Effect of pH on Carbon Dioxide Hydrate Formation in Mixed Soil

Jul 6, 2009 - complexes on soil minerals can make the hydrate gas former stable by .... peak temperature, while neither ice nor CO2 hydrate was observ...
0 downloads 0 Views 1MB Size
Environ. Sci. Technol. 2009, 43, 5908–5914

Effect of pH on Carbon Dioxide Hydrate Formation in Mixed Soil Mineral Suspensions RHEO B. LAMORENA AND WOOJIN LEE* Department of Civil and Environmental Engineering, Korea Advanced Institute of Science and Technology, 335 Gwahangno, Yuseong-Gu, Daejon 305-701, Korea

Received April 15, 2009. Revised manuscript received June 17, 2009. Accepted June 19, 2009.

We investigated the effect of pH on CO2 hydrate formation in the presence of phyllosilicate mixtures. Different pH conditions of phyllosilicate suspensions (Na-montmorillonite-rich and phyllosilicate-rich suspensions) with and without NaCl (3.5%) were prepared and controlled by the addition of an acid or base before the dissolution of CO2. The formation of CO2 hydrates was observed in all phyllosilicate suspensions (30 bar and 273.45 K). The temperature-time plot results showed that hydrate formations were suppressed more in acidic mineral suspensions than in basic suspensions. The fastest hydrate induction time can be observed in Na-montmorillonite-rich and phyllosilicaterich suspensions with and without NaCl at near neutral conditions (pH 6-8), followed by basic (∼pH 12.0) and acidic (∼pH 2.0) pHs. Hydrate induction time can be significantly affected by various chemical species forming under different suspension pHs. The distribution of chemical species in each mineral suspension was estimated by a chemical equilibrium model, PHREEQC, and used for the identification of hydrate formation characteristics in the suspension. Particle-particle and particle-water interactions may possibly contribute to the delay of hydrate formation. NaCl was not an efficient inhibitor but a possible promoter for hydrate formation when pHdependent solid surfaces were present in the system.

Introduction Extensive studies are currently being conducted to develop CO2 sequestration technologies in the ocean using seawater (1), sea bed sediment (2), geological formations (3), and above ground minerals (4). One of the sequestration technologies developed to date is CO2 injection into cold geologic formations where the injected CO2 can form clathrate hydrates in the pores of host rock (5, 6). Clathrate hydrates are a special type of crystalline inclusion compound that cage suitable guest molecules within hydrogen-bonded networks of water molecules under high-pressure and lowtemperature conditions (7, 8). Clathrate hydrate formation is a natural phenomenon that can provide an attractive idea for capturing and storing greenhouse gases in a long-term stable geological formation. However, the mechanism of hydrate formation has not been fully identified. It has been widely accepted that hydrate formations are potentially influenced by temperature, pressure, salinity, gas compositions, and interfacial surface area. The interfacial * Corresponding author phone: +82-42-350-3624; fax: +82-42350-3610; e-mail: [email protected]. 5908

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 15, 2009

surface area enhances the interaction of water and gas hydrate former (7-11). The role of interfacial surface area on hydrate formation was investigated in clay minerals (hydrous aluminum phyllosilicates)-gas hydrate former-water systems. Faster hydrate formation kinetics were found in the systems with clay minerals (11-15). It has been reported that clay mineral surfaces can accelerate hydrate formation by providing nucleation sites and by shifting the pressure and temperature in the system to be more thermodynamically favorable for the stability of the gas hydrate former. Surface complexes on soil minerals can make the hydrate gas former stable by filling void space of the clay bonding network. It provides nucleation sites that result in a rigid and stable clathrate cage or cluster structure (15, 16). Changes in the soil mineral structure and chemical species bound on the interfacial surface may affect hydrate formation kinetics. The distribution of chemical species on soil mineral surfaces in geological formations can be significantly affected by the surrounding pH (∼pH 4-10). The geochemical behavior of sequestered CO2 in geological formations is considerably responsive to changes in pH and salinity. Acid and alkali attacks cause remarkable changes in soil mineral structures due to the dissolution of structural ions and/or rearrangement of the structures (17-19). To understand the potential effect of mineral solubility at different pHs on hydrate formation, it is important to identify the dissolution of soil minerals in acidic and basic conditions. The structural arrangements of phyllosilicates are based on the stacking of tetrahedral sheets of silica and octahedral sheets of alumina (T-O) with ions in interlayers (17). As the ions are dissolved into the bulk solution, the chemical composition, particle association, and structure of the minerals are altered. The hydrate induction time could be significantly influenced by the surrounding pHs of geological formations because the soil mineral structure and its surface chemical species play pivotal roles in the hydrate formation template, and nucleation sites can change under different pH conditions. No significant studies have been conducted to identify the effect of pH changes on the formation of CO2 hydrates in any hydrate formation investigations. We have conducted an experimental study to investigate the effects of pH, electrolyte, and soil mineral mixtures on hydrate induction time and to identify the hydrate formation phenomena observed in an experimental system with various mineral mixtures. Two types of mineral mixture (Namontmorillonite-rich and phyllosilicate-rich) suspensions were prepared with and without an electrolyte (NaCl) at three different pH ranges (acidic, near neutral, and basic). Equilibrium modeling was used to determine the possible chemical species that were being formed under the various conditions.

Experimental Section Chemicals. The mixed soil minerals used in this research have been characterized and are described in the Supporting Information. The phyllosilicate-rich mixture (Pusan, Korea) is composed of several minerals: quartz (27.4%), phyllosilicates [39.2% composed of kaolinite (7.6%), K-mica (12.6), Illite (8.1), Ca-montmorillonite (6.5), chlorite (4.4), alkalifeldspars (9.0%), pyrite (3.3%), plagioclase (18.2%), and calcite (2.9%)]. The Na-montmorillonite-rich mixture (Kyoungju, Korea) is composed of Na-montmorillonite (52.1%), quartz (23.1%), anorthite (19.6%), and kaolinite (5.3%). Details of the X-ray diffraction (XRD) analyses are also described in the Supporting Information. No soil mineral pretreatment with 10.1021/es901066h CCC: $40.75

 2009 American Chemical Society

Published on Web 07/06/2009

acid or reductant was conducted to control the redox state of soil minerals in this study. All soil minerals were screened with sieves. Particles smaller than 250 µm were collected and dried at room temperature (25 ( 1.0 °C). The gas hydrate former used in the experiment was a commercial 99.9% grade compressed CO2 (Sam-O Gas Co., Korea). Zero air (99.999%, Sam-O Gas Co.) was used for control tests. NaCl (99.5%, Junsei, Japan) and ethylene glycol (99.5%, Sam-Chun, Korea) used in the research were reagent grade. Soil minerals and chemicals were used without further purification. An exact amount of electrolyte (NaCl) was added to deionized water (18 MΩ cm) to gravimetrically prepare 3.5% electrolyte solution. Experimental Setup. Hydrate formation experiments were conducted using a simple experimental system that has been previously described in detail (11). A cylindrical 304 stainless steel pressurized vessel (50 cm3) was used as a reactor. The vessel has two tempered sight glasses on opposite sides to allow visual observation during the hydrate formation experiments. Temperature and pressure sensors were connected to a data acquisition unit (Agilent 34970A) with a response time of 20 s. Sample Preparation. Dissolution experiments were carried out in batch reactors at three different pHs (∼2.0, 6-8, and 12.0) and constant room temperature. The suspension pH was monitored until it reached equilibrium. Extremely acidic and basic pH conditions were chosen to alter soil mineral structure by dissolution (17-19). The pH meter was calibrated using standard buffer solutions at room temperature. Measurements of pH were conducted in mixed soil mineral suspensions using a 9272 Orion Ross combination electrode. Samples were prepared in duplicate, and their measurements were accurate within the error range of 0.01 unit. All solutions were made with deionized water. Exact amounts of soil minerals (0.2 g) were added to the deionized water or to a 3.5% NaCl solution, resulting in 30 mL suspensions with a mass solid/liquid ratio of 1:150. Exact amounts (60-1200 uL) of 2 M NaOH and 2 M HCl were added to soil mineral suspensions without NaCl, while 60-400 uL of 10 M NaOH and 10 M HCl were added to soil mineral suspensions with NaCl to obtain the desired equilibrium suspension pHs. The reactors were sealed with Parafilm (ParafilmM, USA) and stirred to enhance the dissolution rate of the soil minerals into the bulk suspensions. As a result, the variations of pH in the suspensions was due mainly to the solid-solution interactions. The suspension was continuously mixed until an equilibrium pH was observed, and its pH was kept constant until it was loaded into the pressurized vessel. The time to reach equilibrium pH was different depending on the mixed soil mineral type and mostly exceeded more than 100 h. Control tests were performed in aqueous solutions with and without NaCl by following the experimental procedures above. To simulate the natural conditions where clathrate hydrates form, we conducted experiments in an isothermal and isobaric environment. The pressurized vessel was filled with 25-30 mL of solid suspension with and without 3.5% NaCl leaving 20-25 mL of headspace. It was then placed into a temperature controlled bath at a constant temperature of 273.45 K. Most researchers conduct hydrate formation experiments by under cooling and/or subcooling to initiate hydrate formation easily; however, a cooling scheme was not applied here (20, 21). The experiments were conducted in a static condition to avoid the effects of continuous mixing on the hydrate formation by a higher gas uptake rate at the water-to-gas interface and local vortices. The vessel was purged with CO2 to remove air from the headspace and dissolved air from the suspension. It was then was pressurized to 30 bar by introducing CO2 into the soil mineral suspension

until the formation of CO2 hydrates. CO2 was continuously supplied to the vessel to maintain a constant pressure. The changes in temperature and pressure in the vessel were monitored during the experiments. The CO2 hydrate was allowed to fully develop in the vessel, and its formation procedure was visually monitored through the sight glasses. The hydrate formation experiments were also carried out in deionized water and NaCl solutions. The experiments were conducted in duplicate by renewing the sample preparation procedure at each experimental run to avoid memory effects. Duplicate control tests were also performed with zero air in the soil mineral suspensions with and without NaCl by following the experimental procedures of hydrate formation above. Chemical Equilibrium Model. PHREEQC (22) with a default database file (PHREEQC.DAT) was used to determine the activity of water and dominant dissolved chemical species in the suspensions with and without NaCl. A set of thermodynamic constants for soil minerals in the suspensions (i.e., quartz, anorthite, kaolinite, K-mica, illite, chlorite, Camontmorillonite, alkali-feldspar (orthoclase), pyrite, plagioclase (albite), and calcite) were chosen from the default database, while those for Na-montmorillonite were obtained from the literature (23). Equilibrium pH, room temperature, ambient pressure (1 atm), type of acid/base, and the maximum concentration (10 mols) from the experiments were used for input data of the equilibrium model as initial and boundary conditions. The suspension pH was fixed by using the equilibrium pH. The initial solution (i.e., deionized water or NaCl solution) was allowed to reach equilibrium with the soil minerals at target pHs. PHREEQC calculated the concentrations of chemical species at equilibrium using the Newton-Raphson method to solve mass balance and mass action relations with the initial and boundary conditions. We used the default ion association and Debye-Huckel equations for all batch reaction simulations.

Results and Discussion Temperature Changes during Hydrate Formation. Temperature changes in control samples (water only and water + zero air) and CO2 hydrate samples (water + CO2 and Namontmorillonite-rich suspension + CO2) during the hydrate formation experiments are shown in Figure 1. The temperature-time profile shows a sharp increase as the hydrate forms. It decreased as the hydrate growth was observed through the tempered sight glasses. The snapshots showing the hydrate formation in each sample are also shown in Figure 1. The sharp temperature peak in the temperature-time plot was used as an indicator of hydrate formation in each sample. Ts and Tp are the starting and peak temperatures of the peak in the temperature-time plot, respectively. Water samples with the CO2 addition at 273.45 K and 30 bar showed a huge temperature peak, while those without CO2 did not show any observable temperature change. CO2 hydrate formation was remarkable in the water sample with CO2 at the time of peak temperature, while neither ice nor CO2 hydrate was observed in the sample without CO2 injection under the same experimental conditions. Hydrate formation was also not observed when zero air was injected instead of CO2 gas at the same experimental conditions. The Na-montmorilloniterich suspension with CO2 showed an earlier abrupt temperature change when the hydrate formed at the time of peak temperature. The peak temperatures observed in this experiment were in the range of 1.4-2.7 °C. After the peak, the temperature decreased by an equivalent amount to the height of the temperature peak (Tp - Ts), indicating the result of heat transfer out of the pressure vessel. A hydrate induction time (hydrate formation period) has been determined by measuring the running time from the introduction of CO2 to the occurrence of the temperature VOL. 43, NO. 15, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5909

FIGURE 1. Temperature responses plotted over time. The start of hydrate formation is detected by a sharp increase in temperature. Starting temperature, Ts, and peak temperature, Tp, are indicated. Photographic images of hydrates in control samples are shown in the insets: water only, water + CO2, and Na-montmorillonite-rich solid suspension + CO2. peak when a thin-layered CO2 hydrate (