Polycarbonate Production and Manufacturing Process


The first aromatic polycarbonates were prepared in the late 1890s by reacting hydroquinone or resorcinol with phosgene in pyridine but the crystalline polymers produced were brittle and difficult to process. In 1941, PPG introduced a crosslinked resin prepared by a peroxide initiated radical polymerisation of the bisallyl carbonate of diethylene glycol. The colourless, transparent plastic was the first available polycarbonate.

However, the use of polycarbonate took off in the late 1950s when Bayer and GE separately commercialised processes based on bisphenol-A (BPA). Early production was based on reacting phosgene with phenol to produce diphenyl carbonate (DPC), which was then reacted with BPA to produce the polymer and liberate phenol for reuse. However, this approach suffered from slow reaction rates and the need for several small-scale batch reactors.

The problem of achieving adequate contacting of solid BPA with gaseous phosgene was solved leading to the current interfacial polymerisation process. Here, alkali salts of BPA in aqueous solution are phosgenated in the presence of an inert solvent. The reaction can be carried out in one or two stages (i.e., phosgenation and polycondensation) in batch or continuous operation.

However, tougher environmental restrictions have given impetus to develop non-phosgene routes to polycarbonate. In addition, the interfacial process uses a chlorinated solvent such as methylene chloride which is subject to exposure limits. There is also an economic penalty in that the chlorine content of phosgene is wasted and converted to sodium chloride. Caustic soda is consumed in the conversion and the disposal of the waste salt solutions presents ecological problems in itself.

Sabic Innovative Plastics (formerly GE Plastics), Bayer, Asahi/Chi Mei and Mitsubishi Chemical/Mitsubishi Gas Chemical have independently developed and are using non-phosgene processes. In addition, Teijin and LG are also developing phosgene-free routes.

They all take the same overall approach where polymerisation relies on the transesterification of DPC with BPA. This is more commonly termed as the melt process as the two-stage polymerisation takes place in the absence of solvents.

In the first stage, BPA and excess DPC are reacted and phenol is removed to produce a prepolymer. Polymerisation to a higher molecular weight product occurs primarily through an ester disproportionation whereby DPC is formed and volatilised from the system.

The melt process has the advantage of making a product in undiluted form that may be palletised directly. Disadvantages include the need for equipment to withstand high temperatures and high vacuum.

Because it is difficult to prepare DPC directly, the new non-phosgene routes make it indirectly by using an intermediate dialkyl carbonate, usually dimethyl carbonate (DMC), as the source of carbonate functionality. The first process step is to react phenol with dimethyl carbonate to make phenyl methyl carbonate. The next step can take one of two routes: either a further reaction of phenyl methyl carbonate with phenol or to convert phenyl methyl carbonate to DPC via disproportionation.

The distinguishing characteristic of the non-phosgene routes is the method to make the dialkyl carbonates.

Methods employed include:

(1) Sabic produces DPC from DMC which is produced from CO, methanol and oxygen using EniChem technology;

(2) Bayer technology reacts NO and methanol oxidatively to give methyl nitrite which undergoes carbonylation to produce DMC;

(3) Mitsuibishi Chemical has technology for making DPC from di-n-butyl carbonate, which is produced from urea and n-butyl alcohol; and

(4) Asahi/Chi Mei produces DPC from DMC which is produced via methanolysis of ethylene carbonate.

(Source: Nexant ChemSystems PERP Programme – Polycarbonate)