Environ. Sci. Technol. 2010, 44, 5313–5317
Investigation of Thermodynamic Parameters in the Thermal Decomposition of Plastic Waste-Waste Lube Oil Compounds Y O N G S A N G K I M , * ,† Y O U N G S E O K K I M , ‡ AND SUNG HYUN KIM§ School of Civil Engineering, Purdue University, West Lafayette, Indiana 47907, Department of Chemical Engineering, University of Massachusetts, Amherst, Massachusetts 01003, and Department of Chemical and Biological Engineering, Korea University, Seoul 136-701, Republic of Korea
Received April 12, 2010. Revised manuscript received May 14, 2010. Accepted June 1, 2010.
Thermal decomposition properties of plastic waste-waste lube oil compounds were investigated under nonisothermal conditions. Polyethylene (PE), polypropylene (PP), polystyrene (PS), and polyethylene terephthalate (PET) were selected as representative household plastic wastes. A plastic waste mixture (PWM) and waste lube oil (WLO) were mixed with mixing ratios of 33, 50, and 67 (w/w) % on a PWM weight basis, and thermogravimetric (TG) experiments were performed from 25 to 600 °C. The Flynn-Wall method and the Ozawa-FlynnWall method were used for analyses of thermodynamic parameters. In this study, activation energies of PWM/WLO compounds ranged from 73.4 to 229.6 kJ/mol between 0.2 and 0.8 of normalized mass conversions, and the 50% PWM/ WLO compound had lower activation energies and enthalpies among the PWM/WLO samples at each mass conversion. At the point of maximum differential mass conversion, the analyzed activation energies, enthalpies, entropies, and Gibbs free energies indicated that mixing PWM and WLO has advantages in reducing energy to decrease the degree of disorder. However, no difference in overall energy that would require overcoming both thermal decomposition reactions and degree of disorder was observed among PWM/ WLO compounds under these experimental conditions.
Introduction Currently, production of biofuels, including ethanol and biodiesel, primarily relies on corn and soybeans. However, demand for farm crops as fuel sources has detrimental effects including fluctuations in grain prices (1). As alternatives to farm crops, plastic waste and waste lube oil are potentially important sources in renewable fuel and energy production (2-5). The main advantages of these waste sources are that 1) massive amounts of plastic waste and waste lube oil are continuously generated in both households and industries (6, 7); as a result, these readily available and inexpensive * Corresponding author phone: (765)494-2189; fax: (765)496-1107; e-mail:
[email protected]. † Purdue University. ‡ University of Massachusetts. § Korea University. 10.1021/es101163e
2010 American Chemical Society
Published on Web 06/11/2010
waste materials can be obtained continuously without fluctuations in price, 2) the reuse of waste prevents contamination of natural environments caused by improper disposal of such waste, and 3) the use of such waste will reduce consumption of other natural resources in fuel production. For example, in the United States, 25 million tons of plastic wastes were discharged in 2001, and less than 4% of the plastic waste was recycled (8). In Western Europe, 48.3 million tons of plastic solid wastes (PSW) were consumed in 2007, and up to 60% of the plastic waste was land-filled or discharged without any treatment (5, 9). To date, an appropriate market for treating PSW, discharged as various types of mixtures, has not been well developed. Application of thermal techniques, such as gasification processes (10), pyrolytic processes (11, 12), and liquid-gas hydrogenation (13), is a potentially productive approach for producing fuel by utilizing plastic waste and waste lube oil as feedstock. Plastic waste and waste lube oil have high heat capacities (e.g., polyolefins: ∼43.3 MJ/kg and general waste lube oil: ∼43.1 MJ/kg) (3, 14), and their heat capacities are closer to petro-disel (∼43 MJ/kg) (15). However, the main problem with these wastes is that they are randomly discharged as various types of mixtures, and the physical properties and thermal degradation properties of each waste are different (16, 17). Thus, in the case of a mixture used as feedstock of fuel production, its efficacy has to be determined mainly by the ratio of the mixed plastic waste and waste lube oil, taking into account their physio-chemical interactions. Several researchers have studied the feasibility of producing fuel from a mixture of plastic waste and waste oil treated through thermolysis, focusing in particular on analysis of regenerated-fuel quality (11, 18) and evaluation of thermal reaction conditions (2, 19). However, fundamental questions that have not been discussed are the physio-chemical interactions of the waste mixture and their effect on energy consumptionandthermodynamicparametersduringthermolysis. In this study, we examined changes in the thermodynamic parameters of plastic waste mixture-waste lube oil (WLO) compounds under thermal decomposition conditions, by applying a nonisothermal thermogravimetry (TG) method. Polyethylene (PE), polypropylene (PP), polystyrene (PS), and polyethylene terephthalate (PET) were selected as commonly discharged household plastic wastes. These four selected plastic wastes, PE, PP, PET, and PS, were mixed by weight in a ratio of 2:2:2:1, respectively, to simulate a commonly discharged household plastic waste mixture (PWM) (16). Subsequently, the PWM and WLO were mixed in ratios of 33 (one-third), 50 (half), and 67 (two-thirds) (w/w) % by PWM weight, and nonisothermal TG experiments were performed. Thermodynamic parameters including activation energy, enthalpy, entropy, and Gibbs free energy were analyzed by the Flynn-Wall method and the Ozawa-Flynn-Wall method.
Experimental Section Materials. A sample of WLO was collected from an auto repair shop located in Seoul, the Republic of Korea. Plastic waste of PE, PP, PET and PS was collected from a plastic waste management facility in the Republic of Korea. The waste of PE, PP and PS was pulverized, washed, dried and pelletized to the size of 2 mm diameter ×3 mm length. Thin PET was washed, dried and cut in a tetragon shape of 2 mm ×2 mm. The physical properties of the WLO including dynamic viscosity, density, carbon number distribution, and analytical methods were previously reported (20, 21). The glass transition temperature and the melting temperature of the PE, PP, VOL. 44, NO. 13, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5313
FIGURE 1. A schematic diagram of a TG analysis system: 1. nitrogen cylinder, 2. heater, 3. sample bowl, 4. reaction chamber, 5. reference bowl, 6. thermocouple, 7. temperature indicator, 8. gas outlet, 9. microbalance, 10. PID controller, 11. data recording computer. PS, and PET, and the qualitative analytic results of the metals and metal oxides in the PE, PP, PS, and PET were also reported (16). Nonisothermal Decomposition Experiment. Thermal decomposition experiments were conducted with a TG analyzer (Thermal Interaction Co. Balance Model TML10001). A schematic diagram of the TG analyzer is shown in Figure 1. A reaction chamber, sample bowl, and reference bowl were made of quartz. The capacity of the sample bowl was 1000 mg, and a microbalance had a readability of 1 µg. The temperature inside the reaction chamber was controlled by a proportional integral derivative (PID) controller. A thermocouple was placed right below the sample bowl to monitor the inside temperature. Nitrogen gas was introduced inside the reaction chamber to maintain an oxygen free environment. 400 mg of each sample was charged to the sample bowl prior to the experiment. In this experiment, PWM was prepared to the mixing ratio of 2:2:2:1 for PE, PP, PET, and PS on a weight basis, respectively. This mixing ratio of the PWM was determined based on the reported statistics on annually discharged household plastic waste from the Republic of Korea (22). More specific information can be found elsewhere (16). Mixture samples of PWM/WLO were prepared to the mixing ratios of 33, 50, and 67 (w/w) % on a PWM weight basis. After a sample charged bowl was placed inside the reaction chamber, any leaking of nitrogen gas was observed, and time was allowed until the microbalance and the TG system were fully stabilized. After stabilization, nonisothermal experiments were performed with different heating rates of 5, 10, and 15 °C/min for each sample, respectively. The experiment started at 25 °C and finished at 600 °C.
Results and Discussion TG Analysis. TG curves of WLO and mixture samples of PWM/ WLO (33, 50, and 67%) obtained at heating rates of 5, 10, and 15 °C/min are shown in Figure 2. Changes in temperature for mass conversion (R) (i.e., the ratio of the amount of a thermally decomposed sample to the amount of an initial sample) at the heating rate of 10 °C are presented in Figure 3. WLO itself started to loss its mass at a lower temperature 5314
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 13, 2010
than the PWM/WLO mixture samples, and the temperature range, from 367 to 442 °C, was broader than that of the PWM/ WLO samples. Previously reported temperature ranges of PS, PE, PP, PET and the PWM at the heating rate of 10 °C between 0.2 and 0.8 of mass conversions were 427-466 °C, 485-509 °C, 450-477 °C, 441-461 °C, and 439-486 °C, respectively (16). At each mass conversion, the temperature was higher in the order of PWM, 67% PWM/WLO, 50% PWM/ WLO, 33% PWM/WLO, and WLO, indicating that PWM/WLO samples containing a larger amount of the PWM were less thermally degraded at the same temperature. The WLO used in this study consisted of various components of paraffin and naphtha in broad carbon ranges of 13 and 42 (20, 21). Because of broad carbon number distribution, structures and covalent bonds of hydrocarbons in WLO have different bond dissociation energies, resulting in diverse reaction products. Therefore, thermal degradation of the WLO sample might be continued at broad temperature ranges that provide appropriate heat energy to break the various chemical bonds of the hydrocarbons (23). On the contrary, each plastic waste in the mixture samples of the PWM/WLO were originally prepared by the polymerization of each monomer; as a result, their molecular structures consist of repeating hydrocarbon functional groups, and they have typical bond cleavage mechanisms and relatively fast thermal decomposition properties at typical temperature ranges. Typical mechanisms of thermal decomposition of PE, PP, PS, and PET are shown in Figure 4. Bond breakages of PE, PP, and PS mainly occur at C-C bonds because bond dissociation energies of C-C bonds are typically weaker than C-H bonds (19, 23, 24). Aromatic rings in PS are relatively stable compared to other C-C bonds because of the π-orbital conjugation effect; hence, PS is mainly decomposed to a styrene monomer, dimer, trimmer, and aromatic derivatives (25). The typical thermal degradation mechanism of PET is that C-O bonds are broken, and its derivatives mainly form a carboxylic acid end group and a hydroxyl-ester end group (26). Thus, the thermal degradation properties of the PWM/WLO samples include both the typical bond dissociation mechanisms of an individual waste and some reactions caused by intermolecular interactions among wastes. A change in activation energies of WLO and PWM/WLO samples from 0.2 to 0.8 of mass conversions is presented in Figure 5(a). In this analysis, mass conversions below 0.2 (459 °C) were not considered for calculating kinetic parameters, because some evaporation of low molecular weight hydrocarbons in WLO at lower temperature (
