Slightly attractive: The attractive and anisotropic nature of the ClCl interaction in C(6)Cl(6) is experimentally demonstrated from an expansion of the electron density rho(r) around the chlorine nuclei. The interaction is explained in a model in which there is a bonding attraction involving electron-deficient (see picture, blue) and electron-rich (red) regions of adjacent Cl atoms.

Chalcogen bonding has been investigated in terms of the electron density distribution ρ(r) around chalcogen atoms. The evolution of ρ(r) along the series of chalcogen atoms is shown based on ab initio calculations on chalcogenophthalic anhydrides C8O2H4Chal (Chal = O, S, Se, and Te), where the Chal atom is in its sp 3 hybridization. From a detailed analysis of the experimental and theoretical electron density and the L(r) = −∇2ρ(r) function in the crystal phase of C8O2H4Se, we characterize directionality and strength of chalcogen bonding (Se···O and Se···Se) and hydrogen bonding (Se···H) interactions. In addition, several isolated dimers and a trimer of C8O2H4Se have been also studied at the X-ray geometry in order to compare interaction energies with those estimated from the measured electron density. Similarly to halogen atoms in halogen bonding interactions, the anisotropic distribution of ρ(r) around the Chal atoms was found to be at the origin of chalcogen bonding. Therefore, the concepts, developed earlier for halogen bonding, are extended here to chalcogen bonding interactions. From the results of this work, the L(r) function proves to be more precise than the σ-hole concept to identify electrophilic sites of Se-atoms in sp 3 hybridization.

Trimers based on intermolecular halogen-bonding interactions (Hal 3 -synthons) have been studied in hexachlorobenzene, hexabromobenzene, pentachlorophenol, and pentabromophenol. Attention is paid to the comparison of Cl 3 -and Br 3 -synthons and to their competition with hydrogen bonds (HBs), based on the experimental and theoretical charge density analyses in crystal and gas phases. The main differences between Cl 3 -and Br 3 -synthons are established coming from the particular structure of the valence shell charge concentration region in Cl and Br atoms. Electrophilic−nucleophilic interactions take place within the intermolecular regions of Hal 3 -synthons by putting face-to-face charge depletion (CD) and charge concentration (CC) regions belonging to the valence shell of the halogen atoms. The electrostatic interaction follows the electrophilic and nucleophilic power of these regions and is monitored by the negative Laplacian values normalized to charge density unit (L/ρ) at the corresponding topological critical points (CPs) of the L(r) = −∇ 2 ρ(r) function. According to the topological and energetic properties at CPs of ρ(r) and L(r), it is observed that Hal 3 -synthons can successfully compete with intermolecular HBs in the analyzed structures. On the basis of the estimated interaction energy and the electrostatic descriptor Δ(L/ρ) = (L/ρ) CC − (L/ρ) CD , we conclude that a strong dispersion contribution assists Hal3-synthons in this competition.

The electron density of l-cystine has been analyzed using 110 K single-crystal Mo Kα X-ray diffraction data to a resolution of (sin θ/λ)max = 1.123 Å-1 with a CCD area detector. Due to the large c-parameter (55.9 Å), a discussion is made for choosing the best experimental data collection strategy and data reduction. A multipolar pseudo-atom density model was fitted against the 2309 observed data with I > 3σ(I), [R(F) = 0.014, R W(F) = 0.019, S = 0.73]. The deformation density distribution and topological analysis of charge density clearly reveals disulfide bridge characteristic features and sulfur lone pair electron regions which validate high-level ab initio calculations. The valence shell charge concentration (VSCC) suggests sp3 hybridization of sulfur atoms.

High-resolution X-ray diffraction experiments and state-of-the-art density functional theory calculations have been performed. The validity of the atoms-in-molecules approach is tested for the neutral-ionic transition of TTF-CA which involves a transfer of less than one electron between the donor and acceptor molecules. Foremost, crystallographical data have been reassessed along the temperature-induced neutral-ionic phase transition undergone by this charge transfer complex. Based on accurate X-ray structures at 105 and 15 K, topological analysis of both DFT and the experimental multipolar electron densities allowed detailed characterization of intra- and interstack intermolecular interactions. Direct quantification of the intermolecular charge transfer and the dipole moment are discussed.

Charge-assisted halogen bonding is unambiguously revealed from structural and electronic investigations of a series of isostructural charge-transfer complexes derived from iodinated tetrathiafulvalene and tetracyanoquinodimethane derivatives, (EDT-TTFI2)2(TCNQF(n)), n=0-2, which exhibit variable degrees of ionicity. The iodinated tetrathiafulvalene derivative, EDT-TTFI2, associates with tetracyanoquinodimethane (TCNQ) and its derivatives of increasing reduction potential (TCNQF, TCNQF2) through highly directional C-I⋅⋅⋅N≡C halogen-bond interactions. With the less oxidizing TCNQ acceptor, a neutral and insulating charge-transfer complex is isolated whereas with the more oxidizing TCNQF2 acceptor, an ionic, highly conducting charge-transfer salt is found, both of 2:1 stoichiometry and isostructural with the intermediate TCNQF complex, in which a neutral-ionic conversion takes place upon cooling. A correlation between the degree of charge transfer and the C-I⋅⋅⋅N≡C halogen-bond strength is established from the comparison of the structures of the three isostructural complexes at temperatures from 300 to 20 K, thus demonstrating the importance of electrostatics in the halogen-bonding interaction. The neutral-ionic conversion in (EDT-TTFI2)2(TCNQF) is further investigated through the temperature dependence of its magnetic susceptibility and the stretching modes of the C≡N groups.

That non-directional pair-wise atom-atom potentials containing only attractive (r À6 dispersion term) and repulsive (inverse-power or exponential term) contributions are insufficient to model Cl···Cl interactions in molecular crystals was appreciated 40 years ago. [1] The layered orthorhombic crystal structures (space group Cmca) of the halogens (Cl 2 , Br 2 , I 2 ) cannot be anticipated with isotropic potentials (which would predict, for example, cubic packings, such as the Pa " 3 3 structure as seen for N 2 , NO, and CO), [2] and consideration of previously postulated quadrupole-quadrupole interactions [3] did not resolve this issue. Two possibilities have commonly been invoked to explain these anomalous halogen-atom contacts. Williams and Hsu proposed that halogen atoms in molecular crystals are weakly bonded (for Cl···Cl interactions, this bonding component was estimated to be around 3 % of the energy of a Cl À Cl covalent bond). [4] Such bonding was then considered to be the origin of the crystal anisotropy. Nyburg and Wong-Ng, on the other hand, proposed a model that assigned anisotropic non-bonded radii (Scheme 1 a) to Cl atoms in crystals. [5] The elliptically shaped atoms were then related to the origin of the anisotropy in the crystal packing. Over the years, all interpretations of Cl···Cl interactions in crystals invoked one or the other of these two models. [6] However, it is difficult to distinguish computationally between them: the Williams model proposes an increased attraction while the Nyburg model proposes a decreased repulsion between non-bonded atoms. Is there any real difference between these situations in an empirical or semiempirical computational approach? An experimental analysis Angewandte Chemie 3897

A new way to synthesize trans-disubstituted cyclam on the X-ray experimental electrostatic potential and molecular modeling of the 1,4,8,11-tetraazatricyclotetraazamacrocycles 1 is reported. The synthesis proceeds in three steps via the tricyclic 1,4,8,11-tetraazatricyclo-[9.3.1.1 4,8 ]hexadecane macrotricycle, has permitted the elucidation of a new reaction pathway leading to the trans-[9.3.1.1 4,8 ]hexadecane system 2, which can be selectively dialkylated and hydrolyzed under basic conditions to give disubstituted cyclam. the final product 1. An understanding of the reactivity, basedThe design and the synthesis of tetraazamacrocycloalk-recrystallisation from a THF/water mixture. The presence of the methylenic bridges was confirmed by 1 H NMR (δ ϭ anes has been the subject of growing interest during past years due to their ability to coordinate different metal cat-5.4) and 13 C NMR (δ ϭ 69). The crystallographic structure of 2 ( Figure 1) shows a trans conformation for the two ions. Among the derivatives of this class of macrocycles, the cyclam (1,4,8,11-tetraazacyclotetradecane) has been exten-methylenic bridges. sively studied, either as a simple ligand or as an N-substi- . Moreover, it is well known that ability thermal ellipsoids for non-H atoms the N,NЈ-functionalized cyclam can lead to hexacoordinated complexes. To date, numerous N,NЈ-functionalized cyclams have been described in the literature [7] [8] , and, recently, we reported the synthesis of the cis-4,8-N-disubstituted cyclam [9] . As the classical synthetic scheme [7] [8] leading to trans-4,8-N-disubstituted cyclam involving 4,8-N-bis-(para-toluenesulfonyl)tetraazacyclotetradecane as intermediate gives low yields, here we propose an alternative synthesis.The described synthetic route proceeds at room temp. to yield the trans-N,NЈ-disubstituted cyclam 1 at a high rate. As represented in Scheme 1, the first step leads to the 1, 4,8,11-tetraazatricyclo[9.3.1.1 4,8 ]hexadecane derivative 2. This macrocyclic compound has already been prepared using formaldehyde and cyclam as starting reagents [10] [11] . In a second step, the 1,4,8,11-tetraazatricyWe have obtained, quantitatively, the macrotricycle 2 by reclo[9.3.1.1 4,8 ]hexadecane ligand was dissolved in CH 3 CN fluxing cyclam in a 30% NaOH aqueous solution in the and two equiv. of an alkyl halide (methyl iodide, benzyl presence of dichloromethane [12] . The pure product was easbromide, or picolyl chloride) were rapidly added to yield the ily isolated by concentration of the organic phase and new disubstituted macrotricycle 4 having two non-adjacent quaternary nitrogen atoms. Due to their ionic character, Supporting information for this article is available on the compounds 4aϪc are insoluble in CH 3 CN and so were iso-WWW under http://www.wiley-vch.de/home/eurjoc or from the author. lated by filtration. High yields were obtained and no cisEur.