6. Common Chain-Growth Polymers

Up to the late 1960s, most low-density polyethylene was produced commercially by high-pressure free-radical polymerization. Much of this has now been replaced by preparation of copolymers of ethylene with α-olefins by coordination polymerization. These preparations are discussed further in this chapter. High-pressure polymerizations of ethylene, however, might still practiced in some places and it is, therefore, discussed here. The reaction requires a minimum pressure of 500 atm [1] to proceed. The branched products contain long and short branches as well as vinylidene groups. With an increase in pressure and temperature of polymerization, there is a decrease in the degree of branching and in the amount of vinylidene groups [2, 3].

Free-radical commercial polymerizations are conducted at 1,000–3,000 atm pressure and 80–300°C. The reaction has two peculiar characteristics: (1) a high exotherm and (2) a critical dependence on the monomer concentration. In addition, at these high pressures oxygen acts as an initiator. At 2,000 atm pressure and 165°C temperature, however, the maximum safe level of oxygen is 0.075% of ethylene gas in the reaction mixture. Any amount of oxygen beyond that level can cause explosive decompositions. History of polyethylene manufacture contains stories of workers being killed by explosions. Yet, the oxygen concentration in the monomer is directly proportional to the percent conversion of monomer to polymer, though inversely proportional to the polymer’s molecular weight. This limits many industrial practices to conducting the reactions below 2,000 atm and below 200°C. These reactions were done, therefore, between 1,000 and 2,000 atm pressures. Small quantities of oxygen, limited to 0.2% of ethylene, are accurately metered in [4, 5]. The conversion per each pass in continuous reactors is usually low, about 15–20%.

There is an induction period that varies inversely with the oxygen concentration to the power of 0.23. During this period oxygen is consumed autocatalytically. This is not accompanied by any significant decrease in pressure. A high concentration of ethylene is necessary for a fast rate of chain growth, relative to the rate of termination. Also, high temperatures are required for practical rates of initiation.

If oxygen is completely excluded and the pressure is raised to between 3,500 and 7,750 atm, while using relatively low temperatures of 50–80°C, linear polyethylene forms [6]. The reactions take about 20 h. Various solvents can be used, like benzene, isooctane, methyl, or ethyl alcohols. Higher ethyl alcohol concentrations and low concentrations of the initiator result in higher molecular weights. The products range from 2,000 for wax-like polymers to 4,000,000 for nearly intractable materials. Favorite free-radical initiators for this reaction are benzoyl peroxide, azobisisobutyronitrile, di-t-butylperoxydicarbonate, di-t-butyl peroxide, and dodecanoyl peroxide. Above conditions differ, however, from typical commercial ones, because such high pressures and long reaction times are not practical.

The actual commercial conditions vary, depending upon location and individual technology of each company. Often, tubular and multiple-tray autoclaves are used [7]. Good reactor design must permit dissipation of the heat of polymerization (800–1,000 cal/g), with good control over other parameters of the reaction. Tubular reactors are judged as having an advantage over stirred autoclaves in offering greater surface-to-volume ratios and better control over residence time [7]. On the other hand, the stirred autoclaves offer a more uniform temperature distribution throughout the reactor.

The tubular reactors have been described as consisting of stainless steel tubes between 0.5 and 1 in. in internal and about 2 in. in the external diameters. The residence time in these tubes is from 3 to 5 min, and they can be equipped with pistons for pressure regulation. Pressure might also be controlled by flow pulses to the reactor [8]. For the oxygen-initiated reactions, the optimum conditions are [7] 0.03–0.1% oxygen at 190–210°C and 1,500 atm pressure. At this pressure, the density of ethylene is 0.46 g/cm³. This compares favorably with the critical density of ethylene that is 0.22 g/cm³. Once the polymerization is initiated, the liquid monomer acts as a solvent for the polymer. Impurities, such as acetylene or hydrogen cause chain transferring and must be carefully removed. In some processes, hindered phenols are added in small quantities (between 10 and 1,000 ppm). This has the effect of reducing long-chain branching and yields film grade resins with better clarity, lower haze, and a reduced amount of microgels. Also, diluents are used in some practices. Their main purpose is to act as heat-exchanging mediums, but they can also help remove the polymer from the reactor. Such diluents are water, benzene, and ethyl or methyl alcohols. Sometimes, chain transferring agents like carbon tetrachloride, ketones, aldehydes, or cyclohexane might also be added to control molecular weight. The finished product (polymer–monomer mixture) is conveyed to a separator where almost all of the unreacted ethylene is removed under high pressure (3,500–5,000 psi) and recycled. The polymer is extruded and palletized. Ethylene conversion per pass is a limiting factor on the economics. A tubular reactor is illustrated in Fig. 6.1.

Polyethylene prepared in this way may have as many as 20–30 short branches per 10,000 carbon atoms in the chain [9] and one or two long-chain branches per molecule, due to “backbiting” [10] (explained in Chap. 3):

The reaction results in predominantly ethyl and butyl branches. The ratio of ethyl to butyl groups is roughly 2:1 [11, 12]. Chain transferring to the tertiary hydrogens at the location of the short branches causes elimination reactions and formation of vinylidene groups [13, 14]. This mechanism also accounts for formation of low molecular weight species.

Commercial grades of low-density polyethylene vary widely in the number of short and long branches, average molecular weights, and molecular weight distributions. M _w/M _n is between 20 and 50 for commercial low-density materials. The short branches control the degree of crystallinity, stiffness, and polymer density. They also influence the flow properties of the molten material.

Low-density polyethylene can be prepared by coordination polymerization through copolymerization of ethylene with α-olefins. This is discussed in the section on copolymers of ethylene. Finding catalytic systems that would allow formation of amorphous, low-density polyethylene from the monomer alone by low-pressure polymerization, however, is an economically worthwhile goal. To this end, considerable research is being carried out to develop such catalytic systems. Particular attention is given to metallocenes and other single-site catalysts for olefin polymerization. Originally, the metallocene catalysts were typical metal complexes with two cyclopentadienyl or substituted cyclopentadienyl groups. Many variations were developed since. These materials are used in combination with methyl aluminoxane and have the potential of forming the polymers with high precision. Nevertheless, at this time it is probably still safe to say that low-density polyethylene is prepared by many but perhaps not by all of the processes and catalytic systems mentioned in this book. This is because the material is manufactured all over the world and different considerations govern the decisions on the processes and catalytic systems. The same is probably true of high-density polyethylene.

New catalysts based on palladium and nickel complexes with bulky α-diimine ligands were developed [17–19]. They can yield highly branched or moderately branched polymers of ethylene, as well as propylene and 1-hexene. The polyolefins produced by such catalysts can contain a considerable amount of branches along the backbone that are randomly distributed throughout the molecules. In these catalysts, the molecular weight-limiting β-hydrogen elimination process that is common to palladium and nickel catalysts has been suppressed through use of bulky α-diamine ligands [19]. This allows formation of high molecular weight polymers from ethylene and α-olefins. A nickel-based catalyst can be illustrated as follows:

It is claimed that the branching of polyethylene can be controlled to the extent that the product can even be more branched than conventional low-density polyethylene (1,2—300 branches/1,000 atoms) [18, 19]. The cationic Ni-diimine catalyst shown above (R = H, CH₃), with the methylaluminoxane analog, has been found to polymerize ethylene in toluene at room temperature at the rate of 110,000 kg/Ni/h. This is comparable to the metallocene rates. The Pd-based catalysts are less active than their Ni analogs [19].

When nickel catalysts are used, the extent of branching is a function of the temperature, ethylene pressure, and catalyst structure. Branching increases as the temperature rises. At higher ethylene pressure less branching occurs. Brookhart et al. illustrate the mechanism of polymerization as follows [18]: where X = CH₃, Br; M = Pd, Ni; R = (CH₂CH₂) _n CH₃; Ar = 2,6-dialkylphenyl.

A patent for the polymerization process of olefins (especially ethylene, α-olefins, cyclopentene, and some fluorinated olefins) describes the above catalytic systems [20]. The hindered diimines stabilize alkyl Ni(II) or Pd(II) with cationic complexes. After preparation, the complexes are reduced with methylaluminoxane and then activated with Lewis acids capable of forming non-coordinating counterions [20].

In addition, preparation of catalysts based on iron and cobalt [21] was also reported. These are complexes of bulky pyridine bis-imine ligands with iron or cobalt that are also activated by methylaluminoxane:

The iron-based catalysts are reported to be considerably more active than the cobalt analogs [21]. The yield of linear, narrow molecular weight distribution polyethylene per gram is reported to be very high [21].

Baugh et al. [22] synthesized and characterized a series of nickel(II) and iron(II) complexes of the general formula [LMX₂] containing bidentate (for M = Ni) and tridentate (for M = Fe) heterocycle-imine ligands. Activation of these pre-catalysts with methyl aluminoxane yields active catalyst systems for the oligomerization/polymerization of ethylene. Compared to α-diimine nickel and bis(imino)pyridine iron catalysts, both metal systems provide only half of the steric protection and consequently the catalytic activities are significantly lower.

Lower activities were attributed to reduced stability of the active species under polymerization conditions. The lower molecular weights of their products were explained to be the result of increased hydrogen transfer rates. Variations within the heterocyclic components of the ligand showed that both steric and electronic factors influence polymerization behavior of such catalysts.

Hanaoka, Oda, and coworkers report [23] that single-site polymerization catalysts are of considerable interest industrially today, because they afford highly controllable polymerization performances based on precise design of catalyst architecture and their industrial applications. Among them, they point to constrained geometry catalyst and phenoxy-induced complex, they call phenics–Ti, that are used together with methyl aluminoxane

These are half-metallocene catalysts with an anionic armed-pendant that have now been well developed for industrial production of copolymers of ethylene with 1-olefins. Modification at the cyclopentadienyl ring system has been mainly tuned to finely control polymerization behaviors such as activity, molecular weight, and regiochemistry. In general, minimizing 2,1-insertion is essential to obtain high molecular weight polyolefins; otherwise, facile 6-elimination occurs, leading to termination of chain growth. Thus, the largely open coordination sites of half-metallocene catalyst systems possess an indispensable problem of irregularity in propagation. Through tuning bulkiness of substituents on the bridged-silicon unit of phenics–Ti, is claimed to have demonstrated that 2,1-insertion of propylene can also controlled by the bridging substituents to produce high molecular weight polypropylene [23].

Hong and coworkers [24] concluded that it is generally desirable to immobilize the single-site metallocene catalysts on a suitable carrier to obtain ideal product morphology. Ultrahigh molecular weight polyethylenes were successfully prepared by them through titanium complexes bearing phenoxy-imine chelate ligands

They demonstrated that the catalyst can be immobilized on silica. The product yields ultrahigh molecular weight polyethylene. Increased polymerization temperature resulted in higher activity, but lower molecular weight of polyethylene.

High-density polyethylene (0.94–0.97 g/cm³) is produced commercially with two types of catalysts:

The two catalytic systems are used at different conditions. Both types have undergone evolution from earlier development. The original practices are summarized in Table 6.1.

The Ziegler process yields polyethylene as low as 0.94/cm³ in density, but process modifications can result in products with a density of 0.965 g/cm³. The transition metal oxide catalysts on support, on the other hand, yield products in the density range of 0.960–0.970 g/cm³.

The original development by Ziegler led to what appears to be an almost endless number of patents for various coordination-type catalysts and processes. As described in Chap. 4, such catalysts have been vastly improved. Progress was made toward enhanced efficiency and selectivity. The amount of polymer produced per gram of the transition metal has been increased manyfold. In addition, new catalysts, based on zirconium compounds complexed with methyl aluminoxane oligomers (sometimes called Kominsky catalysts), were developed. They yield very high quantities of polyethylene per gram of the catalyst. For instance, a catalyst, bis(cyclopentadienyl)-zirconium dichloride combined with methylaluminoxane, is claimed to yield 5,000 kg of linear polyethylene per gram of zirconium per hour [14].

An important factor in the catalysts activity is the degree of oligomerization of the aluminoxane moiety. The catalytic effect is enhanced by increase in the number of alternating aluminum and oxygen atoms. These catalysts have long storage life and offer such high activity that they need not be removed from the product, because the amount present is negligible [14, 15]. This makes the work-up of the product simple.

The continuous solution processes are usually carried out between 120 and 160°C at 400–500 lb/in.² pressure. The diluents may be cyclohexane or isooctane. In one zone reactors, the solid catalyst is evenly dispersed throughout the reactor. In the two zone reactors (specially constructed), the polymerizations are conducted with stirring in the lower zone where the catalysts are present in concentrations of 0.2–0.6% of the diluent. Purified ethylene is fed into the bottom portions of the reactors. The polymers that form are carried with small portions of the catalyst to the top and removed. To compensate for the loss, additional catalysts are added intermittently to the upper “quiescent” zones.

In suspension or slurry polymerizations, various suspending agents, like diesel oil, lower petroleum fractions, heptane, toluene, mineral oil, chlorobenzene, or others, are used. The polymerization temperatures are kept between 50 and 75°C at only slightly elevated pressures, like 25 lb/in.². If these are batch reactions, they last between 1 and 4 h. The slurry reactor is illustrated in Fig. 6.2.

Polymerizations catalyzed by transition metal oxides on support were described variously as employing solid/liquid suspensions, fixed beds, and solid/gas-phase operations. It appears, however, that the industrial practices are mainly confined to use of solid/liquid suspension processes. The polymerization is carried out at the surface of the catalyst suspended in a hydrocarbon diluent.

In continuous slurry processes, the temperatures are kept between 90 and 100°C and pressures between 400 and 450 lb/in.². The catalyst concentrations range between 0.004 and 0.03% and typical diluents are n-pentane and n-hexane. Individual catalyst particles become imbedded in polymer granules as the reaction proceeds. The granules are removed as slurry containing 20–40% solids.

There are variations in the individual processes. In some procedures, the temperature is kept high enough to keep the polymer in solution. In others, it is kept deliberately low to maintain the polymer in slurry. The products are separated from the monomer that is recycled. They are cooled, precipitated (if in solution), and collected by filtration or centrifugation.

Various reactors were developed to handle different slurry polymerization processes. The slurry is maintained in suspension by ethylene gas. The gas rises to the top and maintains agitation while the polymer particles settle to the bottom where they are collected.

Several companies adopted loop reactors. These are arranged so that the flowing reactants and diluents continuously pass the entrance to a receiving zone. The heavier particles gravitate from the flowing into the receiving zone while the lighter diluents and reactants are recycled. To accommodate that, the settling area must be large enough for the heavy polymer particles to be collected and separated.

In addition to suspension, a gas-phase process was developed. No diluent is used in the polymerization step. Highly purified ethylene gas is combined continuously with a dry-powdery catalyst and then fed into a vertical fluidized bed reactor. The reaction is carried out at 270 psi and 85–100°C. The circulating ethylene gas fluidizes the bed of growing granular polymer and serves to remove the heat [15]. Formed polymer particles are removed intermittently from the lower sections of the vertical reactor. The product contains 5% monomer that is recovered and recycled. Control of polymer density is achieved by copolymerization with α-olefins. Molecular weights and molecular weight distributions are controlled by catalyst modifications, by varying operating conditions, and/or use of chain transferring agents [15], such as hydrogen [16]. This is illustrated in Fig. 6.3.

The reactors for the fluidized gas-phase process are simple in design. There are no mechanical agitators and they rely upon blowers to keep the bed fluidized and well mixed. Catalysts and cocatalysts are fed directly to the reactor [25].

Rieger and coworkers [26] investigated gas-phase polymerization of ethylene with supported α-diimine nickel catalysts. The reaction of 2,5 and 2,6 and 1,4 dithiane ligands with Ni(acac)₂ and trityl tetrakis(pentafluorophenyl)borate gave the corresponding Ni(II) complexes in high yields. These complexes were supported on silica without a chemical tether and were used as catalysts for ethylene polymerization reactions in the gas phase. Furthermore, ethylene was polymerized with the unsupported 2,5-complexes in homogeneous solution for comparison. The influence of the ligand structure, hydrogen, and temperature on the polymerization performance was investigated. The supported catalysts showed moderate to high activities and produced polyethylenes ranging from high-density polyethylene to linear low-density polyethylene, without further addition of a α-olefin comonomer.

The weight average molecular weights of most commercial low- and high-density polyethylenes range between 5,000 and 300,000. Very low molecular weight polyethylene waxes and very high molecular weight materials are also available. The molecular weight distributions for high-density polyethylene vary between 4 and 15. The product generally has fewer than three branches per thousand carbon atoms [9].

Table 6.2 summarizes the properties of various polyethylenes.

Materials that are quite similar to polyethylene can be obtained from other starting materials. The most prominent is formation of polymethylene and similar high molecular weight paraffin hydrocarbons from diazoalkanes. The reaction was originally carried out by Pechmann [27] when small quantities of a white flocculent powder formed in an ether solution of diazomethane. Bamberger and Tschirner [28] showed that this white powder is polymethylene –(–CH₂–) _n – that melts at 128°C. The synthesis was improved since by introduction of various catalysts. The reaction can yield highly crystalline polymers that melt at 136.5°C [29] with the molecular weight in millions [30]. Among the catalysts, boron compounds are very efficient [30]. Bawn et al. [31] postulated the mechanism of catalytic action. It consists of initial coordination of a monomer with the initiator, BF₃. This is followed by a loss of nitrogen and a shift of a fluorine atom from boron to carbon. The successive additions of molecules of diazoalkane follow a similar path with a shift of the chain fragment to the electron-deficient carbon:

The resulting macromolecules are still reactive toward additional diazoalkanes. The above step-growth polymerization reactions can also yield block copolymers:

Formation of polymethylene by this reaction is not practical for commercial utilization.

Colloidal gold and fine copper powder also catalyze diazoalkane polymerizations. The reaction appears to precede by formation of alkylidine or carbene species that are bound to the surfaces of metals [31–33]. The initiations are completed by additions of diazoalkanes to the bound carbenes followed by liberation of nitrogen. Termination may take place by chain transfer, perhaps to a monomer, or to the solvent [31–33].

Many different diazoalkanes lend themselves to these polymerization reactions. Polypentylidine, polyhexylidine, polyheptilidine, and polyoctylidine form with a gold complex catalyst, AuCl₃-pyridine [34].

An entirely different route to preparation of macroparaffins is through a high-pressure reaction between hydrogen and carbon monoxide. Transition metals, like finely divided ruthenium, catalyze this reaction. At pressures of about 200 atm and temperatures below 140°C, polymethylene of molecular weight as high as 100,000 forms [35]:

The earliest commercial methods used slurry polymerizations with liquid hydrocarbon diluents, like hexane or heptane. These diluents carried the propylene and the catalyst. Small amounts of hydrogen were fed into the reaction mixtures to control molecular weights. The catalyst system consisted of a deep purple or violet-colored TiCl₃ reacted with diethyl aluminum chloride. The TiCl₃ was often prepared by reduction of TiCl₄ with an aluminum powder. These reactions were carried out in stirred autoclaves at temperatures below 90°C and at pressures sufficient to maintain a liquid phase. The concentration of propylene in the reaction mixtures ranged between 10 and 20%. The products formed in discrete particles and were removed at 20–40% concentrations of solids. Unreacted monomer was withdrawn from the product mixtures and reused. The catalysts were deactivated and dissolved out of the products with alcohol containing some HCl, or removed by steam extraction. This was followed by extraction of the amorphous fractions with hot liquid hydrocarbons.

Later bulk polymerization processes were developed where liquid propylene was either used as the only diluent in a loop reactor or permitted to boil out to remove the heat of reaction. The second was done in stirred vessels with vapor space at the top. More recently, gas-phase polymerizations of propylene were introduced. The technology is similar to the gas-phase technology in ethylene polymerizations [15] described in Sect. 6.1.

Isotactic polypropylene received most attention because it is commercially more desirable. Nevertheless, syndiotactic polypropylene, though less crystalline, has greater clarity, elasticity, and impact resistance. It melts, however, at lower temperature. This isomer was originally prepared with both, heterogeneous, titanium-based catalysts and soluble, vanadium-based ones. The heterogeneous catalysts gave very low yields of the syndiotactic fractions. In fact, original samples contained only a few percent of the desired material, almost an impurity. The yield of syndiotactic polypropylene increased with a decrease in polymerization temperature, but still remained low [65].

Highly syndiotactic polypropylene was prepared by Natta et al. [38] with homogeneous catalysts formed from VCl₄ or from vanadium tri-acetylacetonate, aluminum dialkyl halide, and anisole at −48 to −78°C. No isotactic fractions formed. This led to development of many effective soluble catalysts. The catalyst components and the conditions for their preparation are quite important in maintaining control over syndiotactic placement. For the most effective soluble catalyst the ratio of AIR₂X to the vanadium compound has to be maintained between 3 and 10 [66]. The organic portion of the organoaluminum compound can be either methyl, ethyl, isobutyl, neopentyl, phenyl, or methylstyryl [9, 67]. In addition to VCl₄ and to vanadium tri-acetylacetonate [66], various other vanadates can be used, like [VO(OR) _x Cl_3−x ], where x = 1, 2, or 3 [65]. The exact nature of the vanadium compound, however, is very important to the resultant steric arrangement of the product. For instance, VCl₄ combined with Al(C₂H₅)₂F forms a heterogeneous catalyst that yields the isotactic isomer [65]. Vanadium tri-acetylacetonate, on the other hand, upon reacting with Al(C₂H₅)₂F forms a soluble catalyst that yields the syndiotactic isomer [66]. Addition of certain electron donors increases the amount of syndiotactic placement. These are anisole, furan, diethyl ether, cycloheptanone, ethyl acetate, and thiophene [67]. The optimum results are obtained when an anisole to vanadium ratio is 1:1. Also, the highest amount of syndiotactic polymer is obtained when the soluble catalysts are prepared and used at low temperatures. Even at low temperatures, however, like −78°C, the amount of syndiotacticity that can be obtained with a specific catalyst decreases with time [65, 66, 68]. This indicates a deterioration of the syndiotactic placing sites. On the other hand, polymerization of propylene with soluble vanadium tri-acetylacetonate–Al(C₂C₅)₂Cl system was reported to be a “living” type polymerization [69]. The product has a narrow molecular weight distribution (M _w/M _n = 1.05–1.20). A kinetic study indicates an absence of chain transferring and termination at temperatures below −65°C.

More recent catalysts for syndiotactic polypropylene are complexes, like i-propyl(cyclopentadienyl-1-fluorenyl)hafnium dichloride with methyl aluminoxane [70]. Another, similar catalyst is i-propyl(η⁵-cyclopentadienyl-η³-fluorenyl)zirconium dichloride with methyl aluminoxane. These catalysts yield polymers that are high in syndiotactic material (the zirconium-based compound yields 86% of racemic pentads) [70, 71]. Commercial production of syndiotactic polypropylene is in the early stages. What catalytic system is used, however, is not disclosed at this time. Some of the properties of the two isomers, isotactic and syndiotactic polypropylenes, are compared in Table 6.5.

The molecular weights of syndiotactic polypropylenes can vary from a number average molecular weight of 25,000–60,000, depending upon reaction conditions [70]. Also, in isotactic polypropylene there is less than one double bond per 1,000 carbon atoms [72]. A typical M _w/M _n = 5–12.

Many poly(α-olefin)s reported in the literature are not used commercially for various reasons. Table 6.6 lists some of the olefins polymerized by the Ziegler–Natta catalysts [72, 83].

Isotactic poly(butene-1) is produced commercially with three-component coordination-type catalysts. It is manufactured by a continuous process with simultaneous additions to the reaction vessel of the monomer solution, a suspension of TiCl₂–AlCl₃, and a solution of diethyl aluminum chloride [84]. The effluent containing the suspension of the product is continually removed from the reactor. Molecular weight control is achieved through regulating the reaction temperature. The effluent contains approximately 5–8% of atactic polybutene that is dissolved in the liquid carrier. The suspended isotactic fractions (92–98%) are isolated after catalyst decomposition and removal. The product has a density of 0.92 g/cm³ and melts at 124–130°C.

Isotactic polybutene crystallizes into three different forms. When it cools from the melt, it originally crystallizes into a metastable crystalline one. After several days, however, it transforms into a different form. Noticeable changes in melting point, density, flexural modulus, yield, and hardness accompany this transformation. The third crystalline form results from crystallization from solution. The polymer exhibits good impact and tear resistance. It is also resistant to environmental stress-cracking.

Another commercially produced polyolefin is isotactic poly(4-methyl pentene-1). The polymer carries a trade name of TPX. This material is known for high transparency, good electrical properties, and heat resistance. Poly(4-methyl pentene-1) has a density of 0.83 g/cm³. This polyolefin exhibits poor load-bearing properties and is susceptible to UV degradation. It is also a poor barrier to moisture and gases and scratches readily. This limits its use in many applications.

Poly(4-methyl pentene) is produced by the same process and equipment as polypropylene. A post finishing de-ashing step, however, is required. In addition, aseptic conditions are maintained during manufacture to prevent contamination that may affect clarity.

A number of similar polyolefins with pendant side groups are known. These include poly(3-methyl butene-1), poly(4,4-dimethyl pentene-1), and poly(vinyl cyclohexane). Due to their increased cohesive energy, ability to pack into tight structures, and the effect of increasing stiffness of the pendant groups, some of these polymers have a high melting point. This can be seen from Table 6.9. Many of these polymers, however, tend to undergo complex morphological changes on standing. This can result in fissures and planes of weakness in the structure.

The commercial ethylene-propylene rubbers typically range in propylene content from 30 to 60%, depending upon intended use. Such copolymers are prepared with Ziegler–Natta type catalysts. Soluble catalysts and true solution processes are preferred. The common catalyst systems are based on VCl₄, VOCl₃, V(Acac)₃, VO(OR)₃, VOCl(OR)₂, VOCl₂(OR), etc. with various organoaluminum derivatives. The products are predominantly amorphous. Polymerization reactions are usually carried out at 40°C in solvents like chlorobenzene or pentane. The resultant random copolymers are recovered by alcohol precipitation. Because these elastomers are almost completely saturated, cross-linking is difficult. A third monomer, a diene is, therefore, included in the preparation of these rubbers that carry the trade names EPTR or EPDM. Inclusion of third monomers presents some problems in copolymerization reactions. For instance, it is important to maintain constant feed mixtures of monomers to obtain constant compositions. Yet, two of the three monomers are gaseous and the third one is a liquid. Natta [86] developed a technique that depends upon maintaining violent agitation of the solvent while gaseous monomers were bubbled through the liquid phase. This was referred to as “semi-flow technique.” The process allows the compositions of gaseous and liquid phases to be in equilibrium with each other and to be more or less constant [87]. Other techniques evolved since. All are designed to maintain constant polymerization mixtures.

The vanadium-based catalyst systems deteriorate with time and decrease in the number of catalytic centers as the polymerizations progress. The rate of decay is affected by conditions used for catalyst preparation, compositions of the catalysts, temperature, solvents, and Lewis bases. It is also affected by the type and concentration of the third monomer [88–90]. Additions of chlorinated compounds to the deactivated catalysts, however, help restore activity [91, 92]. Catalyst decay can also be overcome by continually feeding catalyst components into the polymerization medium [93].

While third monomer can be a common diene, like isoprene, more often it is a bridged ring structure with at least one double bond in the ring. In typical terpolymer rubbers with 60–40 ratios of ethylene to propylene the diene components usually comprise about 3% of the total. Some specialty rubbers, however, may contain 10% of the diene or even more. Reaction conditions are always chosen to obtain 1,2 placement of the diene. Dienes in common use are ethylidine norbornene, methylene norbornene, 1,4-hexadiene, dicyclopentadiene, and cycloocatadiene:

In addition to the above, the patent literature describes many other dienes. An idealized picture of a segment of an uncross-linked gum stock might be shown as having the following structure:

Many copolymers of ethylene with α-olefins are prepared commercially. Thus ethylene is copolymerized with butene-1, where a comonomer is included to lower the regularity and the density of the polymer. Many copolymers are prepared with transition metal oxide catalysts on support. The comonomer is usually present in approximately 5% quantities. This is sufficient to lower the crystallinity and to markedly improve the impact strength and resistance to environmental stress-cracking. Copolymers of ethylene with hexene-1, where the hexene-1 content is less than 5%, are also produced for the same reason.

In most cases, the monomers that homopolymerize by Ziegler–Natta coordination catalysts also copolymerize by them [94]. In addition, some monomers that do not homopolymerize may still copolymerize to form alternating copolymers. Because the lifetime of a growing polymer molecule is relatively long (can be as long as several minutes), block copolymerization is possible through changes in the monomer feeds. Also, the nature of the transition metal compound influences the reactivity ratios of the monomers in copolymerizations. On the other hand, the nature of the organometallic compound has no such effect [95]. It also appears that changes in the reaction temperature between 0 and 75°C have no effect on the r values. Copolymers can be formed using either soluble or heterogeneous Ziegler–Natta. One problem encountered with the heterogeneous catalysts is the tendency by the formed polymers to coat the active sites. This forces the monomers to diffuse to the sites and may cause starvation of the more active monomer if both diffuse at equal rates.

Many different block copolymers of olefins, like ethylene with propylene and ethylene with butene-1, are manufactured. Use of the anionic coordination catalysts enables variations in the molecular structures of the products. It is possible to vary the length and stereoregularity of the blocks. This is accomplished by feeding alternately different monomers into the reactor. When it is necessary that the blocks consist of pure homopolymers, then after each addition the reaction is allowed to subside. If any residual monomer remains, it is removed [96]. This requires a long lifetime for the growing chains and an insignificant amount of termination. The stability of the anion depends upon the catalyst system. One technique for catalyst preparation is to form TiCl₃ by reducing TiCl₄ with diethyl aluminum chloride followed by careful washing of the product of reduction to remove the by-product Al(C₂H₅)Cl. Other reports describe using the α-form of TiCl₃ or heat treating it to form the β or γ-forms that yields more stereospecific products.

The transition metal oxide catalysts on support, such as the CrO₃/silica–alumina (Phillips) and MoO₃/Al₂O₃ (Standard Oil), are used to copolymerize minor quantities of α-olefins with ethylene. Such copolymerizations introduce short pendant groups into polymer backbones.

Ethylene and other olefins can also be copolymerized with carbon monoxide to form polymers of aliphatic ketones, using transition metal catalysts, like palladium(II) coupled with non-coordinating anions. There are numerous reports of such catalysts in the literature. One example is a compound composed of bidentate diarylphosphinopropane ligand and two acetonitrile molecules coordinating Pd²⁺ coupled with BF₃ counterions. This compound, bis(acetonitrile)palladium(II)-1,3-bis(diphenyl-phosphino)propane-(tetrafluoborate), can be illustrated as follows [97]:

When the tetrafluoroborate is replaced with a perchlorate, the compound is a very active catalyst [97].

One copolymer of ethylene and carbon monoxide are available commercially. The material offered under the trade name of Carilon is actually a terpolymer, because it contains a small quantity of propylene. It is reported [98] that use of a palladium catalyst permits formation of perfectly alternating interpolymer.⁰⁸ The product is reported to be a tough, chemical resistant material.

Hustad et al. [99] developed a technique to make polydisperse polyethylene diblock copolymers with 1-octene with a distribution of block lengths. When melted and compressed into films, the distinct polymeric segments self-assemble into layered patterns of semi-crystalline and hard and amorphous phases. Because each phase has a different refractive index, the block copolymer, shown below, can function as a photonic crystal and scatter visible light.

Nomura and coworkers [100] studied copolymerization of ethylene with various pentenes: where n = 8, 12. The polymerizations were carried out with titanium catalysts illustrated below. Titanium compound were combined with methyl aluminoxane:

Their results show that the monomer reactivities are influenced not only by substituents on the olefins but also by the nature of the catalytically active species.

Derlin and Kaminsky [101] reported copolymerizations of ethylene and propylene with a sterically hindered monomer, 3-methyl-1-butene, using titanium and zirconium metallocenes with methyl aluminoxane cocatalyst.

Tritto and coworkers [102] reported that the complex [Pd(k²-P,O-{2-(2-MeOC₆H₄)₂P}C₆H₄SO₃)Me(DMSO)] was investigated as a single-component catalyst for the copolymerization of ethylene with norbornene. The catalyst was illustrated as follows:

The copolymers were obtained in very good yields and molar masses were significantly higher than those of polyethylene. Three copolymers were formed:

Determination of microstructure and reactivity ratios revealed a strong inherent tendency to form alternating copolymers.

Jordan [103] described copolymerization of ethylene with vinyl ethers and with vinyl fluoride. The catalyst used was (ortho-phospheno-arenesulfonate)PdMe(pyridine). The reaction was illustrated as follows:

Although presently lacking industrial importance, alternating copolymers can be made from propylene and butadiene [104] and also from propylene and isoprene [105]. Copolymers of propylene and butadiene form with vanadium- or titanium-based catalysts combined with aluminum alkyls. The catalysts have to be prepared at very low temperature (−70°C). Also, it was found that a presence of halogen atoms in the catalyst is essential [75]. Carbonyl compounds, such as ketones, esters, and others, are very effective additives. A reaction mechanism based on alternating coordination of propylene and butadiene with the transition metal was proposed by Furukawa [104].

Various copolymers of ethylene with vinyl acetate are prepared by free-radical mechanism in emulsion polymerizations. Both reactivity ratios are close to 1.0 [106]. The degree of branching in these copolymers is strongly temperature-dependent [107]. These materials find wide use in such areas as paper coatings and adhesives. In addition, some are hydrolyzed to form copolymers of ethylene with vinyl alcohol. Such resins are available commercially in various ratios of polyethylene to poly(vinyl alcohol), can range from 30% poly(vinyl alcohol) to as high as 70%.

Vinyl acetate residues in ethylene–vinyl acetate copolymers reduce regularity of polyethylene. This reduces crystallinity in the polymer. Materials containing 45% vinyl acetate are elastomers and can be cross-linked with peroxides.

Another group of commercial copolymers of ethylene is those formed with acrylic and methacrylic acids, where ethylene is the major component. The copolymerizations are carried out under high pressures. These materials range in comonomer content from 3 to 20%. Typical values are 10%. A large proportion of the carboxylic acid groups (40–50%) are prereacted with metal ions like sodium or zinc. The copolymer salts are called ionomers with a trade name like Syrlin. The materials tend to behave similarly to cross-linked polymers at ambient temperature by being stiff and tough. Yet they can be processed at elevated temperatures, because aggregation of the ionic segments from different polymeric molecules is destroyed. The material becomes mobile but after cooling the aggregates reform. Ionomers exhibit good low temperature flexibility. They are tough, abrasion-resistant resins that adhere well to metal surfaces.

1,3-Butadiene, the simplest of the conjugated dienes, is produced commercially by thermal cracking of petroleum fractions and catalytic dehydrogenation of butane and butene. Polymerization of butadiene can potentially lead to three poly(1,2-butadiene)s, atactic, isotactic, and syndiotactic and two cis and trans forms of poly(1,4-butadiene). This is discussed in Chaps. 3 and 4.

Free-radical polymerizations of 1,3-butadiene usually result in polymers with 78–82% of 1,4-type placement and 18–22% of 1,2-adducts. The ratio of 1,4 to 1,2 adducts is independent of the temperature of polymerization. Moreover, this ratio is obtained in polymerizations that are carried out in bulk and in emulsion. The ratio of trans-1,4 to cis-1,4 tends to decrease, however, as the temperature of the reaction decreases. Polybutadiene polymers formed by free-radical mechanism are branched because the residual unsaturations in the polymeric chains are subjects to free-radical attacks:

Should branching become excessive, infinite networks can form. The products become cross-linked, insoluble, and infusible. Such materials are called popcorn polymers. This phenomenon is more common in bulk polymerizations. The cross-linked polymers form nodules that occupy much more volume than the monomers from which they formed and often clog up the polymerization equipment, sometimes even rupturing it.

High molecular weight homopolymers of 1,3-butadiene formed by free-radical mechanism lack the type of elastomeric properties that are needed from commercial rubbers. Copolymers of butadiene, however, with styrene or acrylonitrile are more useful and are prepared on a large scale. This is discussed in another section.

Polyisoprenes occur in nature. They are also prepared synthetically. Most commercial processes try to duplicate the naturally occurring material.

Copolymerization of butadiene with styrene by free-radical mechanism has been explored very thoroughly [146]. The original efforts started during World War I in Germany. Subsequent work during the 1930s was followed by a particularly strong impetus in the United States during World War II. This led to a development of GR-S rubber in the United States and Buna-S rubber in Germany. After World War II further refinements were introduced into the preparatory procedures and “cold” rubber was developed. Industrially, the copolymer is prepared by emulsion copolymerization of butadiene and styrene at low temperatures in a continuous process. A typical product is a random distribution copolymer, with the butadiene content ranging from 70 to 75%. The diene monomer placement is roughly 18% cis-1,4; 65% trans-1,4; and 17% 1,2. M _n of these copolymers is about 100,000.

A “redox” initiator is used in the cold process, but not in the “hot” one. Also, the “hot” process is carried out at about 50°C for 12 h to approximately 72% conversion. The “cold” process is also carried for 12 h, but at about 5°C to a 60% conversion. The two recipes for preparation of GR-S rubbers are shown in Table 6.10 for comparison of the “hot” and “cold” processes.

In both polymerizations, the unreacted monomer has to be removed. In the “hot” one the reaction is often quenched by addition of hydroquinone, and in the “cold” one by addition of N,N-diethyldithiocarbamate. After the monomers are steam stripped in both processes, an antioxidant like N-phenyl-2-naphthylamine is added. The latex is usually coagulated by addition of a sodium chloride–sulfuric acid solution. The “cold” process yields polymers with less branching than the “hot” one, slightly higher trans to cis ratios.

During the middle 1960s a series of butadiene–styrene and isoprene–styrene block-copolymer-elastomers were developed. These materials possess typical rubber-like properties at ambient temperatures, but act like thermoplastic resins at elevated ones. The copolymers vary from diblock structures of styrene and butadiene to triblock ones, like styrene–butadiene–styrene:

A typical triblock copolymer may consist of about 150 styrene units at each end of the macromolecule and some 1,000 butadiene units in the center. The special physical properties of these block copolymers are due to inherent incompatibility of polystyrene with polybutadiene or polyisoprene blocks. Within the bulk material, there are separations and aggregations of the domains. The polystyrene domains are dispersed in continuous matrixes of the polydienes that are the major components. At ambient temperature, below the T _g of the polystyrene, these domains are rigid and immobilize the ends of the polydiene segments. In effect they serve both as filler particles and as cross-links. Above T _g of polystyrene, however, the domains are easily disrupted and the material can be processed as a thermoplastic polymer. The separation into domains is illustrated in Fig. 6.4.

These thermoplastic elastomers are prepared by anionic solution polymerization with organometallic catalysts. A typical example of such preparation is polymerization of a 75/25 mixture of butadiene/styrene in the presence of sec-butyllithium in a hydrocarbon–ether solvent blend. At these reaction conditions butadiene blocks form first and when all the butadiene is consumed, styrene blocks form. In other preparations, monomers are added sequentially, taking advantage of the “living” nature of these anionic polymerizations.

These block copolymers have very narrow molecular weight distributions. Also, the sizes of the blocks are restricted to narrow ranges to maintain optimum elastomeric properties.

Butadiene–acrylonitrile rubbers are another group of useful synthetic elastomers. These copolymers were originally developed in Germany where they were found superior in oil resistance to the butadiene–styrene rubbers. Commercially, these materials are produced by free-radical emulsion polymerization very similarly to the butadiene–styrene copolymers. Similarly, “hot” and “cold” processes are employed. “Low,” “medium,” and “high” grades of solvent-resistant copolymers are formed, depending upon the amount of acrylonitrile in the copolymer that can range from 25 to 40%. The butadiene placement in these copolymers is approximately 77.5% trans-1,4, 12.5% cis-1,4, and 10% of 1,2-units. Also, the polymers formed by the “cold” process are less branched and have a narrower molecular weight distribution than those formed by the “hot” process.

An interesting alternating copolymer of butadiene and acrylonitrile was developed in Japan [147]. The copolymer is formed with coordination catalysts consisting of AlR₃, AlCl₃, and VOCl₃ in a suspension polymerization process. The product is more than 94% alternate and is reported to have very good mechanical properties and good oil resistance.

Styrene is one of those monomers that lends itself to polymerization by free-radical, cationic, anionic and coordination mechanisms. This is due to several reasons. One is resonance stabilization of the reactive polystyryl species in the transition state that lowers the activation energy of the propagation reaction. Another is the low polarity of the monomer. This facilitates attack by free-radicals, differently charged ions, and metal complexes. In addition, no side reactions that occur in ionic polymerizations of monomers with functional groups are possible. Styrene polymerizes in the dark by free-radical mechanism more slowly than it does in the presence of light [149]. Also, styrene formed in the dark is reported to have greater amount of syndiotactic placement [150]. The amount of branching in the polymer prepared by free-radical mechanism increases with temperature [136]. This also depends upon the initiator used [151].

The following information has evolved about the free-radical polymerization of styrene:

Polystyrene that is manufactured by free-radical polymerization is atactic. Isotactic polystyrene formed with Ziegler–Natta catalysts was introduced commercially in the 1960s, but failed to gain acceptance. Syndiotactic polystyrene is now being produced commercially.

Industrially, free-radical styrene polymerizations are carried out in bulk, in emulsion, in solution, and in suspension. The clear plastic is generally prepared by mass polymerization. Because polystyrene is soluble in the monomer, mass polymerization, when carried out to completion, results in a tremendous increase in melt viscosity. To avoid this, when styrene is polymerized in bulk in an agitated kettle, the reaction is only carried out to 30–40% conversion. After that, the viscous syrup is transferred to another type of reactor for the completion of the reaction. According to one early German patent, polymerization is completed in a plate and frame filter press [157]. Water circulating through the press removes the heat of the reaction, and the solid polymer is formed inside the frames. This process is still used in some places [158].

Another approach is to use adiabatic towers. Styrene is first partially polymerized in two agitated reaction kettles at 80–100°C. The syrup solution of the polymer in the monomer is then fed continually into the towers from the top. The temperatures in the towers are gradually increased from 100–110°C at the top to 180–200°C at the bottom. By the time the material reaches the bottom, in about 3 h, the polymerization is 92–98% complete [159]. The unreacted monomer is removed and recycled. A modification of the process is to remove the monomer vapor at the top of the tower for reuse.

An improvement in the above procedure is the use of agitated towers [160]. To avoid channeling inside the towers and for better heat transfer, three towers are arranged in series. They are equipped with slow agitators and with grids of pipes for cooling and heating [160]. Polymeric melt is heated from 95 to 225°C to reduce viscosity and help heat transfer. A solvent like ethyl benzene may be added. A vacuum devolatalizer removes both monomer and solvent from the product (Fig. 6.5).

Much research was devoted to both cationic and anionic polymerizations. An investigation of cationic polymerization of styrene with Al(C₂H₅)₂Cl/RCl (R = alkyl or aryl) catalyst/cocatalyst system was reported by Kennedy [161, 162]. The efficiency (polymerization initiation) is determined by the relative stability and/or concentration of the initiating carbocations that are provided by the cocatalyst RCl. N-butyl, isopropyl, and sec-butyl chlorides exhibit low cocatalytic efficiencies because of low tendency for ion formation. Triphenylmethyl chloride is also a poor cocatalyst because the triphenylmethyl ion that forms is more stable than the propagating styryl ion. Initiation of styrene polymerizations by carbocations is now well established [163].

Anionic polymerization of styrene with amyl sodium yields an isotactic polymer [164]. Polymerizations catalyzed by triphenylmethylpotassium also yield the same stereospecific polystyrene [165]. The same is true of organolithium compounds [166, 167].

In butyllithium initiated polymerizations of styrene in benzene termination were claimed to occur by association between the propagating anions and the lithium counterions with another butyllithium molecule [168]. This was contradicted by claims that terminations result from association of two propagating chains [169]. Alfin catalysts polymerize styrene to yield stereospecific products [170].

Coordination catalysts based on aluminum alkyls and titanium halides yield isotactic polystyrene [171–174]. The polymer matches isotactic polystyrene formed with amylsodium. It is composed of head-to-tail sequences with the main chain fold being helical. There are three monomer units per each helical fold [171, 172]. The catalyst composition, however, has a strong bearing on the microstructure of the resultant polymer [175, 176].

Ishihara et al. reported in 1986 that syndiotactic polystyrene can be prepared with the aid of organic or inorganic titanium compounds activated with methylaluminoxane [177]. There is much greater incentive to commercialize syndiotactic polystyrene than the isotactic one. This is because isotactic polystyrene crystallizes at a slow rate. That makes it impractical for many industrial applications. Syndiotactic polystyrene, on the other hand, crystallizes at a fast rate, has a melting point of 275°C, compared to 240°C for the isotactic one, and is suitable for use as a strong structural material.

Many catalysts were investigated for the preparation of syndiotactic polystyrene. High activity is claimed for half-sandwich titanocenes of the type CpTiCl₃, IndTiCl₃, and substituted IndTiCl₃ with methylaluminoxane as cocatalyst [178]. It was also reported that BMe(C₆F₅)₃ and other borates can be used as precursors instead of methyl-aluminoxane [178]. In addition, it was disclosed that fluorinated catalysts exhibit very high activities and produce polymers with higher molecular weight polymers [178].

The mechanism of polymerization was investigated by different groups. Grassi and Zambelli claim that the syndiotactic styrene polymerization proceeds through a secondary insertion of styrene into Ti–alkyl (or growing polymer chain) bond. In the half titanocene catalyst, the polymer chain appears to be η⁶-coordinated to the metal of the active species [179]:

Maron and coworkers [180] reported that theoretical methods were used to investigate the syndiospecificity of the styrene polymerization catalyzed by single-site, single-component allyl ansa-lanthanidocenes:

Two limiting chain end stereocontrol mechanisms were studied by them, namely, migratory insertion through a site epimerization and site stereoconfiguration independent of backside insertion on a “stationary” polymer chain. Four consecutive insertions of styrene were computed to reveal that (i) backside insertions are more favorable than, or at least as favorable as, frontside insertions. The formation of a syndiotactic polymer is controlled by the thermodynamics. Moreover, the odd (first and third) insertions are of 2,1-down-si-type and are kinetically favored over the 2,1-up-re-ones. This control is the conjunction of two effects: minimization of styrene–styrene and styrene (phenyl ring)–fluorenyl repulsions. The steric hindrance of the polymer chain induces a fourth insertion by an exocyclic coordination of the fluorenyl ligand that is compensated by the η⁶ coordination of one of the phenyl ring in the growing chain.

Syndiotactic polystyrene is available commercially under the trade name of Questra. This material is produced with the aid of a metallocene catalyst and is sold in several grades [181].

There is a small interest in forming isotactic polystyrenes with vary narrow molecular weight distributions, because of some very limited practical applications, and from purely academic interests. Several preparations of virtually monodisperse polystyrenes of M _w/M _n = 1.06 by anionic polymerizations were developed. The materials are available commercially [181–186], small quantities for use as standards for GPC.

Many derivatives of styrene can be readily synthesized. Some are commercially available. One of them is α-methyl styrene. It is formed from propylene and benzene by a process that is very similar to styrene preparation:

Due to the allylic nature of α-methyl styrene it cannot be polymerized by free-radical mechanism. It polymerizes readily, however, by an ionic one. Resins based on copolymers of α-methyl styrene are available commercially. Other styrene derivatives that can be obtained commercially are:

Vinyl toluene polymerizes readily by free-radical mechanism at 100°C. The absolute rate at that temperature is greater than for styrene. The activation energy for vinyl toluene polymerization is 17–19 kcal per mole while that for styrene is 21 kcal mole. This monomer can also be polymerized by ionic and coordination mechanisms. Earlier attempts at polymerization of α-methyl styrene with Ziegler–Natta catalysts were not successful [187, 188]. Later, however, it was shown that polymerization does take place with TiCl₄/Al(C₂H₅)₃ at −78°C. The activity of the catalyst and the DP depend on the ratio of aluminum to titanium, the nature of the solvent, and on the aging of the catalyst [189]. The optimum ratio of the aluminum alkyl to titanium chloride is 1.0–1.2. Mixing and aging of the catalyst must be done below room temperature, and the valence of titanium must be maintained between 3 and 4 [189]. This led Sakurada to suggest that the reaction actually proceeds via a cationic rather than a coordinated anionic mechanism [189].

Various reports in the literature describe cationic polymerizations of α-methyl styrene with Lewis acids [190–192]. The products are mostly low molecular weight polymers, some containing unsaturation with pendant phenylindane groups. A high molecular weight polymer can be prepared from α-methyl styrene by cationic polymerization at −90 to −130°C with AlCl₃ in ethyl chloride or in carbon disulfide [193]. The product has a narrow molecular weight distribution. Some T _g and T _m values of polystyrene-like materials, including isotactic polystyrene, are presented in Table 6.11.

For many applications, the homopolymer of styrene is too brittle. To overcome that, many different approaches were originally tried. These included use of high molecular weight polymers, use of plasticizers, fillers (glass fiber, wood flour, etc.), deliberate orientation of the polymeric chains, copolymerization and addition of rubbery substances. Effect of plasticizers is too severe for practical use, and use of high molecular weight polymers exhibits only marginal improvement. Use of fillers, though beneficial, is mostly confined to United States. Orientation is limited to sheets and filaments, and copolymerization usually lowers the softening point too much.

Addition of rubbery materials, however, does improve the impact resistance of polystyrene. This is done, therefore, extensively. The most common rubbers used for this purpose are butadiene–styrene copolymers. Some butadiene homopolymers are also used but to a lesser extent. The high-impact polystyrene is presently prepared by dissolving the rubber in a styrene monomer and then polymerizing the styrene. This polymerization is either done in bulk or in suspension. The product contains styrene–butadiene rubber, styrene homopolymer, and a considerable portion of styrene-graft copolymer that forms when polystyrene radicals attack the rubber molecules. The product has very enhanced impact resistance.

Past practices, however, consisted simply in blending a mixture of polystyrene and rubber on a two-roll mill, or in a high shear internal mixer, or passing through an extruder. The impact strength of the product was only moderately better than that of the unmodified polymer. Another procedure was to blend polystyrene emulsion latex with a styrene–butadiene rubber emulsion latex and then to coagulate the two together. The product is also only marginally better in impact strength than styrene homopolymer. This practice, however, may still be in existence in some places.

In high-impact polystyrene, the rubber exists in discrete droplets, less than 50 μm in diameter. In effect the polymerization serves to form an oil in oil emulsion [199] where the polystyrene is in the continuous phase and the rubber is in a dispersed phase. The graft copolymer that forms serves to “emulsify” this heterogeneous polymer solution [200].

Commercial high-impact polystyrene usually contains 5–20% styrene–butadiene rubber. The particle size ranges from 1 to 10 μm. High-impact polystyrene may have as much as seven times the impact strength of polystyrene, but it has only half its tensile strength, lower hardness, and lower softening point.

Styrene–acrylonitrile copolymers are produced commercially for use as structural plastics. The typical acrylonitrile content in such resins is between 20 and 30%. These materials have better solvent and oil resistance than polystyrene and a higher softening point. In addition, they exhibit better resistance to cracking and crazing and an enhanced impact strength. Although the acrylonitrile copolymers have enhanced properties over polystyrene, they are still inadequate for many applications. Acrylonitrile–butadiene–styrene polymers, known as ABS resins, were therefore developed.

Although ABS resins can potentially be produced in a variety of ways, there are only two main processes. In one of them acrylonitrile–styrene copolymer is blended with a butadiene–acrylonitrile rubber. In the other one, interpolymers are formed of polybutadiene with styrene and acrylonitrile.

In the first one, the two materials are blended on a rubber mill or in an internal mixer. Blending of the two materials can also be achieved by combining emulsion latexes of the two materials together and then coagulating the mixture. Peroxide must be added to the blends in order to achieve some cross-linking of the elastomer to attain optimum properties. A wide range of blends are made by this technique with various properties [201]. Most common commercial blends of ABS resins may contain 70 parts of styrene–acrylonitrile copolymer (70/30) and 40 parts of butadiene–nitrile rubber (65/35).

In the second process, styrene and acrylonitrile are copolymerized in the presence of polybutadiene latex. The product is a mixture of butadiene homopolymer and a graft copolymer.

Many styrene–maleic anhydride copolymers are produced commercially for special uses. These are formed by free-radical copolymerization and many commercial grades are partially esterified. Molecular weights of such polymers may range from 1,500 to 50,000, depending upon the source. The melting points of these copolymers can vary from 110 to 220°C, depending upon molecular weight, degree of hydrolysis and esterification, and also the ratio of styrene to maleic anhydride.

Only a small percent of styrene copolymers reported in the literature achieved industrial importance. Some of the interesting copolymers of styrene that were reported but not utilized commercially are copolymers with various unsaturated nitriles. This includes vinylidene cyanide, fumaronitrile, malononitrile, methacrylonitrile, acrylonitrile, and cinnamonitrile [292]. Often, copolymerization of styrene with nitriles yields copolymers with higher heat distortion temperature, higher tensiles, better craze resistance, and higher percent elongation.

Styrene was also copolymerized with many acrylic and methacrylic esters. Products with better weathering properties often form. Copolymerization with some acrylates lowers the value of T _g [203].

Free-radical bulk polymerizations of acrylate esters exhibit rapid rate accelerations at low conversions. This often results in formation of some very high molecular weight polymer and some cross-linked material. The cross-linking is a result of chain transferring by abstractions of labile tertiary hydrogens from already formed “dead” polymeric chains [204]. Eventually, termination by combination of the branched radicals leads to cross-linked structures. Addition of chain transferring agents, like mercaptans (that reduces the length of the primary chains), helps prevent gel formation. There are no labile tertiary hydrogens in methacrylic esters. The growing methacrylate radicals are still capable of abstracting hydrogens from the α-methyl groups. Such abstractions, however, require more energy and are not an important problem in polymerizations of methacrylic esters [205]. Nevertheless, occasional formation of cross-linked poly(alkyl methacrylate)s does occur. This is due to chain transferring to the alcohol moiety [206, 210].

The termination reaction in free-radical polymerizations of the esters of acrylic and methacrylic acids takes place by recombination and by disproportionation [206, 207]. Methyl methacrylate polymerizations, however, terminate at 25°C predominantly by disproportionation [205].

Oxygen inhibits free-radical polymerization of α-methyl methacrylate [208]. The reaction with oxygen results in formation of low molecular weight polymeric peroxides that subsequently decompose to formaldehyde and methyl pyruvate [210]:

Oxygen is less effective in inhibiting polymerizations of acrylic esters. It reacts 400 times faster with the methacrylic radicals than with the acrylic ones. Nevertheless, even small quantities of oxygen affect polymerization rates of acrylic esters [216]. This includes photopolymerizations of gaseous ethyl acrylate that are affected by oxygen and by moisture [217].

Acrylic and methacrylic esters polymerize by free-radical mechanism to atactic polymers. The sizes of the alcohol portions of the esters determine the T _g values of the resultant polymers. They also determine the solubility of the resultant polymers in hydrocarbon solvents and in oils.

Solvents influence the rate of free-radical homopolymerization acrylic acid and its copolymerization with other monomers. Hydrogen bonding solvents slow down the reaction rates [219]. Due to electron withdrawing nature of the ester groups, acrylic and methacrylic ester polymerize by anionic but not by cationic mechanisms. Lithium alkyls are very effective initiators of α-methyl methacrylate polymerization yielding stereospecific polymers [213]. Isotactic poly(methyl methacrylate) forms in hydrocarbon solvents [214]. Block copolymers of isotactic and syndiotactic poly(methyl methacrylate) form in solvents of medium polarity. Syndiotactic polymers form in polar solvents, like ethylene glycol dimethyl ether, or pyridine. This solvent influence is related to Lewis basicity [215] in the following order:

Furthermore, polymerizations in solvating media, like ethylene glycol dimethyl ether, tetrahydrofuran, or pyridine, using biphenylsodium or biphenyllithium yield virtually monodisperse syndiotactic poly(methyl methacrylate) [216].

The nature of the counterion in anionic polymerizations of methyl methacrylate in liquid ammonia with alkali metal amide or alkali earth metal amide catalysts is an important variable [217]. Lithium and calcium amides yield high molecular weight polymers, though the reactions tend to be slow. Sodium amide, on the other hand, yields rapid polymerizations but low molecular weight polymers. Polymers formed with sodium amide, however, have a narrower molecular weight distribution than those obtained with lithium and calcium amides. Calcium amide also yields high molecular weight polymers from ethyl acrylate and methyl methacrylate monomers in aromatic and aliphatic solvents at temperatures from −8 to 110°C. When, however, tetrahydrofuran or acetonitrile is used as solvents much lower molecular weight products form [218].

Products from anionic polymerizations of methyl methacrylate catalyzed by Grignard reagents (RMgX) vary with the nature of the R and X groups, the reaction temperature, and the nature of the solvent [219–221]. Secondary alkyl Grignard reagents give the highest yields and the fastest rates of the reactions. Isotacticity of the products increases with the temperature. When anion-radicals from alkali metal ketyls of benzophenone initiate polymerizations of methyl methacrylate, amorphous polymers form at temperatures from −78 to +65°C [222].

Sodium dispersions in hexane yield syndiotactic poly(methyl methacrylate) [223]. A 60–65% conversion is obtained over a 24-h period at a reaction temperature of 20–25°C. Lithium dispersions [224], butyllithium [203], and Grignard reagents [225, 226] yield crystalline isotactic poly(t-butyl acrylate). The reactions take place in bulk and in hydrocarbon solvents. Isotactic poly(isopropyl acrylate) forms with Grignard reagents [226, 227].

Coordination polymerizations of methyl methacrylate with diethyliron–bipyridyl complex in nonpolar solvents like benzene or toluene yield stereoblock polymers. In polar solvents, however, like dimethylformamide or acetonitrile, the products are rich in isotactic placement [229].

There are many reports in the literature on polymerizations of acrylic and methacrylic esters with Ziegler–Natta catalysts [230–233]. The molecular weights of the products, the microstructures, and rates of the polymerizations depend upon the metal alkyl and the transition metal salt used. The ratios of the catalyst components to each other are also important [234, 235].

In 1992 Yasuda et al. [236, 237] reported that organolanthanide complexes of the type Cp*₂Sm-R (where Cp* is pentamethyl cyclopentadienyl, and R is either an alkyl, alkylaluminum or a hydride) initiate highly syndiotactic, living polymerizations of methacrylates. It was also reported that lanthanide complexes such as Cp*₂Yb(THF)_1–3, Cp*₂Sm(THF)₂, and (indenyl)₂Yb(THF)₂ can also initiate polymerizations of methylmethacrylate [238]. Although very low initiator efficiencies were observed, these were living polymerizations. The polymers that formed had the dispersity of 1.1 and were high in syndiotactic sequences.

Novak and Boffa, in studying lanthanide complexes, observed an unusual facile organometallic electron transfer process takes place that generates in situ bimetallic lanthanide(III) initiators for polymerizations of methacrylates [239]. They concluded that methyl methacrylate polymerizations initiated by the Cp*₂Sm complexes occur through reductive dimerizations of methyl methacrylate molecules to form “bisinitiators” that consists of two samarium(III) enolates joined through their double bond terminally [239]. Their conclusion is based on the tendency of Cp*₂Sm complexes to reductively couple unsaturated molecules:

Montei and coworkers [240] reported that Nickel complexes [(X,O)NiR(PPh₃)] (X = N or P), designed for the polymerization of ethylene, are effective for home- and copolymerization of butyl acrylate, methyl methacrylate, and styrene. Their role as radical initiators was demonstrated from the calculation of the copolymerization reactivity ratios. It was shown that the efficiency of the radical initiation is improved by the addition of PPh₃ to the nickel complexes as well as by increasing the temperature. The dual role of nickel complex as radical initiators and catalysts was exploited to succeed in the copolymerization of ethylene with butyl acrylate and methyl methacrylate.

Multiblock copolymers containing sequences of both ethylene and polar monomers were thus prepared.

Polymers of lower n-alkyl acrylates are used commercially to only a limited extent. Ethyl and butyl acrylates are, however, major components of acrylic elastomers. The polymers are usually formed by free-radical emulsion polymerization. Because acrylate esters are sensitive to hydrolysis under basic conditions, the polymerizations are usually conducted at neutral or acidic pH. The acrylic rubbers, like other elastomers must be cross-linked or vulcanized to obtain optimum properties. Cross-linking can be accomplished by reactions with peroxides through abstractions of tertiary hydrogens with free radicals:

Another way to cross-link acrylic elastomers is through a Claisen condensation:

The above illustrated cross-linking reactions of homopolymers, however, form elastomers with poor aging properties. Commercial acrylic rubbers are, therefore, copolymers of ethyl or butyl acrylate with small quantities of comonomers that carry special functional groups for cross-linking. Such comonomers are 2-chloroethylvinyl ether or vinyl chloroacetate, used in small quantities (about 5%). These copolymers cross-link through reactions with polyamines.

Among methacrylic ester polymers, poly(methyl methacrylate) is the most important one industrially. Most of it is prepared by free-radical polymerizations of the monomer and a great deal of these polymerizations are carried out in bulk. Typical methods of preparation of clear sheets and rods consist of initial partial polymerizations in reaction kettles at about 90°C with peroxide initiators. This is done by heating and stirring for about 10 min to form syrups. The products are cooled to room temperature and various additives may be added. The syrups are solutions of about 20% polymer dissolved in the monomer. They are poured into casting cells where the polymerizations are completed. The final polymers are high in molecular weight, about 1,000,000.

Poly(methyl methacrylate) intended for surface coatings is prepared by solution polymerization. The molecular weights of the polymers are about 90,000 and the reaction products that are 40–60% solutions are often used directly in coatings.

A certain amount of poly(methyl methacrylate) is also prepared by suspension polymerization. The molecular weights of these polymers are about 60,000 and they are used in injection molding and extrusion.

Thermosetting acrylic resins are used widely in surface coatings. Both acrylic and methacrylic esters are utilized and the term is applied to both of them. Often such resins are terpolymers or even tetra polymers where each monomer is chosen for a special function [214]. One is selected for rigidity, surface hardness, and scratch resistance; another for the ability to flexibilize the film, and the third one for cross-linking it. In addition, not all comonomers are necessarily acrylic or methacrylic esters or acids. For instance, among the monomers that may be chosen for rigidity may be methyl methacrylate. On the other hand, it may be styrene instead, or vinyl toluene, etc. The same is true of the other components. Table 6.12 illustrates some common components that can be found in thermoset acrylic resins.

The choice of the cross-linking reaction may depend upon desired application. It may also simply depend upon price, or a particular company that manufactures the resin, or simply to overcome patent restrictions. Some common cross-linking reactions will be illustrated in the remaining portion of this section. If the functional groups are carboxylic acids in the copolymer or terpolymer, cross-linking can be accomplished by adding a diepoxide.

Other reactions can also cross-link resins with pendant carboxylic acid groups. For instance, one can add a melamine formaldehyde condensate:

A diisocyanate, a phenolic, or a melamine–formaldehyde resin can be used as well. Resins with pendant hydroxyl groups can also be cross-linked by these materials. A diisocyanate is effective in forming urethane linkages:

When the pendant groups are epoxides, like glycidyl esters, cross-linking can be carried out with dianhydrides or with compounds containing two or more carboxylic acid groups [241]. Aminoplast resins (urea–formaldehyde or melamine–formaldehyde and similar ones) are also very effective [242].

Pendant amide groups from terpolymers containing acrylamide can be reacted with formaldehyde to form methylol groups for cross-linking [243]:

The product of the above reaction can be thermoset like any urea–formaldehyde resin (see Chaps. 7 and 9). Many cross-linking routes are described in the patent literature, because there are many different functional groups available.

This monomer can be prepared from chloroform [275]:

Tetrafluoroethylene boils at −76.3°C. It is not the only product from the above pyrolytic reaction of difluorochloromethane. Other fluorine by-products form as well and the monomer must be isolated. The monomer polymerizes in water at moderate pressures by free-radical mechanism. Various initiators appear effective [276]. Redox initiation is preferred. The polymerization reaction is strongly exothermic, and water helps dissipate the high heat of the reaction. A runaway, uncontrolled polymerization can lead to explosive decomposition of the monomer to carbon and carbon tetrafluoride [277]:

Polytetrafluoroethylene is linear and highly crystalline [278]. Absence of terminal groups shows that few, if any, polymerization terminations occur by disproportionation but probably all take place by combination [279]. The molecular weights of commercially available polymers range from 39,000 to 9,000,000. Polytetrafluoroethylene is inert to many chemical attacks and is only swollen by fluorocarbon oils at temperatures above 300°C. The T _m of this polymer is 327°C and the T _g is below −100°C.

The physical properties of polytetrafluoroethylene depend upon crystallinity and on the molecular weight of the polymer. Two crystalline forms are known. In both cases the chains assume helical arrangements to fit into the crystalloids. One such arrangement has 15 CF₂ groups per turn and the other has 13.

Polytetrafluoroethylene does not flow even above its melting point. This is attributed to restricted rotation around the C–C bonds and to high molecular weights. The stiffness of the solid polymer is also attributed to restricted rotation. The polymer exhibits high thermal stability and retains its physical properties over a wide range of temperatures. The loss of strength occurs at about the crystalline melting point. It is possible to use the material for long periods at 300°C without any significant loss of its strength.

The monomer can be prepared by dechlorination of trichlorotrifluoroethane with zinc dust and ethanol.

It is a toxic gas that boils at −26.8°C. Polymerization of chlorotrifluoroethylene is usually carried out commercially by free-radical suspension polymerization. Reaction temperatures are kept between 0 and 40°C to obtain a high molecular weight product. A redox initiation based on reactions of persulfate, bisulfite, and ferrous ions is often used. Commercial polymers range in molecular weights from 50,000 to 500,000.

Polychlorotrifluoroethylene exhibits greater strength, hardness, and creep resistance than does polytetrafluoroethylene. Due to the presence of chlorine atoms in the chains, however, packing cannot be as tight as in polytetrafluoroethylene, and it melts at a lower temperature. The melting point is 214°C. The degree of crystallinity varies from 30 to 85%, depending upon the thermal history of the polymer. Polytrifluorochloroethylene is soluble in certain chloro fluoro compounds above 100°C. It flows above its melting point. The chemical resistance of this material is good, but inferior to polytetrafluoroethylene.

The monomer can be prepared by dehydrochlorination of 1,1,1-chlorodifluoroethane: or by dechlorination of 1,2-dichloro-1,1-difluoroethane [280]:

Vinylidine fluoride boils at −84°C. The monomer is polymerized in aqueous systems under pressure. Details of the process, however, are kept as trade secrets. Two different molecular weight materials are available commercially, 300,000 and 6,000,000. Poly(vinylidine fluoride) is crystalline and melts at 171°C. The material exhibits fair resistance to solvents and chemicals, but is inferior to polytetrafluoroethylene and to polytrifluorochloroethylene.

Vinyl fluoride monomer can be prepared by addition of HF to acetylene. The monomer is a gas at room temperature and boils at −72.2°C. Commercially, vinyl fluoride is polymerized in aqueous medium using either redox initiation or one from thermal decomposition of peroxides. Pressures of up to 1,000 atm may be used. Radicals generated at temperatures between 50 and 100°C yield very high molecular weight polymers.

Poly(vinyl fluoride) is moderately crystalline. The crystal melting point, T _m, is approximately 200°C. The high molecular weight polymers dissolve in dimethylformamide and in tetramethyl urea at temperatures above 100°C. The polymer is very resistant to hydrolytic attack. It does, however, loose HF at elevated temperatures.

Mary different copolymers of fluoroolefins are possible and were reported in the literature. Commercial use of fluoroolefin copolymers, however, is restricted mainly to elastomers. Such materials offer superior solvent resistance and good thermal stability.

The elastomers that are most important industrially are vinylidine fluoride–chlorotrifluoroethylene [260] and vinylidine fluoride–hexafluoropropylene copolymers [282]. These copolymers are amorphous due to irregularities in their structures and can range in properties from resinous to elastomeric, depending upon composition [283]. Those that contain 50–70 mole percent of vinylidine fluoride are elastomers. The T _g ranges from 0 to −15°C, also depending upon vinylidine fluoride content [284]. They may be cross-linked with various peroxides, polyamines [284], or ionizing radiation. The cross-linking reactions by peroxides take place through hydrogen abstraction by primary radicals:

Copolymers of vinylidine fluoride with hexafluoropropylene are prepared in aqueous dispersions using persulfate initiators. Hexafluoropropylene does not homopolymerize but it does copolymerize. This means that its content in the copolymer cannot exceed 50%. Preferred compositions appear to contain about 80% of vinylidine fluoride. The cross-linking reactions with diamines are not completely understood. It is believed that the reaction takes place in two steps [285, 286]. In the first one, a dehydrofluorination occurs:

The above elimination is catalyzed by basic materials. These may be in the form of MgO, which is often included in the reaction medium. In the second step the amine groups add across the double bonds:

Free diamines, used for cross-linking, are too reactive and can cause premature gelation. It is common practice, therefore, to add these diamine compounds in the form of carbamates, like ethylenediamine carbamate or hexamethylene diamine carbamate. The above fluoro elastomers exhibit good resistance to chemicals and maintain useful properties from −50 to +300°C.

Copolymers of tetrafluoroethylene with hexafluoropropylene are truly thermoplastic polyperfluoroolefins that can be fabricated by common techniques. Such copolymers soften at about 285°C and have a continuous use temperature of −260 to +205°C. Their properties are similar to, though somewhat inferior to, polytetrafluoroethylene.

One of the miscellaneous fluoroolefin polymers is a copolymer of trifluoronitrosomethane and tetrafluoroethylene [287], an elastomer:

It can be formed by suspension polymerization. One procedure is to carry out the reaction in an aqueous solution of lithium bromide at −25°C with magnesium carbonate as the suspending agent. No initiator is added and the reaction takes about 20 h. Because the reaction in inhibited by hydroquinone and accelerated by ultra-violet light, it is believed to take place by a free-radical mechanism. Whether it is chain-growth polymerization, however, is not certain. A 1:1 copolymer is always formed regardless of the composition of the monomer feed, and the copolymerization takes place only at low temperatures. At elevated temperatures, however, cyclic oxazetidines form instead:

Two polyfluoroacrylates are manufactured on a small commercial scale for some special uses in jet engines. These are poly(1,1-dihyroperfluorobutyl acrylate): and poly(3-perfluoromethoxy-1,1-dihydroperfluoropropyl acrylate):

The polymers are prepared by emulsion polymerization with persulfate initiators.

Although many other fluorine containing polymers were described in the literature, it is not possible to describe all of them here. They are not utilized commercially on a large scale. A few, however, will be mentioned as examples. One of them is polyfluoroprene [288]:

The polymer is formed by free-radical mechanism, in an emulsion polymerization using redox initiation. All three possible placements of the monomer occur [267].

Polyfluorostyrenes are described in many publications. A β-fluorostyrene can be formed by cationic mechanism [289]. The material softens at 240–260°C. An α,β,β-trifluorostyrene can be polymerized by free-radical mechanism to yield an amorphous polymer that softens at 240°C [290]. Ring-substituted styrenes apparently polymerize similarly to styrene. Isotactic poly(o-fluorostyrene) melts at 265°C. It forms by polymerization with Ziegler–Natta catalysts [291]. The meta analog, however, polymerized under the same conditions yields an amorphous material [291].

Poly(vinyl chloride) is used in industry on a very large scale in many applications, such as rigid plastics, plastisols, and surface coatings. The monomer, vinyl chloride, can be prepared from acetylene:

The reaction is exothermic and requires cooling to maintain the temperature between 100 and 108°C.

The monomer can also be prepared from ethylene:

The reaction of dehydrochlorination is carried out at elevated pressure of about 3 atm.

Free-radical polymerization of vinyl chloride was studied extensively. For reactions that are carried out in bulk the following observations were made [292]:

There is autoacceleration in bulk polymerization rate of vinyl chloride [293]. It was suggested by Schindler and Breitenbach [294] that the acceleration is due to trapped radicals that are present in the precipitated polymer swollen by monomer molecules. This influences the rate of the termination that decreases progressively with the extent of the reaction, while the propagation rate remains constant. The autocatalytic effect in vinyl chloride bulk polymerizations, however, depends on the type of initiator used [295]. Thus, when 2, 2′-azobisisobutyronitrile initiates the polymerization, the autocatalytic effect can be observed up to 80% of conversion. Yet, when benzoyl peroxide initiates the reaction, it only occurs up to 20–30% of conversion.

When vinyl chloride is polymerized in solution, there is no autoacceleration. Also, a major feature of vinyl chloride free-radical polymerization is chain transferring to monomer [296]. This is supported by experimental evidence [297, 298]. In addition, the growing radical chains can terminate by chain transferring to “dead” polymer molecules. The propagations then proceed from the polymer backbone [297]. Such new growth radicals, however, are probably short lived as they are destroyed by transfer to monomer [299].

The ¹³C NMR spectroscopy of poly(vinyl chloride), which was reduced with tributyltin hydride, showed that the original polymer contained a number of short four-carbon branches [300]. This, however, may not be typical of all poly(vinyl chloride) polymers formed by free-radical polymerization. It conflicts with other evidence from ¹³C NMR spectroscopy that chloromethyl groups are the principal short chain branches in poly(vinyl chloride) [301, 302]. The pendant chloromethyl groups were found to occur with a frequency of 2–3/1,000 carbons. The formation of these branches, as seen by Bovey and coworkers, depends upon head to head additions of monomers during the polymer formation. Such additions are followed by 1,2 chlorine shifts with subsequent propagations [301, 302]. Evidence from still other studies also shows that some head to head placement occurs in the growth reaction [303]. It was suggested that this may be not only an essential step in formation of branches but also one leading to formation of unsaturation at the chain ends [303, 304]:

Poly(vinyl chloride) prepared with boron alkyl catalysts at low temperatures possesses higher amounts of syndiotactic placement and is essentially free from branches [305–307].

Many attempts were made to polymerize vinyl chloride by ionic mechanisms using different organometallic compounds, some in combinations with metal salts [308–312]. Attempts were also made to polymerize vinyl chloride with Ziegler–Natta catalysts complexed with Lewis bases. To date, however, it has not been established unequivocally that vinyl chloride does polymerize by ionic mechanism. Use of the above catalysts did yield polymers with higher crystallinity. These reactions, however, were carried out at low temperatures where greater amount of syndiotactic placement occurs by the free-radical mechanism [313]. Vinyl chloride was also polymerized by AlCl(C₂H₅)O₂H₅ + VO(C₃H₇O₂) without Lewis bases [312]. Here too, however, the evidence indicates a free-radical mechanism.

On the other hand, butyllithium–aluminum alkyl-initiated polymerizations of vinyl chloride are unaffected by free-radical inhibitors [313]. Also the molecular weights of the resultant polymers are unaffected by additions of CCl₄ that acts as a chain transferring agent in free-radical polymerizations. This suggests an ionic mechanism of chain growth. Furthermore, the reactivity ratios in copolymerization reactions by this catalytic system differ from those in typical free-radical polymerizations [313]. An anionic mechanism was also postulated for polymerization of vinyl chloride with t-butylmagnesium in tetrahydrofuran [314].

Commercially, by far the biggest amount of poly(vinyl chloride) homopolymer is produced by suspension polymerization and to a lesser extent by emulsion and bulk polymerization. Very little polymer is formed by solution polymerization.

One process for bulk polymerization of vinyl chloride was developed in France where the initiator and monomer are heated at 60°C for approximately 12 h inside a rotating drum containing stainless steel balls. Typical initiators for this reaction are benzoyl peroxide or azobisisobutyronitrile. The speed of rotation of the drum controls the particle size of the final product. The process is also carried out in a two-reactor arrangement. In the first one approximately 10% of the monomer is converted. The material is then transferred to the second reactor where the polymerization is continued until it reaches 75–80% conversion. Special ribbon blenders are present in the second reactor. Control of the operation in the second reactor is quite critical [315].

Industrial suspension polymerizations of vinyl chloride are often carried out in large batch reactors or stirred jacketed autoclaves. Continuous reactors, however, have been introduced in several manufacturing facilities [315]. Typical recipes call for 100 parts of vinyl chloride for 180 parts of water, a suspending agent, like maleic acid–vinyl acetate copolymer, a chain transferring agent, and a monomer soluble initiator. The reaction may be carried out at 100 lb/in.² pressure and 50°C for approximately 15 h. As the monomer is consumed the pressure drops. The reaction is stopped at an internal pressure of about 10 lb/in.² and remaining monomer (about 10%) is drawn off and recycled. The product is discharged.

Emulsion polymerizations of vinyl chloride are usually conducted with redox initiation. Such reactions are rapid and can be carried out at 20°C in 1–2 h with a high degree of conversion. Commercial poly(vinyl chloride)s range in molecular weights from 40,000 to 80,000. The polymers are mostly amorphous with small amounts of crystallinity, about 5%. The crystalline areas are syndiotactic [317, 318].

Poly(vinyl chloride) is soluble at room temperature in oxygen-containing solvents, such as ketones, esters, ethers, and others. It is also soluble in chlorinated solvents. The polymer, however, is not soluble in aliphatic and aromatic hydrocarbons. It is unaffected by acid and alkali solutions but has poor heat and light stability. Poly(vinyl chloride) degrades at temperatures of 70°C or higher or when exposed to sun light, unless it is stabilized. Heating changes the material from colorless to yellow, orange, brown, and finally black. Many compounds tend to stabilize poly(vinyl chloride). The more important ones include lead compounds, like dibasic lead phthalate and lead carbonate. Also effective are metal salts, like barium, calcium, and zinc octoates, stearates, and laurates. Organotin compounds, like dibutyl tin maleate or laurate, also belong to that list. Epoxidized drying oils are effective heat stabilizers, particularly in coatings based on poly(vinyl chloride). Some coating materials may also include aminoplast resins, like benzoguanamine–formaldehyde condensate.

The process of degradation is complex. It involves loss of hydrochloric acid. The reactions are free radical in nature, though some ionic reactions appear to take place as well. The process of dehydrochlorination results in formations of long sequences of conjugated double bonds. It is commonly believed that formation of conjugated polyenes, which are chromophores, is responsible for the darkening of poly(vinyl chloride). In addition, the polymer degrades faster in open air than it does in an inert atmosphere. This shows that oxidation contributes to the degradation process. All effective stabilizers are hydrochloric acid scavengers. This feature alone, however, can probably not account for the stabilization process. There must be some interaction between the stabilizers and the polymers. Such interaction might vary, depending upon a particular stabilizer.

Vinylidine chloride homopolymers form readily by free-radical polymerization, but lack sufficient thermal stability for commercial use. Copolymers, however, with small amounts of comonomers find many applications.

The monomer, vinylidine chloride, can be prepared by dehydrochlorination of 1,1,2-trichloroethylene:

It is a colorless liquid that boils at 32°C. Also, it is rather hard to handle as it polymerizes on standing. This takes place upon exposure to air, water, or light. Storage under an inert atmosphere does not completely prevent polymer formation.

Poly(vinylidine chloride) can be formed in bulk, solution, suspension, and emulsion polymerization processes. The products are highly crystalline with regular structures and a melting point of 220°C. The structure can be illustrated as follows:

This regularity in structure is probably due to little chain transferring to the polymer backbone during polymerization. Such regularity of structure allows close packing of the chains and, as a result, there are no effective solvents for the polymer at room temperature.

Copolymerization of vinylidine chloride with vinyl chloride reduces the regularity of the structure. It increases flexibility and allows processing the polymer at reasonable temperatures. Due to extensive crystallization, however, that is still present in 85:15 copolymers of vinylidine chloride with vinyl chloride, they melt at 170°C. The copolymerization reactions proceed at slower rates than do homopolymerizations of either one of the monomers alone. Higher initiator levels and temperatures are, therefore, used. The molecular weights of the products range from 20,000 to 50,000. These materials are good barriers for gases and moisture. This makes them very useful in films for food packaging. Such films are formed by extrusion and biaxial orientation. The main application, however, is in filaments. These are prepared by extrusion and drawing. The tensile strength of the unoriented material is 10,000 lb/in.² and the oriented one 30,000 lb/in.².

Vinylidine chloride is also copolymerized with acrylonitrile. This copolymer is used mainly as a barrier coating for paper, polyethylene, and cellophane. It has the advantage of being heat sealable.