Received 16 September 2013
C-INC halogen bonding in two polymorphs of the mixed-valence 2:1 charge-transfer salt (EDT-TTF-I2)2(TCNQF4), with segregated versus alternated stacks
Julien Lieffrig,a Olivier Jeannin,a Antoine Vacher,a Dominique Lorcy,a Pascale Auban-Senzierb and Marc Fourmiguéa*
Oxidation of diiodoethylenedithiotetrathiafulvalene (EDT-TTF-I2), C8H4I2S6, with the strong oxidizer tetrafluorotetracyanoquinodimethane (TCNQF4), C12F4N4, affords, depending on the crystallization solvent, two polymorphs of the 2:1 charge-transfer salt (EDT-TTF-I2)2(TCNQF4), represented as D2A. In both salts, the TCNQF4 is reduced to the radical anion state, and is associated through short C-INC halogen bonds to two EDT-TTF-I2 molecules. The two polymorphs differ in the solid-state association of these trimeric D-A-D motifs. In polymorph (I) the trimeric motif is located on an inversion centre, and hence both EDT-TTF-I2 molecules have +0.5 charge. Together with segregation of the TTF and TCNQ derivatives into stacks, this leads to a charge-transfer salt with high conductivity. In polymorph (II) two crystallographically independent EDT-TTF-I2 molecules bear different charges, close to 0 and +1, as deduced from an established correlation between intramolecular bond lengths and charge. Overlap interactions between the halogen-bonded D0-A-D motifs give rise, in a perpendicular direction, to diamagnetic A22- and D0-D22+-D0 entities, where the radical species are paired into the bonding combination of respectively the acceptor LUMOs and donor HOMOs. The strikingly different solid-state organization of the halogen-bonded D-A-D motifs provides an illustrative example of two modes of face-to-face interaction between -type radicals, into either delocalized, uniform chains with partial charge transfer and conducting behaviour, or localized association of radicals into face-to-face A22- and D22+ dyads.
Halogen bonding is an effective and reliable tool in solid-state supramolecular chemistry, as described by Metrangolo et al. (2008), and Metrangolo & Resnati (2001, 2012). It has been extensively investigated in supramolecular chemistry (Meyer & Dubois, 2013; Gilday et al., 2013), but also for the elaboration of liquid crystals (Nguyen et al., 2004; Metrangolo et al., 2006; Präsang et al., 2008) and gels (Meazza et al., 2013), anion sensors (Metrangolo et al., 2009; Cavallo et al., 2010), catalytic systems (Bruckmann et al., 2008; Walter et al., 2011; Kniep et al., 2013) nonlinear optical (Cariati et al., 2007) or magnetic/conductive materials (Fourmigué & Batail, 2004; Fourmigué, 2008). Such halogen-bonded conductive materials are most often based on an iodinated tetrathiafulvalene molecule (Gompper et al., 1995; Wang et al., 1994) such as EDT-TTF-I. The electrocrystallization of EDT-TTF-I in the presence of various counter ions affords mixed-valence salts, formulated as (EDT-TTF-I)2X [X = Br-, Ag(CN)2- etc.], where X acts both as an anion to compensate the partial positive charge of the donor molecules, and as a halogen-bond acceptor towards the I atom of EDT-TTF-I (Imakubo et al., 1995; Ueda et al., 2003; Ranganathan et al., 2006). The overlap interactions between TTF molecules, perpendicular to this trimeric motif, lead to the formation of conduction bands. Partial filling, due to the stoichiometry, and large band dispersion are the necessary conditions for metallic behaviour (Fourmigué, 2012).
Besides such cation radical salts with electrochemically innocent anions, it is also possible to consider the chemical formation of charge-transfer salts, where the anion itself is electroactive, from the oxidation of the TTF derivative with an organic oxidant. Within this frame, a series of charge-transfer salts of EDT-TTF-I was recently reported, with a variety of tetracyanoquinodimethane (TCNQ) derivatives of different oxidative ability, i.e. TCNQF4 > TCNQF2 > TCNQF > TCNQ (Lieffrig, Jeannin, Guizouarn et al., 2012). While a neutral charge-transfer complex was isolated with the weaker acceptor TCNQ, an intermediate mixed-valence salt was found with TCNQF2 and a full charge-transfer salt with TCNQF4. Surprisingly, however, while CTTF-INC halogen bonds were observed in the mixed-valence salts, where the donor charge, EDT-TTF-I, amounts to +0.5, and also in the neutral TCNQ complex [EDT-TTF-I = 0], they are surprisingly absent from the full charge-transfer salt with TCNQF4. It was shown that TTF oxidation also activates the TTF sp2 H atom located directly adjacent () to the I atom towards the preferential formation of CTTF-HNC hydrogen bonds. These series provided an opportunity to evaluate the relative strength of competing C-HN hydrogen and C-IN halogen bonds and demonstrated that halogen bonding is not as sensitive as C-HN hydrogen bonding to the charge of the TTF core (Lieffrig, Jeannin, Guizouarn et al., 2012).
In the present work, we wanted to suppress this halogen bond/hydrogen bond competition and accordingly concentrated on the analogous diiodotetrathiafulvalene derivative, namely EDT-TTF-I2. Because of the presence of two electron-withdrawing I atoms, EDT-TTF-I2 oxidizes at a rather high potential (0.57 V versus SCE; SCE = saturated calomel electrode), and only a few charge-transfer salts have been described to date, for example with the strong electron acceptor 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ; Lieffrig, Jeannin, Shin et al., 2012), although many cation radical salts were obtained by electrochemical oxidation (electrocrystallization; Domercq et al., 2001; Devic et al., 2002, 2003; Alberola et al., 2008; Fourmigué & Auban-Senzier, 2008; Shin et al., 2011). In the DDQ charge-transfer salt formulated as (EDT-TTF-I2)2(DDQ), halogen-bonding interactions were observed toward both the O atom of the carbonyl and the N atom of the nitrile groups of reduced DDQ. In addition, a segregation of the partially oxidized EDT-TTF-I2 molecules into layers allowed for the formation of conducting slabs, with a room-temperature conductivity of 0.043 S cm-1. Looking for other possible oxidants comparable to DDQ, we turned our attention to fluorinated TCNQ derivatives. Indeed, because of the electron-withdrawing nature of F, the mono-, di- and tetrafluoro analogues of TCNQ, namely TCNQF, TCNQF2 and TCNQF4, exhibit a reduction potential higher than that of TCNQ itself (+0.14 V versus SCE) and increasing with the number of F atoms, that is +0.26 V in TCNQF, +0.30 V for TCNQF2, +0.53 V for TCNQF4 (all versus SCE). Reaction of EDT-TTF-I2 with TCNQ, TCNQF and TCNQF2 afforded a series of isostructural compounds, where the degree of charge transfer was found to vary with the acceptor ability of TCNQFn (Lieffrig et al., 2013). On the other hand, the stronger oxidant TCNQF4 afforded two charge-transfer salts, both formulated as (EDT-TTF-I2)2(TCNQF4), with strikingly different solid-state organizations of the halogen-bonded EDT-TTF-I and TCNQF radical species. These two polymorphs described herein illustrate the two main modes of solid-state organization of -type radical species (Fourmigué, 2012), either in extended delocalized chains with the possibility for conductivity or into dimeric or tetrameric units with localized bonding interactions.
TCNQF4 was prepared according to a literature procedure (Wheland & Martin, 1975). The neutral EDT-TTF-I2 molecule was prepared as previously described (Domercq et al., 2001), and recrystallized by slow evaporation from a CS2 solution, to provide crystals suitable for X-ray diffraction experiments. Since this reported synthetic procedure provides EDT-TTF-I2 only in low yield, we also explored an alternative synthetic path that had been used for the analogous dibromo derivative (Batsanov et al., 2001; Alberola et al., 2006). Full details of this synthesis are described in the supporting information,1 together with the crystal structure of one of the synthesis intermediates, [NC(CH2)2S]2TTFI2.
In a small glass tube of internal diameter 5 mm, a solution of EDT-TTF-I2 (6.1 mg, 13 × 10-6 mol) in dichloromethane was layered with a solution of TCNQF4 (2.9 mg, 10.5 × 10-6 mol) in acetonitrile (0.5 ml). Slow diffusion at 293 K afforded black needles after 2 weeks.
In a small glass tube of internal diameter 5 mm, a solution of EDT-TTF-I2 (2.2 mg, 4.0 × 10-6 mol) in 1,1,2-trichloroethane (TCE) was layered with a solution of TCNQF4 (1.7 mg, 6.1 × 10-6 mol) in acetonitrile (0.5 ml). Slow diffusion at 293 K afforded black platelets after 2 weeks.
For [NC(CH2)2S]2-TTF-I2 and EDT-TTF-I2, crystals were picked from Paratone® oil with a cryoloop and directly frozen at 150 K under a stream of dry N2. Data were collected on a Bruker APEX-II CCD diffractometer with geometry, using graphite-monochromated Mo K radiation. For the room-temperature data collections of the two TCNQF4 salts, the crystals were mounted on a thin glass fibre and data were collected on a Nonius KappaCCD diffractometer, also using graphite-monochromated Mo K radiation. Details of the data collection and refinement parameters are given in Table 1, and displacement ellipsoid plots are shown in Fig. 1. For EDT-TTF-I2, the molecule lies on a mirror plane, perpendicular to the molecular plane, and the often-encountered disorder of the ethylene group was refined by constraining the site occupancy factors to 0.5 for atoms C5A and C5B (and their associated H atoms), and restraining the S1-C5A and S1-C5B bond lengths to be equivalent with an s.u. of 0.01 Å.
| || Figure 1 |
Molecular structures of (a) EDT-TTF-I2, and (b) polymorph (I) and (c) polymorph (II) of (EDT-TTF-I2)2(TCNQF4), showing displacement ellipsoids at 50% probability for non-H atoms. For EDT-TTF-I2 only one disorder component is shown for the disordered ethylene group (atoms C5A and C5B). Symmetry codes: (i) ; (ii) 2-x, 1-y, 1-z.
To measure the longitudinal resistivity, gold pads were evaporated on the surface of the crystals in order to improve the quality of the contacts. Then, a standard four-points technique was used with a low frequency lock-in detection (IAC = 0.1-1 µA) for measured resistances below 50 k and DC measurement for higher resistances (IDC = 0.1-0.01 µA). Low temperatures (down to 25 K) were provided by cryocooler equipment.
The tight-binding intermolecular interactions were calculated with the effective one-electron Hamiltonian of the extended Hückel method (Whangbo & Hoffmann, 1978), as implemented in the Caesar1.0 chain of programs (Ren et al., 1998). The off-diagonal matrix elements of the Hamiltonian were calculated according to the modified Wolfsberg-Helmholz formula (Ammeter et al., 1978). All valence electrons were explicitly taken into account in the calculations and the basis set consisted of double- Slater-type orbitals for all atoms except H ( Slater-type orbital) using the Roothaan-Hartree-Fock wavefunctions (Clementi & Roetti, 1974).
The structure of the neutral molecule EDT-TTF-I2 is of interest to determine its geometrical characteristics, particularly the intramolecular bond lengths within the TTF core, as they are known to evolve with the oxidation state of the molecule, with a lengthening of the C=C and concomitant shortening of the C-S bonds upon oxidation. Although many cation radical salts of EDT-TTF-I2 have been described so far, the structure of the neutral donor molecule was previously unknown. EDT-TTF-I2 crystallizes in space group Pnma, with the crystallographic mirror plane lying perpendicular to the molecular plane (Fig. 1). In the crystal (Fig. 2) the EDT-TTF-I2 molecules stack into uniform chains along the c axis. No short II intermolecular contacts are identified. Note that this structure differs from that reported for the analogous dichloro and dibromo derivatives, EDT-TTF-Cl2 and EDT-TTF-Br2 (Kux et al., 1995), where head-to-tail dyads were reported to organize into a herringbone structure.
| || Figure 2 |
Projection view along b of one layer of molecules in the crystal structure of EDT-TTF-I2.
Polymorph (I) crystallizes in space group , with the EDT-TTF-I2 molecule in a general position and the TCNQF4 molecule on an inversion centre, giving the 2:1 stoichiometry (Fig. 1b). The charge of TCNQF4 can be estimated by the Kistenmacher relationship first established for TCNQ salts (Kistenmacher et al., 1982) and adapted to TCNQF4 (Miyasaka et al., 2010). It writes as TCNQF4 = A[c/(b + d)] + B, where b-d are the intramolecular distances of TCNQF4 defined in Table 2, and A and B are parameters determined by assuming a completely neutral (Emge et al., 1981) and a fully ionized (i.e. 1 e- reduced) form of TCNQF4, such as in (nBu4N)(TCNQF4) (O'Kane et al., 2000), giving A = -46.729 and B = 22.308. Applied to polymorph (I), it gives an approximate calculated charge of -0.85, confirming that the TCNQF4 is reduced to the radical anion state, i.e. TCNQF. As a consequence, the EDT-TTF-I2 molecule is only partially oxidized (TTF = +0.5). Comparison of the intramolecular bond distances within the TTF core (Table 3) confirms this assumption, with lengthening of the C=C bonds and shortening of the C-S bonds associated with the partial oxidation, when compared to the neutral donor molecule.
In the solid state the donor and acceptor molecules stack along the a axis, giving rise to the formation of (011) layers where a TCNQF4 stack alternates with two neighbouring EDT-TTF-I2 stacks (Figs. 3 and 4). These layers are interconnected through a particularly short C-INC halogen-bonding interaction. Indeed, the IN distance of 2.934 (6) Å is notably shorter than the sum of the van der Waals radii of I (1.98 Å) and N (1.52 Å), hence a reduction parameter, defined as 100× the sum of van der Waals radii over the actual IN distance, of 84%. Its strength is also demonstrated by its linearity, with a C-IN angle of 177.3 (2)°, while the INC angle at 163.9 (5)° shows that the N atom acts as a halogen-bond donor through its lone pair. The IN distance also compares with those reported (Lieffrig et al., 2013) for the ionic charge-transfer salts isolated with TCNQF2 and TCNQF (2.93-2.95 Å), while the neutral charge-transfer complexes present longer IN contacts (> 3.0 Å). This confirms the ionic character of polymorph (I). Note also that this halogen-bonding interaction, combined with the centrosymmetric character of TCNQF4, most probably favours the 2:1 stoichiometry, at variance with the most common charge-transfer salts such as TTF·TCNQ or TTF·chloranil, which have 1:1 stoichiometry.
| || Figure 3 |
Projection view along a of the unit cell of (EDT-TTF-I2)2(TCNQF4), polymorph (I). Symmetry codes: (i) 2-x, 1-y, 1-z; (ii) 1-x, 1-y, 1-z.
| || Figure 4 |
A view along the long molecular axes of both donor and acceptor molecules in (EDT-TTF-I2)2(TCNQF4), polymorph (I), showing alternation of DDA columns in the (011) plane.
The mixed-valence character of the EDT-TTF-I2 stacks suggests that this salt could be conducting. Indeed, it exhibits a room-temperature conductivity (RT) of 0.4 S cm-1, while the temperature dependence of the resistivity (Fig. S2 in the supporting information ) shows a semiconducting behaviour with a large activation energy, Ea = 0.146 eV (1700 K). Under pressure, RT increases up to 0.7 S cm-1 at 0.25 GPa (Fig. S3 in the supporting information ). Calculations of the HOMO-HOMO interaction energies between neighbouring EDT-TTF-I2 molecules within these twin stacks show that the strongest interaction is found within the stack, with intra = -0.18 eV, while the lateral interaction between coplanar EDT-TTF-I2 molecules, inter, amounts to +0.091 eV. The system can thus be described as two interacting, non-dimerized, 3/4-filled chains. Such regular chains are currently the subject of strong interest as model systems, since most one-dimensional cation radical salts are slightly dimerized (Hünig & Herverth, 2004; Auban-Senzier et al., 2009; Foury-Leylekian et al., 2011). However, the inter-stack coupling between the two inversion-related chains gives rise here to two non-degenerate bands, with the upper one only half-filled (Fig. 5). Under those circumstances, we are brought back to the half-filled systems such as the semiconducting TMTTF series (Pouget, 2012). Note also that TCNQF most probably does not contribute to the crystal conductivity as the overlap interaction LUMO-LUMO is very weak, +0.005 eV, a consequence of a strongly shifted overlap between molecules (Fig. S4 in the supporting information ).
| || Figure 5 |
Calculated band structure for the one-dimensional EDT-TTF-I2 system in polymorph (I). The dotted line indicates the Fermi level for the hypothetical metallic state.
Polymorph (II) of (EDT-TTF-I2)2(TCNQF4) also crystallizes in space group , but with two crystallographically independent donor molecules, together with a TCNQF4 molecule in a general position (Figs. 1c and 6). The TCNQF4 molecules are organized into centrosymmetric dimers, stacking with centrosymmetric EDT-TTF-I2 tetramers along the  direction (Fig. 7). Those layers are connected to each other by the halogen-bonding interactions shown in Fig. 5. This motif is quite similar to that observed in polymorph (I), but its supramolecular organization differs completely (see below). Note, however, that the IN distances, notably shorter than 3 Å, are in accordance with the ionic nature of this compound.
| || Figure 6 |
Halogen-bonded motifs in (EDT-TTF-I2)2(TCNQF4) polymorph (II). Halogen bonds are shown as purple dotted lines and H atoms are omitted. Symmetry codes: (i) 1-x,-y, 1-z; (ii) 2-x,2-y,1-z.
| || Figure 7 |
View along the long molecular axes of both donor and acceptor molecules in (EDT-TTF-I2)2(TCNQF4) polymorph (II), showing the alternating ...D4A2D4A2... columns of molecules stacking along .
As already detailed for polymorph (I), the analysis of the intramolecular bond distances within both donor and acceptor molecules helps to assess their charge. For TCNQF4, application of the formula used in Table 2 shows that the acceptor molecule is again fully reduced to the radical anion state, TCNQF. At variance with polymorph (I), however, this does not imply a +0.5 charge for the two EDT-TTF-I2 molecules, there are now two crystallographically independent donor molecules. Inspection of the central C=C bond length (Table 3), shows that molecule B [C11=C12 1.393 (12) Å] is probably more oxidized than molecule A, which exhibits a shorter C3=C4 bond [1.344 (12) Å]. In order to assess in a more precise way the actual charge of both donor molecules, we have developed a correlation between intramolecular bond length and charge, by analogy with those reported for BEDT-TTF (Guionneau et al., 1997). This becomes possible with the structural characterization of the neutral molecule reported above, complemented by structural data for selected EDT-TTF-I2 salts (Table 3) where the oxidation state of the molecule is unambiguous, that is within those structures with only one crystallographically independent molecule. Our correlation between bond distances and charges, based on neutral EDT-TTF-I2, partially oxidized ( = +0.5) and fully oxidized ( = +1) EDT-TTF-I2 salts, is cal = A + B[(b + c) - (a + d)], with A = 6.8617 and B = -8.1332. Applied to the two crystallographically independent EDT-TTF-I2 molecules in polymorph (II), it gives calc(A) = +0.21 and calc(B) = +1.12, confirming that molecule A is essentially neutral while molecule B is fully oxidized to the radical cation state.
This strong charge ordering has a striking influence on the solid-state association of the EDT-TTF-I2 molecules in the salt. As shown in Fig. 8, the two singly oxidized B molecules form an almost perfectly eclipsed face-to-face dyad (Fig. 8b). This leads to strong overlap between the -type SOMO, with the bonding combination occupied by two electrons, a prototype of a single 2e- bond (Fourmigué, 2012), delocalized between the two molecules. These diamagnetic dyads are surrounded by two essentially neutral EDT-TTF-I2 molecules, alternating then along the stacking direction with TCNQF4 dyads (Fig. 7). In the latter, the face-to-face association of TCNQF4 radical anions detailed in Fig. 9 corresponds most probably (see below) to a strong interaction between LUMOs.
| || Figure 8 |
(a) Side view of the EDT-TTF-I2 tetramers, showing the distortions from planarity of the essentially neutral donors A (in red), interacting with oxidized B molecules (in blue). (b) Projection view of the AB and BB overlaps within the tetramer, showing the eclipsed BB overlap characteristic of dicationic dimers.
| || Figure 9 |
(a) Side view of the TCNQF4 dimer in polymorph (II). (b) Projection view of the TCNQF4 dimer in polymorph (II), showing the bond-over-ring overlap pattern.
Calculations of the interaction energies, HOMO-HOMO and LUMO-LUMO, confirm these assumptions. For the donor's HOMO, the BB interaction energy amounts to -0.95 eV while by contrast AB is only -0.35 eV. A large LUMO-LUMO for the bond-over-ring overlap is also calculated for the TCNQF4 dimers, with LUMO-LUMO = +0.46 eV. This strong pairing of both radical species into bonding combinations of the singly occupied molecular orbitals lets us infer a diamagnetic, insulating character for polymorph (II).
Comparison of the two polymorphs shows that the IN distances in both cases are essentially the same, whatever the charge of the EDT-TTF-I2 molecule: 2.934 (6) Å for a charge +0.5 in polymorph (I), and 2.933 (9) and 2.949 (9) Å, respectively, for the EDT-TTF-I2 molecules with charges 0 and +1 in polymorph (II). It has recently been shown, however, that in isostructural systems with variable charge, there is indeed a direct correlation between charge and halogen-bond lengths (Lieffrig et al., 2013). The similar halogen-bond distances observed here might therefore be only a consequence of the different crystal packing. It also indicates that it is essentially the negative charge of the TCNQF radical anion which contributes strongly to the halogen-bond strength, while the I atoms are only weakly activated upon TTF oxidation. This feature was already noted in the charge-transfer salts of the monoiodio TTF derivative, EDT-TTF-I (Lieffrig, Jeannin, Guizouarn et al., 2012; Lieffrig, Jeannin, Shin et al., 2012), The strengthening of the halogen bond in such systems when turning from neutral to ionic is therefore essentially attributable to the negative charge which develops on the halogen-bond acceptor, here the nitrile substituent of TCNQF4.
We have shown that halogen bonding with iodinated tetrathiafulvalenes is a very efficient structural tool to associate these donor molecules, not only with counter ions in cation radical salts but also with electroactive anions in charge-transfer salts. The symmetry of the TCNQ derivatives, here TCNQF4, favours the formation of 2:1 salts, at variance with the prototypical TTF·TCNQ or TTF·chloranil where both partners have the same symmetry. With strong acceptors such as TCNQF4, it leads to a mixed-valence state for the donor molecules, a radical state which is stabilized through two very different ways. In polymorph (I) of (EDT-TTF-I2)2(TCNQF4), the formation of segregated stacks of donor and acceptor molecules allows the radicals to delocalize into partially filled bands. On the other hand, in polymorph (II), we observe a complete charge separation, with one neutral (D0) and one fully oxidized () donor molecule, together with the formation of localized two-electron bonds, in D22+ and A22- species. This very different solid-state organization of the halogen-bonded D-A-D trimetric motifs provides a textbook example of two modes of face-to-face interaction between -type radicals, either delocalized, uniform chains with partial charge transfer and conducting behaviour, or localized association of radicals into face-to-face dyads. It also demonstrates that charge-assisted halogen bonding relies essentially on the negative charge of the halogen-bond acceptor (the Lewis base), rather than on the partial positive charge of the I atoms, at least in these TTF-based systems. Comparison with halogenated cations such as iodopyridinium (Derossi et al., 2009) or diiodopyridinium (Derossi et al., 2009; Kosaka et al., 2007, 2013) salts could provide interesting answers to this point.
Financial support from the ANR (Paris, France) under contract No. ANR-08-BLAN-0091-02 is acknowledged. We also thank the X-ray facility in Rennes (Th. Roisnel, CDIFX) for providing access to diffractometers.
Alberola, A., Collis, R. J., García, F. & Howard, R. E. (2006). Tetrahedron, 62, 8152-8157.
Alberola, A., Fourmigué, M., Gómez-García, C. J., Llusar, R. & Triguero, S. (2008). New J. Chem. 32, 1103-1109.
Altomare, A., Burla, M. C., Camalli, M., Cascarano, G. L., Giacovazzo, C., Guagliardi, A., Moliterni, A. G. G., Polidori, G. & Spagna, R. (1999). J. Appl. Cryst. 32, 115-119.
Ammeter, J. H., Buergi, H. B., Thibeault, J. C. & Hoffmann, R. (1978). J. Am. Chem. Soc. 100, 3686-3692.
Auban-Senzier, P., Pasquier, C. R., Jérome, D., Suh, S., Brown, S. E., Mézière, C. & Batail, P. (2009). Phys. Rev. Lett. 102, 257001.
Batsanov, A. S., Bryce, M. R., Chesney, A., Howard, J. A. K., John, D. E., Moore, A. J., Wood, C. L., Gershtenman, H., Becker, J. Y., Khodorkovsky, V. Y., Ellern, A., Bernstein, J., Perepichka, I. F., Rotello, V., Gray, M. & Cuello, A. O. (2001). J. Mater. Chem. 11, 2181-2191.
Brandenburg, K. & Brendt, M. (2001). DIAMOND, Release 2, 1e. Crystal Impact GbR, Bonn, Germany.
Bruckmann, A., Pena, M. A. & Bolm, C. (2008). Synlett, 6, 900-902.
Bruker (2003). SADABS and SAINT. Bruker AXS Inc., Madison, Wisconsin, USA.
Bruker (2005). APEX2. Bruker AXS Inc., Madison, Wisconsin, USA.
Cariati, E., Forni, A., Biella, S., Metrangolo, P., Meyer, F., Resnati, G., Righetto, S., Tordin, E. & Ugo, R. (2007). Chem. Commun. 25, 2590-2592.
Cavallo, G., Metrangolo, P., Pilati, T., Resnati, G., Sansotera, M. & Terraneo, G. (2010). Chem. Soc. Rev. 39, 3772-3783.
Clementi, E. & Roetti, C. (1974). At. Data Nucl. Data Tables, 14, 177-478.
Derossi, S., Brammer, L., Hunter, C. A. & Ward, M. D. (2009). Inorg. Chem. 48, 1666-1677.
Devic, T., Domercq, B., Auban-Senzier, P., Molinié, P. & Fourmigué, M. (2002). Eur. J. Inorg. Chem. pp. 2844-2849.
Devic, T., Evain, M., Moëlo, Y., Canadell, E., Auban-Senzier, P., Fourmigué, M. & Batail, P. (2003). J. Am. Chem. Soc. 125, 3295-3301.
Domercq, B., Devic, T., Fourmigué, M., Auban-Senzier, P. & Canadell, E. (2001). J. Mater. Chem. 11, 1570-1575.
Duisenberg, A. J. M. (1992). J. Appl. Cryst. 25, 92-96.
Duisenberg, A. J. M., Kroon-Batenburg, L. M. J. & Schreurs, A. M. M. (2003). J. Appl. Cryst. 36, 220-229.
Emge, T., Maxfield, M., Cowan, D. & Kistenmacher, T. (1981). Mol. Cryst. Liq. Cryst. 65, 161-178.
Farrugia, L. J. (2012). J. Appl. Cryst. 45, 849-854.
Fourmigué, M. (2008). Struct. Bond. 126, 181-207.
Fourmigué, M. (2012). The Importance of -Interactions in Crystal Engineering: Frontiers in Crystal Engineering, edited by E. Tiekink & J. Zukerman-Schpector, 2nd ed, Ch. 6, pp. 143-162. New York: John Wiley and Sons Ltd.
Fourmigué, M. & Auban-Senzier, P. (2008). Inorg. Chem. 47, 9979-9986.
Fourmigué, M. & Batail, P. (2004). Chem. Rev. 104, 5379-5418.
Foury-Leylekian, P., Auban-Senzier, P., Coulon, C., Jeannin, O., Fourmigué, M., Pasquier, C. & Pouget, J.-P. (2011). Phys. Rev. B, 84, 195134.
Gilday, L. C., Lang, T., Caballero, A., Costa, P. J., Félix, V. & Beer, P. D. (2013). Angew. Chem. Int. Ed. 52, 4356-4360.
Gompper, R., Hock, J., Polborn, K., Dormann, E. & Winter, H. (1995). Adv. Mater. 7, 41-43.
Guionneau, P., Kepert, C. J., Bravic, G., Chasseau, D., Truter, M. R., Kurmoo, M. & Day, P. (1997). Synth. Met. 86, 1973-1974.
Hervé, K., Cador, O., Golhen, S., Costuas, K., Halet, J.-F., Shirahata, T., Muto, T., Imakubo, T., Miyazaki, A. & Ouahab, L. (2006). Chem. Mater. 18, 790-797.
Hünig, S. & Herverth, E. (2004). Chem. Rev. 104, 5535-5564.
Imakubo, T., Sawa, H. & Kato, R. (1995). Synth. Met. 73, 117-122.
Imakubo, T., Shirahata, T., Hervé, K. & Ouahab, L. (2006). J. Mater. Chem. 16, 162-173.
Kistenmacher, T. J., Emge, T. J., Bloch, A. N. & Cowan, D. O. (1982). Acta Cryst. B38, 1193-1199.
Kniep, F., Jungbauer, S. H., Zhang, Q., Walter, S. M., Schindler, S., Schnapperelle, I., Herdtweck, E. & Huber, S. M. (2013). Angew. Chem. Int. Ed. 52, 7028-7032.
Kosaka, Y., Yamamoto, H. M., Nakao, A., Tamura, M. & Kato, R. (2007). J. Am. Chem. Soc. 129, 3054-3055.
Kosaka, Y., Yamamoto, H. M., Tajima, A., Nakao, A., Cui, H. & Kato, R. (2013). CrystEngComm, 15, 3200-3211.
Kux, U., Suzuki, H., Sasaki, S. & Iyoda, M. (1995). Chem. Lett. pp. 183-184.
Lieffrig, J., Jeannin, O., Frackowiak, A., Olejniczak, I., Swietlik, R., Dahaoui, S., Aubert, E., Espinosa, E., Auban-Senzier, P. & Fourmigué, M. (2013). Chem. Eur. J. 19, 14804-14813.
Lieffrig, J., Jeannin, O., Guizouarn, T., Auban-Senzier, P. & Fourmigué, M. (2012). Cryst. Growth Des. 12, 4248-4257.
Lieffrig, J., Jeannin, O., Shin, K.-S., Auban-Senzier, P. & Fourmigué, M. (2012). Crystals, 2, 327-337.
Meazza, L., Foster, J. A., Fucke, K., Metrangolo, P., Resnati, G. & Steed, J. W. (2013). Nat. Chem. 5, 42-47.
Metrangolo, P., Meyer, F., Pilati, T., Resnati, G. & Terraneo, G. (2008). Angew. Chem. Int. Ed. 47, 6114-6127.
Metrangolo, P., Pilati, T., Terraneo, G., Biella, S. & Resnati, G. (2009). CrystEngComm, 11, 1187-1196.
Metrangolo, P., Präsang, C., Resnati, G., Liantonio, R., Whitwood, A. C. & Bruce, D. W. (2006). Chem. Commun. 31, 3290-3292.
Metrangolo, P. & Resnati, G. (2001). Chem. Eur. J. 7, 2511-2519.
Metrangolo, P. & Resnati, G. (2012). Cryst. Growth Des. 12, 5835-5838.
Meyer, F. & Dubois, P. (2013). CrystEngComm, 15, 3058-3071.
Miyasaka, H., Motokawa, N., Matsunaga, S., Yamashita, M., Sugimoto, K., Mori, T., Toyota, N. & Dunbar, K. R. (2010). J. Am. Chem. Soc. 132, 1532-1544.
Nguyen, H. L., Horton, P. N., Hursthouse, M. B., Legon, A. C. & Bruce, D. W. (2004). J. Am. Chem. Soc. 126, 16-17.
Nonius (1998). COLLECT. Nonius BV, Delft, The Netherlands.
O'Kane, S. A., Clérac, R., Zhao, H., Ouyang, X., Galán-Mascarós, J. R., Heintz, R. & Dunbar, K. R. (2000). J. Solid State Chem. 152, 159-173.
Pouget, J.-P. (2012). Crystals, 2, 466-520.
Präsang, C., Whitwood, A. C. & Bruce, D. W. (2008). Chem. Commun. 18, 2137-2139.
Ranganathan, A., El-Ghayoury, A., Meziere, C., Harté, E., Clérac, R. & Batail, P. (2006). Chem. Commun. pp. 2878-2880.
Ren, J., Liang, W. & Whangbo, M.-H. (1998). CAESAR. PrimeColor Software, Inc. Cary, North Carolina, USA.
Sheldrick, G. M. (2008). Acta Cryst. A64, 112-122.
Shin, K.-S., Brezgunova, M., Jeannin, O., Roisnel, T., Camerel, F., Auban-Senzier, P. & Fourmigué, M. (2011). Cryst. Growth Des. 11, 5337-5345.
Ueda, K., Sugimoto, T., Faulmann, C. & Cassoux, P. (2003). Eur. J. Inorg. Chem. pp. 2333-2338.
Walter, S. M., Kniep, F., Herdtweck, E. & Huber, S. M. (2011). Angew. Chem. Int. Ed. 50, 7187-7191.
Wang, C., Ellern, A., Khodorkovsky, V., Bernstein, J. & Becker, J. Y. (1994). J. Chem. Soc. Chem. Commun. pp. 983-984.
Whangbo, M. & Hoffmann, R. (1978). J. Am. Chem. Soc. 100, 6093-6098.
Wheland, R. C. & Martin, E. L. (1975). J. Org. Chem. 40, 3101-3109.