Tetra-urea

Tetra-urea Calix[4]arenes - From Dimeric Capsules to Novel Catenaries and Rotaxanes... 

Hydrogen-bonded dimers of tetra-urea calix[4]arenes offer a unique, unprecedented preorganization [28], Before we describe how this can be used for the rational construction of molecules with novel topologies, it is reasonable to discuss the properties of such dimeric capsules and the conditions under which they exist. 

Starting from tert-butylcalix[4]arcnc [29], the preparation of a typical tetra-urea derivative usually requires four steps (Scheme 5.8) ... 

The isocyanate can be replaced by the corresponding activated urethane (as a milder reagent), and this activation can also involve the p-amino functions of the calix[4]arene (step e) and their subsequent reaction with a suitable amine (step f). Multigram quantities of simple tetra-urea calix[4]arenes are easily available in this way. 

Scheme 5.8 Synthesis of tetra-urea calix[4]arenes. (a) Alkylating reagents, NaH, DMF, rt (b) HN03/H0Ac, CH2CI2, rt (c) H2, Raney-Ni, toluene, rt (d) R-NCO, rt (e) p-nitrophenyl chloro-formate (f) R-NH2, rt.

Scheme 5.9 Synthesis of tetra-urea calix[4]arenes bearing two different urea residues in the sequence ABAB (a) and AABB (b), using the protection of amino functions by trityl- or Boc-groups. (Y = C5Hln).

Figure 5.1 Sections of nH NMR spectra (400 MHz, 25 °C) of tetra-urea 1 (a) in DMSO-d6 (monomers) (b) in CDCI3 (dimeric capsules 1CDCI31) (c) in CDCI3 in the presence of Et4N+-BF4 (dimeric capsules [lEt4N+l]BF4 ). The solvent signal is marked by an asterisk.

Figure 5.2 Molecular structure of a tetra-urea dimer 2-2 determined by X-ray analysis [34], (a) Side view (b) top view. Alkyl residues (in b), hydrogen atoms, and solvent molecules are omitted for clarity.

A dimeric capsule can also be formed by two different tetra-urea molecules, and in aprotic solvents a mixture of two tetra-urea derivatives usually contains not only the two homodimers but also the heterodimer in a more or less statistical ratio. This formation of heterodimers is an additional proof of the dimerization [39], which is also valid when other indications (e.g. m-coupled doublets for aryl protons) fail. As an example, sections of the spectra of tetra-tolylurea (1), tetra-hexylurea (3), and of a mixture of the two are shown in Figure 5.3a-c [37a],... 

However, the statistical formation of heterodimers is not the rule. Examples are known where no heterodimers are formed, e.g. between 1 and tetra-ureas derived from a rigid bis-crown-3 [40] (see Section 7). Further examples will be shown in the next sections. In such cases, the solution of the two tetra-urea calix[4]arenes contains exclusively the two homodimers. 

Figure 5.3 Sections of NMR spectra (CDCI3, 400 MHz, 25 °C) of tetra-urea dimers with tetraethylammonium as guest (a) [l-Et4N+-l]BF4 (b) [3-Et4N+-3]BF4 (c) 1 1 mixture of [l-Et4N+-l]BF4 , and [3-Et,N+-3]BFr. In addition to the signals for the two homodimers (a, b), the mixture (c) contains a double set of signals for the two parts of the heterodimer.The signals of free Et4N+are marked by a circle.

On the other hand, there are examples of exclusive heterodimerization. Both tetra-aryl- and tetra-tosylureas readily form homodimers when they are dissolved alone in an apolar solvent. However, heterodimers are the only detectable species in a solution containing both compounds in a 1 1 ratio. An example of this selective heterodimerization [41] is demonstrated by the NMR-spectra in Figure 5.4. Only when one of these tetra-ureas is present in excess can signals for its homodimer be seen by NM R. This early observation by Rebek and Castellano [42] is especially important for the template syntheses described below. Other examples of exclusive formation of heterodimers will be discussed later. 

A homodimer of a tetra-urea calix[4]arene consisting of identical phenolic units A is composed of two enantiomers with C4-symmetry, which results in overall S8-symmetry. Consequently, a heterodimer with a second calixarene consisting of four units B must be chiral, but this chirality is due only to the directionality of the hydrogen-bonded belt or (in other words) to the orientation of the carbonyl groups [42,43]. Rotation around the (four) aryl-NH bonds leads to the opposite enantiomer (conformational chirality). 

Preorganization in Dimers of Tetra-urea Calix[4]arenes 153... 

Figure 5.5 Survey of the symmetry properties for selected types of homo- and heterodimers of tetra-urea calix[4]arenes, represented by squares with the phenolic units (A, B) on the corners. Symmetry elements and symmetry classes (with and without directionality of the hydrogen bonds, shown by arrows) are indicated.

The inclusion of guests is not the only interesting aspect of the dimers formed by tetra-urea calixarenes. Numerous other self-assembled capsules are known, larger in volume, different in shape, and able to include more than one guest molecule, and various aspects of guest inclusion have been extensively studied [45]. 

To obtain selectively a single, well-defined product, it is obviously not sufficient to arrange the reacting groups in an appropriate mutual position. A perfect preorganization also demands the separation of those functional groups which should not be involved in the reaction. For an intramolecular connection between reactive functions attached to the urea residues, this preorganization is possible in heterodimers with a non-reactive tetra-urea calix[4]arene. 

Figure 5.8 Formula survey of tetra-urea calix[4]arenes used as precursors for bis-, tris-, and tetraloop tetra-ureas (Y = alkyl).

The inability of bis-, tris-, or tetraloop compounds to form homodimers and the general tendency of tetra-urea calix[4]arenes to dimerize can be further exploited for the synthesis of mechanically interlocked molecules. A 1 1 mixture of tetra-ureas 5 or 6 with bis- or tetraloop compounds 8 or 9 (in practice the non-reactive loop component is added in a small excess) contains exclusively heterodimers (e.g., 5-8, 5-9, or 6-9), since this is the only possibility, to have all the urea functions involved in the favorable hydrogen-bonded belt.5 Again this is easily evidenced by the complete absence of peaks for the homodimer 5-5, or 6-6 in the H NMR spectra (for an example see Figure 5.11). 

Molecular models suggest that such a P -connection between remote alkenyl residues in a dimer is possible. To check whether it really occurs under the conditions of the metathesis reaction, we synthesized the monoloop tetra-urea compounds 16, in which one of the non-cyclic urea residues is substituted by a bulky group which cannot penetrate the loop (Scheme 5.18). (The synthesis is analogous to that of 15, introducing in the final two steps the bulky residue first.). 

Formally compound 16 belongs to the AABC-type of tetra-ureas (compare Section 5.3.2) for which numerous regioisomeric dimers are possible. However, these substituents at the urea residues ensure that selectively only one dimer is formed, in which no overlap of the loops occurs and no penetration of the loop by the bulky group takes place. NMR spectra are in agreement with the formation of a single C2-symmetrical dimer, which is necessarily composed of the same enantiomer of 16. 

Bulky residues can be attached to the urea groups of a tetra-urea under conditions where it forms a heterodimer (a pseudoro-taxane) with a bis-, tris-, or tetraloop tetra-urea. Often this strategy is called stoppering (see Scheme 5.7), because stoppers are attached to the axles. 

Multiple ring-closure reactions between adjacent urea functions of a suitably functionalized tetra-urea in a heterodimer with a second tetra-urea bearing bulky groups can also create the structural elements of a rotaxane. This strategy is usually called clipping (see Scheme 5.7). 

To realize the first possibility, the tetra-ureas 20 bearing maleic imide functions at all four urea arms were synthesized. These dienophiles react with anthracene derivatives, here 1,4,5,8-tetrapentoxyanthracene 21, in a Diels-Alder cydoaddition (Scheme 5.21). 

Tetra-ureas 20 form homodimers in solvents such as CDC13 or C6D6, but they are completely converted to heterodimers upon addition of an equimolar amount of a tetra-loop compound 9. When a 5-10% excess of the anthracene derivative 21 was applied, a quantitative conversion was obtained in refluxing toluene after 72 h (as judged by NMR), and the tetra[2]rotaxane 22 was isolated in 40-50% yield. 

The rotaxane structure of 22 can be demonstrated by ESI-MS or MALDI-TOF-MS. It remains intact in TH F-d8, a solvent in which the tetra-urea dimers usually dissociate. However, an upfield shift of the NH-signals in 1 H NMR indicates that the... 

The synthesis of fourfold [2]rotaxanes by clipping requires the efficient formation of heterodimers between a tetra-urea substituted by bulky stopper groups and an octaalk-enyl urea 6. We initially hoped that the steric crowding in homodimers of a tetra-tritylphenylurea calix[4]arene would be sufficient to shift the equilibrium toward the heterodimers in a mixture with 6. However, the distribution of the dimers was close to the statistical ratio and the desired rotaxane could be obtained in only 5% yield [59]. 

---