A New Hydrate-Based Recovery Process for Removing Chlorinated Hydrocarbons from Aqueous Solutions

Transcription

1 Environ. Sci. Technol. 2001, 35, A New Hydrate-Based Recovery Process for Removing Chlorinated Hydrocarbons from Aqueous Solutions YONGWON SEO AND HUEN LEE* Department of Chemical Engineering, Korea Advanced Institute of Science and Technology, Kusong-dong, Yusong-gu, Taejon , Korea The main objective of this study was to check the feasibility of the newly proposed hydrate-based chlorinated hydrocarbon (CHC) recovery process for removing chlorinated hydrocarbons from aqueous solutions. Two key process variables of hydrate phase equilibria and formation kinetics were closely examined to develop the overall conceptual design of this technology. First, the ternary four-phase (H-L W -L CHC -V) hydrate equilibria of aqueous solutions containing methylene chloride (CH 2 Cl 2 ), carbon tetrachloride (CCl 4 ), 1,2-dichloroethane (CH 2 ClCH 2 Cl), 1,1,1- trichloroethane (CH 3 CCl 3 ), and 1,1-dichloroethylene (CH 2 d CCl 2 ) were measured at various temperature and pressure conditions using three different types of help gases (CO 2,N 2,CH 4 ). The help gas + water + chlorinated hydrocarbons systems greatly reduced the hydrateforming pressure, which confirmed the mixed hydrates with chlorinated hydrocarbons more stabilized than the simple hydrates consisting of a help gas and water. The degree of stabilization was found to follow the order of 1,2- dichloroethane < 1,1-dichloroethylene < methylene chloride < 1,1,1-trichloroethane < carbon tetrachloride. For the N 2 + water + carbon tetrachloride system, the formation pressure reduction as much as 96% was observed at K. Second, the formation kinetic experiments of carbon dioxide hydrates containing chlorinated hydrocarbons were conducted under isothermal and isobaric conditions. The consumption rate of carbon dioxide gas became fast at the early time of the growth period, gradually decreased, and finally went to the complete hydration. The proposed hydrate-based recovery process appears to be very simple from the operational point of view because no special facilities requiring sensitive and complex function are needed. Another advantage is that this process only requires carbon dioxide as a hydrate former. Best of all, this process can be applied to separation and recovery of other organic pollutants dissolved in aqueous solutions without changing the basic concept. Introduction Gas hydrates are nonstoichiometric crystalline compounds formed when guest molecules of suitable size and shape are incorporated in the well-defined cages in the host lattice made up of hydrogen-bonded water molecules. These * Corresponding author phone: (82) ; fax: (82) ; hlee@mail.kaist.ac.kr. compounds exist in three distinct structures termed as structure I, II, and H. Among the common substances, CO 2 and CH 4 have been known to form structure I hydrate while N 2 forms structure II hydrate. Industrial interest was placed on the avoidance of solid hydrates formed in pipelines and process equipments. In particular, many researches have been mainly focused on understanding the fundamental mechanism of hydrate formation behavior. Naturally occurring gas hydrates in the earth consisting mostly of methane are regarded as a future energy resource, since each volume of hydrate can contain as much as 184 volumes of gas (STP) (1). Another major application of gas hydrates lies with the recent plan for sequestering liquid carbon dioxide in the ocean (2). A new hydrate-based gas separation (HBGS) process was developed for recovering carbon dioxide from flue gas (3, 4). Best of all, this process can be applied to separation and recovery of other greenhouse gases such as methane and Freons without changing the basic concept. However, general reviews on gas hydrates can be found from Sloan (1), Holder et al. (5), and Englezos (6). The volatile organic compounds (VOCs) are among the most common pollutants emitted by the chemical process industries. Chlorinated hydrocarbons such as methylene chloride (CH 2Cl 2), carbon tetrachloride (CCl 4), 1,2-dichloroethane (CH 2ClCH 2Cl), 1,1,1-trichloroethane (CH 3CCl 3), and 1,1-dichloroethylene (CH 2dCCl 2) are designated as VOCs. They are the common contaminants in many aquifers. Several methods have been suggested for removing chlorinated hydrocarbons such as thermal oxidation, biofiltration, and membrane separation (7). Recently, experimental works on the formation of solid hydrates by certain chlorinated hydrocarbon compounds and water were attempted with help gases such as H 2S and Xe for the purpose of immobilizing or recovering these contaminants from groundwater (8). The separation efficiency of CCl 4 in the presence of Xe as a help gas was found to be as high as 94%. However, these two help gases are not suitable for the recovery of chlorinated hydrocarbons because of their high cost and toxicity. The overall results of this study only provided a proof-of-concept suggesting that the in situ immobilization or ex situ recovery of some common hydrophobic groundwater contaminants may be possible in a wide range of subsurface environments. In the present work, the hydrate-based recovery process using the nontoxic and more common help gases (CO 2,N 2, and CH 4) was developed for removing chlorinated hydrocarbons from the aqueous solutions. To check thermodynamic feasibility of this process the pressure and temperature ranges of hydrate stability region were carefully determined through measurements of four phase equilibria (hydrate (H), liquid water (L W), chlorinated hydrocarbon (L CHC), and vapor (V)) for the help gases + water + chlorinated In addition, the kinetic experiments of hydrate formation were performed to obtain the required time taken for complete conversion to hydrate. These two key pieces of information, one thermodynamic phase equilibria and the other the kinetic behavior, play the fundamental role in designing the conceptual process for removing the chlorinated hydrocarbons. Experimental Section Apparatus. A schematic diagram and detailed description of the experimental apparatus for hydrate phase equilibria is given in the previous paper (4). The apparatus was constructed to measure the hydrate dissociation pressures through the visual observation of phase transitions. The equilibrium cell was made of 316 stainless steel and had an ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 16, /es010528j CCC: $ American Chemical Society Published on Web 07/18/2001

2 FIGURE 1. Experimental apparatus for hydrate formation kinetics. internal volume of about 50 cm 3. Two sapphire windows equipped at the front and back of the cell allowed the visual observation of phase transitions occurred inside the equilibrium cell. The cell content was vigorously agitated by a magnetic spin bar with an external magnet immersed in a water bath. The kinetic measurements of hydrate formation were carried out by using the experimental apparatus schematically shown in Figure 1. The primary function of this specifically designed apparatus was to determine the volumetric rate of a hydrate-forming gas consumed during hydrate formation that might be the most important parameter in the kinetic study. The hydrate-forming reactor with an internal volume of about 140 cm 3 was made of a type 316 stainless steel and had the inner part of a cylindrical shape. It had two sight glasses equipped at the bottom and one side of the reactor that allowed a visual observation of hydrate-forming features occurring in the reactor. The magnetically driven mechanical stirrer was used for uniform mixing during kinetic measurement. The stirring rate was stably controlled within the range of rpm using the built-in frequency generator. The kinetic reactor was kept in the water-ethanol mixture bath, and its temperature was controlled by an externally circulating refrigerator/heater (JEIO TECH, RBC-20). The actual temperature in the reactor was measured by the K-type thermocouple with a digital temperature readout (Cole- Parmer, ) having a resolution of (0.1 K. A Heise gauge (CMM 44307) ranged bar was used to measure the reactor pressure with the maximum error of (0.2 bar. The reactor was initially pressurized with gaseous carbon dioxide supplied from a gas cylinder and then was maintained at a constant pressure condition by using a micro-flow syringe pump (ISCO, Model 260D) operated in the constant pressure mode. As the system pressure decreased due to gas consumption during hydrate formation, the syringe pump with an operating condition of constant pressure supplemented the gas to the reactor to maintain the system pressure constant. The mass flowmeter (BROOKS, Model 5850E) allowed for measuring accurately the flow rate of gases and finally provided the total amount of a hydrate-forming gas consumed during hydrate formation. A PC-LabCard PCL- 711B data acquisition system and a PC486 computer were used to obtain the flow-rate signals from the mass flowmeter. Reagents. The carbon dioxide and nitrogen gases used for the present study were supplied by World Gas Co. and had a stated purity of 99.9 mol %. The methane gas with a minimum purity of mol % was supplied by Linde Gas UK Ltd. The water (ultrahigh purity) was supplied from Merck Co. Methylene chloride, carbon tetrachloride, 1,2-dichloroethane, 1,1,1-trichloroethane, and 1,1-dichloroethylene with a purity of 99.9 mol % were purchased from Aldrich Co. All materials were used without further purification. Procedure. The experiment for hydrate-phase equilibrium measurements began by charging the equilibrium cell with about 20 cm 3 of the aqueous solutions containing chlorinated hydrocarbons. After the cell was pressurized to a desired pressure with the help gas, the system was cooled to about 5 K below the expected hydrate-forming temperature. Once the system temperature became constant, hydrate nucleation was induced by agitating the magnetic spin bar with the immersed magnet in the water bath. When hydrate formed and the system pressure reached a steady-state condition, the cell temperature was increased at a rate of about 1 K/h until the hydrate phase was in coexistence with the liquid and vapor phases. The system temperature was then slowly raised at a rate of 0.1 K/h. The nucleation and dissociation steps were repeated at least two times in order to reduce the hysteresis phenomenon. When a very small amount of crystals existed by visual observation without significantly increasing or decreasing its size, the system temperature was kept constant at least for 8 h after stabilizing the system pressure. Then the pressure was considered as an equilibrium hydrate dissociation pressure at the specified temperature. The formation kinetic experiments of the carbon dioxide hydrate containing chlorinated hydrocarbons were conducted under isothermal and isobaric conditions. The experimental temperatures and pressures were selected in the region where the three phases (H-V-L CHC) coexist and no condensation of carbon dioxide occurs. The reactor was initially evacuated and then charged with 50 cm 3 of the aqueous solution. The reactor was pressurized to the experimental pressure (20 bar), and the temperature was set to a value slightly higher than the four-phase (H-L W-L CHC-V) equilibrium temperature. The solution was agitated by the mechanical stirrer for nearly 3-4 h so that the aqueous solution was saturated with carbon dioxide gas. When no pressure change was detected at the fully saturated state of a solution, the stirrer was stopped, and the temperature was lowered to the desired one. Once the specified pressure and temperature conditions were attained, the agitation was restarted. The syringe pump was then set to a constant pressure mode and started its self-operating pumping in order to maintain the system pressure constant. At the same time, the data acquisition system scanned the flow-rate signals detected by the mass flowmeter. The carbon dioxide gas consumed during hydrate-formation process was instantaneously supplemented to the reactor by a syringe pump. The flow rate signals from the mass flowmeter were obtained at intervals of 0.5 s by the online data acquisition system. The experiment proceeded through the induction, nucleation, and growth periods, successively, and each period was confirmed carefully by visual observation. When the mass flowmeter decreased to the zero point after a sufficient time passed, the formation kinetic experiment was completed by stopping stirring the reactor. The formation reaction proceeded for about 180 min after the nucleation began. Results and Discussion Chlorinated hydrocarbons are generally immiscible with water, and therefore two liquid phases form when two components are mixed together. In cases that the concentration of chlorinated hydrocarbon exceeds the saturation in the water-rich phase, the resulting equilibrium temperature and pressure must be necessarily the same. First of all, seven different types of chlorinated hydrocarbons were tested with/ without using three help gases of CO 2,N 2, and CH 4, and the overall results were summarized in Table 1. However, trichloroethylene (CHCldCCl 2) and tetrachloroethylene (CCl 2d VOL. 35, NO. 16, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

3 TABLE 1. Hydrate Formation of Chlorinated Hydrocarbons CO 2 N 2 CH 4 CH 2Cl 2 (methylene chloride) a O b O b O b CCl 4 (carbon tetrachloride) O b O b O b CH 2ClCH 2Cl (1,2-dichloroethane) X c O b O b CH 3CCl 3 (1,1,1-trichloroethane) O b O b O b CH 2dCCl 2 (1,1-dichloroethylene) O b O b O b CHCldCCl 2 (trichloroethylene) X c X c X c CCl 2)CCl 2 (tetrachloroethylene) X c X c X c a Hydrate formation without help gas. b O: hydrate formation with help gas. c X: no hydrate formation with help gas. FIGURE 3. Hydrate phase equilibria for the N 2 + H 2O + chlorinated FIGURE 2. Hydrate phase equilibria for the CO 2 + H 2O + chlorinated CCl 2) failed to form hydrates for all cases. Therefore, methylene chloride (CH 2Cl 2), carbon tetrachloride (CCl 4), 1,2- dichloroethane (CH 2ClCH 2Cl), 1,1,1-trichloroethane (CH 3- CCl 3), and 1,1-dichloroethylene (CH 2dCCl 2) were chosen in this study. Among these chlorinated hydrocarbons methylene chloride was found to be a unique substance for forming hydrates with/without a help gas. Four-phase equilibria (H-L W-L CHC-V) for the CO 2 + water + chlorinated hydrocarbons, N 2 + water + chlorinated hydrocarbons, and CH 4 + water + chlorinated hydrocarbons systems were measured at the fixed concentration of 3 mol %, and the overall results were shown in Figures 2-4. Four chlorinated hydrocarbons successfully formed hydrates with water and CO 2. However, 1,2-dichloroethane did not form hydrate in the presence of CO 2 as a help gas. At any given pressure, the stabilizing region of hydrates formed by chlorinated hydrocarbons was extended to higher temperature than the corresponding simple hydrate formed only by water and CO 2. The hydrates composed of CO 2 + water + chlorinated hydrocarbons become structure II. Carbon tetrachloride and other large molecules cannot form hydrates on their own and require FIGURE 4. Hydrate phase equilibria for the CH 4 + H 2O + chlorinated the help gas molecules to occupy the smaller cavities resulting in stabilizing the structure (9). The upper end point of four phase (H-L W-L CHC-V) curve for the CO 2 + water + chlorinated hydrocarbons systems becomes the quintuple point at which five phases (H-L W-L CHC-L CO2-V) coexist. The phase behavior indicates that the five different four-phase curves merge at this quintuple point. To confirm the exact location of the upper quintuple point, we measured additionally another ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 16, 2001

4 FIGURE 5. Formation kinetics of carbon dioxide hydrates containing chlorinated hydrocarbons at 20 bar and K. four-phase (L W-L CHC-L CO2-V) boundary. Two different fourphase equilibrium curves were intersected at the corresponding quintuple point without any noticeable difference. From the comparison of Figures 2 and 3, N 2 appeared to be stronger than CO 2 for stabilizing chlorinated hydrocarbon hydrates even though the general trend for both help gases are the same. The degree of stabilization was found to follow the order of 1,2-dichloroethane < 1,1-dichloroethylene < methylene chloride < 1,1,1-trichloroethane < carbon tetrachloride. The N 2 + water + chlorinated hydrocarbons systems greatly reduced the hydrate-forming pressures, which makes the formed hydrates more stable than the N 2 + water system. For example, in the case of the N 2 + water + carbon tetrachloride system the pressure reduction as much as 96% was observed at K. The stabilization phenomenon of the CH 4 + water + chlorinated hydrocarbon systems shown in Figure 4 was the same as the ternary systems using CO 2 and N 2 as guest molecules. From the experimental results stated above it must be, however, noted that the equilibrium dissociation temperatures of the CO 2 + water + chlorinated hydrocarbons system increase to a certain temperature with increasing pressure, but a further increase of pressure lowered the corresponding dissociation temperature as compared with the retrograde condensation behavior which generally appeared in the high-pressure vapor-liquid equilibria. This peculiar turnover of equilibrium dissociation temperature was not observed in the N 2 + water + chlorinated hydrocarbons and CH 4 + water + chlorinated hydrocarbons systems possessing no upper quintuple point. To examine the applicability of hydrate-based separation process for the removal of chlorinated hydrocarbons from aqueous solutions the hydrate formation kinetics of the ternary CO 2 + water + chlorinated hydrocarbons systems was also determined at K and 20 bar as the first attempt, and the results are presented in Figure 5. The accumulated moles of carbon dioxide consumed during hydrate formation were based on 1 mole of water and plotted against time. The formation rate was presented in the figure directly after the turbidity point occurred. No stirring effects were observed over 400 rpm, and thus, all the experiments were performed at 600 rpm. The consumption rate of carbon dioxide gas became fast at the early time of the growth period, gradually decreased, and finally went to the complete hydration. An abrupt increase of the consumption rate at the initial stage might be considered due to the direct hydration of guest FIGURE 6. Formation kinetics of carbon dioxide hydrates containing carbon tetrachloride at 20 bar and and K. FIGURE 7. Schematic diagram of the hydrate-based chlorinated hydrocarbon recovery process. molecules with host water molecules. After the short period of time the increment of carbon dioxide consumption rate became low and finally almost flat because carbon dioxide molecules were only consumed during further reaction with the remaining free water. In addition the formation kinetics of the CO 2 + water + CCl 4 system were measured at 20 bar and two different temperatures of and K, and the results were presented in Figure 6. As expected, the consumption rate was more favorable at the low-temperature condition because of its positive effect on hydrate formation. One hour was found to be enough to attain complete hydration. The conceptual diagram of the hydrate-based chlorinated hydrocarbon recovery process is schematically demonstrated in Figure 7. The proposed hydrate-based recovery process seems to be very simple from the operational point of view because no special facilities requiring sensitive and complex function are needed. First, a help gas is injected into a hydration vessel containing aqueous solution of chlorinated hydrocarbons. If temperature and pressure conditions are suitable for forming hydrates, solid hydrates containing chlorinated hydrocarbons settle to the bottom of the hydration vessel. The precipitated solid hydrates can be readily separated from the coexisting liquid phase. By following this procedure a large amount of chlorinated hydrocarbons can be removed from aqueous solutions, and the dissociated help gas can be recycled to the hydrate formation unit. The volumes of hydrator and dissociator can be approximated from the preliminary kinetic results of this study. Among three help gases (CO 2,N 2,CH 4) tested in this study, N 2 gave a little higher equilibrium pressure at a given temperature than other two help gases. Although CH 4 + water + chlorinated hydrocarbons system provided favorable VOL. 35, NO. 16, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

5 equilibrium condition of low pressure and high temperature, CH 4 is expensive and flammable. Accordingly, CO 2 might be the best potential help gas for recovering chlorinated hydrocarbons from aqueous solutions. Moreover, a pressure of only about 10 bar is required for forming CO 2 hydrates containing chlorinated hydrocarbons in the temperature range of K. Several process advantages can be ascribed to this hydrate-based recovery technology. The operation temperature for hydration is low, in the range of K. The required heat for dissociation can be provided from ambient surroundings. A continuous operation permits this process to treat a large volume of aqueous solution and to compete with the commonly used recovery processes. The dissociated CO 2 can be easily recycled to the hydrator for further use. Best of all, this process can be applied to separation and recovery of other organic pollutants dissolved in aqueous solutions without changing the basic concept. Acknowledgments This research was performed for the Greenhouse Gas Research Center, one of the Critical Technology-21 Programs, funded by the Ministry of Science and Technology of Korea and also partially supported by the Brain Korea 21 Project. Literature Cited (1) Sloan, E. D. Clathrate Hydrates of Natural Gas, 2nd edition, revised and expanded; Dekker: New York, (2) Teng, H.; Yamasaki, A.; Chun, M. K.; Lee, H. Energy 1997, 22, (3) Yoon, J.-H.; Lee, H. AIChE J. 1997, 43, (4) Kang, S.-P.; Lee, H. Environ. Sci. Technol. 2000, 34, (5) Holder, G. D.; Zetts, P.; Pradhan, N. Rev. Chem. Eng. 1988, 5, (6) Englezos, P. Ind. Eng. Chem. Res. 1993, 32, (7) Moretti, E. C.; Mukhopadhyay, N. Chem. Eng. Prog. 1993, 89, (8) Bontha, J. R.; Kaplan, D. I. Environ. Sci. Technol. 1999, 33, (9) Jeffrey, G. A.; McMullan, R. K. Prog. Inorg. Chem. 1967, 8, Received for review January 11, Revised manuscript received May 24, Accepted May 31, ES010528J ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 16, 2001