当前位置:首页 >> 自然科学 >>

Determination of Thermal Decomposition Products from a Phenolic Urethane Resin by PyGC-MS

Determination of Thermal Decomposition Products from a Phenolic Urethane Resin by Pyrolysis-Gas Chromatography-Mass Spectrometry
Cory A. Lytle, Wolfgang Bertsch*
University of Alabama, Department of Chemistry, Box 870336, Tuscaloosa, AL 35487-0336, USA

Marvin D. McKinley
University of Alabama, Department of Chemical Engineering, Box 870203, Tuscaloosa, AL 35487-0203, USA Ms received: January 5, 1998; accepted: January 5, 1998

Key Words: Pyrolysis; GC/MS; phenolic urethane resin; thermal decomposition; metal casting 1 Introduction
Analytical pyrolysis is a powerful yet under-utilized tool in polymer characterization [1]. When combined with separation methods, in particular with GC/MS, it can provide answers to many questions which are difficult to get from other analytical methods. Synthetic polymers have been subjected to Py-GC-MS to obtain information of polymer microstructure [2], determine kinetic parameters [3], and evaluate degradation in different environments [4]. The determination of potentially toxic decomposition products from the thermal decomposition of industrially important synthetic polymers is of particular interest, for obvious reasons [5, 6]. Data from such studies allow the assessment of potential health risks and provide the necessary baseline information for the implementation of emissions control standards. Industries in the United States are subject to legislation which limits emissions of hazardous air pollutants (HAPs). A wide variety of analytical methods are available to ensure compliance at the work place [7]. The metal casting industry produces a significant quantity of volatiles. Metal casting is a process where hot metal is poured into molds of various shapes [8]. A critical issue in the metal casting industry is the production of hollow parts. Components such as automobile engine casings, which must be hollow to function, can be made inexpensively by casting. Sand cores, held together by various resins and shaped into the design of the hollow part of the casting are used inside of sand molds during the pouring and casting process. Synthetic organic polymers are applied as resin binders to form sand into hardened molds and cores [7]. Due to the large amount of heat given off during the casting process, the resin decomposes, allowing the sand to fall from the casting. The initial composition of the binder and the temperature to which it is exposed are the major factors influencing the nature and quantity of volatiles being emitted from the resin. It is well known that a variety of compounds are produced when resins decompose thermally [8]. Some of these components are toxic and are therefore of considerable interest [9]. As of today, no detailed data are available on the quantity of HAPs generated by specific resins. The objective of this research was to develop analytical methodology for determining products emitted during decomposition of an industrially important resin binder. Data from Py-GC-MS and thermogravimetric analysis (TGA) were used to assess the volatile components released during decomposition of a phenolic resin which has found wide commercial use. The data from studies such as ours are combined with existing solidification codes and incorporated into computer modeling programs to calculate emissions [10]. This study thus serves as a model for the foundry industry to predict the amounts of HAPs produced by the phenolic urethane cold box PUCB resin binder system. Phenolic urethane resin binder systems can be classified as either no-bake or cold-box systems. The chemistry behind these two systems is similar. The difference lies in the nature and use of the catalyst. In order to be considered a cold-box system, the catalyst, a gaseous amine, is blown through the core box to achieve immediate cure. An amine catalyst is also used in the no-bake system; however, it is added as a liquid. This allows for a slower rate of cure, thus increasing the work time of the resin binder system [11]. The phenolic urethane system is comprised of three parts [12] (see Table 1). In essence, it consists of a phenolic resin, a diisocyanate, and a tertiary amine catalyst. The resin is formed by the reaction between diisocyanates and phenolic precursors as shown in Figure 1. It is essentially a nucleophilic attack of the oxygen atom of the phenol on the carbon atom of the diisocyanate group. Phenolic urethane resins are becoming the most widely used binders in the foundry industry due to many favorable characteristics [13, 14].

2 Experimental 2.1 Chemicals
The PUCB resin was prepared by adding 0.6 g of the formaldehyde precursor (Isocurem I) to 0.5 g of the diisocyanate (Isocurem II, both from Ashland Chemical Co., Columbus, OH). To this mixture, 50 lL of the catalyst triethylamine, TEA (Sigma
Table 1. Composition of the three part phenolic urethane system. Part I (Isocurem I) phenol formaldehyde resin aromatic petroleum distillates dioctyl adipate phenol dimethyl glutarate dimethyl adipate dimethyl succinate naphthalene 1,2,4-trimethylbenzene methylene diphenyldiisocyanate polymeric MDI aromatic petroleum distillates aliphatic hydrocarbon methylene diphenylisocyanate naphthalene triethylamine

Part II (Isocurem II)

Part III


VOL. 21, FEBRUARY 1998

J. High Resol. Chromatogr.

Thermal Decomposition Products from a Phenolic Urethane Resin

with a coil probe. The pyrolyzer used a CDS Pyroprobe 100 interface to the GC/MS. Probe temperatures of 500, 700, and 900 8C were employed, each held for 1, 5, 10, and 20 s. The probe was cleaned at 1000 8C for 10 s between samples. The background was checked periodically by running blanks. Samples were pre-weighed, loaded into quartz tubes and held in place by small amounts of silanized glass wool.

2.5 GC-MS Analysis
The analysis of the volatiles from the pyrolyzate was carried out on a quadrupole type GC/MS instrument (HP GCD, Avondale, PA), using a 30 m 60.25 mm 60.25 lm fused silica capillary column coated with 5% diphenyl/95% dimethylpolysiloxane stationary phase (HP-5, Hewlett Packard, Avondale, PA). A temperature program of 40 – 280 8C at a rate of 10 8/min was found to give the best separation within a reasonable amount of time. A 30 m 6 0.32 mm fused silica PLOT type column coated with a porous divinylbenzene homopolymer (GS-Q, J & W Scientific, Folsom, CA), and a 30 m 60.32 mm 610 lm fused silica capillary molsieve absorbent type column (CP-Molsieve 5A, Chrompack Inc., Raritan, NJ) was used for the separation of low molecular weight components. Standard EI conditions were used.

Figure 1. Structure of phenolic urethane resin.

Aldrich Co., St. Louis, MO) was added using nitrogen to blow the TEA through the core. The PUCB resin coated sand was prepared by adding the resin mixture to 75 g of Wedron washed sand before the TEA was added. Wedron washed sand is a high purity sand commonly used in the foundry industry. The coated sand contained 1.5% binder in the final product.

3 Results and Discussion
TGA was performed on the uncoated sand and the PUCB resin. The resin coated sand was then investigated to determine how the resin behaved on the sand matrix. Data produced by repetitive sampling showed a high degree of variation, indicating that the uncoated sand was rather inhomogeneous. Volatile organic compounds (VOCs) emitted from the uncoated sand consist mainly of benzene, toluene, and cresols and ranged from 0.3 – 1.5% (wt/wt). These compounds are probably “left over” from the washing process the sand endures before being used. The thermal profiles of the PUCB resin and the PUCB resin coated sand are shown in Figure 2a and Figure 2b, respectively. They behave very similar up to around 300 8C. With the PUCB resin, the majority of the volatiles have evolved by 450 8C. However, with the PUCB resin coated sand, a large portion of the volatiles remains after approximately 600 8C. The discrepancy can be explained by the high heat capacity of the sand, which is capable of absorbing a large amount of heat before transferring it to the resin. This creates a “lag” in the volatilization of the resin. Low molecular weight gases were released from both the resin and resin coated sand starting at around 200 8C. They were analyzed on a PLOT column. This column separated ethene, propene, butene, 1,3-butadiene, 2-propenenitrile, 1-pentene, and 1,3-cyclopentadiene. Mass spectrometry and retention time matching were used for identification. This column was not suitable for the separation of carbon monoxide from air. In this case the molsieve column produced very good results. It was also used for determining carbon monoxide and methane. The origin of the water in the PUCB resin was of interest. A sample of the coated sand weighing about 10.0 mg was baked in an oven at 100 8C for 30 min, then pyrolyzed. The results indicated that the water is a product of pyrolysis and originates from an elimination reaction taking place at high temperature.
VOL. 21, FEBRUARY 1998

2.1 Standards
n-Pentadecane, ACS grade (Polyscience Corporation, Niles, IL) was diluted with n-hexane, to produce a 1000 ppm stock solution which was further diluted, as needed. A gas standard which contained methane, ethane, ethene, acetylene, carbon monoxide, and carbon dioxide, each at 1% in nitrogen was purchased from Supelco Inc. (Bellefonte, PA). A 0.1% propane standard was produced in house using grade HD-5 propane from a propane torch purchased locally. cis – 2-Butene was purchased from Matheson Co. Inc. (Atlanta, GA), and a 0.1% standard, in nitrogen was prepared. Dilute HCl was added to a small amount of NaCN in a 150 mL Erlenmeyer flask. Headspace was withdrawn using a gas tight syringe to produce an in house retention time standard of HCN.

2.3 Thermogravimetric Analysis
Weight loss analysis was performed on the uncoated sand, PUCB resin and on the resin coated sand using a model TA 2950 thermogravimetric analyzer (TA Instruments, New Castle, DE). The standard temperature program was from room temperature to 1000 8C at a rate of 150 8/min. Nitrogen was used as the purge gas at a rate of 40 mL/min to the balance, and 60 mL/min to the furnace, as recommended by the instrument manufacturer.

2.4 Pyrolysis
Pyrolysis was performed using a standalone pyrolyzer, CDS Pyroprobe 2000 (Chemical Data Systems, Oxford, PA) equipped
J. High Resol. Chromatogr.


Thermal Decomposition Products from a Phenolic Urethane Resin

Figure 2. First derivative curve of weight loss with respect to temperature of (a) PUCB resin and (b) PUCB resin coated sand.

Figure 3. Pyrograms of PUCB resin coated sand at (top) 500 8C, (middle) 700 8C, and (bottom) 900 8C.

Mass spectrometry was the primary tool used to determine the identity of VOCs. Standards and retention indices were also used to identify and quantitate the VOCs produced by the PUCB resin. Table 2 lists the major volatile components emitted from the PUCB resin coated sand at 900 8C. These conditions most closely resemble the pouring environment. Table 3 shows the distribution of pyrolysis products of the PUCB resin coated sand at 500 8C, 700 8C, and 900 8C. Average RSD values for Table 2 and 3 ranged from 10 to 15%. Due to the many variables involved in sample manipulation and pyrolysis, large variations are expected. The amount of VOCs produced as a function of probe temperature was also examined. Figure 3 show programs at 500 8C (top), 700 8C (middle) and 900 8C (bottom). The amount of VOCs increases as the pyrolysis temperature increases. The evolution of HCN, carbon monoxide and methane as a function of temperature is shown in Figure 4. The amount of HCN produced at 700 8C is minimal. At 900 8C, the presence of HCN is small but distinct. CO and CH4 are both present in noticeable quantities at 700 8C, and increase in amount at 900 8C. The reasons for the relatively high temperature required for the emission for these low molecular weight components are not fully understood.

Figure 4. Evolution of HCN, CO, and CH4 as a function of temperature.

4 Conclusions
Conditions typical of pouring and casting were modeled in the laboratory. Pyrolysis of PUCB resin results in the formation of a complex mixture of volatile compounds. Except for fixed gases which are the major pyrolysis products at all pyrolysis temperaJ. High Resol. Chromatogr.


VOL. 21, FEBRUARY 1998

Thermal Decomposition Products from a Phenolic Urethane Resin

Table 2. Major volatile components emitted from the PUCB resin coated sand at 900 8C. Compound Fixed Gases methane carbon monoxide Light Gases, Volatiles ethene propene HCN 1-butene 1,3-butadiene 2-propenenitrile 1-pentene 1,3-pentadiene Semi-Volatiles benzene toluene 3-methylene heptane ethyl benzene p-xylene propyl benzene benzene isocyanato 3-ethyl toluene 1,3,5-trimethyl benzene phenol benzonitrile 1,2,3-trimethyl benzene benzofuran 1,2,4-trimethyl benzene 2-ethyl-1-hexanol dimethyl ester butanedioic acid indene 1-methyl-3-propyl benzene 2-methyl phenol 1.99 acridine 0.19 0.37 0.15 phenanthrene anthracene 0.04 0.03 3.08 1.68 0.04 0.17 0.39 0.04 0.25 0.30 0.20 3.78 0.41 0.42 0.33 0.20 0.10 0.90 3042 9200 3478 2782 723 2138 2919 1613 4337 Conc. lg/g Compound 1-ethyl-3,5-dimethyl benzene 1-isocyanate-4-methyl benzene 4-ethyl-1,2-dimethyl benzene 2-methyl benzofuran 2,6-dimethyl phenol 1,2,4,5-tetramethyl benzene 1,2,3,5-tetramethyl benzene 4-ethenyl-1,2-dimethyl phenol dimethyl ester pentanedioic acid 1,1-dimethyl propyl benzene naphthalene 1,3-dimethyl-5-(1-methylethyl)benzene 2,3,5-trimethyl phenol ethyl-1,2,4-trimethyl benzene dimethyl ester hexanedioic acid 2,3-dihydro-4,7-dimethyl-1H-indene pentamethyl benzene 1-methyl naphthalene 2-methyl naphthalene biphenyl 1-ethyl naphthalene 2-ethyl naphthalene 2,6-dimethyl naphthalene 1,5-dimethyl naphthalene 2,3-dimethyl naphthalene 2-ethenyl naphthalene 1,3-dimethyl naphthalene 1,5-dimethyl naphthalene 2,6-dimethyl naphthalene 2-(1-methylethyl)-naphthalene 1,4,6-trimethyl naphthalene 1,4,5-trimethyl naphthalene 1,6,7-trimethyl naphthalene 2,3,6-trimethyl naphthalene Conc. lg/g 0.09 0.06 0.46 0.10 0.59 0.35 0.51 0.06 2.99 0.07 2.87 0.11 0.03 0.12 0.96 0.11 0.26 5.61 2.97 0.16 0.63 0.13 1.49 1.13 0.37 0.28 0.44 0.03 0.26 0.22 0.19 0.18 0.05 0.07

Table 3. Distribution of pyrolysis products of PUCB resin coated sand. Temperature, 8C Fixed Gases 500 700 900 trace 32% 19% Light Gases, Volatiles trace 67% 80% Semi-Volatiles 100% 0.7% 0.1%

tures studied, phenol is a major pyrolysis product. At 700 8C benzene and toluene begin to evolve. At 900 8C phenol is no longer the major component. 1-Methyl naphthalene becomes the major pyrolysis product. HCN is produced in negligible amounts. The amount of VOCs determined in the model studies appears to be high. It is important to note that foundry workers are not exposed to these high concentrations, because the molds are continually flushed with fresh air.
VOL. 21, FEBRUARY 1998

J. High Resol. Chromatogr.


Thermal Decomposition Products from a Phenolic Urethane Resin

[1] S.A. Liebman, E.J. Levy, Pyrolysis and GC in Polymer Analysis.Marcel Dekker, Inc., New York 1985. [2] S. Tsuge, H. Ohtani, Pyrolysis-Gas Chromatographic Studies on the Micro-structures of Stereospecific Polypropylenes, in Analytical Pyrolysis, Butterworth, 1984, p. 407. [3] J.J.R. Mertens, E. Jacobs, A. Callaerts, A. Buekeus, Macromol. Chem. Rapid Commun. 1982, 3, 349. [4] A. Alajbeg, B. Stripak. J. Anal. Appl. Pyrol. 1985, 7, 283. [5] M.M. Shapi, A. Hesso. J. Anal. Appl. Pyrol. 1990, 18, 143. [6] S.L. Hathcock, W. Bertsch, J. High Resol. Chromatogr. 1993, 16, 609.

[7] S.A. Ness, Air Monitoring for Toxic Exposures. An Integrated Approach. Nostrand Reinhold, New York, NY 1991. [8] J.S. Cambell, Casting and Forming Process in Manufacturing. McGraw Hill, New York 1950. [9] M. McKinley, A. Jefcoat, W.J. Herz, C. Frederick, AFS Transactions 1993, 101, 979. [10] M. McKinley, Modeling of Casting Process Combustion Products. Technical Management Concepts, Inc. Beavercreek, OH, 1997. [11] P.A. Blackburn, C. M. Henry, AFS Transactions 1996, 104, 945. [12] M.M. Geoffrey. AFS Transactions 1995, 103, 587. [13] J.L. Archibald. Modern Casting 1994, 3, 35. [14] D.M. Churches; K.B. Rundman. AFS Transactions 1995, 103, 587.


VOL. 21, FEBRUARY 1998

J. High Resol. Chromatogr.