Secondary Oxonium Ions

With secondary and tertiary alcohols Ihis slage is an 8 1 reaclion m which Ihe alkyl oxonium ion dissociates to a carbocalion and water... 

A protonic acid derived from a suitable or desired anion would seem to be an ideal initiator, especially if the desired end product is a poly(tetramethylene oxide) glycol. There are, however, a number of drawbacks. The protonated THF, ie, the secondary oxonium ion, is less reactive than the propagating tertiary oxonium ion. This results in a slow initiation process. Also, in the case of several of the readily available acids, eg, CF SO H, FSO H, HCIO4, and H2SO4, there is an ion—ester equiUbrium with the counterion, which further reduces the concentration of the much more reactive ionic species. The reaction is illustrated for CF SO counterion as follows ... 

Entry 4 shows that reaction of a secondary 2-octyl system with the moderately good nucleophile acetate ion occurs wifii complete inversion. The results cited in entry 5 serve to illustrate the importance of solvation of ion-pair intermediates in reactions of secondary substrates. The data show fiiat partial racemization occurs in aqueous dioxane but that an added nucleophile (azide ion) results in complete inversion, both in the product resulting from reaction with azide ion and in the alcohol resulting from reaction with water. The alcohol of retained configuration is attributed to an intermediate oxonium ion resulting from reaction of the ion pair with the dioxane solvent. This would react until water to give product of retained configuratioiL When azide ion is present, dioxane does not efiTectively conqiete for tiie ion-p intermediate, and all of the alcohol arises from tiie inversion mechanism. ... 

Contained within intermediate 25 is an acid-labile mixed acetal group and it was found that treatment of 25 with camphorsulfonic acid (CSA) results in the formation of dioxabicyclo[3.3.0]octane 26 in 77 % yield. Acid-induced cleavage of the mixed cyclic acetal function in 25, with loss of acetone, followed by intramolecular interception of the resultant oxonium ion by the secondary hydroxyl group appended to C leads to the observed product. Intermediate 26 clearly has much in common with the ultimate target molecule. Indeed, the constitution and relative stereochemistry of the dioxabicyclo[3.3.0]octane framework in 26 are identical to the corresponding portion of asteltoxin. 

The use of iodotrimethylsilane for this purpose provides an effective alternative to known methods. Thus the reaction of primary and secondary methyl ethers with iodotrimethylsilane in chloroform or acetonitrile at 25—60° for 2—64 hours affords the corresponding trimethylsilyl ethers in high yield. The alcohols may be liberated from the trimethylsilyl ethers by methanolysis. The mechanism of the ether cleavage is presumed to involve initial formation of a trimethylsilyl oxonium ion which is converted to the silyl ether by nucleophilic attack of iodide at the methyl group. tert-Butyl, trityl, and benzyl ethers of primary and secondary alcohols are rapidly converted to trimethylsilyl ethers by the action of iodotrimethylsilane, probably via heterolysis of silyl oxonium ion intermediates. The cleavage of aryl methyl ethers to aryl trimethylsilyl ethers may also be effected more slowly by reaction with iodotrimethylsilane at 25—50° in chloroform or sulfolane for 12-125 hours, with iodotrimethylsilane at 100—110° in the absence of solvent, " and with iodotrimethylsilane generated in situ from iodine and trimcthylphenylsilane at 100°. ... 

Trimethylsilyl iodide (TMSI) cleaves methyl ethers in a period of a few hours at room temperature.89 Benzyl and f-butyl systems are cleaved very rapidly, whereas secondary systems require longer times. The reaction presumably proceeds via an initially formed silyl oxonium ion. 

Thus the quantity of EtOH in the hydrolysate is equivalent to the number of tertiary oxonium ions any secondary oxonium ions react with EtO to give EtOH which is removed before hydrolysis of the polymer. 

Hetero-cations, such as secondary or tertiary oxonium ions can be formed easily by the addition of a proton or a carbenium ion to an ether ... 

One must treat a reaction mixture with the sodium salt RO-Na+ of an alcohol or phenol, and search for compounds of the type H-[0(CH2)20CH2] -0R, which would be formed from homologues of (VII) (Mainz theory), but not from secondary oxonium ions such as (V) or (VI) (Y = H) (Keele theory), which would form ROH. 

The test (b) we carried out typically as follows but we have also used many variations of this procedure [18]. We used an assembly of connected reaction tubes attached to the vacuum line. In one tube we polymerised (I) by perchloric acid in methylene dichloride. Reaction was stopped by adding sodium phenate, and any phenol formed from secondary oxonium ions was neutralised with sodium hydride. The volatile compounds were distilled into a second tube where the same experiment was repeated. This technique is based on that of Saegusa and Matsumoto [19] phenol and phenyl ethers can be estimated separately by their UV spectra. 

Although this work is still incomplete, it shows that some tertiary oxonium ions are formed in the reaction, but that by far the greater part of the active species are secondary oxonium ions. The origin of the tertiary oxonium ions, which yield the involatile phenyl ether by reaction with C6H5CT, is not at all clear at present. Some may be formed from an impurity in the monomer and others may arise from a slow side-reaction. 

It follows that when the water content of the reaction mixtures is less than ca. 10 4 M the propagation is principally by secondary oxonium ions, and that it must therefore go by the ring-expansion mechanism of Plesch and Westermann. 

The results obtained by Jaacks s ethoxide method, shown in Tables 1 and 2, prove that for all systems the concentrations of tert-oxonium ions are very considerably smaller than those of the perchloric acid. In view of the correlations shown above, they must also be much smaller than the concentrations of ions in the polymerising solutions. We conclude, therefore, that the principal growing ions are not tertiary, and that they must, therefore, be secondary. 

Since most types of experimental inadequacy or incompetence produce more ethanol in the hydrolysate than could be derived from fert.-oxonium ions, one must conclude from this evidence that the propagating species is a secondary oxonium ion. It follows necessarily that the ring-expansion mechanism is the best representation of the propagation reaction, that tert.-oxonium ions do not play an essential role in the polymerisation of DCA by perchloric acid, and that therefore this part of the old controversy appears now to be settled. 

Strong protonic acids such as trifluoroacetic, fluorosulfonic, and trifluoromethanesulfonic (triflic) acids initiate polymerization via the initial formation of a secondary oxonium ion... 

This type of initiation is limited hy the nucleophilicity of the anion A derived from the acid. For acids other than the very strong acids such as fluorosulfonic and triflic acids, the anion is sufficiently nucleophilic to compete with monomer for either the proton or secondary and tertiary oxonium ions. Only very-low-molecular-weight products are possible. The presence of water can also directly dismpt the polymerization since its nucleophilicity allows it to compete with monomer for the oxonium ions. 

Combinations of a Lewis acid, protogen or cationogen, and a reactive cyclic ether (e.g., oxirane or oxetane) have been used to initiate the polymerization of less reactive cyclic ethers such as tetrahydrofuran [Saegusa and Matsumoto, 1968]. Initiation occurs by formation of the secondary and tertiary oxonium ions of the more reactive cyclic ether, which then act as initiators for polymerization of the less reactive cyclic ether. The reactive cyclic ether, referred to as a promoter, is used in small amounts relative to the cyclic ether being polymerized and increases the ability of the latter to form the tertiary oxonium ion. 

Resonance, similar to that in pyrylium salts, was shown594,595 to exist between oxonium ion (299a) and carbenium ion (299b) forms in alkylated ketones, esters, and lactones that were obtained via alkylation with trimethyl- or triethyloxonium tetra-fluoroborates596 [Eq. (3.78)]. Ramsey and Taft597 used H NMR spectroscopy to investigate the nature of a series of secondary and tertiary carboxonium ions (300-302). 

Secondary Oxonium Ions [RR OH ]. 1H NMR and IR investigation75 of protonated ether salts (hexachloroantimonates) in dichloromethane solutions showed the formation of both (a) dialkyloxonium ions in which the proton is bound to only one oxygen (21) and (b) bidentate complexes in which the proton is shared between two ether molecules (22). Structural analysis of such a bidentate complex of diethyl ether with a complex anion shows a broad H+ resonance at 81 H 16.3 and unequal O—H bond distances (1.39 and 1.11 A).76... 

Trialkyloxonium salts can also be prepared by the reaction of secondary oxonium ion salts with diazoalkanes93 [Eq. (4.16)]. 

---