On the Way to “Solid Nitrogen” at Normal Temperature and Pressure? Binary Azides of Heavier Group 15 and 16 Elements

Abstract

Caution explosive! Homoleptic azides are of great interest as polynitrogen compounds particularly because of their high energy content. A recent breakthrough is the first synthesis and characterization of the highly sensitive neutral tellurium azide [Te(N3)4] and of ionic (N5)+[P(N3)6]− containing 23 nitrogen atoms. The first synthesis of a covalent azide HN3 was reported by Curtius more than 100 years ago.1 However, the chemistry of coordinated azides received little attention until the early 1960s presumably because of their explosive nature and shock sensitivity. Today the high potential energy content of these compounds is the driving force to synthesize new and even more endothermic compounds that nevertheless have some kinetic stability. The relatively stable azide anion (I) contains two double bonds, but the covalent-bonded azides (II) are polarized towards structures containing a single and a triple bond, which facilitate the elimination of dinitrogen.21 Every azido group adds about 70 kcal mol−1 to the energy content of the molecule, and so the synthesis of molecules with a high number of azido groups is a useful but challenging way to synthesize high-energy materials. Azidometalates in salts containing large counter cations (e.g. [EPh4]+, E=P, As) and multiple azido groups can often be stabilized and safely handled.3d The salt is stabilized by its crystal lattice enthalpy, and the dissociation energy into highly unstable covalent species by N3− transfer is minimized with large non-electrophilic counterions. In addition the activation energy barrier of the decomposition may be raised by separation of the azido groups with large counterions (dilution effect). Therefore the covalent azides incorporated into a salt containing large ions were often characterized prior to the corresponding neutral species. If there is an electronegativity difference between M and the azido group in the formally covalent M(N3) bond, then the molecule will be stabilized by ionic resonance. However the M(N3) bond will be particularly weak where M is large and the electronegativities are similar.4 Neutral covalent azides can be stabilized by adduct formation with a Lewis base (e. g. pyridine), which results in surprisingly thermally stable compounds, as was successfully shown for azides of Group 133f and 14.13a Thus despite many difficulties, many fascinating inorganic azides have been prepared and structurally characterized over the last decades.3 Klapötke and co-workers as well as Christe and co-workers have made some very significant advances in the field of highly energetic polynitrogen compounds. Christe’s group discovered the novel N5+ ion,5 and Klapötke’s group synthesized and characterized a number of main group azides.3b–3e One of last highlights in this race was the back-to back publication of the syntheses and structure determinations of the first neutral and anionic binary tellurium azides.6 This is remarkable, because binary tellurium–nitrogen compounds and homoleptic azides of Group 16 are rare. For many years the only structurally characterized binary tellurium azide was ionic lattice stabilized [Te(N3)3][SbF6] containing [Te(N3)3]+,7 except for the few reports on extremely labile tellurium nitrides. The chemistry of tellurium azides commenced in 1972 with the synthesis of TeCl3(N3) and TeCl2(N3)2 by the reaction of TeCl4 with Me3SiN3 by Wiberg and co-workers.8 Wiberg et al. predicted that TeF4 would react with Me3SiN3 to give [Te(N3)4] in a straightforward reaction, but at the same time warned of the likely potential explosive nature of the product. The ionic, lattice-stabilized tellurium azides are relatively stable, whereas the neutral [Te(N3)4] is very sensitive and can explode under various conditions.6 These reactions appear to be very simple, however very special experimental knowledge and expertise is necessary to do chemistry of this kind.11 Potential safety hazards are reduced by preparing very small amounts of material and by the skilful use of various analytical tools (multinuclear NMR, Raman, and IR spectroscopy). The more stable ionic species were also characterized by single-crystal X-ray diffraction, whereas the complete structural determination of the neutral [Te(N3)4] remains a challenge. In all three known binary tellurium azide structures the free electron pair is sterically active. [Te(N3)3]+ has a pseudo-tetrahedral AX3E structure,7 and [Te(N3)5]− (Figure 1 a), one of only two examples of a pentacoordinate azido species,12 has a pseudo-octahedral AX5E geometry.6a The free electron pair is sterically active (AX6E) even in [Te(N3)6]2− (Figure 1 b) and distorts the geometry from ideal S6 symmetry,6b which is found in the Group 14 azido AX6 metalates [M(N3)6]2− (M=Si, Ge).13 Therefore all three tellurium ions have geometries in agreement with a simple VSEPR model. The as yet experimentally unknown structure of [Te(N3)4], is predicted by theoretical calculations to have an AX4E type structure like SF4, with two isomers having energies differing only by 1.8 kcal mol−1 (Figure 2).6b Structure of a) the [Te(N3)5]− (intermolecular TeN interactions between [Te(N3)5]− are not shown) ion and b) the [Te(N3)6]2− ions.6 Calculated minimum-energy structures of [Te(N3)4].6b Similar binary azides of the lighter homologues selenium and sulfur are still unknown. Over 30 years ago Wiberg et al. reacted SCl4 and SeCl4 with Me3SiN3 and observed the reduction of the chalcogen chlorides.8 Two years ago the first ionic selenonium azide [R3Se](N3) (R=H3C, C6H5) was prepared by the reaction of [R3Se]I with AgN3.14 Very recently the first covalent selenium azide 2-Me2NCH2C6H4Se(N3) was synthesized and structurally characterized.15 The latter compound is stabilized by an intramolecular dative bond formed by donation of the nitrogen lone pair into the σ* orbital of the SeN bond, which inhibits the elimination of dinitrogen and the formation of the corresponding stable diselane (Figure 3). The structure of the azide 2-Me2NCH2C6H4Se(N3) stabilized by an intramolecular dative bond.15 In contrast, the binary azides of phosphorus ([P(N3)3], [P(N3)4]+, [P(N3)5], [P(N3)6]−) were prepared before the azides of the heavier homologues arsenic and antimony, although the crystal structures of the binary phosphorus azides are still to be determined.16 The ionic arsenic azide salts containing [As(N3)4]−, [As(N3)4]+, and [As(N3)6]− are stabilized by their corresponding lattice energies and the factors referred to in the introduction.17 The neutral arsenic and antimony azides are similar to the neutral tellurium azides but are much more explosive; however, they are stabilized by forming donor–acceptor adducts. This allows the isolation of the binary arsenic and antimony azides in the oxidation state +5 as Lewis base adducts [E(N3)5]LB (E=As, Sb; LB=Lewis base, for example, py). The free azides [E(N3)5] can be handled safely in solution, but could not be isolated even at −70 °C.17b, 18 Very recently the structures of the less explosive binary AsIII and SbIII azides have been determined.19 Both have AX3E geometries in which the free valence electron pair is sterically active and the structure of [Sb(N3)3] (isoelectronic to [Te(N3)3]+)7 presents an example of perfect C3 symmetry. The phosphorus compound (N5)[P(N3)6] contains 23 nitrogen atoms and has a nitrogen content of 91 wt %. In the corresponding boron compound (N5)[B(N3)4] the nitrogen content is even higher (95.7 wt %). Both compounds are extremely sensitive and explode upon the slightest provocation. These recent developments show that some of the most outrageous compounds imaginable can be prepared and characterized. We anticipate that in the next few years many other thermodynamically unstable and increasingly marginally kinetically unstable simple species with even greater nitrogen content will be isolated and characterized as real compounds that can be seen and weighed, and not just observed as images on a computer screen.