Basic reactions in the synthesis of PET

Poly(ethylene terephthalate) (PET) is formed by step growth reaction of ethylene glycol with either dimethyl terephthalate (DMT) or purified terephthalic acid (PTA). The chemical structure of monomers and compounds used in PET synthesis is shown below in table 1.3:

When DMT is used as the monomer, it is known as the DMT process. This was the first to be commercialized because DMT was available in required purity, but terephthalic acid was not. Later the purified terephthalic acid became available and PTA process was introduced. Now both processes are used.

The strategy used to synthesize polymeric chains of PET is very different to that used for nylon 6,6. The ethylene glycol monomer is not a strong enough base to give rise to a salt with the other monomer, terephthalic acid, therefore the route to exact stoichiometry is not available for PET synthesis. Therefore, the synthesis is started with an excess of ethylene glycol and later on removed by evaporation as polymerization proceeds so that ultimately the monomer stoichiometry of 1:1 is achieved.

First Step of polymerization

(a) DMT route: Ester Interchange (or transesterification) Reaction

In the ester interchange (EI) process, excess of ethylene glycol is reacted with dimethyl terephthalate in the presence of catalysts at temperatures starting from 150°C and slowly increasing to 210°C as shown in reaction [36]. The heat of reaction is ~14 kcal/mol. The reaction is reversible and is driven to completion by distilling off the byproduct methanol. This distillation is carried out using fractionating column so that a minimum of ethylene glycol is removed. In practice, the ethylene glycol to DMT ratio of 1.7 - 2 is used so that the temperature at completion does not rise too high.

This pre-polycondensation reaction leads to products which mainly contain bis(hydroxyethyl) terephthalate (BHET). The product also contains some unreacted ethylene glycol together with linear oligoesters of the general formula as shown in structure [37]:

The average degree of polymerization (n) of oligomers in the reaction mixture depends on the ratio of ethylene glycol to DMT. If number of moles of EG is twice that of DMT, only BHET is obtained. If it is lower than 2 times the DMT, an higher value of n is obtained. This mixture is called prepolymer and is used in second step known as polycondensation (discussed later).

The addition of catalyst, most often metallic salts of a weak or volatile acid, such as Pb, Zn, Mn, Ca or Cd acetate, is preferred.

b) TPA Route: Direct Esterification Reaction

In the direct esterification process (DE) as shown in reaction [38], due to the low solubility of terephthalic acid in boiling glycol, it is necessary to raise the reaction temperature and pressure to 240-260 °C and 4 x 10⁵ Pa, respectively, to obtain an adequate reaction rate. The reaction is self-catalyzed by the carboxylic acid groups. However, additional catalysts such as stronger acids and esters of titanic acid are also used. The volatile byproduct of direct esterification is water, which is distilled out of the reactor to force the reversible reaction towards completion.

Esterification reaction

One of the side products of this reaction is the formation of diethyl glycol (DEG) as shown in reaction [39].

The catalytic action of acid groups and the high temperature employed during the esterification increases the formation of DEG. DEG formation can be reduced by reducing the EG/TPA ratio during the start of the reaction and by addition of chemicals which reduce or prevent DEG formation. Therefore an EG/TPA ratio of 1.1 to 1.3 is used and consequently, the initial reaction product is a mixture of linear oligomers of a somewhat higher degree of polymeriza­tion (n) than that from ester interchange reaction discussed above.

A very small amount of a strong base, such as sodium hydroxide or an organic quaternary hydroxide is added as effective catalysts to reduce DEG formation. The formation of DEG during pre-polycondensation step results in decrease of the melting point of the polymer obtained from TPA route, and therefore, its melting point is generally lower than that obtained by DMT route. After esterification is complete, a phosphite or phosphate ester may be added to inactivate the basic additive and stabilize the polymer.

TPA route has several economical advantages over DMT route, which are given below:

Therefore, the PTA- based production saves 8% of total capital investment and 15 % of feedstock cost.

Melt polycondensation (PC) reaction

The product from first step is subjected to a further reaction stage known as polycondensation (PC) which produces poly(ethylene tereph­thalate) of fibre-forming molecular weight. This reaction takes place in melt phase and the byproduct, EG is removed from melt using high vacuum. Reaction [40] shows stoichiometric equation for the synthesis of PET. High viscosity PET grades for bottles or technical yarns are typically produced by further polycondensation in an additional solid state process (SSP) under vacuum or in an inert gas atmosphere.

At this stage of polymer synthesis, the BHET molecules or its short oligomers undergo ester interchange reactions (i.e. transesterification reaction) at their reactive ends to join up into longer chains (in presence of suitable catalyst). One mole of glycol is eliminated in each reaction step.

As the polycondensation reaction is reversible, and the whole process is in complex equilibrium of forward and backward reactions. Therefore, it is important to remove the gly­col produced from the system as quickly as possible. As the glycol is removed by high vacuum; the reaction is forced to move towards higher and higher molecular weight. With increase in molecular weight, the melt becomes increasingly viscous; the melt viscosity of the fibre grade PET polymer at 285°C is of the order of 3000 poise under the shear conditions of roughly 1000 s^-1. It is difficult to continue to drive the reaction forward as the viscosity becomes very high and it becomes difficult to turn the reactant mixture. This prevents EG from evaporating further and makes difficult for any two reactive ends of polymer chains to find each other for further reaction. This stage of the reaction is called diffusion controlled.

The polymerization may be carried out in a semibatch reactor or a continuous reactor.

In this step, a further catalyst (explained later) is added to the mixture of linear oligo­mers, free glycol is distilled out, and the temperature is raised until it reaches 280-­290°C. The pressure is reduced as rapidly as possible without causing carry-over due to foaming, and polymerization is carried out at a pressure below 25 Pa (0.2 mm Hg) until the required molecular weight is attained. In the melt phase, a molecular weight (number average) of typically between 16000 and 19000, which is equivalent to Intrinsic viscosity (IV) of 0.58-0.68, is reached.

What are the Catalysts used for polycondensation

Polycondensation reaction also requires a catalyst, popular catalyst for transesterification reaction at this stage is antimony trioxide, Sb₂O₃, although antimony pentaoxide and germanium dioxide have also been used. The proposed catalysis mechanism of titanium compounds is illustrated in reaction [41].

The primary step in catalysis is the coordination of metal ion to the ester carbonyl bond to increase its polarity to facilitate nucleophilic attack. In the early stages of polycondensation, when the hydroxyl group concentration is high, the antimony catalysts coordinate with hydroxyl groups, and therefore, the catalytic activity of antimony is lower in the initial stages of polycondensation. As the EG is evaporated, the effectiveness of antimony catalysts improves towards the later part of the polycondensation reaction.

How to obtain high molecular weight?

(a) Reactor design

As stated above, the rate of attainment of high molecular weight is diffusion-controlled rather than chemically controlled. It is useful to minimize the diffusion path by maximizing the surface to volume ratio and the rate of generation of free surface. This helps in escape of EG under vacuum and high temperature. This may be facilitated by feeding the melt into rotating disc ring reactors that have multiple discs that create thin and renewable film of the polymer melt, thus significantly increasing the available surface area, and decreasing the diffusion path for condensate (EG) removal.

(b) Chain extenders

The later stages of the polycondensation can be chemically modified to overcome the limitation imposed by the diffusion rate of EG. At an intermediate stage in the polycondensation, 0.5-2% of a diphenyl ester, preferably phenyl carbonate or diphe­nyl terephthalate is introduced into the reaction mixture. These molecules act as chain extenders and react rapidly to couple polymer chains with evolution of phenol. It has been reported that PET formation proceeds at two to three times faster and the product colour is improved when TPA is added as a chain extender. The amount of TPA to be added depends on the DP of the polymer at the time of addition.

The equilibrium in this reaction is much more favorable to the formation of polymer and free phenol than it is in usual reaction which forms polymer and free EG. This helps in greatly accelerating the polymerization reactions and substantially reducing the reaction times. A further result is that polymers of lower carboxyl end-group content and higher molecular weight can be obtained. This modification is now little used because it is less advantageous in a continuous polymerization reactor.