Evidence for the Template Effect

It is not clear exactly when the association illustrated above actually takes place. It is certainly involved by the final ring closure stage, but it seems reasonable to assume that there is some cation-glyme type interaction taking place from the instant of solution. The fact that wrapping occurs in such a way that the two ends of the molecule are held in proximity, allows the reaction to be conducted at much higher concentrations than might otherwise be practical. Tlie evidence for the operation of such a template effect is presented and considered below. 

One of the prime tenets of the template effect is that a cation of appropriate size coordinates with the heteroatom binding sites in the incipient ring system. Evidence to be discussed below indicates that there is a clear relationship between the size of the coordinating cation and the size of the ring formed. Before presenting this evidence, it is useful to note the approximate ring sizes of common crowns and the known ionic diameters of sundry metal cations. This is done in Table 2.1. 

Tlie first suggestion of a template effect which was offered in the literature was made by Greene in 1972 °. Tlie illustration of this concept is approximately that shown in Eq. (2.2), above. Greene presented several pieces of evidence in addition to the concept itself. First, he noted that when the final concentration of 18-crown-6 in a reaction mixture (MejSO) was increased from 0.04 At to 0.09 Af, the yield of crown dropped only slightly (84% to 75%). In a competition experiment, equal amounts of 18-crown-6 and 21-crown-7 were formed when one molar equivalent each of triethylene glycol and tetra-ethylene glycol were allowed to compete for triethylene glycol ditosylate (KO-t-Bu/THF). 

Crown Ether (hole size, A) Favored Cation (ionic diameter, A)  

Equal amounts of the two crowns were also formed when BU4NOH was used as base, but both rate and yield were considerably reduced. Greene noted that such a template effect would actually involve binding between the open-chained intermediates and cations and these interactions are known to be weak relative to crown-cation interactions. Nevertheless, evidence for such interactions exists (see also Chap. 7)  

---

The addition of various alkali and alkaline earth cations to the cyclooligomerization of ethylene oxide by Dale and Daasvatn also provides strong presumptive evidence for the template effect. Recently, Reinhoudt, de Jong and Tomassen utilized several metal fluorides to effect crown formations . The reaction rates were found to be in the order Cs > Rb > > Na LE. Such an order would be expected on the basis of binding... 

Whereas Gold and Sghibartz showed that cation complexation depressed the rate of crown-ether disrupture, there is convincing evidence that crown ether formation is facilitated by the presence of cations. The template effect, presumably due to complexation of the open-chain precursor and formation of a crown-type conformation, clearly emerged from studies in which the yield of crown ethers was related to the type of cations present (Reinhoudt et al., 1976). Kinetic evidence for the template effect was presented by Mandolini and Masci (1977), who showed that the rate of cyclization of the precursor of benzo-18-crown-6 [2061 decreased in the order Ba2+ > SrJ+ > K+ > Na+ > Li+. This sequence is the same as the one found for the stability constants of the 1 1 complexes of these cations with 18-crown-6 in water (Table 3). 

Mandolin , L. Masci. B. Kinetic evidence for the template effect of added cations on the rate of formation of benzo-18-crown-6 in water solution. J. Am. Chem. Soc. 1977, 99. 7709-7710. 

There is probably no better evidence for a template effect than its application directly in the solution of a synthetic problem. Rastetter and PhiUion have utilized a substituted 19-crown-6 compound (shown below in Eq. 2.9) in the formation of macrocyclic lactones. Although there were certain experimental variations and the the possibility of intermolecular potassium ion complexation, the overall formation of lactone was favorable. 

However, studies with self-supported and much thinner membranes by Sergeyeva et al. [90,91] (cf. Section III.D.l) provided very clear evidence for the gate effect a most remarkable template specificity of conductivity response could be observed (seeTable 1). Similar, supporting results hadbeen obtainedby another group [104]. One more convincing proof for the gate effect was results of Hattori et al. [101] with pore-filled MIP composite membranes (cf. Section III.E) The transport rate of another solute (creatinine) increased 1.23-fold in the presence of the template (Tho) while without any additive and with Caf the flux was the same (seeTable 1). 

The template effect has long been accepted prima facie by workers in the crown field because of the obvious relationship between ring size and complexation constant. In fact, Cram stated in 1975 that the templating properties of for preparing crown ethers is well established.. . In fact, the template effect was widely acknowledged and has always received overwhelming support. Nevertheless, few direct comparisons are available in the literature and we have restricted ourselves m this discussion only to direct rather than the large body of presumptive evidence which is currently available. 

One of the most spectacular and useful template reactions is the Curtis reaction , in which a new chelate ring is formed as the result of an aldol condensation between a methylene ketone or inline and an imine salt. The initial example of this reaction was the formation of a macrocyclic nickel(II) complex from tris(l,2-diaminoethane)nickel(II) perchlorate and acetone (equation 53).182 The reaction has been developed by Curtis and numerous other workers and has been reviewed.183 In mechanistic terms there is some circumstantial evidence to suggest that the nucleophile is an uncoordinated aoetonyl carbanion which adds to a coordinated imine to yield a coordinated amino ketone (equation 54). If such a mechanism operates then the template effect is largely, if not wholly, thermodynamic in nature, as described for imine formation. Such a view is supported by the fact that the free macrocycle salts can be produced by acid catalysis alone. However, this fact does not... 

Taking into account that such a perfect correlation with the ultramicropore volume is not so well observed for all kinds of porous carbons [131], it is once more obvious that the electrochemical performance for hydrt en storage is strongly controlled by the presence of mesopores connected to the microporcs. The mesopores are essential to Teed the micropores with hydrated ions or water molecules which will be further reduced giving hydrogen able to be fixed in the ultramicropores. A direct evidence of the beneficial effect of mesopores has been demonstrated by the carbonization of sucrose in the SBA-15 template... 

Further evidence for the importance of the template molecule in effecting the selectivity of the reaction was that more harsh reaction conditions that disrupted... 

It should be noted that uncoordinated allylphosphine does not cyclise when treated with initiator, This gives evidence for the kinetic coordination template effect operating in this process. In conditions of kinetic control (enforced by spatial effects resulting from coordination to the metal), macrocyclic product formation by addition of P-H to the C=C double bond of a neighbouring coordinated allylphosphine proceeds contrary to Markovnikov s rule (phosphorus as a more electronegative element bonded to a less substituted carbon atom). This template process is notable for its exceptional regiospecific character in that the formation of Markovnikov addition products is not detected. 

Although it was stated above that the templating effect of aniline does not appear to be a strong one, it must nevertheless have some influence, given the evidence for complexation of Ce(IV) by aniline, leading to the precipitation of a complex. In addition, FTIR spectra of samples prepared in the presence and absence of aniline present differences in both the OH stretch peaks at about 3400 cm as well the Ce-O-Ce bands at 1300 cm. ... 

---