A detailed mass spectrometric characterization of self‐assembling polynuclear metal complexes is described. The complexes can only be ionized as intact species under a surprisingly narrow range of conditions by electrospray ionization. Comparison with the results from NMR experiments shows that several solution‐phase features of these squares and triangles (such as trends in bond energies, ligand‐exchange reactions, or square–triangle equilibria) are qualitatively reflected in the gas‐phase data. Consequently, mass spectrometry represents a valuable method for the characterization of these compounds. Nevertheless, the formation of unspecific aggregates during the ionization process occurs and its implications are discussed. Beyond the chemistry in solution, the fragmentation pathways of these complexes in the gas phase have been studied by infrared multiphoton dissociation (IRMPD) experiments. The results of IRMPD studies allow us to draw conclusions with respect to the structure and energetics of fragmentation products. In this tandem MS experiment, reaction pathways can be observed directly which can hardly be analyzed in solution. According to these results, the equilibration of triangles and squares involves the supramolecular analogue of a neighboring‐group effect.

A series of rotaxanes has been synthesized which contain two ester groups in their axles. All rotaxanes bear the same tetralactam wheel. The kinetics of the de-slipping reaction of these rotaxanes were monitored in tetrachloroethane (TCE) and dimethyl sulfoxide (DMSO) resulting in the observation of a significant solvent effect. In TCE, two isomeric rotaxanes that differ merely with respect to the orientation of the ester groups show a remarkable difference in their deslipping behavior. When the ester carbonyl group is directly attached to the axle center piece, the rotaxane decomposes with a half life of ca. 10 h at 100°C. The reverse orientation with the

A series of new rotaxanes with axles different in length was prepared. Following the synthetic protocol utilizing a known anion template effect (Scheme 1), surprisingly low yields in the order of 2 ± 5% were obtained (Scheme 3), which furthermore significantly depended on the nature of the stopper (Fig. 1). Variations in the synthetic procedures and computational results from Monte Carlo simulations allowed us to analyze the origin of these findings: The rotaxane wheel 3 acts as a noncovalently bound −protecting group× for the stopper nucleophile. The protection of the nucleophilic phenolate O-atom depends much on the steric demands of the stoppers (see 2 vs. 10) which induce different conformations of the wheel. Based on this model, an improved synthetic scheme is suggested. 1. Introduction. ± The synthesis of rotaxanes, catenanes, and other types of mechanically interlocked molecules [1] strongly relies on the operation of efficient template effects [2]. There exist different apparoches including inter alia those based on the tetrahedral [3] or octahedral [4] coordination geometry of metal ions, p-donor/pacceptor interactions [5], and H-bonding involving ammoniums ions [6] or neutral amides [7]. The latter is believed to play a crucial role in the formation of a trefoil knot with twelve amide bonds [8]. Recently, Vˆgtle and co-workers reported [9] the highyield rotaxane synthesis shown in Scheme 1, which is based on the recognition of phenolate stopper 2 À within the macrocyclic rotaxane wheel 3. Two H-bonds bind the trityl phenolate with a surprisingly high binding constant of K > 10 5 m À1 [9] [10] so that the equilibrium is shifted far to the side of the stopper-wheel complex 2 À ¥ 3. If the axlecenter piece 1 is added to the reaction mixture, the semi-axle 4 is formed ± either in a direct reaction of 1 with free 2 À , or in a reaction of 1 and 2 À ¥ 3 followed by deslipping that occurs due to the much lower strength of the H-bonds formed between the wheel and the semi-axle ether O-atom. Finally, the semi-axle 4 reacts with 2 À ¥ 3 to yield the rotaxane 5 ¥ 3 in up to 95% yield.For deslipping experiments [11], we attempted to synthesize rotaxanes with axlecenter pieces of different lengths, i.e., 9a ± j (obtained from 6 and 7a ± j via 8a ± j, see Scheme 2), and stoppers of intermediate size such as 10 (Scheme 3). Surprisingly, the rotaxane yields decreased dramatically for the rotaxanes discussed here, even to below 5%. Instead, large amounts of the free axle were isolated as the by-product. Three questions arise from these findings: i) Why does the yield of rotaxane depend so much on the nature of the stopper? ii) If this is due to an unfavorable competition between axle and rotaxane formation, why is the free axle formed so much faster than the rotaxane if 10 is applied as the stopper instead of 2? iii) What is the influence of the center piece?

No abstract

A series of rotaxanes equipped with tetralactam wheels and di-t-butyl stopper groups, which differ only with respect to the length of the alkyl chain serving as the axle center piece, are examined with respect to their deslipping behavior at elevated temperatures. 1 H NMR experiments are used to follow the deslipping reactions kinetically, and it is found that the axle length does not have a significant influence on the rate of deslipping. In accordance with expectation, this result confirms that the rate-determining step of the reaction is the passage of the wheel over the stopper group. It also sheds new light on earlier results on rotaxanes bearing ester groups in the axle center piece. The deslipping reaction rate for these rotaxanes was extremely dependent on the choice of solvent and the orientation of the ester groups. The rotaxanes presented here do not contain functional groups in their center pieces and no such effects are observed. For these rotaxanes, the simple model of a thin thread inside a macrocycle mechanically trapped by bulky stopper groups is valid.