General procedures in chain-growth polymerization

This chapter is intended to provide a general introduction to the laboratory techniques used in polymer synthesis, by focusing on some relatively well-known polymerizations that occur by chain-growth processes. In this way some of the more commonly used procedures in polymer chemistry are described. Due to the nature of the intermediates produced, such as free radicals, carbanions, carbocations, together with a range of organometallic species, the techniques often involve handling compounds in the complete absence of oxygen and moisture. Because of this the best results may require quite sophisticated equipment and glassware; however, it is our intention to show that the general procedures are accessible to any reasonably equipped laboratory, and indeed some of the techniques are suitable for use in an undergraduate teaching laboratory. Chain-growth polymerization involves the sequential step-wise addition of monomer to a growing chain. Usually, the monomer is unsaturated, almost always a derivative of ethene, and most commonly vinylic, that is, a monosubstituted ethane, 1 particularly where the growing chain is a free radical. For such monomers, the polymerization process is classified by the way in which polymerization is initiated and thus the nature of the propagating chain, namely anionic, cationic, or free radical; polymerization by coordination catalyst is generally considered separately as the nature of the growing chain-end may be less clear and coordination may bring about a substantial level of control not possible with other methods. Ring-opening polymerizations exhibit many of the features of chain-growth polymerization, but may also show some of the features expected from stepgrowth polymerizations. However, it is probably fair to say that from a practical point of view the techniques involved are rather similar or the same as those used in chain-growth processes and consequently some examples of ring-opening processes are provided here. It is particularly instructive to consider the requirements of chain-growth compared to step-growth processes in terms of the demands for reagent purity and reaction conditions.

Polymer characterization

Polymer science is, of course, driven by the desire to produce new materials for new applications. The success of materials such as polyethylene, polypropylene, and polystyrene is such that these materials are manufactured on a huge scale and are indeed ubiquitous. There is still a massive drive to understand these materials and improve their properties in order to meet material requirements; however, increasingly polymers are being applied to a wide range of problems, and certainly in terms of developing new materials there is much more emphasis on control. Such control can be control of molecular weight, for example, the production of polymers with a highly narrow molecular weight distribution by anionic polymerization. The control of polymer architecture extends from block copolymers to other novel architectures such as ladder polymers and dendrimers. Cyclic systems can also be prepared, usually these are lower molecular weight systems, although these also might be expected to be the natural consequence of step-growth polymerization at high conversion. Polymers are used in a wide range of applications, as coatings, as adhesives, as engineering and structural materials, for packaging, and for clothing to name a few. A key feature of the success and versatility of these materials is that it is possible to build in properties by careful design of the (largely) organic molecules from which the chains are built up. For example, rigid aromatic molecules can be used to make high-strength fibres, the most highprofile example of this being Kevlar®; rigid molecules of this type are often made by simple step-growth polymerization and offer particular synthetic challenges as outlined in Chapter 4. There is now an increasing demand for highly specialized materials for use in for example optical and electronic applications and polymers have been singled out as having particular potential in this regard. For example, there is considerable interest in the development of polymers with targeted optical properties such as second-order optical nonlinearity, and in conducting polymers as electrode materials, as a route towards supercapacitors and as electroluminescent materials. Polymeric materials can also be used as an electrolyte in the design of compact batteries.

Step-growth polymerization—basics and development of new materials

Step-growth polymerization is often referred to as condensation polymerization, since often—but by no means always—small molecules such as water are released during the formation of the polymer chains. There are a number of differences in the way polymerization occurs in step-growth polymerization compared to chain-growth processes, and these have marked practical implications. The most obvious difference is that, as the name implies, the polymer chain grows in a step-wise fashion; the initial stage of the reaction involves the conversion of monomers to dimers and from these other lower molecular weight oligomers. It is only as the reaction nears completion that significant quantities of higher molecular weight material can be formed. Thus, in order to obtain effective molecular weights, the reaction must proceed almost to completion, indeed the molecular weight (in terms of the number average degree of polymerization xn) of the polymer can be linked to the extent of reaction (p) using eqn (1). Thus, in the simplest case of a difunctional (AB) monomer, when 50% of the available groups have reacted, the number average degree of polymerization is only 2. The consequence of eqn (1) is that high molecular weights in step-growth polymerizations are associated with highly efficient reactions that do not have side-reactions. Notwithstanding this, the types of molecular weights associated with chain-growth processes are not encountered in these processes (except in the case of monomers with more than two reactive groups where hyper-branched or even cross-linked polymers are possible). There is an additional complication, namely the role of cyclization. Kricheldorf has recently shown that under perfect conditions cyclization is the ultimate fate of any polymerization reaction. Thus, under extremely high conversions the prediction given by eqn (1) would overestimate the actual molecular weights produced. Molecules that undergo step-growth polymerization must have at least two reactive functional groups. If the functionality is greater than this, for example, trifunctional, then hyperbranched polymers or even cross-linked systems can be formed. Commonly, this involves the reaction of two different reactive difunctional monomers.

Controlled/‘living’ polymerization methods

Chain-growth polymerizations such as free-radical polymerizations are characterized by four key processes:(i) initiation, (ii) propagation, (iii) chain transfer, and (iv) termination. If it is possible to minimize the contribution of chain transfer and termination during the polymerization, it is possible to achieve a level of control over the resulting polymer and achieve a predetermined number average molecular weight and a narrow molecular weight distribution (polydispersity). If such an ideal scenario can be created, the number of polymer chains that are produced is equal to the number of initiator groups; the polymerization will proceed until all of the monomer has been consumed and the polymer chain ends will remain active so that further addition of monomer will lead to continued polymerization. This type of polymerization was termed a ‘living’ polymerization by Szwarc in 1956 and represents one of the ultimate goals of synthetic polymer chemists. Flory determined that in the absence of termination, the number of propagating polymer chains must remain constant and that the rate of polymerization for each growing chain must be equal. In this situation, the number average degree of polymerization (DPn) and hence the molecular weight of the polymer can be predicted by simple consideration of the monomer to initiator ratio (see eqns (1) and (2), respectively). Several key criteria are used to elucidate the ‘living’ nature of a polymerization. For a polymerization to be considered ‘living’, the rate of initiation must exceed the rate of propagation. Therefore, all the propagating polymer chains are formed simultaneously and grow at the same rate. If this situation did not occur, the first chains formed would be longer than those initiated later and the molecular weight distribution of the propagating chains would broaden. In addition, an ideal ‘living’ or ‘immortal’ polymerization must not exhibit any termination of the propagating polymer chains over the lifetime of the reaction. Consequently, ‘living’ polymerizations are characterized by very narrow molecular weight distributions (Mw/Mn < 1.2).

Liquid crystalline polymers

The idea of combining the anisotropic behaviour of liquid crystalline materials with the properties of macromolecular systems was first suggested by Onsanger and subsequently Flory. The actual realization that such systems could exist came from studies of natural polymers such as the tobacco mosaic virus. Interest in these systems intensified with the development of highstrength systems, based on rigid-rod systems, notably the aramid fibres, however, liquid crystallinity in such systems occurs only at high temperatures, usually close to the decomposition point of the polymer. It was only in the late 1970s that the design criteria for liquid crystalline polymers became apparent, the secret being largely in the decoupling of the rigid aromatic groups which give rise to the anisotropic behaviour. As a result of these ideas two classifications of liquid crystalline materials were described. Main-chain liquid crystalline polymers, are those in which rigid aromatic molecules form part of the polymer backbone, either as a continuous chain or separated by a series of methylene groups in order to lower temperature at which liquid crystalline phase behaviour is observed. Side-chain systems resemble the comb-like systems studied by Shibaev and Plate, and have the rigid aromatic groups attached as a side-chain. In general, the monomer systems required for main-chain liquid crystalline polymers are relatively simple; synthetically these systems are prepared by step-growth methods and the main challenge is often maintaining sufficient solubility to allow suitable chain-lengths to be grown (an example of how such problems might be overcome is given in Chapter 4). Side-chain systems tend to be produced from more complex structural sub-units, and may be produced either by polymerization of the appropriate monomer or by functionalization of a preformed polymer backbone. Examples of both approaches are given in this chapter. From a practical viewpoint, the advantage of side-chain systems is that they tend to be much more soluble in common organic solvents and also that thermal phase transitions occur at reasonable temperatures (reasonable being well below the temperature at which the polymer decomposes). A further advantage of such side-chain systems is that the phase behaviour can be effectively tuned through the chemical modifications of the three components, namely the side-group, the flexible coupling chain and the polymer backbone.

New methodologies in the preparation of imprinted polymers

Molecular imprinting is a rapidly emerging method for the creation of recognition sites in synthetic polymers, and the resultant materials offer considerable promise as selective adsorbents in a number of applications. The technique exploits the principle of using elements of a target molecule to create its own recognition site. This is achieved by the formation of a highly cross-linked polymeric matrix around a template, which can be the target molecule itself or a close structural analogue. The key to this procedure is to ensure that, during the polymerization, functional groups of the template molecule are fully engaged in interactions with ‘complementary functionality’ of polymer-forming components. These interactions are then ‘locked in’ by the incorporation of the whole assembly into the polymer structure. Subsequent removal of the template reveals the newly created binding sites containing functional groups in the precise stereochemical arrangement to ensure recognition of the target in a highly selective manner (Scheme 1). The first reports of molecular imprinting in organic polymers involved the templating of protected sugars, in the form of esters with a polymerizable boronic acid (however, see Ref. 18 for an earlier example of the imprinting concept) into a cross-linked polymer ‘scaffold’, and variations of the basic technique have now been adopted by many research groups around the world. In general, molecularly imprinted polymers (MIPs) are prepared by thermal or photochemical free-radical routes, employing acrylic or vinylic monomers in a solvent chosen to ensure that the final matrix is microporous. The numbers and types of molecules which have now been imprinted is very large, but a key factor in the preparation of MIP materials with the desired recognition properties is still the chemical nature of the link between the template and the polymer backbone. Consequently, strategies by which the template can be securely fixed in space as the growing matrix forms around it, yet be readily removed to generate the recognition site after polymer synthesis is complete, are of particular interest.

The synthesis of conducting polymers based on heterocyclic compounds

Polymers are best known for their effectiveness as electrical insulators, indeed electrical wiring throughout the world is now sheathed in plastic. However, it was recognized early on that polymers with an appropriate structure ought to be able to conduct electricity. Unfortunately, the same features that might allow this phenomenon also introduce intractability and processing difficulties. As a consequence, it was not until the mid-1970s that the potential of these materials was explored and better-defined materials started to be made. There are now numerous polymers with substantial electrical conductivities and the topic of electrically conducting polymers still continues to excite with many hundreds of new publications printed each year. The backbone structures of some of conjugated polymers are given in Table 6.1. In this chapter we shall deal with electrochemical and chemical syntheses of some relatively simple examples. For electrical conductivity, it is necessary to transfer charge along a conjugated chain, between chains, and also along grain boundaries or between particles. The most energetically difficult process will control the rate of charge transport and this will vary with nature of the polymer, its physical form, and other parameters, but in all cases conjugation along the chain is necessary although it is not sufficient for carbonaceous polymers to simply possess a conjugated chain. To promote conductivity π-overlap along the entire polymer chain length is required to give a half-filled band of delocalized π -electrons. In real systems, distortions of the bonds disrupt the conjugation, and the materials are generally semiconductors. The higher metallic conduction can be achieved by a process known as doping in which electrons are added or more generally removed from the conjugated system (although this is not same as the doping process found in semiconductor technology) The simplest conjugated polymer chain is a polyacetylene chain. Such materials can be prepared by coordination polymerization, or using a sophisticated route involving the degradation of a soluble precursor polymer.

The formation of cyclic oligomers during step-growth polymerization

Step-growth polymerization is controlled both by the efficiency of the synthetic routes chosen (as indicated in Chapter 4) and by statistical considerations. In particular, the formation of the desired polymer is almost always accompanied by a cyclic oligomer fraction. As the dilution increases, the chances of cyclization also increase, since polymerization is a second-order process involving the reaction between linear species, whereas cyclization, involving the (intramolecular) reaction between the two ends of a linear molecule, is inherently a first-order process. Cyclization is a particular feature of the early stages of a step-growth polymerization (up to extents of reaction of 98–99%), where a proportion of the end groups that react are on the same molecule. Hence, cyclics form. Since the chances of meeting of the end groups decrease rapidly as the distance between them increases, the cyclics are of relatively low molecular weight, that is, they are oligomers. Further reaction leads mainly to linear molecules, although at extremely high conversions the number of end groups is quite small and intramolecular reactions essentially terminate the process, such that it might be expected that all chains ultimately cyclize. Practically though, the levels of conversion necessary to obtain these very large rings are extremely high and difficult to obtain (either by virtue of side reactions, monomer imperfections, or simply the level of viscosity of high molecular weight polymer solutions). What is usually obtained, therefore, is a mixture of cyclics and linear molecules. However, since cyclic oligomers often differ considerably in, for example, solubility compared to their high molar mass linear homologues, separation is often relatively straightforward. The commercial importance of polymers produced by step-growth polymerization gives a particular significance to understanding the nature of such materials. The presence of cyclic oligomers can be detrimental to the polymer properties since their presence could cause problems during processing. For instance, cyclic oligomers of polyethylene terephthalate (PET) tend to migrate to the surface of spun fibres and, under certain conditions, they crystallize to produce a surface ‘bloom’ which interferes with subsequent dyeing. More recently, it is the reverse of cyclization, namely ring-opening polymerization, which has been a particular focus of attention.

Some examples of dendrimer synthesis

Dendrimers are highly branched macromolecules with unique structural properties. They may be thought of as core–shell type macromolecules wherein they amplify their mass and terminal groups as a function of growth stages. These growth stages are referred to as generations (i.e. G= 0, 1, 2, . . .). They possess three key architectural features: (i) a core region; (ii) interior shell zones containing cascading tiers of branch cells (generations) with radial connectivity to the initiator core; and (iii) an exterior or surface region of terminal moieties attached to the outermost generation. With this architecture, a careful choice of building blocks and functional groups can provide control over shape, dimensions, density, polarity, reactivity, and solubility. One of the earlier dendrimers made, using a divergent strategy, is the Starburst® poly(amidoamine) (PAMAM) dendrimer family (Scheme 1). This method involved assembling repeat units to introduce branch cells around the initiator core through successive chemical reactions at the periphery of the growing macromolecule. The first step of PAMAM synthesis involves Michael addition of four moles of methyl acrylate to the nucleophilic ethylenediamine core. This leads to an electrophilic carbomethoxy surface, which is then allowed to react with an excess of 1,2-diaminoethane to give a nucleophilic surface at generation zero. Reiteration of these two steps now involves addition of 8 mol of methyl acrylate to give G = 0.5 (electrophilic, carbomethoxy surface). This is followed by amidation to return to a nucleophilic surface at G = 1.0. As a result of this reiterative branch cell assembly, it is apparent that these constructions follow systematic dendritic branching rules, with radial symmetry giving a well-defined three-dimensional geometry to the final dendritic product. In general, the placement of reactive functionalities on the exterior surface of the dendrimers allows introduction of a wide variety of terminal moieties. In alternate synthetic approaches, spacer groups have been deliberately introduced to relieve the steric hindrance in order to facilitate construction of the next generation. This may provide the possibility of enhancing interior cargo spaces for ‘guest–host’ type chemistry.