Effects of Pressure on the Recovery of CO2 by Phase Transition from a

Dec 29, 2009 - Luzerne County Community College. This article is ... C2-C5+ olefin production from CO2 hydrogenation using ceria modified Fe/Mn/K cata...
1 downloads 0 Views 406KB Size
Download PDF
J. Phys. Chem. A 2010, 114, 4003–4008

4003

Effects of Pressure on the Recovery of CO2 by Phase Transition from a Seawater System by Means of Multilayer Gas Permeable Membranes† Heather D. Willauer,*,‡ Dennis R. Hardy,‡ M. Kathleen Lewis,§,| Ejiogu C. Ndubizu,‡ and Frederick W. Williams‡ NaVal Research Laboratory, Chemistry DiVision, Code 6180, 4555 OVerlook AVenue, SW, Washington, DC 20375, Office of NaVal Research, 875 North Randolph Street Suite 260, Arlington, Virginia 22203, and Luzerne County Community College, 1333 South Prospect Street, Nanticoke, PennsylVania 18634 ReceiVed: NoVember 6, 2009; ReVised Manuscript ReceiVed: December 1, 2009

Using seawater doped with sodium bicarbonate and Celgard 2400 gas permeable membranes, bicarbonate ion disproportionates to carbon dioxide and carbonate when gaseous carbon dioxide is first removed from the seawater solution by diffusion through gas permeable membranes at elevated water pressures. The permeability of CO2 by phase transition from bicarbonate solutions at pressures above 100 psi is only possible due to the use of multiple gas permeable membrane layers. The multiple layers minimize water permeability at pressures below and above the Young-Laplace bubble point of single membrane layers, however the gas permeability efficiency and rate are greatly decreased. 1. Introduction The total carbon content of the world’s oceans is roughly 38 000 GtC (gigaton of carbon). Over 95% of this carbon is in the form of dissolved bicarbonate ion (HCO3-).1 This ion along with carbonate ion (CO32-) is responsible for buffering and maintaining the ocean’s pH, which is relatively constant below the first 100 m. These dissolved bicarbonate and carbonate are essentially bound CO2, and the sum of these species along with gaseous CO2, shown in eq 1, represents the total carbon dioxide concentration [CO2]T, of seawater.

∑ [CO2]T ) [CO2(aq)] + [HCO3-(aq)] + [CO32-(aq)]

(1)

At a typical ocean pH of 7.8, [CO2]T is about 2000 µmol/kg near the surface, and 2400 µmol/kg at all depths below 300 m.2,3 This equates to approximately 100 mg/L of [CO2]T of which 2-3% is [CO2 (aq)] (eq 1), 1% is carbonate, and the remainder is dissolved bicarbonate. The only available carbon source in the atmosphere is CO2 gas. Its concentration is approximately 370 ppm (v/v),which is 0.7 mg/L (w/v) or 780 GtC. Comparing this value on a w/v basis to that found in the ocean (100 mg/L), it is readily apparent that CO2 in seawater is about 140 times greater than air.4 Thus if carbon dioxide could be economically and efficiently extracted from the ocean, then marine engineering processes such as ocean thermal energy conversion (OTEC)5,6 could be proposed to utilize this carbon as a chemical feedstock in processes such as catalytic polymerization with hydrogen. The difficulty is that seawater is a very complex buffered system in equilibrium with the atmosphere above it. Thus under equilib† Part of the special issue “Green Chemistry in Energy Production Symposium”. * To whom correspondence should be addressed. E-mail: Heather. [email protected]. ‡ Naval Research Laboratory. § Office of Naval Research. | Luzerne County Community College.

10.1021/jp910603f

rium conditions, dissolved [CO2(aq)] in eq 1 is actually hydrated with one mole of water in the form known as carbonic acid (eq 2). The dissolved bicarbonate and carbonate are in equilibrium with this dissolved carbonic acid species as shown in eq 3 H2O

H2O

[CO2(g)]air {\} [CO2(aq)] {\} [H2CO3(aq)]

H2O

(2)

H2O

CO2(g) {\} CO2(aq) {\} H2CO3(aq) a 2HCO3-(aq) a CO32-(aq) + H2O(l) + CO2(g)v

(3)

The carbonic acid species is in equilibrium with the gas phase above the water as carbon dioxide, and so any reduction of carbon dioxide partial pressure in the gas phase will cause the equilibria of the entire system to shift to the left (eq 3).7,8 Kinetically, this is a very slow and not well-understood process.9 From an ocean engineering perspective, the rate of carbon recovery by vacuum extraction or vacuum degassing of the 2-3% of the dissolved CO2(aq) would be entirely too slow on the order of 10 kg/sec to obtain useful amounts of carbon. In addition to the complex equilibrium buffer system affecting CO2 recovery from seawater, there is a potential for the excessive salinity of seawater to interfere with developing fast and efficient approaches to CO2 capture. The salinity is on the order of 35 g/L and is attributed primarily to sodium chloride. The chloride content is approximately 240 times more concentrated than that of bicarbonate.7 In efforts to overcome these challenges, Willauer et al. and Hardy et al. investigated the use of simple ion-exchange resin systems as direct and indirect methods of CO2 recovery from seawater.10,11 Strong base anion exchange resins’ selectivity and capacity was proven inefficient to acquire the 10 kg/sec of carbon needed for a catalytic process.10 Strong acid cation exchange resins were successful at acidifying seawater to a pH less than 6 so the total CO2 existed in the dissolved gas form.11,12 However that approach was

This article not subject to U.S. Copyright. Published 2010 by the American Chemical Society Published on Web 12/29/2009

4004

J. Phys. Chem. A, Vol. 114, No. 11, 2010

Willauer et al.

deemed impractical from a regeneration perspective.11 Thus as an alternative avenue of exploration to increase carbon yield from the sea, gas permeable membranes are being investigated.13,14 Gas permeable membranes are available commercially for the removal or addition of gases from liquids. Most of these applications are near atmospheric pressure and include water purification, blood oxygenation, and artificial lung devices.15–18 However some are operated at higher pressures such as beverage carbonation.19,20 It is well-known that these membranes work on the principle of dissolved gases such as carbon dioxide diffusing across the membrane through the pores as a function of differential partial gas pressures. Therefore it has been assumed until recently for gas/ liquid systems that only the dissolved carbon dioxide gas is removed while the bound carbon dioxide in the ionic form of bicarbonate and carbonate is not involved.21 The rate and yield of carbon dioxide for a given separation is dependent on several parameters which include the degasification matrix, membrane material and geometry, temperature, and pressure.22 Using simple model systems of aqueous sodium bicarbonate, Willauer et al. showed the feasibility of extracting bound CO2 from the ionic form of bicarbonate and carbonate in addition to the dissolved gas in solution using a multiple layer gas permeable membrane configuration in a closed system near or above the bubble point pressure of a single layer.13,14 CO2 in the ionic form of bicarbonate underwent a phase transition as the bicarbonate disproportionated to carbon dioxide and carbonate, when gaseous carbon dioxide was removed from the water by diffusion through the multiple gas permeable membrane layers. The results also suggested that at pressures below 100 psi the bound CO2 in the ionic form could not be extracted from the aqueous bicarbonate system. This is because the equilibrium of the buffer anions is an opposing force that limits the amount of CO2 released.13,14 Because of the unique opposing forces that limit CO2 release in multiple membrane layer systems and the potential to operate gas permeable membranes at elevated pressures for recovery of carbon from the ocean, this work was performed using seawater to determine and establish a relationship between pressure at, below, and above the bubble point of a single layer membrane and CO2 selectivity and rate of CO2 flux across the membrane. The bubble point pressure by the Young-Laplace equation for the membranes used in this work is approximately 500 psi and it is defined as

P ) 4γ/D

(4)

where P is the bubble point pressure, γ is the surface tension of the liquid in dynes/centimeter, and D is the effective diameter of the membrane pore in micrometers.23 Furthermore, the carbon content of the seawater was increased by the addition of sodium bicarbonate to concentrations about five times those seen in typical seawater. This was done to enhance the measurable loss in bound CO2 in the ionic form of bicarbonate and carbonate and to overcome the limitations of the very short experiments and very low surface to volume ratios required by the experimental high pressure apparatus.

ments. The pH of the seawater/bicarbonate solution measured 8.0, and all pH measurements were conducted with a standardized Fisher combination glass electrode. A microporous polypropylene membrane commercially designated as 2400 Microporous Membrane with a 37% porosity, 0.117 × 0.042 µm rectangular pore dimensions (with effective pore area of 0.005 µm2, which is equivalent to a circular pore size of 0.08 µm diameter), 25 µm thickness, 24 s gurley, and 1300 kg/cm2 tensile strength was used in the studies, and it was obtained from Celgard (Charlotte, NC). The membrane is known to preferentially permeate CO2 gas and not water at low pressures.24 The membranes were never reused throughout the experiments. Figure 1 is a schematic of the high pressure membrane extractor. A Monel pressure cylinder held approximately 150 cm3 of sample volume and the membrane filter holder approximately 45 cm3. A stainless steel frit and rubber O-ring was used to support the membranes (Phenomenex, 2.54 cm diameter, 1.5 mm thickness). Stainless steel fittings held the frit, a 2.54 cm outer diameter rubber O-ring (inner diameter of 1.5 cm), and membranes in place. In Figure 1, the seawater/bicarbonate solution was pumped into the top of the high pressure cylinder and pressurized using the stainless steel syringe liquid pump from Teledyne ISCO, Inc. (Lincoln, NE). The seawater/ bicarbonate solution was pressurized and maintained at given pressure (100, 250, 500, and 1000 psi) in the cylinder and membrane holder. The pressurized solution was in constant contact with the membrane. A stir bar was used in the Monel cylinder and membrane filter holder to stir the solution to increase mass transfer. In each experiment, the seawater/bicarbonate solution was in contact with seven layers of membrane at a given pressures (100, 250, 500, and 1000 psi) for 5 h. Seven layers of membrane were used to minimize water permeation in the system. At the higher pressures (500 and 1000 psi), approximately 5 mL of the 150 mL seawater/bicarbonate solution permeated over 5 h. Even though a small amount of solution permeated through the membranes, the solution in the cylinder was being replaced by the syringe pump to maintain constant pressure. After each experiment, the seawater/bicarbonate solution was depressurized and the entire 150 mL of the seawater/bicarbonate solution sample was collected and titrated with 0.2 M HCl.25,26 The titration end points for the pressurized solutions in contact with the membranes were compared to the end point measured for the same solution at atmospheric pressure and temperature that was never in contact with the membranes. The lower values from the titrations represent the µmol/kg of CO2 gas released from the seawater/bicarbonate solution under pressure (after contact with membrane) compared to the seawater/bicarbonate solution not subjected to pressure/time regimes (inlet sample). 3. Results and Discussion Facilitated transport membrane (FTM) systems using simple concentrated water/bicarbonate solutions is one of the most viable approaches for CO2 separation when the CO2 partial pressures are low and high CO2 selectivity is desired. One such example is the removal of CO2 from breathing atmospheres. In the FTM systems, bicarbonate-carbonate acts as carrier species that reversibly react with CO2, as shown in eqs 3 and 5, to increase CO2 gas selectivity and flux8,22,27–30

2. Experimental Section Reagent grade sodium bicarbonate was added to Key West seawater at a concentration of 1 g per liter with the total CO2 concentration approximately 14200 µmol/kg as measured by strong acid titration with 0.2 M HCl. This concentration is about 5 times higher than the initial seawater concentration. The reason for using higher salt concentrations is to compensate for the small experimental sample sizes used in the pressure experi-

CO2(g) v + CO32-(aq) + H2O(l) a 2HCO3-(aq)

(5) In the present work the recovery of bound CO2 in ionic form (eqs 3 and 5) from seawater/bicarbonate solutions by gas permeable membranes as a function of pressure is being discussed.

Effects of Pressure on the Recovery of CO2

J. Phys. Chem. A, Vol. 114, No. 11, 2010 4005

Figure 1. Membrane extractor.

TABLE 1: CO2 Reduction Results As Measured by Acid Titration Using Potentiometry sample

pressure (psi)

conditions

µmols/kg CO2

1 g/L NaHCO3 in Key West seawater 1 g/L NaHCO3 in Key West seawater 1 g/L NaHCO3 in Key West seawater 1 g/L NaHCO3 in Key West seawater 1 g/L NaHCO3 in Key West seawater 1 g/L NaHCO3 in Key West seawater 1 g/L NaHCO3 in Key West seawater 1 g/L NaHCO3 in Key West seawater 1 g/L NaHCO3 in Key West seawater 1 g/L NaHCO3 in Key West seawater 1 g/L NaHCO3 in Key West seawater mean standard deviation standard error

0 100 100 250 250 500 500 500 1000 1000 1000

inlet sample outlet sample outlet sample outlet sample outlet sample outlet sample outlet sample outlet sample outlet sample outlet sample outlet sample

14200 14200 14200 11786 12212 13206 13561 12638 13614 12354 12638 12751 12751 ( 656 12751 ( 232

In Table 1, an initial seawater/sodium bicarbonate solution was measured to determine the µmol/kg of CO2 in the sample as measured by acid titration using potentiometry before contact with the membrane layers at different pressures.25,26 A sample containing approximately 14 200 µmol/kg of CO2 was then pumped into the high pressure cylinder and the membrane holder where it was in contact with seven layers of membrane at a given pressure (100, 250, 500, and 1000 psi) for 5 h. After contact with the membrane for 5 h, the sample was depressurized and measured to determine total CO2 reduction. During the initial experimental runs, the [CO2]T at pressures of 100 psi showed no reduction after 5 h. In Table 1, as the pressure is increased from 100 to 250 psi, a significant reduction in [CO2]T is observed and this is a much greater percentage than the fraction of simple dissolved carbon dioxide in the

Q-Test

0.23

0.03

% reduction 0 0 17 14 7 4.5 11 4 13 11 10 10 ( 4.6 10 ( 1.6

system (2 to 3%). The table shows an average reduction of about 15% under these experimental conditions. This is significant as it would represent an increase of about an order of magnitude in recovery rates as compared to those rates obtained by just recovering the dissolved CO2. Thus the yield for a very large shipboard constrained system could be as high as 1 kg/sec compared to 0.1 kg/sec achievable by only recovering dissolved CO2. When the pressure is raised to pressures corresponding to that of the bubble point pressure of a single membrane layer (500 psi), there is still a measurable loss in [CO2]T; however, this reduction appears to be less and difficult to repeat. Similar findings are shown for the results measured at 1000 psi. Pressures above 100 psi appear to shift the equilibrium conditions shown in eqs 3 in favor of bicarbonate re-equilibrating to gaseous carbon dioxide; however, any increase in the

4006

J. Phys. Chem. A, Vol. 114, No. 11, 2010

Willauer et al.

Figure 2. Schematic diagram of gas permeable membrane process. The equations discussed in the text are placed either to the left or right (or directly over the membrane) to show the loss of CO2 from the bulk liquid (on the left of the membrane) into air (on the right of the membrane). The actual effects of the CO2 loss on the species remaining in the bulk liquid are shown. Initially in seawater [CO2(g)]l is hydrated by water to form undissociated carbonic acid (equation a). The decomposition of H2CO3 at the membrane surface to CO2 causes an increase in pH (loss in acidic species) as shown in (equation b). To compensate for an increase in pH, the bicarbonate begins to dissociate contributing a proton and a carbonate ion (equation c). The bicarbonate ions that were once in equilibrium, as shown in equation a, also begin to disproportionate to carbonate, and CO2 is lost through the membrane (equation d). The ratio of carbonate to bicarbonate limits pH change in the solution. The initial process is a to b. As this occurs the process of a to c begins to occur followed by the process a to d.

CO2 flux or amount of CO2 recovered by this process has been affected by the bubble point of a single membrane layer. Below 500 psi CO2 gas passes through the membrane pores while the liquid is prohibited from passing. At or above this pressure, the liquid fills the pores and permeates completely, preventing the gas permeability effect from operating. The multiple layers of membranes were used to prevent water permeability at higher pressures but they actually decreased the rate of CO2 gas permeability that would be expected by the use of a single layer. Thus the results suggest that under these experimental conditions no measurable pressure effect is observed below 250 psi. To further substantiate these findings, a Q-test was performed using the eight measured results for CO2 loss at pressures of 250, 500, and 1000 psi. The Q-test is the preferred method for data rejection when less than 10 measurements are involved. The standard Q for eight values is Q ) 0.47.31 Table 1 shows that the Q-values (0.23 and 0.03) obtained for the highest and lowest CO2 values (11 786 and 13 614) measured in the test series is less than the standard Q-value for eight measurements. Thus all eight measurements are valid and none of the measurements can be rejected from the data set assuming no negligible effects of pressure below 250 psi. Therefore all eight values were averaged to determine a standard deviation and standard error as shown at the bottom of Table 1. While the standard deviation is an expression of the precision of the results obtained from the method, the standard error is an expression of limits that represents the true value with a confidence of 68%.31 The mean, standard deviation, and standard error were also determined and reported for the percent reduction of CO2. The table shows an average 10% loss in CO2 which compares well to previous results obtained for simple water/bicarbonate systems pressurized to 500 psi. Figure 2 is a schematic diagram showing the equilibrium equations that describe the gas permeable membrane process for a seawater solution. In seawater, [CO2(g)]1 is hydrated by water to form undissociated carbonic acid as shown in equation a, Figure 2, and this carbonic acid species is in equilibrium with the dissolved bicarbonate and carbonate species. The decomposition of undissociated carbonic acid at the membrane surface as shown in equation b accounts for up to 2-3% percent of [CO2]T in seawater. As the CO2 passes through the membrane, the bulk liquid experiences an increase in pH (loss of an acidic species) as shown in equation b of Figure 2. Seawater is buffered by bicarbonate and carbonate and in order to compensate

for this initial pH increase, the bicarbonate begins to dissociate contributing a proton and a carbonate ion (equation c Figure 2). In addition, bicarbonate ions that were once in equilibrium, as shown in equation a, also begin to disproportionate (pKa ) 10.3) according to equation d of Figure 2, resulting in the loss of an acidic species through the membrane (CO2) and the formation of the additional strong base carbonate ions (pKa ) 3.7). The ratio of carbonate to bicarbonate limits the pH change. Carbonate is the stronger base and so bicarbonate concentrations decrease as illustrated in equations c and d. The process occurring in equations a and b is essentially physical in nature and initiates a second process that is represented by moving from equation a to c and finally a to d, which is essentially chemical in nature. In addition to these two aqueous phase buffer anions, we must also consider the phase transfer of CO2. It is reasonable to expect a welldesigned buffer system that works in equilibrium with all three species to begin to replace removed CO2 as a new equilibrium is approached (equation d). Time is now a factor to consider in that the dissolution and hydration of CO2 (equations a and b) is relatively slow compared to concentration changes in the anions (equations c and d).7 This shift has previously been shown to be independent of solution ionic strength. The final solution pH after the experiment is always lower (more acidic) because the bulk of the CO2 comes from the loss of the basic bicarbonate/carbonate system. All the data have been measured by standard acid titrations using potentiometery. This is possible since the acid titrates the bicarbonate and carbonate ions that are changed when bicarbonate is removed through re-equilibration to gaseous carbon dioxide and subsequent transfer to the gas phase on the air side of the membrane as illustrated in Figure 2. Similar methods could not be used to measure reduction in [CO2]T for bicarbonate/water solutions that had been in contact with seven layers of membranes at 500 psi for 5 h.13,14 In deionized water, the bicarbonate species forms a basic solution as shown in the following equations (6 and 7):

NaHCO3 + H2O a NaOH + H2CO3

(6)

H2CO3 + H2O a H3O+ + HCO3-

(7)

Titration with acid only titrates the hydroxide ion, which is unchanged by any loss of CO2.14 Thus when a simple aqueous

Effects of Pressure on the Recovery of CO2

J. Phys. Chem. A, Vol. 114, No. 11, 2010 4007

TABLE 2: Gas Permeability in Membrane Configurations material Celgard 2400 Celgard X30-240 Celgard X30-240 KPF 280E Mitsubishi MHF 200 L

system gas-gas gas-gas gas-gas

rate gas permeance mL/ sec cm2cm Hg 1.5 × 10-2 CO2 gas 1.6 × 10-2 CO2 gas 1.3 × 10-2 CO2 gas

solution is titrated with HCl it appears that no change in pH has been produced even though [CO2]T has been decreased by 10%. Since the reduction in carbon dioxide has been measured, the rate of CO2 permeability across the membrane for this system may be determined by Fick’s law shown in eq 8

F ) n/At

(8)

where F is the CO2 flux, n is the moles of CO2, A is the membrane area and thickness, and t is the time.7 Assuming no measurable pressure effects above 100 psi, the CO2 permeability for these system is identical to that measured for a simple bicarbonate/water systems previously reported (4.0 × 10-7 mL/ sec cm2 cm Hg).13,14 Table 2 shows that in the context of typical micorporous membrane systems, the membrane permeability in gas-liquid systems has been shown to be on the order of 102-105 times slower than gas-gas systems (Table 2). This difference has been attributed to liquid infiltration into the membrane pores along with slower gas diffusion in a liquid medium.16–18,32 However, since CO2 is in the ionic form as carbonate and bicarbonate in this gas-liquid membrane system, the degree of flux across the membrane will also be dependent on the rate of CO2 disproportionation in the system and the relatively higher ion mobility to the membrane surface. Disproportionation occurs at the surface of the membrane and contributes to increasing CO2 flux across the membrane as the removed CO2 is continually being replenished as carbonic acid, carbonate, and bicarbonate re-equilibrate. In typical gas-liquid systems CO2 gas diffusion is slow. The addition of carrier species described in FTM systems enhances ion diffusion mobility to increase the flux of CO2 across the membrane. Thus ion mobility will contribute to increasing CO2 flux in this gas-liquid membrane system. From Table 2 we can see the rates of CO2 permeance using different materials in both a gas-gas and gas-liquid system. Regardless of membrane material and treatment, the range of CO2 permeabilities in a gas-liquid system is from about 840 to 3200 × 10-7 mL/sec cm2 cm Hg from literature values. Thus values from this work are about 100 times slower. This is attributed to the fact that the bulk water pressures of our system were above 100 psi and at these higher pressures additional membrane layers are needed to minimize water permeation. Similarly Lund et al. put into perspective the difficulty of eliminating water permeability in microporous fiber gas-liquid environments that led to lower than expected CO2 permeance rates.16 The multilayer gas permeable membrane system employed in these studies has two effects on reducing gas permeability. The first is that the multilayers function as a series of delta pressure gradients from the initial bulk pressure at the top surface of the top membrane (∼ 500 psi) to near atmospheric pressure on the effluent side of the bottom membrane. The second effect of the multiple membrane system is to decrease the water permeability that effectively decreased the effective porosity

system

rate gas permeance mL/ sec cm2cm Hg

gas-liquid

7 layers 4.0 × 10-7 CO2 gas

gas-liquid gas-liquid

3.2 × 10-4 CO2 gas 8.4 × 10-5 CO2 gas

ref past/present work 11 13 12 12

of the entire stack of membranes by providing a more torturous path for the liquid. Though higher pressure causes permeance of water into the membrane that effectively lowers the rate of gas permeability, the effect bulk liquid pressure has on the phase transition of CO2 in the ionic form of bicarbonate as it disproportionates to CO2 and carbonate in a closed system was found to be significant. The results verify that at pressures below 100 psi, bound CO2 in the ionic form cannot be extracted from both aqueous bicarbonate and seawater/bicarbonate systems due to the opposing forces caused by the equilibrium of the buffer anions. At pressures near or above the bubble point pressure of a single membrane layer any increase in the phase transition of CO2 in the ionic form of bicarbonate is reduced by the permeance of the water into the membrane pores. 4. Conclusions Seawater/bicarbonate systems have been used to investigate the feasibility of extracting bound CO2 in the ionic form of bicarbonate and carbonate (primarily through bicarbonate disproportionation) using a unique multiple layer gas permeable membrane configuration below, near, and above the bubble point pressure of a single membrane layer (100, 250, 500, 1000 psi). The results indicate that bicarbonate does appear to equilibrate (through disproportionation) and that carbon dioxide permeance rates may be high enough to warrant further investigation of gas permeable membranes in open ocean systems. The advantage of operating gas permeable membranes above 100 psi in a closed system is the ability to extract the bound CO2 in addition to the dissolved gas in the solution, and thus obtaining 5 times more CO2. The disadvantage is a decrease in gas permeance rates. In the future, research will be directed toward operating gas permeable membranes in the open ocean. The advantage is that the system is not limited by a fixed volume as it was in these studies. Thus the effect of continuous cycling of seawater on extracting bound CO2 in the ionic form is of interest. Acknowledgment. This work was supported by the Office of Naval Research both directly and through the Naval Research Laboratory. The authors acknowledge the valuable input from Professor Kathleen Hardy of St. Mary’s College of Maryland. References and Notes (1) Cline, W. The Economics of Global Warming; Institute for International Economics: Washington DC, 1992. (2) Takahasi, T.; Broecker, W. S.; Bainbridge, A. E. The Alkalinity and Total Carbon Dioxide Concentration in the World Oceans. In Carbon Cycle Modelling; SCOPE: New York, 1981; Vol. 16, pp 271-286. (3) Takahasi, T.; Broecker, W. S.; Werner, S. R.; Bainbridge, A. E. Carbonate Chemistry of the Surface of the Waters of the World Oceans. In Isotope Marine Chemistry; Goldberg, E. D., Horibe, Y., Katsuko, S., Eds.; Uchida Rokakuho: Tokyo, Japan, 1980; pp 291-326. (4) Coffey, T.; Hardy, D. R.; Besenbruch, G. E.; Schultz, K. R.; Brown, L. C.; Dahlburg, J. P. Hydrogen as a Fuel for DOD. Defense Horizons 2003, 36, 1. (5) Mohanasundaram, S. Renewable Power Generation-Utilising Thermal Energy From Oceans. EnViron. Sci. Eng. 2007, 4, 35.

4008

J. Phys. Chem. A, Vol. 114, No. 11, 2010

(6) Avery, W. H.; Wu, C. Renewable Energy From The Ocean; Oxford University Press: New York, 1994. (7) Werner, S.; Morgan, J. J. Aquatic Chemistry: An introduction emphasizing chemical equilibrium in natural waters; Wiley-Interscience: New York, 1970. (8) Glade, H.; Al-Rawajfeh, A. E. Modeling of CO2 release and the carbonate system in multiple-effect distillers. Desalination 2008, 222, 605. (9) Wilson, E. K. A Renaissance for Hofmeister. Chem. Eng. News 2007, 85, 47. (10) Willauer, H. D.; Hardy, D. R.; Lewis, M. K.; Ndubizu, E. C.; Williams, F. W. Extraction of CO2 From Seawater By Ion Exchange Resin Part II: Using a Strong Base Anion Exchange Resins; Memorandum Report 6180-09-9211; Naval Research Laboratory: Washington, DC, September 29, 2009. (11) Hardy, D. H.; Zagrobelny, M.; Willauer, H. D.; Williams, F. W. Extraction of Carbon Dioxide From Seawater by Ion Exchange Resin Part I: Using a Strong Acid Cation Exchange Resin; Memorandum Report 618007-9044; Naval Research Laboratory: Washington, DC, April 20, 2007. (12) Johnson, K. M.; King, A. E.; Sieburth, J. Coulometric TCO2 Analyses for Marine Studies: An Introduction. Marine Chem. 1985, 16, 61. (13) Willauer, H. D.; Hardy, D. R.; Lewis, M. K.; Ndubizu, E. C.; Williams, F. W. Recovery of CO2 from Aqueous Bicarbonate Using a Gas Permeable Membrane. Energy Fuels 2009, 23, 1770–1774. (14) Willauer, H. D.; Hardy, D. R.; Lewis, M. K.; Ndubizu, E. C.; Williams, F. W. RecoVery of [CO2]T from Aqueous Bicarbonate Using a Gas Permeable Membrane; NRL Memorandum Report, 6180-08-9129; Naval Research Laboratory: Washington, DC, June 25, 2008. (15) Bhaumik, D.; Majumdar, S.; Fan, Q.; Sirkar, K. K. Hollow Fiber Membrane Degassing in Ultrapure Water and Microbiocontamination. J. Membr. Sci. 2004, 235, 31. (16) Lund, L. W.; Hattler, B. G.; Federspiel, W. J. Gas Permeance Measurement of Hollow Fiber Membranes in Gas-Liquid Environment. AIChE J. 2002, 48, 635. (17) Lund, L. W.; Federspiel, W. J.; Hattler, B. G. Gas Permeability of Hollow Fiber Membranes in a Gas-Liquid System. J. Membr. Sci. 1996, 117, 207. (18) Eash, H. J.; Jones, H. M.; Hattler, B. G.; Federspiel, W. J. Evaluation of Plasma Resistant Hollow Fiber Membranes for Artificial Lungs. ASAIO J. 2004, 50, 491.

Willauer et al. (19) Bosko, R. S. Hollow Fiber Carbonation. U.S. Patent 6,712,342, March 20, 2004. (20) Gabelman, A.; Hwang, S.-T. Hollow Fiber Membrane Contactors. J. Membr. Sci. 1999, 159, 61. (21) Wiesler, F.; Sodaro R. Deaeration: Degasification of Water Using Novel Membrane Technology. Ultrapure Water, UP130653; Tall Oaks Publishing Inc.: Littleton, CO, 1996; pp 53-56. (22) Matson, S. L.; Lopez, J.; Quinn, J. A. Review Article Number 13: Separation of Gases with Synthetic Membranes. Chem. Eng. Sci. 1983, 38, 503. (23) Roma´n, F. L.; Faro, J.; Velasco, S. A Simple Experiment for Measuring the Surface Tension of Soap Solutions. Am. J. Phys. 2001, 69, 920. (24) Bhave, R. R.; Sirkar, K. K. Gas Permeation and Separation by Aqueous Membranes Immobilized Across the Whole Thickness of in a Thin Section of Hydrophobic Microporous Celgard Films. J. Membr. Sci. 1986, 27, 41. (25) Millero, F. J.; Zhang, J-Z; Lee, K.; Campbell, D. M. Titration Alkalinity of Seawater. Marine Chem. 1993, 44, 153. (26) Edmond, J. M. High Precision Determination of Titration Alkalinity and Total Carbon Dioxide Content of Sea Water by Potentiometric Titration. Deep-Sea Res. 1970, 17, 737. (27) Ward, W. J.; Robb, W. L. Carbon Dioxide-Oxygen Separation: Facilitated Transport of Carbon Dioxide across a Liquid Film. Science 1967, 156, 1481. (28) Kovvali, A. S.; Sirkar, K. K. Carbon Dioxide Separation with Novel Solvents as Liquids Membranes. Ind. Eng. Chem. Res. 2002, 41, 2287. (29) Chen, H.; Obuskovic, G.; Majumdar, S.; Sirkar, K. K. Immobilized Glycerol-Based Liquid Membranes in Hollow Fibers for Selective CO2N2 mixtures. J. Membr. Sci. 2001, 183, 75. (30) Yang, H.; Xu, Z.; Fan, M.; Gupta, R.; Slimane, R. B.; Bland, A. E.; Wright, I. Progress in Carbon Dioxide Separation and Capture; A Review. J. EnViron. Sciences 2008, 20, 14. (31) Bauer, L. A Statistical Manual for Chemists; Academic Press: New York, 1971. (32) Yasuda, H.; Lamaze, C. E. Transfer of Gas to Dissolved Oxygen in Water Via Porous and Nonporous Polymer Membranes. J. Appl. Polym. Sci. 1972, 16, 595.

JP910603F