Lactams dominate the group of bicyclic monomers which undergo uncoordinated anionic polymerization (Table 1).

Lactam synthesis usually involves removal of a small molecule from an amino acid derivative, often by simply heating the amino acid under vacuum to remove water. Cyclization of an aminoacyl chloride hydrochloride with triethylamine has also been employed. The polymerizations are catalyzed by the lactam anion, assisted by cocatalyst N-acyllactam. Bicyclic amides 1–15 shown below (except the one in brackets) have been converted to polymers; the polymer is shown for monomer 1 only.

When the carbonyl group is attached to the ring in the initial polymer, it may be epimerized by the strongly basic lactam anion (Scheme I).

Oxabicyclic lactam 16 (Scheme II) and derivatives were synthesized from starting material 16_sm by internal addition of the carboxamide group to the dihydropyran group. Although monomer 16 is both a bicyclic lactam and a bicyclic ether, it has only been polymerized as the former by an anionic mechanism; cationic ether polymerization is excluded because of the basic amide group. This topic has been reviewed [14,15,16].

An especially interesting class of lactam and related N-containing monomers contains a bridgehead N next to a carbonyl group. These are greatly destabilized relative to normal amides because the usual N–CO resonance is absent according to Bredt’s Rule, which forbids the occurrence of double bonds at bridgeheads as pointed out by Lukes [5]. Yakhontov also discussed this problem [6]. Pracejus [7] synthesized 1-aza-6,6-dimethylbicyclo[2.2.2]octan-2-one, while recent workers synthesized the parent structure as a salt by a nitrene insertion reaction; the free lactam could not be generated from the salt [8]. Although polymerization studies were not reported, these anti-Bredt lactams undoubtedly polymerize, being destabilized by their boat form as well as Bredt’s Rule violation. The effect can be offset with an adjacent heteroatom, or in ureas and urethanes; the results of these studies are discussed below.

Bicyclic ureas (Table 2) are prepared by carbonylation of the corresponding diamine [3]. They polymerize anionically. The bicyclo[3.3.1]nonane-N-bridgehead ureas 22 and 23 were stable because of electron donation to the carbonyl by the non-bridgehead N atom [17,18]. The molecules adopted a chair-boat conformation, as does bicyclo[3.3.1]non-1-ene.

Bicyclic urethanes (Table 3) are synthesized from amino alcohols with carbonylating agents such as phosgene and diphenyl carbonate [3]. Attempted cyclizations of 3-hydroxypyrrolidine and 4-hydroxypiperidine failed to yield bicyclic urethanes 24 and 29, respectively. However, the N-bridgehead urethanes 25 and 27 were sufficiently stable to be isolated because of the resonance stabilization provided by electron donation from the oxygen adjacent to the carbonyl [19,20,21]. This topic has been reviewed [22].

Lactones (Table 4) can undergo “uncoordinated” anionic polymerization, but “coordinated” anionic polymerization, in which the addition of a Lewis acid of a metal like aluminum, tantalum, or titanium provides simultaneous Lewis acid-base activation, is much more selective and does not lead to reshuffling of the polymer chains. The lactone monomers are synthesized by loss of small molecules from hydroxyacids or hydroxyesters. Also in this group are cyclic carbonates, prepared by carbonylation of alicyclic diols.

The polymer from monomer 31 retains the cis form when the polymerization is carried out under mild conditions [3,24]. Monomer 33 readily gave high polymer while monomer 32 gave lower molecular weight polymer. Monomer 36 showed a propensity for macrocyclic oligomerization. The bicyclic oxalactone polymers, also stemming from dihydropyrans, have been reviewed [14,15,16].

Ethers (Table 5)comprise the first group of oxygen-containing monomers we will consider which undergo cationic polymerization. Their polymerization involves oxonium ion intermediates. This topic has been reviewed [31].

The stereochemistry of these polymerizations, involving nucleophilic displacement on the propagating oxonium ion, is trans. The isomeric 2-methyl-7-oxabicyclo[2.2.1]heptanes gave stereoregular polymers [32,33,34].

Unsubstituted acetal monomers (Table 6) were prepared by syntheses involving malonic ester alkylation followed by acid-catalyzed bicyclization with loss of methanol. Crucial to the success of these reactions was removal of the sensitive acetal by distillation under reduced pressure as it formed [37,38,39,40]. This topic has been reviewed [41,42,43].

Unsymmetrical acetals like 45 can cleave in either or both of two ways (Scheme III).

Although atom-bridged monomers have generally been unsubstituted, the acetals represent a marked exception. Numerous atom-bridged bicyclic acetals like 50–52 derived from sugars (Scheme IV) have been used to synthesize unnatural polysaccharides. Their hydroxyl groups are protected by benzylation or acylation before polymerization and can be removed after polymerization [46,47,48,49]. These substituents can produce stereoregularity in the oxacarbenium ion polymerization.

Orthoesters (Table 7) are the third member of this oxygen group. They are synthesized from condensation of triols with ethyl orthoformate, again removing the bicyclic monomer as it forms [60,61,62,63,64,65]. This topic has been reviewed [41,42,43,66].

For the bicyclic acetals and orthoesters, oxonium ion propagation is still predominant. However, for these monomers, cleavage of the oxonium ion to oxacarbenium ion can lead to loss of stereoregularity [66]. Moreover, in the many sugar-derived monomers, neighboring group participation by adjacent benzoate or benzyloxy groups plays a role in determining polymer stereochemistry.

Finally, bicyclic amines 57 and 58 (Table 8) have also been polymerized cationically via S_N2 displacements on ammonium ion intermediates [66]. 1,4-Diazabicyclo[2.2.2]octane (57), widely used as a catalyst, is now seen to be a monomer itself.