Emulsion polymerisation is one of the most common industrial manufacturing processes, with some 1.5 million tonnes of polymer produced per annum in Europe alone. If we assume, in the resulting polymer dispersions, a surfactant level of 0.5% on polymer weight, it can be seen that the polymerisation industry is also an appreciable consumer of surfactants. Generally, the polymer manufacturer would prefer to omit surfactants, which are usually more expensive than the monomers and can have deleterious effects on the properties of the final polymer. This chapter sets out to show how and why emulsion polymer manufacturers overcome their inhibitions over surfactant use and to discuss some of the factors relevant to the choice of surfactants and surfactant levels.
Emulsion polymerisation is reviewed extensively in Chapter 2 and in books by Gilbert [1] and Lovell and El-Aasser [2]. This chapter makes no contribution to the theory of emulsion polymerisation but concentrates on the practical aspects.
From an industrial viewpoint, the principal advantages of the emulsion polymerisation process are rapid polymerisation, low viscosities and the presence of a dispersing medium that permits better heat dissipation during manufacture. In addition, the final product is in a usable and relatively environmentally friendly form for direct supply to customers. For those end applications using solid polymer, the product may be coagulated or spray dried.
A large variety of monomers are polymerised by emulsion polymerisation. Most commercial products are copolymers and can have as many as six or more different monomers. The backbone will usually consist of 2–3 monomers and smaller amounts of other monomers will be included to impart functionality or stability to the polymer emulsion. The backbone monomer ratio will govern the stiffness of the polymer; the choice of monomers will be determined by the particular application and cost constraints.
A list of some commercially used monomers is given in Table 1.1, together with homopolymer, glass transition temperature (Tg) and water solubility data. From the diversity of water solubilities and reactivity ratios, it will be apparent that the application of simple, or even sophisticated, theory to an industrial mix would be very difficult.
Table 1.1 Typical monomers found in industrial copolymer emulsions Monomer | Function | Glass transition temperature, Tg (°C) | Solubility (%) |
Butadiene | Backbone | −90 | 0.08 |
Styrene | Backbone | 100 | 0.01 |
Methyl methacrylate | Backbone | 105 | 1.5 |
Butyl acrylate | Backbone | −56 | 0.13 |
Ethyl acrylate | Backbone | −22 | 1.5 |
Ethylhexyl acrylate | Backbone | −50 | 0.01 |
Vinyl acetate | Backbone | 29 | 2.8 |
Acrylonitrile | Backbone | 104 | 8.5 |
Vinyl neodecanoate (VeoVa) | Backbone | −3 | 0.01 |
Acrylic acid | Stabiliser | 106 | |
Methacrylic acid | Stabiliser | 200 | 100 |
Itaconic acid | Stabiliser | | 5 |
Acrylamide | Stabiliser | 165 | 144 |
N-Methylolacrylamide | Post-crosslinker | | >50 |
Divinylbenzene | Crosslinker | | <0.1 |
AMPS | Stabiliser | | >10 |
In addition to the monomers, a formulation will include a chain-transfer agent to control molecular weight, surfactants, an initiator and a chelating agent to mop up any metal ions arising from pipework, etc. Metal ions can affect the initiator decomposition, the stability of the mix and the ageing of the final product. The most common initiator is a persulfate salt as they readily decompose in the 60–95°C range within which the bulk of polymerisations are carried out. For lower temperatures, peroxides or hydroperoxides activated by reducing agents are used. Thiols are the most widespread chain-transfer agents used. Some earlier formulations included carbon tetrachloride for this purpose on account of lower product odour, but its ozone-depleting properties have led to its demise. Other minor ingredients are likely to be electrolytes, buffers and pH-adjusting agents.
The other major component is water, which may not be as simple an ingredient as may be expected, as the treatment of water to remove impurities/hardness varies between industrial sites: water may be used untreated, dealkylated, in which divalent ions are replaced by sodium ions, or fully deionised. Bearing in mind the differences in mains water purity between different areas, recipes are not necessarily directly transferable from one site of manufacture to another.
A typical industrial copolymer will have a solids content greater than 45%, particle size 100–500 nm and be reacted to full conversion. It should not contain significant levels of coagulum and the final product should contain low levels of volatiles. It is expected to have a shelf-life of at least 6 months and preferably 1–2 years.
1.3 The industrial process
Polymers are made by batch, semi-batch and continuous processes. The same reactors are often used to make a variety of products.
Water, monomers and surfactant are charged to a stirred reactor (capacity from 5 to 40 tonnes) and some form of purging carried out to remove oxygen, which inhibits the polymerisation. Styles of agitator vary from one plant to another and may be critical for certain recipes [3]. Turbine-type agitators will give the best mixing but a flat-blade anchor type will give better heat transfer at the walls of the reactor. The contents are heated to initiation temperature and the initiator is injected. For energysaving reasons, the initiation temperature is frequently below the polymerisation temperature, thus enabling the exotherm to take the temperature to the desired value. The reactor is then kept to the desired temperature profile, usually by adjustment of the jacket temperature or by the use of external heat exchangers. When the monomer has suitable volatility, reflux can also be employed. Reaction times will usually lie in a 1–24 h period. During the reaction it may be necessary to add further surfactant or initiator.
Reactions are usually run to conversions in excess of 99%, but in some cases this is not possible, because of rate problems and consequent excessive reaction times, or not desirable, on account of crosslink formation in the polymer caused by transfer reactions. In the latter case, final conversions may be as low as 60% where highperformance materials are required. Unconverted monomer and any other volatile organic matter must be removed, and recovered in the case of the low-conversion products, in a process called stripping. This normally takes place in a different vessel. Two methods are employed, often in combination: chemical treatment or vacuum distillation with high-temperature steam injection. Chemical treatment uses a high free radical flux produced by a redox pair to convert the residual monomer to low molecular weight polymer. Steam and high temperatures are more effective for harder polymers, where transport of the monomer is difficult, and for less polymerisable monomers. Chemical stripping can cause crosslinking with certain polymers. Foam control can be a problem in steam-stripped grades and it is normal to add a short-life antifoam. These antifoams are based on oils and gradually become less effective as they either emulsify or absorb into the polymer. In order to increase stability, the pH is normally raised prior to stripping and further surfactant may also be added. This will be discussed later. Finally, the product is concentrated to the desired solids content, cooled, sieved and compounded with additional ingredients such as surfactant, biocide or antioxidant.
Concentration should be avoided, if at all possible, as water evaporation is expensive in both energy and time. It is therefore desirable to run polymerisations at their maximum practicable solids content. Similarly, it is desirable to minimise reaction times, the limiting factor being the rate of heat removal from the reaction. One of the disadvantages of the batch technique is the potential for runaway reactions due to the high monomer levels in the reactor in the initial stages of reaction.
As bio plants now treat most industrial wastes, any surfactant used must be biodegradable.
The semi-batch process is similar to the batch process except that all or the bulk of the monomers is added over several hours. Often other materials, including surfactant, are fed concurrently. Usually a small portion of the reactants is polymerised at a very low monomer to water ratio and then the feeding commenced. This first stage is referred to as the seed stage.
The semi-batch process gives several differences in performance to a batch polymerisation. Typical conversion curves are shown in Figure 1.1 for batch and semi-batch reactions. Immediately apparent are the differences in monomer to polymer ratios as the reaction progresses for the two different methods. In the batch case, there is a relatively slow start during the nucleation period followed by a steady polymerisation period and finally a tailover period as the monomer becomes depleted. For the semi-batch case, the graph shows the conversion of the monomer in the reactor. The overall conversion is much lower than this and only coincides with the reactor conversion at the end of the addition period.
The seed stage takes the conversion in the reactor to 70% or more before the monomer, which can be fed as an emulsion in water, and initiator are fed in. The rate of polymerisation compared with the rate of feed will govern the in-reactor conversion and a range of conversion profiles are possible. Two possible profiles are shown on the graph. Reactions are frequently run under so-called monomerstarved conditions whereby the in-reactor conversion is kept high at 80–90%. This has the effect of reducing the polymerisation rate. To compensate for this, reactions are normally carried out at higher temperature and consequently the initiator depletes faster. Hence higher levels of initiator, which is usually fed during the reaction, are added in semi-batch reactions than in batch reactions. The net result is a polymer of shorter chain length but with greater branching due to transfer to polymer. The latter arises from the higher polymer to monomer ratio in the semi-batch process. A useful feature of the semi-batch process is the reduction in side reactions between the monomers and, in particular, reduced transfer to monomer and the formation of odorous by-products arising from the Diels-Alder reactions of diene monomers. For those copolymers that show a drift in composition during polymerisation in the batch process, the semi-batch route can lead to polymers with a much more uniform composition. This can show itself as sharper glass transition temperatures or differences in stiffness of the final products. For example, styrene–butadiene copolymers of t...