Environ. Sci. Technol. 2004, 38, 1066-1072
Hydrolytic Stability of Toluene Diisocyanate and Polymeric Methylenediphenyl Diisocyanate Based Polyureas under Environmental Conditions VAHID SENDIJAREVIC, AISA SENDIJAREVIC, AND IBRAHIM SENDIJAREVIC Troy Polymers, Inc., 330 East Maple Road, Suite L, Troy, Michigan 48083 ROBERT E. BAILEY* Bailey Associates, 424 Little Lake Drive, No. 13, Ann Arbor, Michigan 48103 DENIS PEMBERTON Gilbert International Ltd., Bridgewater House, Whitworth Street, Manchester, M1 6LT, U.K. KURT A. REIMANN BASF Corporation, 1609 Biddle Avenue, Wyandotte, Michigan 48192-3736
Polyureas were prepared by reacting toluene diisocyanate (TDI) or polymeric methylenediphenyl diisocyanate (PMDI) with water under prolonged vigorous mixing at room temperature. Hydrolytic degradation of these (powdered) TDI- and PMDI-based polyureas was studied by measuring the rates of formation of free toluenediamine (TDA) or methylenedianiline (MDA) in water as a function of time by utilizing HPLC. The heterogeneous hydrolysis reactions were carried out in glass reaction tubes under nitrogen with initial polyurea loadings of 2 g/L in deionized water or buffer solution. In the case of TDI-polyurea, concentrations of free 2,4- and 2,6-toluenediamine in water were measured after hydrolysis at 120, 140, and 160 °C. The hydrolysis of PMDI-polyurea was carried out at 150, 160, and 170 °C, and concentrations of 2,4′-methylenedianiline (2,4′MDA), 4,4′-methylenedianiline (4,4′-MDA), and 2,4-bis(paminobenzyl)aniline were measured. In both cases the rate of formation of diamine was well represented by both a pseudo-first-order reaction and a zero-order reaction. The temperature dependence of rate constants fit Arrhenius behavior. The half-time for hydrolysis of TDI-polyurea at 25 °C was calculated to range from about 18 000 to 300 000 years and that of PMDI-polyurea was estimated to range from about 110 000 to 12 million years, depending on the kinetic assumptions made. The half-times for hydrolysis at buffered pH levels of 4, 7, and 9 were within a factor of 2 of those in deionized water. These results are of importance in understanding the fate of polyureas formed in the event of a release of TDI or PMDI into the environment.
Introduction Toluene diisocyanate (TDI) and polymeric methylenediphenyl diisocyanate (PMDI) are used in the manufacture of * Corresponding
[email protected]. 1066
9
author
phone:
(734)994-8989;
e-mail:
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 4, 2004
polyurethanes. Approximately 4 million metric tons of these two reactive materials are transported around the world annually. They are among the major isocyanates used in the manufacture of polyurethane resins and products. TDI is normally used as a mixture of the 2,4- and 2,6-isomers, 80% and 20%, respectively. PMDI, commonly known as “polymeric” MDI, contains about 50% of the “monomer”, 4,4′-MDI (n ) 0 in the structure shown in Figure 1). It has been well established that isocyanates, especially aromatic isocyanates, are highly reactive with water (1). Solid insoluble polyureas have been reported to be the predominant products in these reactions (1). The reaction of isocyanates with water is shown in simplified form for the 2,6-TDI isomer (Scheme 1). Hence, environmental contact with water, as in an accidental spill and subsequent cleanup, leads to formation of polyureas. These polyureas formed in the reaction of TDI and MDI with water have not been observed to degrade significantly by hydrolysis under normal environmental conditions (temperatures up to 40 °C and pH 4-9) (2, 3). They are hydrophobic and insoluble in water and organic solvents. Yakabe (2) reported that the polyurea from TDI in water was not completely soluble in DMF containing LiCl with the molecular weight of the soluble portion approximately 5600. Only a minor fraction of the polyurea from PMDI was soluble in DMF containing LiCl. Results of studies carried out on spill sites (4) and model spillage experiments of PMDI indicate that the environmental effect is minimal (5). No toluenediamine (TDA) or methylenedianiline (MDA) was detected in the water at a spill site (4) or in the Heimbach et al. study (5). However, to determine the long-term environmental stability and potential impact of these polyureas, quantitative data on their hydrolysis rates and the potential for TDA or MDA formation under environmental conditions are required. A simplified representation of the complete hydrolysis of a polyurea based on 2,6-TDI is shown in Scheme 2. For much of the course of this reaction, shorterchain polyureas rather than diamine will be the major products of hydrolysis. Most of the hydrolysis studies reported in the literature were carried out on smaller molecules (mono- and bis-ureas) in solution utilizing a mixture of water with organic solvents. The reported hydrolysis rate for the diphenyl urea made from 4,4′-methylenediphenyl diisocyanate (4,4′-MDI) at 70 °C in DMSO/water is 6.43 × 10-4 h-1, a half-life of about 45 days (6). Hydrolysis of a series of substituted ureas over a range of temperatures was studied by Audu and Heyn (7). Using the activation energy of Audu and Heyn, approximately 100 kJ/mol, and correcting the results of Chapman for the solvent composition, the neutral hydrolysis half-life of the diphenyl urea of 4,4′-MDI in water at 20 °C is estimated to be 15-20 years. These data for soluble ureas cannot be used to quantitatively extrapolate the environmental hydrolytic stability of solid insoluble polyureas derived from the reaction of aromatic diisocyanates with water. A previous study (8) of the preparation of polyureas from TDI and PMDI with water and their hydrolysis in water showed no significant hydrolysis, and no TDA or MDA detected, after 85 days in water at 70 °C. The solid polyureas formed were found to be porous and contained unreacted isocyanate functions bound to the polyurea. Studies on hydrolysis of polyureas are complicated by the fact that these reactions are heterogeneous. Thus, the rate of hydrolysis may depend on particle size and surface area, and on efficiency of mixing. The focus of the present study was on the rate of TDA and MDA formation by hydrolysis of polyureas prepared 10.1021/es035072f CCC: $27.50
2004 American Chemical Society Published on Web 01/16/2004
FIGURE 1. Structures of PMDI and TDI.
TABLE 1. Materials Other than Those Obtained from Laboratory Suppliers name
chemical composition
supplier
Lupranate M-20S
Polymethylene polyphenyl isocyanate: 4,4′-diphenylmethane diisocyanate, 39.4%; 2,4′-diphenylmethane diisocyanate, 3.4%; 2,4-bis(p-isocyanatobenzyl)phenyl isocyanate, Figure 1, n ) 1, 18.9%; other congeners of polymethylene polyphenyl isocyanate, balance 2,6-toluene diisocyanate, 20.0%; 2,4-toluene diisocyanate, 80.0% 2,4′-methylenedianiline, 98.21% 2,4-bis(p-aminobenzyl)aniline, 95.1%
BASF
Lupranate T80-1 2,4′-MDA “Triamine”
SCHEME 1. Formation of polyurea from 2,6-TDI
BASF BASF BASF
TABLE 2. Composition of Eluents Used in the HPLC Analyses water ammonium carbamate acetonitrile
eluent A
eluent B
850 mL 1.3 g 150 mL
100 mL 0.2 g 900 mL
SCHEME 2. Hydrolysis of Polyurea Based on 2,6-TDI TABLE 3. Composition of Mobile Phases Used in HPLC Analyses TDA analysis
from commercial samples of TDI and PMDI to estimate any potential risk to the environment.
Experimental Section Preparation of Polyureas. Hydrolysis studies were carried out on polyureas prepared from TDI (Lupranate T80-1), and PMDI (Lupranate M-20S) with the compositions shown in Table 1. Polyureas were prepared by reacting isocyanates with deionized water at room temperature in a covered glass reaction kettle with vigorous mixing. The isocyanate, up to 50 g in 1 L of water, was added dropwise for about an hour while the reaction was stirred at about 300 rpm with a glass rod equipped with a Teflon paddle. The isocyanate conversion was monitored by the measurement of the isocyanate content of both the solid polyurea product (separated from the suspension by filtration) and in the filtrate by titration (9). Mixing was continued for 2 weeks. To look for any free TDI remaining in the solid polyurea, 0.6 g of ground solid TDIpolyurea with an isocyanate concentration of approximately 7% was mixed with 50 mL of toluene for 30 min at 60 °C. The toluene filtrate was treated with excess di-n-butylamine and back-titrated with hydrochloric acid for isocyanate determination (9). Polyurea master batches were prepared for use in hydrolysis studies. The suspension of TDI-polyurea was filtered under vacuum, washed several times with deionized water, and dried in an oven at 60 °C for 24 h. The dried polyurea was ground by hand in a mortar and sieved through standard sieves. In the case of TDI-polyurea, the fraction with particles
