Cyclic formal

Cyclic formals (formaldehyde-cyclic ether dimer) ... 

Despite the statement above concerning the acid lability of cyclic formals, Gold and Sghibartz have shown that the acid catalyzed hydrolysis of these compounds is markedly depressed by some metal ions . Although the smaller cyclic formals did not exhibit a substantial rate reduction even in the presence of small cations like lithium, in certain larger systems the rate reduction was more than an order of magnitude. 

The propagating species involved in the polymerization of cyclic formal seem to resemble carbocations, and random copolymers are formed in the copolymerization of cyclic formals with styrene. For the copolymerization of DOL with styrene, the DOL-St cross-sequence was estimated, by NMR or by chemical methods, from the decrease of the formal unit in the copolymer and the formation of nearly random copolymer was confirmed132. ... 

Temperature programmed GC (Fig. 2) separates these components as well as a cyclic formal. The mono, di and tri brominated products of 1 require higher temperatures to elute in a reasonable time more than the column can withstand. TMS derivatives do not require temperatures quite so high (Fig. 3). Using this technique for quantitation, however, is complicated by the decreasing sensitivity of the FID to increasing bromine content. 

Reaction of trans-1,2-cyclohexanediol with para-formaldehyde in the presence of Indion 130 as catalyst to yield the corresponding cyclic formal has been successfully carried out (Matkar and Sharma, 1995). A conversion of 52% was realized with 70% selectivity towards cyclohexanediol formal the other side products are rrans-hexahydrobenzo-l,3,5-trioxypin, di(tra i -2-hydroxycyclo-hexyloxy) methane. 

Studies on the cationic polymerization of cyclic ethers, cyclic formals, lactones and other heterocyclic compounds have proliferated so greatly in the last few years that a detailed review of the evidence concerning participation of oxonium and analogous ions in these reactions cannot be given here. Suffice it to say that there is firm evidence for a few, and circumstantial evidence for many such systems, that the reactive species are indeed ions and there appears to be no evidence to the contrary. A few systems will be discussed in sub-sections 3.2 and 4.4. 

The propagation for cyclic formals also involves a solvated oxonium ion and a highly polar monomer [10, 12]. Since under optimum conditions the polymers are essentially free from any kind of end-group (except very small amounts of -OH) Plesch and Westermann [36, 37] concluded that they must be formed by a ring-expansion mechanism, involving a 4-centred transition state ... 

However, for a variety of reasons it seems extremely unlikely that the same mechanism is applicable to the polymerisation of cyclic formals and acetals. One reason is that these compounds cannot be co-polymerised with cyclic ethers another is that the polymers are predominantly cyclic, with the number of end-groups far smaller than the number of growing chains. One mechanism which has been proposed and which accounts for most of the observations involves formation of an oxonium ion (X) from the initiator and the monomer, and a subsequent propagation by a ring-expansion reaction (see 13). 

The Mechanism of the Polymerisation of Cyclic Formals, P.H. Plesch, IUPAC International Symposium on Macromolecular Chemistry, 1969, Plenary and Main Lectures, Budapest, 1971, 213-221. 

There are many studies of the mechanism and kinetics of the polymerisation of cyclic oxygen compounds, but only relatively few of these are concerned with cyclic formals. In the present paper I will review this field, and I hope to show that formals have some special characteristics which distinguish their polymerisations from those of other cyclic oxygen compounds. Since it is not possible to deal with all aspects of this group of reactions in one lecture, I will concentrate attention on questions of chemistry and mechanism, and I will not deal with other aspects, such as the thermodynamics and kinetics of these polymerisations. Most of the published work has been done with 1,3-dioxolan (I) and there are only very few papers on any other cyclic formals, although the patent literature on the homo- and copolymerisation of (I) and other cyclic formals is quite extensive. 

The polymerisation of (I) has been studied over a wide temperature range, in bulk and in solution, and with several different catalysts. All these are protonic acids, or metal halides, or carbonium, carboxonium, or oxonium compounds. None of the cyclic formals tested can be polymerised by radicals or anions. 

Since the carboxonium ion has been eliminated as a possible propagating species, one is left with two alternatives which we may call the Keele and the Mainz theories. Plesch and Westermann [6, 8] have suggested that the cyclic formals polymerise by a ring-expansion mechanism, in which no free end is ever formed. This is illustrated in Reaction (B), where Y = H if the initiator is a protonic acid, and Y = Et if the initiator is a... 

The ring expansion mechanism is of course only a special case of the well-known mechanism by which dioxolan reacts with non-cyclic formals e.g., (I) and CH2-(OMe)2 give (MeOCH2OCH2-)2 in this way. It also accounts in a simple manner for the cleanness of the monomer-polymer equilibrium and for the high yields of cyclic dimer (without any detectable linear fragments) which are obtainable from 1,3-dioxane and 1,3-dioxepan [8]. 

From the foregoing discussion it appears that the mechanism of polymerisation of cyclic formals may differ considerably from that by which cyclic ethers polymerise,... 

Of course, even for the cyclic formals there are still unresolved problems, especially the mechanism by which metal halides initiate their polymerisation [16]. Once again it has become evident that attention to impurity effects and side-reactions is of paramount importance if the conclusions from chemical studies of catalytic reactions are to be valid. 

I cannot conclude without referring to the numerous papers dealing with the kinetics of cyclic formal polymerisations. Some of these show considerable mathematical virtuosity in interpreting the kinetic results, but very regrettably they contribute relatively little to our understanding, because the mechanisms used are completely unsubstantiated by any chemical evidence and the nature and concentration of growing centres is unknown. We see here again that kinetics without chemistry is little more than a mathematical exercise. 

We must also remember that so far only one single cyclic formal has been studied in detail. It seems to me very likely that this class of compounds, which are so easily prepared from cheap materials, will eventually repay a much more extensive study. 

In concluding this paper dedicated to our distinguished octogenarian, it is especially appropriate to mention the heuristic practical value which our ring-expansion theory has had, since he has always been intent upon applications and practical uses. It was the ring-expansion theory which led the senior author to imagine that the acid-catalysed reactions of cyclic formals with olefins would be insertion reactions [21]. In his view they involve the insertion of an olefin into the 0-1, C-2 bond of a protonated (or alkylated) 1,3-dioxacycloalkane ... 

While a variety of techniques are available for the monoprotection of symmetrical diols, there are few methods that allow for the chemoselective functionalization of the more hindered hydroxyl in an unsymmetrical 1,3-diol.5 The acid-catalyzed reaction of an unsymmetrically substituted cyclic formal with acetyl chloride described here invariably proceeds via preferential rupture of the less congested C(2)-0 bond to give a product having an acetate at the less congested site... 

Cyclic formals react with dinitrogen pentoxide in chlorinated solvent to yield unstable but interesting ring-opened products, including hemiformal nitrates 1,3-dioxolane (44) reacts to yield a mixture of hemiformal nitrate (45) and formal ether (46) products. Similar products are formed from acyclic formals and dinitrogen pentoxide. ... 

Some of the more important monomers whose ring opening polymerisations have been induced by stable cation salts include, 1,4-epoxides, notably tetra-hydrofuran (20,112,113), 1,2-epoxides (114), 1,3-episulphides (thietans) (33,53), 1,2-episulphides (thiiranes) (53), azetidines (115,116), aziridines (117), the cyclic formals, 1,3-dioxolan (23,54, 118-120), and 1,3-dioxepan (118,119), trioxane (121,122) and more recently lactones (123). Aldehydes (124) may also be included since these molecules can be regarded as the smallest possible oxygen hetero-... 

THF readily copolymerizes with cyclic formals also (25, 104). Okada et al. (104) have determined reactivity ratios for copolymerization of THF with 1,3-dioxolane using BFg-etherate catalyst at 0°C. The values found are (THF) = 28 4, and r2 (dioxolane) = 0.25 0.05. 

DOL (6), although we could obtain only a homopolymer mixture in the copolymerization of -PL and St. We intend to obtain a high molecular weight polymer with such a random sequence of cyclic ether and vinyl monomer by using cyclic formals as intermediates. 

Cleavage of oxacyclic compounds (cyclic formals) from the cationic chain ends (9,18). 

The structure of the cationic chain ends is not clear in the polymerization of cyclic formals. Two different kinds of active centers and hence two types of propagation reactions have been proposed (9) ... 

Hill and Carothers (4) investigated the acetal interchange reaction with di-n-butyl formal and glycols (Equation 4). When n was 3 or 4, cyclic formals were the principal products. Fentamethylene glycol (n = 5) gave a sirupy liquid polymer. The reaction with decamethylene gly-... 

---