Conventional Production of TDI
A simplified block diagram for production of TDI is given in Figure 2.50. Most TDI produced in the world is from a three-step process as shown. It begins with nitration of toluene, which is subsequently hydrogenated to the polyamine. The last step is reaction of the amine with phosgene to make TDI. Innovations have centered on increasing unit efficiency/reduction of waste and minimization of process phosgene [117-119]. Additionally, there has been ongoing research into nonphosgene routes to polyisocyanates . In Figure 2.50, blocks associated with waste treatment, intermediate purification steps, hydrogen generation, solvent handling, and handling of bottoms products are not included. The
FIGURE 2.50 Simplified block diagram for TDI production.
FIGURE 2.51 Preparation of mtrotoluenes from toluene and nitric acid with a sulfuric acid catalyst. Reprinted with permission from Ref. . © John Wiley & Sons, Inc.
first step of the process is nitration of toluene (Fig. 2.51). This simple reaction vastly understates the complexity of pathways leading from reactants to products. Several by-products are formed including cresol derivatives.
In practice, and as shown in the block diagram, nitration is handled in two steps. Formation of the mononitrotoluene is performed with recycled acids from previous production batches at approximately a 30/55 (wt/wt%) ratio. Along with enhancing overall reagent efficiency, it has the secondary effect of reducing nitric acid in the sulfuric acid prior to sulfuric acid regeneration steps. Formation of dinitrotoluene is performed with nearly the same acid concentrations but at higher temperatures. The yield of the desired 2,4- and 2,6-dinitro positional isomers is approximately 96% and about 4% in other isomers. Controls against formation of trinitro isomers are provided by careful control of acid concentrations and nitration conditions . Figure 2.52 provides explicit pathways of mono- and dinitrotoluene formation. The initial mono-nitration products consist of 57.5% ortho (substitutes at the 2-position), 38.5% para (substitution at the 4-position), and about 4% meta (substitution at the 3-position). Sulfuric acid is regenerated by removal of water with hot air stripping .
The production of nitrotoluenes that avoids use of sulfuric acid and also the expensive sulfuric acid regeneration step has been reported, but is not yet implemented commercially. Problems with the "single acid" nitric acid-only process include creation of more unwanted by-products, slower rates, difficulty in nitrotoluene recovery due to solubility of the product in nitric acid, and an expensive nitric acid regeneration procedure .
The second step to production of TDI is hydrogenation of dinitrotoluene to toluene diamine (Fig. 2.53). The hydrogenation is accomplished in a continuous process using a palladium-platinum on carbon or other metal heterogeneous catalyst system [124, 125]. The hydrogenation is thus a multiphase system of dinitrotoluene melt, catalyst, and toluene diamines in water and hydrogen gas. Hydrogen pressure can vary from ca. 100 psig to over 500 psig. The hydrogenation process is exothermic and temperature is carefully controlled (ca 120 °C). Water and orthoisomers are removed by distillation. As with nitration, the mechanism of hydrogenation to the
FIGURE 2.52 Pathways and product distributions of dintrotoluene from monomtrotoluene. Reprinted with permission from Ref. . © John Wiley & Sons, Inc.
FIGURE 2.53 Preparation of toluene diamines from nitrotoluenes by catalytic hydrogénation. Reprinted with permission from Ref. . © John Wiley & Sons, Inc.
polyamines is very complex, going potentially through hydroxylamine, azo, and hydrazine intermediates among several others.
The third and final step (apart from distillation) is phosgenation of the toluene diamine isomers to diisocyanates (Fig. 2.54).
It is generally accepted that the reaction to isocyanate is a two-step process. First, phosgene reacts with the diamine to form a dicarbamoyl chloride (Fig. 2.55). The dicarbamoyl chloride is not isolated and is directly converted to diisocyanates at elevated temperature (Fig. 2.56) .
In a conventional batch process, TDA is dissolved in monochlorobenzene and mixed with liquid phosgene. The mixture is heated to effect reaction, and HC1 and excess phosgene are purged with nitrogen. The amine and HC1 form an acid-base pair (sometimes referred to as "slurry"), which must be thermally broken, and operationally requires excess phosgene for efficiency. In a continuous process, the carbamoyl chloride is formed at ambient conditions and then pumped to a reaction vessel maintained at an elevated temperature (65-80°C) to form the diisocyanates . A so-called hot phosgenation will gradually increase temperature until there is no more HC1 evolved at which point the reaction is considered complete. The hot process has been
FIGURE 2.54 Phospgenation of toluene diamines to produce toluene diisocyanates. Reprinted with permission from Ref. . © John Wiley & Sons, Inc.
FIGURE 2.55 First step of the phosgeneation process forming the dicarbamoyl chloride. The acid product can complex the free toluene diamine requiring application of excess phosgene and heat to complete the reaction.
FIGURE 2.56 Thermolysis of the dicarbamoyl chloride to form the isocyanate and acid co-product. The acid is recovered by distillation.
run up to nearly 90% conversion efficiency of toluene diamine to TDI. Unreacted diamine can be recycled back to the phosgenation step, and partially converted isocyanate can be reverted back to the diamine by hydrolysis and sent back for further phosgenation. Reactions of diisocyanates with amines occurring under concentrated conditions, such as urea, biuret, and carbodiimide formation, are additional routes for loss of final product. These reactions will be covered in detail in Chapter 3.
A significant innovation that has developed commercial reality has been the gas-phase phosgenation process originally developed by Bayer and described in a series of patents [118, 128]. The main innovation allows for the efficient formation of diisocyanates without the need for solvent such as monochlorobenzene. This may not sound to the ear like a major breakthrough, but production of commodity chemicals is a very conservative endeavor, especially when there are toxic reagents involved. Although the gas-phase process does incorporate some monochlorobenzene solvent in the phosgenation, the significant reduction reduces the need for handling, heating, and recovering solvents representing savings in capital as well as process expense. Furthermore, the shorter contact time of toluene diamine with phosgene results in a reduction of phosgene inventory required to operate the plant reliably—a considerable health and safety outcome.
In gas-phase phosgenation, TDA, phosgene, and a small amount of chlorinated benzene solvent are heated to very high temperatures (>300 °C). The amine and phosgene feeds are mixed in a reactor tube in a manner that maximizes intermolecular interaction . The molar ratio of phosgene to toluene diamine is high to maintain a high reaction rate reflecting second-order reaction kinetics. Residence times and process are optimized for decomposition of the carbamoyl chloride intermediate resulting in yields on the order of 99% of that expected from theory. Solvent, unreacted phosgene, HC1, and by-products are separated as a vapor phase and subsequently quenched and separated for recycle or disposal.
The attractive features of this process are resulting in several announcements of new TDI plant construction based on the gas-phase phosgenation technology and perhaps have contributed to closing of older less profitable assets by other companies. The economic advantage in cost of manufacture using gas-phase technology for a large (660 million pound/year) TDI plant is on the order of 10%—a large difference for commodity chemicals .