
J. H.
Williams
,
J. K.
Cockcroft
, and
A. N.
Fitch
, “
Structure of the lowest temperature phase of the solid benzene-hexafluorobenzene adduct
,”
Angew. Chem., Int. Ed. Engl.
31
,
1655
1657
(
1992
).
9.
J.
Vrbancich
and
G. L. D.
Ritchie
, “
Quadrupole moments of benzene, hexafluorobenzene and other non-dipolar aromatic molecules
,”
J. Chem. Soc., Faraday Trans. 2
76
,
648
659
(
1980
).
10.
M. R.
Battaglia
,
A. D.
Buckingham
, and
J. H.
Williams
, “
The electric quadrupole moments of benzene and hexafluorobenzene
,”
Chem. Phys. Lett.
78
,
421
423
(
1981
).
11.
O. R.
Lozman
,
R. J.
Bushby
, and
J. G.
Vinter
, “
Complementary polytopic interactions (CPI) as revealed by molecular modelling using the XED force field
,”
J. Chem. Soc., Perkin Trans 2
2001
,
1446
1452
.
12.
M. W.
Feyereisen
,
D.
Feller
, and
D. A.
Dixon
, “
Hydrogen bond energy of the water dimer
,”
J. Phys. Chem.
100
,
2993
2997
(
1996
).
13.
A. D.
Buckingham
, “
Molecular quadrupole moments
,”
Q. Rev., Chem. Soc.
13
,
183
214
(
1959
).
14.
N. M. D.
Brown
and
F. L.
Swinton
, “
The importance of quadrupolar interactions in determining the structure of solid hydrocarbon–fluorocarbon compounds
,”
J. Chem. Soc., Chem. Commun.
1974
,
770
771
.
15.
J. H.
Williams
, “
The molecular electric quadrupole moment and solid-state architecture
,”
Acc. Chem. Res.
26
,
593
598
(
1993
).
16.
J. H.
Williams
, “
Modeling the vibrational dynamics of solid benzene: Hexafluorobenzene. The anatomy of a phase transition
,”
Chem. Phys.
172
,
171
186
(
1993
).
17.
J. M.
Steed
,
T. A.
Dixon
, and
W.
Klemperer
, “
Molecular beam studies of benzene dimer, hexafluorobenzene dimer, and benzene-hexafluorobenzene
,”
J. Chem. Phys.
70
,
4940
4946
(
1979
).
18.
J. D.
Laposa
,
M. J.
McGlinchey
, and
C.
Montgomery
, “
Infrared and Raman spectra of C6H6C6F6
,”
Spectrochim. Acta
39A
,
863
866
(
1983
).
19.
V. B.
Smith
and
A. G.
Massey
, “
Perfluorophenyl derivatives of the elements—XXII Perfluorotriphenylene
,”
Tetrahedron
25
,
5495
5501
(
1969
).
20.
M.
Weck
,
A. R.
Dunn
,
K.
Matsumoto
,
G. W.
Coates
,
E. B.
Lobkovsky
, and
R. H.
Grubbs
, “
Influence of perfluoroarene⋯arene interactions on the phase behavior of liquid crystalline and polymeric materials
,”
Angew. Chem., Int. Ed.
38
,
2741
2745
(
1999
).
21.
T.
Clark
,
M.
Hennemann
,
J. S.
Murray
, and
P.
Politzer
, “
Halogen bonding: The σ-hole
,”
J. Mol. Model.
13
,
291
296
(
2007
).
22.
P.
Politzer
,
J. S.
Murray
, and
T.
Clark
, “
Halogen bonding: An electrostatically-driven highly directional noncovalent interaction
,”
Phys. Chem. Chem. Phys.
12
,
7748
7757
(
2010
).
23.
J. S.
Murray
,
P.
Lane
,
T.
Clark
,
K. E.
Riley
, and
P.
Politzer
, “
σ-holes, π-holes and electrostatically-driven interactions
,”
J. Mol. Model.
18
,
541
548
(
2012
).
24.
P.
Politzer
,
J. S.
Murray
,
T.
Clark
, and
G.
Resnati
, “
The σ-hole revisited
,”
Phys. Chem. Chem. Phys.
19
,
32166
32178
(
2017
).
25.
P.
Politzer
and
J. S.
Murray
, “
An overview of strengths and directionalities of noncovalent interactions: σ-holes and π-holes
,”
Crystals
9
,
165
(
2019
).
26.
P.
Politzer
,
J. S.
Murray
, and
T.
Clark
, “
The π-hole revisited
,”
Phys. Chem. Chem. Phys.
23
,
16458
16468
(
2021
).
27.
M. J.
Frisch
,
G. W.
Trucks
,
H. B.
Schlegel
et al, Gaussian 16, Revision C.01; Gaussian, Inc., Wallingford, CT,
2019
.
28.
R. F.
Stewart
, “
On the mapping of electrostatic properties from Bragg diffraction data
,”
Chem. Phys. Lett.
65
,
335
342
(
1979
).
29.
J. S.
Murray
and
P.
Politzer
, “
The electrostatic potential: An overview
,”
WIREs Comput. Mol. Sci.
1
,
153
163
(
2011
).
30.
S. E.
Wheeler
and
K. N.
Houk
, “
Through-space effects of substituents dominate molecular electrostatic potentials of substituted arenes
,”
J. Chem. Theory Comput.
5
,
2301
2312
(
2009
).
31.
P.
Politzer
and
J. S.
Murray
, “
σ-hole interactions: Perspectives and misconceptions
,”
Crystals
7
,
212
(
2017
).
32.
J. S.
Murray
and
P.
Politzer
, “
Interaction and polarization energy relationships in σ-hole and π-hole bonding
,”
Crystals
10
,
76
(
2020
).
33.
X.
Pang
,
H.
Wang
,
X. R.
Zhao
, and
W. J.
Jin
, “
Co−crystallization turned on the phosphorescence of phenanthrene by C−Br···π halogen bonding, π-hole···π bonding and other assisting interactions
,”
CrystEngComm
15
,
2722
2730
(
2013
).
34.
V. d P. N.
Nziko
and
S.
Scheiner
, “
Comparison of π-hole tetrel bonding with σ-hole halogen bonds in complexes of XCN (X = F, Cl, Br, I) and NH3
,”
Phys. Chem. Chem. Phys.
18
,
3581
3590
(
2016
).
35.
H.
Wang
,
W.
Wang
, and
W. J.
Jin
, “
σ-hole bond vs π-hole bond: A comparison based on halogen bond
,”
Chem. Rev.
116
,
5072
5104
(
2016
).
36.
T.
Clark
,
P.
Politzer
, and
J. S.
Murray
, “
Correct electrostatic treatment of noncovalent interactions: The importance of polarization
,”
WIREs Comput. Mol. Sci.
5
,
169
177
(
2015
).
37.
P.
Politzer
and
J. S.
Murray
, “
σ-holes and π-holes: Similarities and differences
,”
J. Comput. Chem.
39
,
464
471
(
2018
).
38.
P.
Gamez
,
T. J.
Mooibroek
,
S. J.
Teat
, and
J.
Reedijk
, “
Anion binding involving π-acidic heteroaromatic rings
,”
Acc. Chem. Res.
40
,
435
444
(
2007
).
39.
B. L.
Schottel
,
H. T.
Chifotides
, and
K. R.
Dunbar
, “
Anion-π interactions
,”
Chem. Soc. Rev.
37
,
68
83
(
2008
).
40.
H. T.
Chifotides
and
K. R.
Dunbar
, “
Anion-π interactions in supramolecular architectures
,”
Acc. Chem. Res.
46
,
894
906
(
2013
).
41.
S. E.
Wheeler
and
K. N.
Houk
, “
Are anion/π interactions actually a case of simple charge−dipole interactions?
,”
J. Phys. Chem. A
114
,
8658
8664
(
2010
).
42.
S. E.
Wheeler
and
J. W. G.
Bloom
, “
Anion-π interactions and positive electrostatic potentials of N-heterocycles arise from the positions of the nuclei, not changes in the π-electron distribution
,”
Chem. Commun.
50
,
11118
11121
(
2014
).
43.
R.
Wysokiński
,
W.
Zierkiewicz
,
M.
Michalczyk
, and
S.
Scheiner
, “
Anion···anion attraction in complexes of MCl3 (M=Zn, Cd, Hg) with CN
,”
ChemPhysChem
21
,
1119
1125
(
2020
).
44.
R.
Wysokiński
,
W.
Zierkiewicz
,
M.
Michalczyk
, and
S.
Scheiner
, “
Crystallographic and theoretical evidences of anion···anion interaction
,”
ChemPhysChem
22
,
818
821
(
2021
).
45.
R.
Wysokiński
,
W.
Zierkiewicz
,
M.
Michalczyk
, and
S.
Scheiner
, “
Anion···anion (MX3)2 dimers (M=Zn, Cd, Hg; X=Cl, Br, I) in different environments
,”
Phys. Chem. Chem. Phys.
23
,
13853
13861
(
2021
).
46.
A.
Grabarz
,
M.
Michalczyk
,
W.
Zierkiewicz
, and
S.
Scheiner
, “
Anion–anion interactions in aerogen-bonded complexes. Influence of solvent environment
,”
Molecules
26
,
2116
(
2021
).
47.
B. V.
Pandiyan
,
P.
Deepa
, and
P.
Kolandaivel
, “
Does the occurrence of resonance (by the delocalization of radical/cationic/anionic charges) induce the existence of intramolecular halogen–halogen contacts?
,”
RSC Adv.
6
,
66870
66878
(
2016
).
48.
M.
Mascal
,
A.
Armstrong
, and
M. D.
Bartberger
, “
Anion−aromatic bonding: A case for anion recognition by π−acidic rings
,”
J. Am. Chem. Soc.
124
,
6274
6276
(
2002
).
49.
B. L.
Schottel
,
H. T.
Chifotides
,
M.
Shatruk
,
A.
Chouai
,
L. M.
Pérez
,
J.
Bacsa
, and
K. R.
Dunbar
, “
Anion–π interactions as controlling elements in self-assembly reactions of Ag(I) complexes with π-acidic aromatic rings
,”
J. Am. Chem. Soc.
128
,
5895
5912
(
2006
).
50.
P.
Ballester
, “
Anions and π-aromatic systems. Do they interact attractively?
,”
Struct. Bonding
129
,
127
174
(
2008
).
51.
S. M.
Tekarli
,
T. R.
Cundari
, and
M. A.
Omary
, “
Rational design of macrometallocyclic trinuclear complexes with superior π-acidity and π-basicity
,”
J. Am. Chem. Soc.
130
,
1669
1675
(
2008
).
52.
O. B.
Berryman
,
V. S.
Bryantsev
,
D. P.
Stay
,
D. W.
Johnson
, and
B. P.
Hay
, “
Structural criteria for the design of anion receptors:  The interaction of halides with electron-deficient arenes
,”
J. Am. Chem. Soc.
129
,
48
58
(
2007
).
53.
D. X.
Wang
,
Q. Y.
Zheng
,
Q. Q.
Wang
, and
M. X.
Wang
, “
Halide recognition by tetraoxacalix[2]arene[2]triazine receptors: Concurrent noncovalent halide–π and lone-pair–π interactions in host–halide–water ternary complexes
,”
Angew. Chem., Int. Ed.
47
,
7485
7488
(
2008
).
54.
G. R.
Desiraju
,
P. S.
Ho
,
L.
Kloo
,
A. C.
Legon
,
R.
Marquardt
,
P.
Metrangolo
,
P.
Politzer
,
G.
Resnati
, and
K.
Rissanen
, “
Definition of the halogen bond
,”
Pure Appl. Chem.
85
,
1711
1713
(
2013
).
55.
C. B.
Aakeroy
,
D. L.
Bryce
,
G. R.
Desiraju
,
A.
Frontera
,
A. C.
Legon
,
F.
Nicotra
,
K.
Rissanen
,
S.
Scheiner
,
G.
Terraneo
,
P.
Metrangolo
, and
G.
Resnati
, “
Definition of the chalcogen bond
,”
Pure Appl. Chem.
91
,
1889
1892
(
2019
).
56.
M.
Giese
,
M.
Albrecht
, and
K.
Rissanen
, “
Anion–π interactions with fluoroarenes
,”
Chem. Rev.
115
,
8867
8895
(
2015
).
57.
M.
Savastano
,
C.
García-Gallarín
,
M. D.
López de la Torre
,
C.
Bazzicalupi
,
A.
Bianchi
, and
M.
Melguizo
, “
Anion-π and lone pair-π interactions with s-tetrazine-based ligands
,”
Coord. Chem. Rev.
397
,
112
137
(
2019
).
58.
D. X.
Wang
and
M. X.
Wang
, “
Exploring anion–π interactions and their applications in supramolecular chemistry
,”
Acc. Chem. Res.
53
,
1364
1380
(
2020
).
59.
M.
Giese
,
M.
Albrecht
,
A.
Valkonen
, and
K.
Rissanen
, “
The pentafluorophenyl group as π-acceptor for anions: A case study
,”
Chem. Sci.
6
,
354
359
(
2015
).
60.
R.
Plais
,
G.
Clavier
,
J.-Y.
Salpin
,
A. A.
Gaucher
, and
D.
Prim
, “
Anion-π interaction for molecular recognition of anions: Focus on cooperativity with hydrogen bonding
,”
Eur. J. Org. Chem.
26
,
e202201281
(
2023
).
61.
I.
Alkorta
,
I.
Rozas
, and
J.
Elguero
, “
Interaction of anions with perfluoro aromatic compounds
,”
J. Am. Chem. Soc.
124
,
8593
8598
(
2002
).
62.
C.
Garau
,
A.
Frontera
,
D.
Quiñonero
,
P.
Ballester
,
A.
Costa
, and
P. M.
Deyà
, “
A topological analysis of the electron density in anion-π interactions
,”
ChemPhysChem
4
,
1344
1348
(
2003
).
63.
D.
Kim
,
P.
Tarakeshwar
, and
K. S.
Kim
, “
Theoretical investigations of anion-π interactions: The role of anions and the nature of π systems
,”
J. Phys. Chem. A
108
,
1250
1258
(
2004
).
64.
M.
Egli
and
S.
Sarkhel
, “
Lone pair–aromatic interactions: To stabilize or not to stabilize
,”
Acc. Chem. Res.
40
,
197
205
(
2007
).
65.
T. J.
Mooibroek
,
P.
Gamez
, and
J.
Reedijk
, “
Lone pair–π interactions: A new supramolecular bond?
,”
CrystEngComm
10
,
1501
1515
(
2008
).
66.
J.
Novotný
,
S.
Bazzi
,
R.
Marek
, and
J.
Kozelka
, “
Lone-pair–π interactions: Analysis of the physical origin and biological implications
,”
Phys. Chem. Chem. Phys.
18
,
19472
19481
(
2016
).
67.
J.
Kozelka
, “
Lone pair–π interactions in biological systems: Occurrence, function, and physical origin
,”
Eur. Biophys. J.
46
,
729
737
(
2017
).
68.
G. Y.
Lee
,
E.
Hu
,
A. L.
Rheingold
,
K. N.
Houk
, and
E. M.
Sletten
, “
Arene-perfluoroarene interactions in solution
,”
J. Org. Chem.
86
,
8425
8436
(
2021
).
69.
Z.
Huang
,
X.
Chen
,
G.
Wu
,
P.
Metrangolo
,
D.
Whitaker
,
J. A.
McCune
, and
O. A.
Scherman
, “
Host-enhanced phenyl-perfluorophenyl polar-π interactions
,”
J. Am. Chem. Soc.
142
,
7356
7361
(
2020
).
70.
M. J. S.
Phipps
,
T.
Fox
,
C. S.
Tautermann
, and
C.-K.
Skylaris
, “
Energy decomposition analysis approaches and their evaluation on prototypical protein–drug interaction patterns
,”
Chem. Soc. Rev.
44
,
3177
3211
(
2015
).
71.
J.
Andrés
,
P. W.
Ayers
,
R. A.
Boto
,
R.
Carbó-Dorca
,
H.
Chermette
,
J.
Cioslowski
,
J.
Contreras-García
,
D. L.
Cooper
,
G.
Frenking
,
C.
Gatti
,
F.
Heidar-Zadeh
,
L.
Joubert
,
Á. M.
Pendás
,
E.
Matito
,
I.
Mayer
,
A. J.
Misquitta
,
Y.
Mo
,
J.
Pilmé
,
P. L. A.
Popelier
,
M.
Rahm
,
E.
Ramos-Cordoba
,
P.
Salvador
,
W. H. E.
Schwarz
,
S.
Shahbazian
,
B.
Silvi
,
M.
Solà
,
K.
Szalewicz
,
V.
Tognetti
,
F.
Weinhold
, and
É.-L.
Zins
, “
Nine questions on energy decomposition analysis
,”
J. Comput. Chem.
40
,
2248
2283
(
2019
).
72.
K.
Kitaura
and
K.
Morokuma
, “
A new energy decomposition scheme for molecular interactions within the Hartree–Fock approximation
,”
Int. J. Quantum Chem.
10
,
325
340
(
1976
).
73.
K.
Szalewicz
and
B.
Jeziorski
, “
Symmetry-adapted double-perturbation analysis of intramolecular correlation effects in weak intermolecular interactions
,”
Mol. Phys.
38
,
191
208
(
1979
).
74.
M.
Rahm
and
R.
Hoffmann
, “
Distinguishing bonds
,”
J. Am. Chem. Soc.
138
,
3731
3744
(
2016
).
75.
I.
Mayer
, “
Towards a “chemical” Hamiltonian
,”
Int. J. Quantum Chem.
23
,
341
363
(
1983
).
76.
M.
Rahm
and
R.
Hoffmann
, “
Toward an experimental quantum chemistry: Exploring a new energy partitioning
,”
J. Am. Chem. Soc.
137
,
10282
10291
(
2015
).
77.
B.
Jeziorski
,
R.
Moszynski
, and
K.
Szalewicz
, “
Perturbation theory approach to intermolecular potential energy surfaces of van der Waals complexes
,”
Chem. Rev.
94
,
1887
1930
(
1994
).
78.
K.
Szalewicz
, “
Symmetry-adapted perturbation theory of intermolecular forces
,”
WIREs Comput. Mol. Sci.
2
,
254
272
(
2012
).
79.
E. G.
Hohenstein
and
C. D.
Sherrill
, “
Density fitting and Cholesky decomposition approximations in symmetry-adapted perturbation theory: Implementation and application to probe the nature of π-π interactions in linear acenes
,”
J. Chem. Phys.
132
,
184111
(
2010
).
80.
E. G.
Hohenstein
and
C. D.
Sherrill
, “
Density fitting of intramonomer correlation effects in symmetry-adapted perturbation theory
,”
J. Chem. Phys.
133
,
014101
(
2010
).
81.
T. H.
Dunning
, Jr.
, “
Gaussian basis sets for use in correlated molecular calculations. I. The atoms boron through neon and hydrogen
,”
J. Chem. Phys.
90
,
1007
1023
(
1989
).
82.
C. D.
Sherrill
, “
Energy component analysis of π interactions
,”
Acc. Chem. Res.
46
,
1020
1028
(
2013
).
83.
W.
Wang
,
Y.
Zhang
, and
Y.-B.
Wang
, “
Noncovalent π···π interaction between graphene and aromatic molecule: Structure, energy, and nature
,”
J. Chem. Phys.
140
,
094302
(
2014
).
84.
W.
Wang
,
T.
Sun
,
Y.
Zhang
, and
Y.-B.
Wang
, “
The benzene···naphthalene complex: A more challenging system than the benzene dimer for newly developed computational methods
,”
J. Chem. Phys.
143
,
114312
(
2015
).
85.
G.
Jansen
, “
Symmetry-adapted perturbation theory based on density functional theory for noncovalent interactions
,”
WIREs Comput. Mol. Sci.
4
,
127
144
(
2014
).
86.
K.
Patkowski
, “
Recent developments in symmetry-adapted perturbation theory
,”
WIREs Comput. Mol. Sci.
10
,
e1452
(
2020
).
87.
M. O.
Sinnokrot
and
C. D.
Sherrill
, “
Highly accurate coupled cluster potential energy curves for the benzene dimer: Sandwich, T-shaped, and parallel-displaced configurations
,”
J. Phys. Chem. A
108
,
10200
10207
(
2004
).
88.
S.
Tsuzuki
,
T.
Uchimaru
, and
M.
Mikami
, “
Intermolecular interaction between hexafluorobenzene and benzene: Ab initio calculations including CCSD(T) level electron correlation correction
,”
J. Phys. Chem. A
110
,
2027
2033
(
2006
).
89.
W.
Wang
,
Y.
Zhang
, and
Y.-B.
Wang
, “
Highly accurate benchmark calculations of the interaction energies in the complexes C6H6···C6X6 (X = F, Cl, Br, and I)
,”
Int. J. Quantum Chem.
117
,
e25345
(
2017
).
90.
A. J.
Stone
, “
Distributed polarizabilities
,”
Mol. Phys.
56
,
1065
1082
(
1985
).
91.
K.
Pluháčková
,
P.
Jurečka
, and
P.
Hobza
, “
Stabilisation energy of C6H6···C6X6 (X = F, Cl, Br, I, CN) complexes: Complete basis set limit calculations at MP2 and CCSD(T) levels
,”
Phys. Chem. Chem. Phys.
9
,
755
760
(
2007
).
92.
S.
Ehrlich
,
J.
Moellmann
, and
S.
Grimme
, “
Dispersion-corrected density functional theory for aromatic interactions in complex systems
,”
Acc. Chem. Res.
46
,
916
926
(
2013
).
93.
M.
Linnemannstöns
,
J.
Schwabedissen
,
A. A.
Schultz
,
B.
Neumann
,
H.-G.
Stammler
,
R. J. F.
Berger
, and
N. W.
Mitzel
, “
London dispersion-driven hetero-aryl⋯aryl interactions in 1,2-diaryldisilanes
,”
Chem. Commun.
56
,
2252
2255
(
2020
).
94.
C. A.
Hunter
and
J. K. M.
Sanders
, “
The nature of π–π interactions
,”
J. Am. Chem. Soc.
112
,
5525
5534
(
1990
).
95.
K.
Carter-Fenk
and
J. M.
Herbert
, “
Electrostatics does not dictate the slip-stacked arrangement of aromatic π–π interactions
,”
Chem. Sci.
11
,
6758
6765
(
2020
).
96.
J. M.
Herbert
, “
Neat, simple, and wrong: Debunking electrostatic fallacies regarding noncovalent interactions
,”
J. Phys. Chem. A
125
,
7125
7137
(
2021
).
97.
K.
Carter-Fenk
,
K. U.
Lao
, and
J. M.
Herbert
, “
Predicting and understanding non-covalent interactions using novel forms of symmetry-adapted perturbation theory
,”
Acc. Chem. Res.
54
,
3679
3690
(
2021
).
98.
J.
Henrichsmeyer
,
M.
Thelen
,
M.
Bröckel
,
M.
Fadel
,
S.
Behnle
,
M.
Sekkal-Rahal
, and
R. F.
Fink
, “
Rationalizing aggregate structures with orbital contributions to the exchange-repulsion energy
,”
ChemPhysChem
24
,
e202300097
(
2023
).
99.
B. W.
Gung
and
J. C.
Amicangelo
, “
Substituent effects in C6F6–C6H5X stacking interactions
,”
J. Org. Chem.
71
,
9261
9270
(
2006
).
100.
T. G.
Beaumont
and
K. M. C.
Davis
, “
Multiple charge transfer bands in complexes of hexafluorobenzene with aromatic amines
,”
Nature
218
,
865
(
1968
).
101.
B. W.
Gung
,
M.
Patel
, and
X.
Xue
, “
A threshold for charge transfer in aromatic interactions? A quantitative study of π-stacking interactions
,”
J. Org. Chem.
70
,
10532
10537
(
2005
).
102.
X.
Pang
,
H.
Wang
,
W.
Wang
, and
W. J.
Jin
, “
Phosphorescent π-hole···π bonding cocrystals of pyrene with haloperfluorobenzenes (F, Cl, Br, I)
,”
Cryst. Growth Des.
15
,
4938
4945
(
2015
).
103.
L.
Li
,
H.
Wang
,
W.
Wang
, and
W. J.
Jin
, “
Interactions between haloperfluorobenzenes and fluoranthene in luminescent cocrystals from π-hole···π to σ-hole···π bonds
,”
CrystEngComm
19
,
5058
5067
(
2017
).
104.
W.
Wang
,
T.
Sun
,
Y.
Zhang
, and
Y.-B.
Wang
, “
Benchmark calculations of the adsorption of aromatic molecules on graphene
,”
J. Comput. Chem.
36
,
1763
1771
(
2015
).
105.
P.
Politzer
,
J. S.
Murray
, and
T.
Clark
, “
Mathematical modeling and physical reality in noncovalent interactions
,”
J. Mol. Model.
21
,
52
(
2015
).
106.
T.
Clark
, “
Polarization, donor–acceptor interactions, and covalent contributions in weak interactions: A clarification
,”
J. Mol. Model.
23
,
297
(
2017
).
107.
J. S.
Murray
and
P.
Politzer
, “
Molecular electrostatic potentials and noncovalent interactions
,”
WIREs Comput. Mol. Sci.
7
,
e1326
(
2017
).
108.
T.
Clark
,
J. S.
Murray
, and
P.
Politzer
, “
A perspective on quantum mechanics and chemical concepts in describing noncovalent interactions
,”
Phys. Chem. Chem. Phys.
20
,
30076
30082
(
2018
).
109.
R. F. W.
Bader
, “
Comment on the comparative use of the electron density and its Laplacian
,”
Chem. - Eur. J.
12
,
7769
7772
(
2006
).
110.
G.
Frenking
,
C.
Esterhuysen
, and
A.
Kovacs
, “
Reply to “Comment on the comparative use of the electron density and its Laplacian
,”
Chem. - Eur. J.
12
,
7773
7774
(
2006
).
111.
H.
Hellmann
,
Einführung in Die Quantenchemie
(
Deuticke
,
Leipzig
,
1937
), p.
285
.
112.
R. P.
Feynman
, “
Forces in molecules
,”
Phys. Rev.
56
,
340
343
(
1939
).
113.
S. M.
Huber
,
E.
Jimenez-Izal
,
J. M.
Ulgalde
, and
I.
Infante
, “
Unexpected trends in halogen-bond based noncovalent adducts
,”
Chem. Commun.
48
,
7708
7710
(
2012
).
114.
J.
Thirman
,
E.
Engelage
,
S. M.
Huber
, and
M.
Head-Gordon
, “
Characterizing the interplay of Pauli repulsion, electrostatics, dispersion and charge transfer in halogen bonding with energy decomposition analysis
,”
Phys. Chem. Chem. Phys.
20
,
905
915
(
2018
).
115.
C.
Wang
,
D.
Danovich
,
S.
Shaik
,
W.
Wu
, and
Y.
Mo
, “
Attraction between electrophilic caps: A counterintuitive case of noncovalent interactions
,”
J. Comput. Chem.
40
,
1015
1022
(
2019
).
116.
T.
Clark
and
A.
Heβelmann
, “
The coulombic σ-hole model describes bonding in CX3I···Y complexes completely
,”
Phys. Chem. Chem. Phys.
20
,
22849
22855
(
2018
).
117.
J. S.
Murray
,
P. G.
Seybold
, and
P.
Politzer
, “
The many faces of fluorine: Some noncovalent interactions of fluorine compounds
,”
J. Chem. Thermodyn.
156
,
106382
(
2021
).
118.
J. C.
Collings
,
K. P.
Roscoe
,
E. G.
Robins
,
A. S.
Batsanov
,
L. M.
Stimson
,
J. A. K.
Howard
,
S. J.
Clark
, and
T. B.
Marder
, “
Arene-perfluoroarene interactions in crystal engineering 8: Structures of 1:1 complexes of hexafluorobenzene with fused-ring polyaromatic hydrocarbons
,”
New J. Chem.
26
,
1740
1746
(
2002
).
119.
J. C.
Collings
,
K. P.
Roscoe
,
R. L.
Thomas
,
A. S.
Batsanov
,
L. M.
Stimson
,
J. A. K.
Howard
, and
T. B.
Marder
, “
Arene-perfluoroarene interactions in crystal engineering. Part 3. Single-crystal structures of 1:1 complexes of octafluoronaphthalene with fused-ring polyaromatic hydrocarbons
,”
New J. Chem.
25
,
1410
1417
(
2001
).
120.
P.
Politzer
,
J. S.
Murray
, and
T.
Clark
, “
Explicit inclusion of polarizing electric fields in σ- and π-hole interactions
,”
J. Phys. Chem. A
123
,
10123
10130
(
2019
).
121.
P.
Politzer
and
J. S.
Murray
, “
Electrostatics and polarization in σ- and π-hole noncovalent interactions: An overview
,”
ChemPhysChem
21
,
579
588
(
2020
).
122.
S. E.
Wheeler
, “
Local nature of substituent effects in stacking interactions
,”
J. Am. Chem. Soc.
133
,
10262
10274
(
2011
).
123.
R. K.
Raju
,
J. W. G.
Bloom
, and
S. E.
Wheeler
, “
Broad transferability of substituent effects in π-stacking interactions provides new insights into their origin
,”
J. Chem. Theory Comput.
9
,
3479
3490
(
2013
).
124.
S. E.
Wheeler
and
J. W. G.
Bloom
, “
Toward a more complete understanding of noncovalent interactions involving aromatic rings
,”
J. Phys. Chem. A
118
,
6133
6147
(
2014
).
125.
S. E.
Wheeler
, “
Unraveling the origin of substituents effects in π-stacking interactions
,” in
Noncovalent Forces, Challenges and Advances in Computational Chemistry and Physics 19
, edited by
S.
Scheiner
(
Springer International Publishing
,
Cham
,
2015
), Chap. 14, pp.
421
442
.
126.
S. E.
Wheeler
, “
Understanding substituent effects in noncovalent interactions involving aromatic rings
,”
Acc. Chem. Res.
46
,
1029
1038
(
2013
).
127.
J.-I.
Seo
,
I.
Kim
, and
Y. S.
Lee
, “
π–π Interaction energies in monosubstituted-benzene dimers in parallel- and antiparallel-displaced conformations
,”
Chem. Phys. Lett.
474
,
101
106
(
2009
).
128.
E. C.
Lee
,
D.
Kim
,
P.
Jurečka
,
P.
Tarakeshwar
,
P.
Hobza
, and
K. S.
Kim
, “
Understanding of assembly phenomena by aromatic–aromatic interactions: Benzene dimer and the substituted systems
,”
J. Phys. Chem. A
111
,
3446
3457
(
2007
).
129.
M.
Watt
,
L. K. E.
Hardebeck
,
C. C.
Kirkpatrick
, and
M.
Lewis
, “
Face-to-face arene⋯arene binding energies: Dominated by dispersion but predicted by electrostatic and dispersion/polarizability substituent constants
,”
J. Am. Chem. Soc.
133
,
3854
3862
(
2011
).
130.
R. M.
Parrish
and
C. D.
Sherrill
, “
Quantum-mechanical evaluation of π–π versus substituent–π interactions in π stacking: Direct evidence for the Wheeler–Houk picture
,”
J. Am. Chem. Soc.
136
,
17386
17389
(
2014
).
131.
T.
Dahl
, “
Crystal structure of the 1:1 complex between mesitylene and hexafluorobenzene
,”
Acta Chem. Scand.
25
,
1031
1039
(
1971
).
132.
T.
Dahl
, “
Crystallographic studies of molecular complexes containing hexafluorobenzene
,”
Acta Chem. Scand., Ser. A
42
,
1
7
(
1988
).
133.
J. C.
Collings
,
A. S.
Batsanov
,
J. A. K.
Howard
, and
T. B.
Marder
, “
Octafluoronaphthalene–diphenyl-acetylene (1/1)
,”
Acta Crystallogr., Sect. C
57
,
870
872
(
2001
).
134.
A. S.
Batsanov
,
J. A. K.
Howard
,
T. B.
Marder
, and
E. G.
Robins
, “
Arene–perfluoroarene interactions in crystal engineering. 4. Hexafluorobenzene–trans-stilbene (1/1)
,”
Acta Crystallogr., Sect. C
57
,
1303
1305
(
2001
).
135.
A. S.
Batsanov
,
J. C.
Collings
,
J. A. K.
Howard
, and
T. B.
Marder
, “
Octafluoronaphthalene–1,8-diaminonaphthalene (1/1)
,”
Acta Crystallogr., Sect. E
57
,
o950
o952
(
2001
).
136.
J. C.
Collings
,
A. S.
Batsanov
,
J. A. K.
Howard
, and
T. B.
Marder
, “
Arene-perfluoroarene interactions in crystal engineering 7: Single-crystal structures of 1:1 complexes of octafluoronaphthalene and hexafluorobenzene with acenaphthene
,”
Cryst. Eng.
5
,
37
46
(
2002
).
137.
C. E.
Smith
,
P. S.
Smith
,
R. Ll.
Thomas
,
E. G.
Robins
,
J. C.
Collings
,
C.
Dai
,
A. J.
Scott
,
S.
Borwick
,
A. S.
Batsanov
,
S. W.
Watt
,
S. J.
Clark
,
C.
Viney
,
J. A. K.
Howard
,
W.
Clegg
, and
T. B.
Marder
, “
Arene-perfluoroarene interactions in crystal engineering: Structural preferences in polyfluorinated tolans
,”
J. Mater. Chem.
14
,
413
420
(
2004
).
138.
J. C.
Collings
,
P. S.
Smith
,
D. S.
Yufit
,
A. S.
Batsanov
,
J. A. K.
Howard
, and
T. B.
Marder
, “
Arene–perfluoroarene interactions in crystal engineering. Part 10. Crystal structures of 1:1 complexes of octafluoronaphthalene with biphenyl and biphenylene
,”
CrystEngComm
6
,
25
28
(
2004
).
139.
S. W.
Watt
,
C.
Dai
,
A. J.
Scott
,
J. M.
Burke
,
R. L.
Thomas
,
J. C.
Collings
,
C.
Viney
,
W.
Clegg
, and
T. B.
Marder
, “
Structure and phase behavior of a 2:1 complex between arene- and fluoroarene-based conjugated rigid rods
,”
Angew. Chem., Int. Ed.
43
,
3061
3063
(
2004
).
140.
J. C.
Collings
,
A. S.
Batsanov
,
J. A. K.
Howard
,
D. A.
Dickie
,
J. A. C.
Clyburne
,
H. A.
Jenkins
, and
T. B.
Marder
, “
1:1 Complexes of octafluoronaphthalene with trans-stilbene and trans-azobenzene
,”
J. Fluorine Chem.
126
,
513
517
(
2005
).
141.
J. C.
Collings
,
A. S.
Batsanov
,
J. A. K.
Howard
, and
T. B.
Marder
, “
Arene–perfluoroarene interactions in crystal engineering. Part 14. 1:1 Complexes of octafluoronaphthalene with fluorene and 9,10-dihydrophenanthrene
,”
Can. J. Chem.
84
,
238
242
(
2006
).
142.
A. S.
Batsanov
,
J. C.
Collings
, and
T. B.
Marder
, “
Arene–perfluoroarene interactions in crystal engineering. XV. Ferrocene–decafluorobiphenyl (1/1)
,”
Acta Crystallogr., Sect. C
62
,
m229
m231
(
2006
).
143.
A. S.
Batsanov
,
J. A. K.
Howard
,
D.
Albesa-Jové
,
J. C.
Collings
,
Z.
Liu
,
I. A. I.
Mkhalid
,
M.-H.
Thibault
, and
T. B.
Marder
, “
Structural versatility of pyrene-2–(4,4,5,5-tetramethyl-[1,3,2]dioxaborolane) and pyrene-2,7-bis(4,4,5,5-tetramethyl-[1,3,2]dioxaborolane)
,”
Cryst. Growth Des.
12
,
2794
2802
(
2012
).
144.
G. W.
Coates
,
A. R.
Dunn
,
L. M.
Henling
,
D. A.
Dougherty
, and
R. H.
Grubbs
, “
Phenyl−perfluorophenyl stacking interactions: A new strategy for supermolecule construction
,”
Angew. Chem., Int. Ed.
36
,
248
251
(
1997
).
145.
T.
Steiner
, “
The hydrogen bond in the solid state
,”
Angew. Chem., Int. Ed.
41
,
48
76
(
2002
).
146.
S.
Meejoo
,
B. M.
Kariuki
, and
K. D. M.
Harris
, “
The interplay of aryl⋯perfluoroaryl stacking interactions and interstack hydrogen bonding in controlling the structure of a molecular cocrystal
,”
ChemPhysChem
4
,
766
769
(
2003
).
147.
J.
Alfuth
,
J.
Chojnacki
,
T.
Połoński
,
A.
Herman
,
M. J.
Milewska
, and
T.
Olszewska
, “
Interplay between aryl⋅⋅⋅perfluoroaryl and hydrogen bonding interactions in cocrystals of pentafluorophenol with molecules of trigonal symmetry
,”
Cryst. Growth Des.
22
,
3493
3504
(
2022
).
148.
M.
Gdaniec
,
W.
Jankowski
,
M. J.
Milewska
, and
T.
Połoński
, “
Supramolecular assemblies of hydrogen-bonded carboxylic acid dimers mediated by phenyl–pentafluorophenyl stacking interactions
,”
Angew. Chem., Int. Ed.
42
,
3903
3906
(
2003
).
149.
B.
Piotrkowska
,
M.
Gdaniec
,
M. J.
Milewska
, and
T.
Połoński
, “
Aryl–perfluoroaryl stacking interactions, hydrogen bonding and steric effects in controlling the structure of supramolecular assemblies of N,N'-diaryloxalamides
,”
CrystEngComm
9
,
868
872
(
2007
).
150.
J.
Alfuth
,
K.
Kazimierczuk
,
T.
Połoński
, and
T.
Olszewska
, “
Intermolecular hydrogen bonding directed by aryl⋯perfluoroaryl π–π stacking interactions
,”
Cryst. Growth Des.
23
,
6830
6839
(
2023
).
151.
J.
Alfuth
,
K.
Kazimierczuk
,
T.
Połoński
, and
T.
Olszewska
, “
Self-assembly of secondary aryl amides in cocrystals with perfluoroaryl coformers: Structural analysis and synthon preferences
,”
CrystEngComm
25
,
1403
1410
(
2023
).
152.
G.
Campillo-Alvarado
,
M. M.
D'mello
,
M. A.
Sinnwell
,
H.
Höpfl
,
H.
Morales-Rojas
, and
L. R.
MacGillivray
, “
Channel confinement of aromatic petrochemicals via aryl⋯perfluoroaryl interactions with a B←N host
,”
Front. Chem.
7
,
695
(
2019
).
153.
M. P.
Zhuo
,
G. P.
He
,
Y.
Yuan
,
Y. C.
Tao
,
G. Q.
Wei
,
X. D.
Wang
,
S. T.
Lee
, and
L. S.
Liao
, “
Super-stacking self-assembly of organic topological heterostructures
,”
CCS Chem.
2
,
413
424
(
2020
).
154.
W.
Wang
,
M.
Pitoňák
, and
P.
Hobza
, “
C–H stretching vibrational shift of benzene dimer: Consistency of experiment and calculation
,”
ChemPhysChem
8
,
2107
2111
(
2007
).
155.
M. P.
Zhuo
,
G. P.
He
,
X. D.
Wang
, and
L. S.
Liao
, “
Organic superstructure microwires with hierarchical spatial organization
,”
Nat. Commun.
12
,
2252
(
2021
).
156.
Q.
Lv
,
X. D.
Wang
,
Y.
Yu
,
M. P.
Zhuo
,
M.
Zheng
, and
L. S.
Liao
, “
Lattice-mismatch-free growth of organic heterostructure nanowires from cocrystals to alloys
,”
Nat. Commun.
13
,
3099
(
2022
).