- Email: [email protected]
Adsorption of sodium dodecyl sulfate onto clathrate hydrates in the presence of salt
Journal of Colloid and Interface Science 386 (2012) 333–337
Contents lists available at SciVerse ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Adsorption of sodium dodecyl sulfate onto clathrate hydrates in the presence of salt O. Salako a, C. Lo a,b, J.S. Zhang a, A. Couzis a, P. Somasundaran b, J.W. Lee a,⇑ a b
Department of Chemical Engineering, The City College of New York, New York, NY 10031, USA Department of Earth and Environmental Engineering, Columbia University, New York, NY 10027, USA
a r t i c l e
i n f o
Article history: Received 10 May 2012 Accepted 7 July 2012 Available online 20 July 2012 Keywords: Adsorption SDS Clathrate hydrates NaCl Salt effect
a b s t r a c t This work presents the effect of NaCl on the adsorption of sodium dodecyl sulfate (SDS) at the cyclopentane (CP) hydrate–water interface. The adsorption isotherms and the SDS solubility in NaCl solutions are obtained using liquid–liquid titrations. The solubility data are determined at typical hydrate forming temperatures (274–287 K) to ensure that the adsorption isotherms are obtained within SDS solubility limits in NaCl solutions. The isotherms show L–S (Langmuir–Step) type behaviors with 1 mM and 10 mM NaCl solutions while L type isotherm is determined for 25 mM NaCl solutions due to the low SDS solubility in this salt concentration. Zeta potentials of CP hydrate particles in the aqueous solutions support the shape of the adsorption isotherm with the 1 mM NaCl solution. The 1 mM NaCl case shows the highest SDS adsorption amount among the cases with 0 mM, 10 mM, and 25 mM NaCl solutions. In this case, the competition for adsorption between Cl and DS is not as strong compared to the 10 and 25 mM NaCl cases and the presence of Na+ ions may reduce the repulsion between DS ions, which results in a higher adsorption of DS ions and enhanced enclathration. Published by Elsevier Inc.
1. Introduction Clathrate hydrates are ice-like non-stoichiometric crystalline compounds in which water molecules form cages that trap molecules of low molecular weight [1]. Clathrate hydrates have attracted lots of attention since the discovery of large reserves of natural gas hydrates in oceanic deposits and permafrost areas [1]. These natural gas hydrates can be tapped as a future energy resource. In addition, a clathrate form of natural gas can provide a high density of gas storage up to 170 vol./vol. of clathrate hydrate and be utilized as a natural gas storage medium [1,2]. Clathrate hydrates also have negative aspects in that they can form inside gas and oil delivery lines, which have clogged of the pipelines and hindered continuous production of oil and gas. Understanding hydrate formation kinetics is essential for both cases. The kinetics of hydrate formation is unfavorable due to diffusion limitations of guest molecules to liquid–solid growth fronts, but can be accelerated by applying mechanical agitations or adding surfactants to reaction systems [2–5]. Employing surfactants might be more promising option for accelerating hydrate formation rates. This is due to the high energy costs of mechanical agitation for the solidification process and the low surfactant dosage (tens or hundreds of ppm) requirement. Sodium dodecyl sulfate (SDS) has been ⇑ Corresponding author. Address: Department of Chemical Engineering, The City College of the CUNY, 140th Street and Convent Ave., New York, NY 10031, USA. Fax: +1 212 650 6660. E-mail address: [email protected] (J.W. Lee). 0021-9797/$ - see front matter Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jcis.2012.07.017
found to be one of the most potent enclathration promoters for low molecular weight hydrocarbon hydrate formation like methane and ethane hydrates [2]. The mechanism behind which the surfactant speeds up the hydrate formation rate has not been well understood. It has been suggested that the nucleation sites formed by surfactant micelles enhance the kinetics of hydrate formation [2]. However, the temperatures at which enclathration takes place are lower than the Krafft point for surfactants [6–8] at which the critical micelle concentration and the surfactant solubility are equal. The adsorption of DS ions at the hydrate–water interface has also been proposed as a reason for SDS’s ability to accelerate enclathration kinetics [9]. Lo et al. [10] studied the adsorption of SDS on the CP hydrate–water interface through adsorption isotherms and they proposed the orientation of SDS at the hydrate– water interface. Zhang et al. [11] studied the effect of NaCl on the formation of methane hydrate in the presence of SDS and cyclopentane (CP) using a high-pressure cell. It was experimentally determined that the rate and amount of methane enclathration were highest in the presence of 1 mM NaCl, with the salt concentration ranging from 0 to 100 mM and 20 ppm SDS in the cell. However, it has not been understood as to why the highest enclathration rate was observed for the case of 1 mM NaCl and how this is related to the SDS adsorption behavior. Thus, this study aims at determining the adsorption isotherm of SDS on CP hydrates in the presence of NaCl solution and utilizing the isotherm data to explain the salt effect on the SDS adsorption. The isotherms will be determined using liquid–liquid titrations.
334
O. Salako et al. / Journal of Colloid and Interface Science 386 (2012) 333–337
Zeta potential measurements will be carried out on CP hydrate particles in NaCl aqueous solutions. These measurements will be used as supporting information to verify the shape of the adsorption isotherm and to identify the adsorption preference between Cl and DS on the hydrate–water interface. CP hydrates will be used as a model system for natural gas hydrates, since CP is a hydrophobic guest molecule and forms the same sII clathrate structure as natural gas at atmospheric pressures [1,12–15]. This work will show the dependence of SDS adsorption at the CP hydrate–water interface on different NaCl concentrations. The solubility of SDS in salt solutions is determined at low temperatures (typical hydrate forming temperature: 274–287 K) to confirm that the isotherms are obtained within the SDS solubility limitation. This work will support the hypothesis that adsorption of DS ions is responsible for accelerated enclathration even with a salt in solutions. Using the SDS adsorption isotherm, the adsorption mechanism of SDS at the hydrate water interface will be proposed while accommodating the effect of salt. 2. Experimental section 2.1. Materials and methods Cyclopentane (CP) and sodium dodecyl sulfate (SDS) with 99% purity were purchased from Sigma–Aldrich. Methylene blue with indicator purity was supplied by Sigma–Aldrich. Sulfuric acid and sodium sulfate, with purity of 96% and 99%, respectively, were also purchased from Sigma–Aldrich. The deionized (D.I.) water used was produced from our lab with a resistivity of 18 MO cm 1. 2.2. Preparation of CP slurry A 300 mL mixture of CP and water with 10 wt.% CP was charged into a 1 L bottle. The bottle was completely sealed, vigorously shaken and immediately transferred into a freezer with a temperature of 263 K. After ice formation, the bottle was vigorously shaken under ambient conditions and the ice melted as a result. CP enclathrated as the ice melted and enclathration was confirmed by the appearance of white particles in the bottle. The bottle was transferred into a chiller at 275 ± 0.2 K for a week, during which the bottle was shaken at least five times per day to accelerate enclathration. Calorimetric measurements were carried out on the CP hydrate slurry and the CP hydrate concentration was found to be 51 wt.%. CP hydrate slurries were employed for the experiment to minimize variations of the total surface area during the isotherm and zeta-potential experiments and to avoid moisture condensation on hydrates. 2.3. Adsorption isotherm 10 g of surfactant solutions were charged to 25 mL vials. The vials were transferred into the chiller at 275 K for about 12 h. 10 g of CP hydrate slurry was quickly transferred into the vial. The vial was tightly sealed and then returned into the chiller at 275 ± 0.2 K for one week to allow adsorption to reach equilibrium. The vials were shaken periodically in the course of the week to accelerate the surfactant adsorption. After one week, several milliliters of surfactant solution were extracted from the lower portion of the vial using 6 mL syringes. To avoid hydrate melting, sample extraction inside the chiller usually lasted for less than a minute. The SDS concentrations in solutions prepared before adsorption and solutions extracted after adsorption were analyzed. Thus, the difference between the concentrations of the supernatant before and after the adsorption experiment gives the amount of SDS adsorbed onto the hydrate–water interface.
2.4. SDS solubility in NaCl solutions A 450 mL high-pressure stirred reactor customized by Parr Instruments was used for the solubility measurement. The reactor temperature was controlled by circulating a coolant from an Isotemp 3006P thermostat (Fisher Scientific) with a stability of ±0.01 K within the jacket around the cell. Two type-T thermocouples (Omega Engineering) were used to monitor the reactor temperature with one immersed in the solution and the other placed in the reactor head space. The precision of the temperature measurement is ±0.5 K. The temperature of the liquid and gas phase was sampled every 20 s by Labview interface. 17.3 mM SDS solutions were prepared with 1, 10, and 25 mM NaCl solutions. About 250 mL of one of these solutions, at a time, was charged into the high-pressure stirred reactor. Then the reactor was pressurized up to 7.5 MPa of natural gas to maintain meta-stable regions of natural gas hydrate formation as described in our previous work [8]. The temperature of the reactor was first set to 274 K and was kept at this temperature for about 12 h to allow the system reach equilibrium after which samples are collected. The same procedure was observed for the temperatures of 275, 277, 281, 284, and 287 K for all of the NaCl solutions. The SDS solubility is also determined by liquid–liquid titrations as described below. 2.5. Liquid–liquid titrations One milliliter of SDS solution of unknown concentration was pipetted into a 20 mL test-tube followed by the addition of 2.5 mL of methylene blue solution (0.003 wt.% methylene blue, 1.2 wt.% H2SO4 and 5 wt.% Na2SO4) and 2.5 mL chloroform. After the test-tube was vigorously shaken, hyamine (titrant) was added to the tube. Before the introduction of hyamine into the test-tube, the upper layer of the solution in the test-tube was clear while the lower layer was blue. The end-point was reached when the lower layer and the upper layer had the same blue color. This titration procedure was adapted from Epton [16]. The accuracy of concentration measured using this analysis was within 2% from triplicate measurements. 2.6. Zeta potential measurement 10 g of solution was charged to 25 mL vials and the vials were transferred into the chiller at 275 K for about 12 h. 1 g of CP hydrate slurry was quickly transferred into the vial. The vial was tightly sealed and then returned into the chiller at 275 ± 0.2 K for one week to allow adsorption to reach equilibrium. CP hydrate particles suspended in aqueous solution were quickly transferred into a chill folded capillary cell using a chill pipette. The folded capillary cell was loaded into the zeta potential machine (Zetasizer Nano ZS, Malvern Instruments) at 277 K and then zeta potential data were recorded. 3. Results and discussion It was observed that addition of a small amount of NaCl (1 mM) into 20 ppm SDS aqueous phase enhances methane hydrate formation [11]. We conjecture concurrently with our previous work [10,17,18] that the adsorption of SDS onto the hydrate–water interface is a main cause for accelerated hydrate formation and that the presence of salt can affect the SDS adsorption behavior. 3.1. Proposed adsorption scheme Fig. 1 proposes the SDS adsorption scheme onto the CP hydrate– water interface in the presence of NaCl and also present the
335
O. Salako et al. / Journal of Colloid and Interface Science 386 (2012) 333–337
0 mM NaCl 1 mM NaCl 10 mM NaCl 25 mM NaCl
0.025
Adsorption (mM/g)
0.020
0.015
0.010
0.005
0.000 0.0
0.5
1.0
1.5
2.0
2.5
3.0
SDS Equilibrium Conc (mM) Fig. 1. Adsorption isotherms of SDS with 0, 1, 10 and 25 mM NaCl solutions and proposed adsorption schematics. The circles in the schematics represent Cl ions.
-30 -40 -50
Zeta Potential (mV)
adsorption isotherm of SDS at different salt concentrations. At the initial stages of adsorption, before the monolayer coverage (0– 0.005 mM SDS/g of hydrate), depicted by (1) in Fig. 1, the DS ions adsorb onto the CP hydrate–water interface with their tails in a ‘‘lying down’’ configuration because the SDS tail has hydrophobic interaction with guest molecules (CP) while the headgroup form hydrogen bonds with pendant hydrogen of water cages [10,18]. The location of Na+ ions is assumed to be close to DS and Cl ions as shown in a previous work [19] and Cl ions compete with DS ions for adsorption sites. Fig. 1(2) depicts a pseudo-monolayer coverage (at approximately 0.01 mM SDS/g of hydrate) with the tails of the adsorbed DS ions still lying down at the interface with some of available hydrogen bonding sites occupied. Since the monolayer saturation appears in a narrow SDS concentration between 0.75 and 1.5 mM as shown in Fig. 1, the pseudo monolayer may easily steer to bi-layer formation to stabilize the adsorption process with headgroups exposed to the bulk aqueous phase. At the beginning of the bi-layer formation (0.012–0.016 mM SDS/g of hydrate) depicted by (3), the tails of DS ions in the bulk and the tails of the DS ions adsorbed on the CP hydrate–water interface begin to interact, resulting in the tails of the adsorbed DS ions having a ‘‘stand-up’’ configuration [10,18]. The vertical positioning of the tails that were previously in ‘‘lying down’’ configurations on the CP hydrate–water interface exposes more DS ions to hydrogen bonding sites at the CP hydrate–water interface. Fig. 1(4) depicts the completion of bi-layer formation (0.018 mM/ g of hydrate). The isotherms quantify the SDS adsorption with different NaCl concentrations. From the adsorption isotherms, the effect of NaCl on SDS adsorption and the role of SDS in enclathration can be better understood. Lo et al. [10] determined the adsorption isotherm of SDS onto the CP hydrate–water interface without any salt and they identified the type of isotherm as Langmuir–Step (L–S) type with the amount of SDS adsorbed at the first step of the L isotherm and the second step of the S-isotherm reported as 0.01 mM and 0.02 mM SDS/g of CP hydrate, respectively. Fig. 1 shows that the SDS isotherms with 1 and10 mM NaCl solutions still follow the L–S type adsorption behavior. However, the shape of isotherm is not clearly pictured from Fig. 1. For example, the 1 mM NaCl case levels off at approximately 0.018 mM SDS/g CP hydrate. The isotherm might be either L-type or L–S type. The L-type signifies that the adsorption is one step (monolayer step) while the L–S type signifies that the adsorption is in two steps (monolayer and bilayer of SDS). In order to determine the
-60 -70 -80 -90 -100 -110 -120 0
1
2
3
4
SDS Concentration (mM) Fig. 2. Zeta potential measurements of CP hydrate particles in the presence of 1 mM NaCl at different SDS concentrations.
shape of the isotherm, we analyze the changes in the surface charge of CP hydrate particles in the presence of 1 mM NaCl at different SDS concentrations using zeta potential measurements. The negative surface charge of CP hydrate particles in the presence of 1 mM NaCl (at 0 mM SDS) can be attributed to the adsorption of Cl ions onto the CP hydrate–water interface as shown in Fig. 2. At relatively low SDS concentration (0.10–0.5 mM SDS), the zeta potential of CP hydrate particles linearly increases as shown in Fig. 2. At these low SDS concentrations, the adsorption sites occupied by Cl ions are displaced by DS ions. This is because DS ions adsorb more favorably than Cl , which is indicated by higher zetapotential values at the same concentrations for SDS and NaCl. For example, the 1 mM SDS case in Fig. 2 gives more negative values ( 50 to 55 mV) than the 1 mM NaCl case in Fig. 3 ( 32 mV). The same applies for both of 4 mM cases (4 mM SDS: 80 mV in Fig. 2 versus 4 mM NaCl in Fig. 3: 35 mV). Thus, DS ions can replace Cl ions from adsorption sites previously occupied by Cl ions in the low SDS concentrations of Fig. 2. The zeta potential of CP hydrate particles stays approximately constant for SDS concentrations from 0.50 to 1.70 mM in Fig. 2. This region of SDS concentration can be interpreted as a monolayer
336
O. Salako et al. / Journal of Colloid and Interface Science 386 (2012) 333–337
-25
Zeta Potential (mV)
-35 -40 -45
SDS Concentration (mM)
0 mM NaCl 1 mM NaCl 3 mM NaCl 4 mM NaCl 6 mM NaCl 8 mM NaCl 10 mM NaCl
-30
-50 -55 -60
20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1
1 mM NaCl 10 mM NaCl 25 mM NaCl
272
-65 0
2
4
6
8
274
10
276
278
280
282
284
286
288
Temperature (K)
NaCl Concentration (mM) Fig. 3. Zeta potentials of CP hydrate particles at different NaCl concentrations without any SDS. Zeta-potential measurements above 8 mM NaCl are difficult to perform due to the reaction between the cell electrode and Cl . The electrode turns to brown (refer to pictures in the supplementary data).
step. The zeta potential again increases around 2.5 mM SDS and stays fairly constant from 3.30 to 4.00 mM SDS concentration. This second plateau in Fig. 2 can be interpreted as a bi-layer coverage at 1 mM NaCl. Therefore, we can infer from Fig. 2 that the isotherm in Fig. 1 is L–S type isotherms for 1 mM NaCl. However, for the 10 mM and the 25 mM cases in shown in Fig. 1, they seem to form the SDS monolayer but not to complete the bi-layer. The case of 1 mM NaCl shows relatively higher SDS adsorption amount as compared to the 0, 10 mM, and 25 mM NaCl cases shown in Fig. 1. The monolayer coverage at the CP hydrate–water interface for all of the isotherms is observed at 0.01 mM (SDS)/g of hydrate. However, the monolayer formation delays at the SDS equilibrium concentration from 0.75 to 1.0 mM SDS for 1 mM NaCl and 1.25– 1.5 mM SDS for 10 mM and 25 mM NaCl. This shift in the equilibrium SDS concentration is insignificant compared to one order of magnitude increase in the NaCl concentration. It is an indicator of ion exchange strength between Cl and DS of which DS is the greater. In other words, DS is more favorably adsorbed onto the interface than Cl as also evidenced in the zeta-potential measurements in Figs. 2 and 3. Fig. 1 shows bi-layer coverage at the hydrate–water interface for 0 and 1 mM NaCl curves occurs at 0.018 mM (SDS)/g of hydrate. However, the isotherm with 10 mM NaCl reaches the solubility limit of SDS (2.5 mM SDS at 275 K as shown in Fig. 4) as the bilayer coverage of SDS is approached, and the bi-layer is not completely formed. For the 25 mM NaCl case, the SDS solubility is around 1.5 mM in Fig. 4 and thus the bi-layer is not formed as shown in Fig. 1. 3.2. Proposed adsorption mechanism DS ions adsorb onto the hydrate–water interface dominantly via hydrogen bonding [10,17,18]. Although the surface area of the CP hydrates is presently unknown, the number of hydrogen bonding sites per gram of hydrate is assumed to be equal in the hydrate particles. Therefore, the adsorption amount for the monolayer coverage is statistically the same for all the NaCl concentrations studied in this experiment (0.01 mM (SDS)/g of hydrate in Fig. 1), despite the higher adsorption amount observed for 1 mM NaCl. The only difference is that the equilibrium SDS concentrations (x-axis in Fig. 1) corresponding to the monolayer coverage increase due to the
Fig. 4. SDS solubility at low hydrate-forming temperatures and different NaCl solutions.
competitive adsorption between DS with Cl as the NaCl concentration moves up from 1 to 25 mM. The reasons why the 1 mM NaCl case has an adsorption amount of DS ion higher than that for all NaCl concentrations are (1) the competitive adsorption between DS and Cl is not serious due to the low NaCl concentration and (2) at the same time, Na+ ions may provide charge shielding for the repulsion of DS ions at the CP hydrates–water interface. It should be noted that evidence of Cl ion adsorption on the ice– water interface was previously reported [20]. Then, this work firstly presents the adsorption of Cl ions onto the hydrate–water interface as shown in the zeta-potential measurements of Figs. 2 and 3. At NaCl concentrations of 150 mM or above, much higher than those used in this study, there was evidence that NaCl solutions shrink the size of DS ion head group [19]. Despite this evidence of shrinking DS ion head group size, the amount DS ions adsorbed on the hydrate–water interface at the monolayer coverage remains the same for all of the NaCl concentrations studied in this work as mentioned before. It could be hypothesized that the NaCl concentrations employed in our experiments are too low to significantly shrink the size of DS ion head group. Thus, a main driving force for adsorption of DS– ions at the CP hydrate–water interface before the monolayer coverage is hydrogen bonding [17,18]. Otherwise, the shrinkage of SDS head group due to Na+ ions should change the amount of DS– ions adsorbed at the monolayer or bilayer coverage with different NaCl concentrations.
4. Conclusions This work has investigated the effect of NaCl on adsorption of SDS at the CP hydrate–water interface and analyzed the adsorption behavior using adsorption isotherms, zeta-potential, and SDS solubility. The zeta-potential information supports the shape of L–S type isotherm at 1 mM NaCl, while the SDS solubility data were used to ensure that the isotherms are obtained within the SDS solubility limitation. The highest amount of SDS adsorption was observed for the 1 mM NaCl case at the same SDS bulk equilibrium concentration, which accounts for fast methane hydrate formation in our previous study. The same L–S type isotherm was observed for SDS adsorption in all of the NaCl concentrations except for 25 mM NaCl, in which case the bilayer is not formed due to the low solubility of SDS. Although some previous studies have shown that the head of DS ions shrinks due to the presence of NaCl, it does not have significant effect on the amount of DS ions
O. Salako et al. / Journal of Colloid and Interface Science 386 (2012) 333–337
adsorbed at both the monolayer and the bi-layer coverage for NaCl concentrations up to 25 mM; thus making hydrogen bonding the dominant interaction between SDS and water–hydrate interface. Acknowledgment The authors are grateful for the support of the National Science Foundation for this work (under Grant Numbers CBET-0854210 and HRD-0833180). This research was also made possible in part by a grant from BP/GoMRI. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2012.07.017. References [1] E.D. Sloan, C. Koh, Clathrate Hydrates of Natural Gas, third ed., CRC Press, Boca Raton, FL, 2008. [2] Y. Zhong, R.E. Roger, Chem. Eng. Sci. 55 (2000) 4175–4187.
337
[3] P. Englezos, N. Kalogerakis, P.D. Dholabhai, P.R. Bishnoi, Chem. Eng. Sci. 42 (1987) 2659–2666. [4] W.X. Wang, C.L. Bray, D.J. Adams, A.I. Cooper, J. Am. Chem. Soc. 230 (2008) 11608–11609. [5] W.X. Wang, B.O. Carter, C.L. Bray, A. Steiner, J. Bacsa, J.T. Jones, C. Croper, Y.Z. Khimyak, D.J. Adams, A.I. Cooper, Chem. Mater. 21 (2009) 3810–3815. [6] K. Watanabe, S. Imai, Y.H. Mori, Chem. Eng. Sci. 60 (2005) 4846–4857. [7] P. Gayet, C. Dicharry, G. Marion, A. Graciaa, J. Lachaise, A. Nesterov, Chem. Eng. Sci. 60 (2005) 5751–5758. [8] J.S. Zhang, S.Y. Lee, J.W. Lee, J. Colloid Interface Sci. 315 (2007) 313–318. [9] J.S. Zhang, S.Y. Lee, J.W. Lee, Ind. Eng. Chem. Res. 46 (2007) 6353–6359. [10] C. Lo, J.S. Zhang, J. Phys. Chem. C 114 (2010) 13385–13389. [11] S. Zhang, J.A. Salera, J.W. Lee, Ind. Eng. Chem. Res. 49 (2010) 8267–8270. [12] G.N. Werezak, Chem. Eng. Prog. Symp. Ser. 65 (1969) 6–18. [13] Y. Zhang, P.G. Debenedetti, R.K. Rrud’homme, B.A. Pethica, J. Phys. Chem. B 108 (2004) 16717–16722. [14] J.S. Zhang, J.W. Lee, J. Chem. Eng. Data 54 (2009) 659–661. [15] M. Nakajima, R. Ohmura, Y.H. Mori, Ind. Eng. Chem. Res. 47 (2008) 8933–8939. [16] S.R. Epton, Trans. Faraday Soc. 44 (1948) 226–230. [17] J.S. Zhang, C. Lo, P. Somasundaran, S. Lu, A. Couzis, J.W. Lee, J. Phys. Chem. C 112 (2008) 12381–12385. [18] C. Lo, J.S. Zhang, P. Somasundaran, A. Couzis, J.W. Lee, Langmuir 24 (2008) 12723–12726. [19] G.P. Funkhomer, M.P. Arevalo, D.T. Glatzhofer, E.A. O’Reart, Langmuir 11 (1995) 1443–1447. [20] J. Drzymala, Z. Sadowski, L. Holysz, E. Chibowski, J. Colloid Interface Sci. 220 (1999) 229–234.
Contents lists available at SciVerse ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Adsorption of sodium dodecyl sulfate onto clathrate hydrates in the presence of salt O. Salako a, C. Lo a,b, J.S. Zhang a, A. Couzis a, P. Somasundaran b, J.W. Lee a,⇑ a b
Department of Chemical Engineering, The City College of New York, New York, NY 10031, USA Department of Earth and Environmental Engineering, Columbia University, New York, NY 10027, USA
a r t i c l e
i n f o
Article history: Received 10 May 2012 Accepted 7 July 2012 Available online 20 July 2012 Keywords: Adsorption SDS Clathrate hydrates NaCl Salt effect
a b s t r a c t This work presents the effect of NaCl on the adsorption of sodium dodecyl sulfate (SDS) at the cyclopentane (CP) hydrate–water interface. The adsorption isotherms and the SDS solubility in NaCl solutions are obtained using liquid–liquid titrations. The solubility data are determined at typical hydrate forming temperatures (274–287 K) to ensure that the adsorption isotherms are obtained within SDS solubility limits in NaCl solutions. The isotherms show L–S (Langmuir–Step) type behaviors with 1 mM and 10 mM NaCl solutions while L type isotherm is determined for 25 mM NaCl solutions due to the low SDS solubility in this salt concentration. Zeta potentials of CP hydrate particles in the aqueous solutions support the shape of the adsorption isotherm with the 1 mM NaCl solution. The 1 mM NaCl case shows the highest SDS adsorption amount among the cases with 0 mM, 10 mM, and 25 mM NaCl solutions. In this case, the competition for adsorption between Cl and DS is not as strong compared to the 10 and 25 mM NaCl cases and the presence of Na+ ions may reduce the repulsion between DS ions, which results in a higher adsorption of DS ions and enhanced enclathration. Published by Elsevier Inc.
1. Introduction Clathrate hydrates are ice-like non-stoichiometric crystalline compounds in which water molecules form cages that trap molecules of low molecular weight [1]. Clathrate hydrates have attracted lots of attention since the discovery of large reserves of natural gas hydrates in oceanic deposits and permafrost areas [1]. These natural gas hydrates can be tapped as a future energy resource. In addition, a clathrate form of natural gas can provide a high density of gas storage up to 170 vol./vol. of clathrate hydrate and be utilized as a natural gas storage medium [1,2]. Clathrate hydrates also have negative aspects in that they can form inside gas and oil delivery lines, which have clogged of the pipelines and hindered continuous production of oil and gas. Understanding hydrate formation kinetics is essential for both cases. The kinetics of hydrate formation is unfavorable due to diffusion limitations of guest molecules to liquid–solid growth fronts, but can be accelerated by applying mechanical agitations or adding surfactants to reaction systems [2–5]. Employing surfactants might be more promising option for accelerating hydrate formation rates. This is due to the high energy costs of mechanical agitation for the solidification process and the low surfactant dosage (tens or hundreds of ppm) requirement. Sodium dodecyl sulfate (SDS) has been ⇑ Corresponding author. Address: Department of Chemical Engineering, The City College of the CUNY, 140th Street and Convent Ave., New York, NY 10031, USA. Fax: +1 212 650 6660. E-mail address: [email protected] (J.W. Lee). 0021-9797/$ - see front matter Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jcis.2012.07.017
found to be one of the most potent enclathration promoters for low molecular weight hydrocarbon hydrate formation like methane and ethane hydrates [2]. The mechanism behind which the surfactant speeds up the hydrate formation rate has not been well understood. It has been suggested that the nucleation sites formed by surfactant micelles enhance the kinetics of hydrate formation [2]. However, the temperatures at which enclathration takes place are lower than the Krafft point for surfactants [6–8] at which the critical micelle concentration and the surfactant solubility are equal. The adsorption of DS ions at the hydrate–water interface has also been proposed as a reason for SDS’s ability to accelerate enclathration kinetics [9]. Lo et al. [10] studied the adsorption of SDS on the CP hydrate–water interface through adsorption isotherms and they proposed the orientation of SDS at the hydrate– water interface. Zhang et al. [11] studied the effect of NaCl on the formation of methane hydrate in the presence of SDS and cyclopentane (CP) using a high-pressure cell. It was experimentally determined that the rate and amount of methane enclathration were highest in the presence of 1 mM NaCl, with the salt concentration ranging from 0 to 100 mM and 20 ppm SDS in the cell. However, it has not been understood as to why the highest enclathration rate was observed for the case of 1 mM NaCl and how this is related to the SDS adsorption behavior. Thus, this study aims at determining the adsorption isotherm of SDS on CP hydrates in the presence of NaCl solution and utilizing the isotherm data to explain the salt effect on the SDS adsorption. The isotherms will be determined using liquid–liquid titrations.
334
O. Salako et al. / Journal of Colloid and Interface Science 386 (2012) 333–337
Zeta potential measurements will be carried out on CP hydrate particles in NaCl aqueous solutions. These measurements will be used as supporting information to verify the shape of the adsorption isotherm and to identify the adsorption preference between Cl and DS on the hydrate–water interface. CP hydrates will be used as a model system for natural gas hydrates, since CP is a hydrophobic guest molecule and forms the same sII clathrate structure as natural gas at atmospheric pressures [1,12–15]. This work will show the dependence of SDS adsorption at the CP hydrate–water interface on different NaCl concentrations. The solubility of SDS in salt solutions is determined at low temperatures (typical hydrate forming temperature: 274–287 K) to confirm that the isotherms are obtained within the SDS solubility limitation. This work will support the hypothesis that adsorption of DS ions is responsible for accelerated enclathration even with a salt in solutions. Using the SDS adsorption isotherm, the adsorption mechanism of SDS at the hydrate water interface will be proposed while accommodating the effect of salt. 2. Experimental section 2.1. Materials and methods Cyclopentane (CP) and sodium dodecyl sulfate (SDS) with 99% purity were purchased from Sigma–Aldrich. Methylene blue with indicator purity was supplied by Sigma–Aldrich. Sulfuric acid and sodium sulfate, with purity of 96% and 99%, respectively, were also purchased from Sigma–Aldrich. The deionized (D.I.) water used was produced from our lab with a resistivity of 18 MO cm 1. 2.2. Preparation of CP slurry A 300 mL mixture of CP and water with 10 wt.% CP was charged into a 1 L bottle. The bottle was completely sealed, vigorously shaken and immediately transferred into a freezer with a temperature of 263 K. After ice formation, the bottle was vigorously shaken under ambient conditions and the ice melted as a result. CP enclathrated as the ice melted and enclathration was confirmed by the appearance of white particles in the bottle. The bottle was transferred into a chiller at 275 ± 0.2 K for a week, during which the bottle was shaken at least five times per day to accelerate enclathration. Calorimetric measurements were carried out on the CP hydrate slurry and the CP hydrate concentration was found to be 51 wt.%. CP hydrate slurries were employed for the experiment to minimize variations of the total surface area during the isotherm and zeta-potential experiments and to avoid moisture condensation on hydrates. 2.3. Adsorption isotherm 10 g of surfactant solutions were charged to 25 mL vials. The vials were transferred into the chiller at 275 K for about 12 h. 10 g of CP hydrate slurry was quickly transferred into the vial. The vial was tightly sealed and then returned into the chiller at 275 ± 0.2 K for one week to allow adsorption to reach equilibrium. The vials were shaken periodically in the course of the week to accelerate the surfactant adsorption. After one week, several milliliters of surfactant solution were extracted from the lower portion of the vial using 6 mL syringes. To avoid hydrate melting, sample extraction inside the chiller usually lasted for less than a minute. The SDS concentrations in solutions prepared before adsorption and solutions extracted after adsorption were analyzed. Thus, the difference between the concentrations of the supernatant before and after the adsorption experiment gives the amount of SDS adsorbed onto the hydrate–water interface.
2.4. SDS solubility in NaCl solutions A 450 mL high-pressure stirred reactor customized by Parr Instruments was used for the solubility measurement. The reactor temperature was controlled by circulating a coolant from an Isotemp 3006P thermostat (Fisher Scientific) with a stability of ±0.01 K within the jacket around the cell. Two type-T thermocouples (Omega Engineering) were used to monitor the reactor temperature with one immersed in the solution and the other placed in the reactor head space. The precision of the temperature measurement is ±0.5 K. The temperature of the liquid and gas phase was sampled every 20 s by Labview interface. 17.3 mM SDS solutions were prepared with 1, 10, and 25 mM NaCl solutions. About 250 mL of one of these solutions, at a time, was charged into the high-pressure stirred reactor. Then the reactor was pressurized up to 7.5 MPa of natural gas to maintain meta-stable regions of natural gas hydrate formation as described in our previous work [8]. The temperature of the reactor was first set to 274 K and was kept at this temperature for about 12 h to allow the system reach equilibrium after which samples are collected. The same procedure was observed for the temperatures of 275, 277, 281, 284, and 287 K for all of the NaCl solutions. The SDS solubility is also determined by liquid–liquid titrations as described below. 2.5. Liquid–liquid titrations One milliliter of SDS solution of unknown concentration was pipetted into a 20 mL test-tube followed by the addition of 2.5 mL of methylene blue solution (0.003 wt.% methylene blue, 1.2 wt.% H2SO4 and 5 wt.% Na2SO4) and 2.5 mL chloroform. After the test-tube was vigorously shaken, hyamine (titrant) was added to the tube. Before the introduction of hyamine into the test-tube, the upper layer of the solution in the test-tube was clear while the lower layer was blue. The end-point was reached when the lower layer and the upper layer had the same blue color. This titration procedure was adapted from Epton [16]. The accuracy of concentration measured using this analysis was within 2% from triplicate measurements. 2.6. Zeta potential measurement 10 g of solution was charged to 25 mL vials and the vials were transferred into the chiller at 275 K for about 12 h. 1 g of CP hydrate slurry was quickly transferred into the vial. The vial was tightly sealed and then returned into the chiller at 275 ± 0.2 K for one week to allow adsorption to reach equilibrium. CP hydrate particles suspended in aqueous solution were quickly transferred into a chill folded capillary cell using a chill pipette. The folded capillary cell was loaded into the zeta potential machine (Zetasizer Nano ZS, Malvern Instruments) at 277 K and then zeta potential data were recorded. 3. Results and discussion It was observed that addition of a small amount of NaCl (1 mM) into 20 ppm SDS aqueous phase enhances methane hydrate formation [11]. We conjecture concurrently with our previous work [10,17,18] that the adsorption of SDS onto the hydrate–water interface is a main cause for accelerated hydrate formation and that the presence of salt can affect the SDS adsorption behavior. 3.1. Proposed adsorption scheme Fig. 1 proposes the SDS adsorption scheme onto the CP hydrate– water interface in the presence of NaCl and also present the
335
O. Salako et al. / Journal of Colloid and Interface Science 386 (2012) 333–337
0 mM NaCl 1 mM NaCl 10 mM NaCl 25 mM NaCl
0.025
Adsorption (mM/g)
0.020
0.015
0.010
0.005
0.000 0.0
0.5
1.0
1.5
2.0
2.5
3.0
SDS Equilibrium Conc (mM) Fig. 1. Adsorption isotherms of SDS with 0, 1, 10 and 25 mM NaCl solutions and proposed adsorption schematics. The circles in the schematics represent Cl ions.
-30 -40 -50
Zeta Potential (mV)
adsorption isotherm of SDS at different salt concentrations. At the initial stages of adsorption, before the monolayer coverage (0– 0.005 mM SDS/g of hydrate), depicted by (1) in Fig. 1, the DS ions adsorb onto the CP hydrate–water interface with their tails in a ‘‘lying down’’ configuration because the SDS tail has hydrophobic interaction with guest molecules (CP) while the headgroup form hydrogen bonds with pendant hydrogen of water cages [10,18]. The location of Na+ ions is assumed to be close to DS and Cl ions as shown in a previous work [19] and Cl ions compete with DS ions for adsorption sites. Fig. 1(2) depicts a pseudo-monolayer coverage (at approximately 0.01 mM SDS/g of hydrate) with the tails of the adsorbed DS ions still lying down at the interface with some of available hydrogen bonding sites occupied. Since the monolayer saturation appears in a narrow SDS concentration between 0.75 and 1.5 mM as shown in Fig. 1, the pseudo monolayer may easily steer to bi-layer formation to stabilize the adsorption process with headgroups exposed to the bulk aqueous phase. At the beginning of the bi-layer formation (0.012–0.016 mM SDS/g of hydrate) depicted by (3), the tails of DS ions in the bulk and the tails of the DS ions adsorbed on the CP hydrate–water interface begin to interact, resulting in the tails of the adsorbed DS ions having a ‘‘stand-up’’ configuration [10,18]. The vertical positioning of the tails that were previously in ‘‘lying down’’ configurations on the CP hydrate–water interface exposes more DS ions to hydrogen bonding sites at the CP hydrate–water interface. Fig. 1(4) depicts the completion of bi-layer formation (0.018 mM/ g of hydrate). The isotherms quantify the SDS adsorption with different NaCl concentrations. From the adsorption isotherms, the effect of NaCl on SDS adsorption and the role of SDS in enclathration can be better understood. Lo et al. [10] determined the adsorption isotherm of SDS onto the CP hydrate–water interface without any salt and they identified the type of isotherm as Langmuir–Step (L–S) type with the amount of SDS adsorbed at the first step of the L isotherm and the second step of the S-isotherm reported as 0.01 mM and 0.02 mM SDS/g of CP hydrate, respectively. Fig. 1 shows that the SDS isotherms with 1 and10 mM NaCl solutions still follow the L–S type adsorption behavior. However, the shape of isotherm is not clearly pictured from Fig. 1. For example, the 1 mM NaCl case levels off at approximately 0.018 mM SDS/g CP hydrate. The isotherm might be either L-type or L–S type. The L-type signifies that the adsorption is one step (monolayer step) while the L–S type signifies that the adsorption is in two steps (monolayer and bilayer of SDS). In order to determine the
-60 -70 -80 -90 -100 -110 -120 0
1
2
3
4
SDS Concentration (mM) Fig. 2. Zeta potential measurements of CP hydrate particles in the presence of 1 mM NaCl at different SDS concentrations.
shape of the isotherm, we analyze the changes in the surface charge of CP hydrate particles in the presence of 1 mM NaCl at different SDS concentrations using zeta potential measurements. The negative surface charge of CP hydrate particles in the presence of 1 mM NaCl (at 0 mM SDS) can be attributed to the adsorption of Cl ions onto the CP hydrate–water interface as shown in Fig. 2. At relatively low SDS concentration (0.10–0.5 mM SDS), the zeta potential of CP hydrate particles linearly increases as shown in Fig. 2. At these low SDS concentrations, the adsorption sites occupied by Cl ions are displaced by DS ions. This is because DS ions adsorb more favorably than Cl , which is indicated by higher zetapotential values at the same concentrations for SDS and NaCl. For example, the 1 mM SDS case in Fig. 2 gives more negative values ( 50 to 55 mV) than the 1 mM NaCl case in Fig. 3 ( 32 mV). The same applies for both of 4 mM cases (4 mM SDS: 80 mV in Fig. 2 versus 4 mM NaCl in Fig. 3: 35 mV). Thus, DS ions can replace Cl ions from adsorption sites previously occupied by Cl ions in the low SDS concentrations of Fig. 2. The zeta potential of CP hydrate particles stays approximately constant for SDS concentrations from 0.50 to 1.70 mM in Fig. 2. This region of SDS concentration can be interpreted as a monolayer
336
O. Salako et al. / Journal of Colloid and Interface Science 386 (2012) 333–337
-25
Zeta Potential (mV)
-35 -40 -45
SDS Concentration (mM)
0 mM NaCl 1 mM NaCl 3 mM NaCl 4 mM NaCl 6 mM NaCl 8 mM NaCl 10 mM NaCl
-30
-50 -55 -60
20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1
1 mM NaCl 10 mM NaCl 25 mM NaCl
272
-65 0
2
4
6
8
274
10
276
278
280
282
284
286
288
Temperature (K)
NaCl Concentration (mM) Fig. 3. Zeta potentials of CP hydrate particles at different NaCl concentrations without any SDS. Zeta-potential measurements above 8 mM NaCl are difficult to perform due to the reaction between the cell electrode and Cl . The electrode turns to brown (refer to pictures in the supplementary data).
step. The zeta potential again increases around 2.5 mM SDS and stays fairly constant from 3.30 to 4.00 mM SDS concentration. This second plateau in Fig. 2 can be interpreted as a bi-layer coverage at 1 mM NaCl. Therefore, we can infer from Fig. 2 that the isotherm in Fig. 1 is L–S type isotherms for 1 mM NaCl. However, for the 10 mM and the 25 mM cases in shown in Fig. 1, they seem to form the SDS monolayer but not to complete the bi-layer. The case of 1 mM NaCl shows relatively higher SDS adsorption amount as compared to the 0, 10 mM, and 25 mM NaCl cases shown in Fig. 1. The monolayer coverage at the CP hydrate–water interface for all of the isotherms is observed at 0.01 mM (SDS)/g of hydrate. However, the monolayer formation delays at the SDS equilibrium concentration from 0.75 to 1.0 mM SDS for 1 mM NaCl and 1.25– 1.5 mM SDS for 10 mM and 25 mM NaCl. This shift in the equilibrium SDS concentration is insignificant compared to one order of magnitude increase in the NaCl concentration. It is an indicator of ion exchange strength between Cl and DS of which DS is the greater. In other words, DS is more favorably adsorbed onto the interface than Cl as also evidenced in the zeta-potential measurements in Figs. 2 and 3. Fig. 1 shows bi-layer coverage at the hydrate–water interface for 0 and 1 mM NaCl curves occurs at 0.018 mM (SDS)/g of hydrate. However, the isotherm with 10 mM NaCl reaches the solubility limit of SDS (2.5 mM SDS at 275 K as shown in Fig. 4) as the bilayer coverage of SDS is approached, and the bi-layer is not completely formed. For the 25 mM NaCl case, the SDS solubility is around 1.5 mM in Fig. 4 and thus the bi-layer is not formed as shown in Fig. 1. 3.2. Proposed adsorption mechanism DS ions adsorb onto the hydrate–water interface dominantly via hydrogen bonding [10,17,18]. Although the surface area of the CP hydrates is presently unknown, the number of hydrogen bonding sites per gram of hydrate is assumed to be equal in the hydrate particles. Therefore, the adsorption amount for the monolayer coverage is statistically the same for all the NaCl concentrations studied in this experiment (0.01 mM (SDS)/g of hydrate in Fig. 1), despite the higher adsorption amount observed for 1 mM NaCl. The only difference is that the equilibrium SDS concentrations (x-axis in Fig. 1) corresponding to the monolayer coverage increase due to the
Fig. 4. SDS solubility at low hydrate-forming temperatures and different NaCl solutions.
competitive adsorption between DS with Cl as the NaCl concentration moves up from 1 to 25 mM. The reasons why the 1 mM NaCl case has an adsorption amount of DS ion higher than that for all NaCl concentrations are (1) the competitive adsorption between DS and Cl is not serious due to the low NaCl concentration and (2) at the same time, Na+ ions may provide charge shielding for the repulsion of DS ions at the CP hydrates–water interface. It should be noted that evidence of Cl ion adsorption on the ice– water interface was previously reported [20]. Then, this work firstly presents the adsorption of Cl ions onto the hydrate–water interface as shown in the zeta-potential measurements of Figs. 2 and 3. At NaCl concentrations of 150 mM or above, much higher than those used in this study, there was evidence that NaCl solutions shrink the size of DS ion head group [19]. Despite this evidence of shrinking DS ion head group size, the amount DS ions adsorbed on the hydrate–water interface at the monolayer coverage remains the same for all of the NaCl concentrations studied in this work as mentioned before. It could be hypothesized that the NaCl concentrations employed in our experiments are too low to significantly shrink the size of DS ion head group. Thus, a main driving force for adsorption of DS– ions at the CP hydrate–water interface before the monolayer coverage is hydrogen bonding [17,18]. Otherwise, the shrinkage of SDS head group due to Na+ ions should change the amount of DS– ions adsorbed at the monolayer or bilayer coverage with different NaCl concentrations.
4. Conclusions This work has investigated the effect of NaCl on adsorption of SDS at the CP hydrate–water interface and analyzed the adsorption behavior using adsorption isotherms, zeta-potential, and SDS solubility. The zeta-potential information supports the shape of L–S type isotherm at 1 mM NaCl, while the SDS solubility data were used to ensure that the isotherms are obtained within the SDS solubility limitation. The highest amount of SDS adsorption was observed for the 1 mM NaCl case at the same SDS bulk equilibrium concentration, which accounts for fast methane hydrate formation in our previous study. The same L–S type isotherm was observed for SDS adsorption in all of the NaCl concentrations except for 25 mM NaCl, in which case the bilayer is not formed due to the low solubility of SDS. Although some previous studies have shown that the head of DS ions shrinks due to the presence of NaCl, it does not have significant effect on the amount of DS ions
O. Salako et al. / Journal of Colloid and Interface Science 386 (2012) 333–337
adsorbed at both the monolayer and the bi-layer coverage for NaCl concentrations up to 25 mM; thus making hydrogen bonding the dominant interaction between SDS and water–hydrate interface. Acknowledgment The authors are grateful for the support of the National Science Foundation for this work (under Grant Numbers CBET-0854210 and HRD-0833180). This research was also made possible in part by a grant from BP/GoMRI. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2012.07.017. References [1] E.D. Sloan, C. Koh, Clathrate Hydrates of Natural Gas, third ed., CRC Press, Boca Raton, FL, 2008. [2] Y. Zhong, R.E. Roger, Chem. Eng. Sci. 55 (2000) 4175–4187.
337
[3] P. Englezos, N. Kalogerakis, P.D. Dholabhai, P.R. Bishnoi, Chem. Eng. Sci. 42 (1987) 2659–2666. [4] W.X. Wang, C.L. Bray, D.J. Adams, A.I. Cooper, J. Am. Chem. Soc. 230 (2008) 11608–11609. [5] W.X. Wang, B.O. Carter, C.L. Bray, A. Steiner, J. Bacsa, J.T. Jones, C. Croper, Y.Z. Khimyak, D.J. Adams, A.I. Cooper, Chem. Mater. 21 (2009) 3810–3815. [6] K. Watanabe, S. Imai, Y.H. Mori, Chem. Eng. Sci. 60 (2005) 4846–4857. [7] P. Gayet, C. Dicharry, G. Marion, A. Graciaa, J. Lachaise, A. Nesterov, Chem. Eng. Sci. 60 (2005) 5751–5758. [8] J.S. Zhang, S.Y. Lee, J.W. Lee, J. Colloid Interface Sci. 315 (2007) 313–318. [9] J.S. Zhang, S.Y. Lee, J.W. Lee, Ind. Eng. Chem. Res. 46 (2007) 6353–6359. [10] C. Lo, J.S. Zhang, J. Phys. Chem. C 114 (2010) 13385–13389. [11] S. Zhang, J.A. Salera, J.W. Lee, Ind. Eng. Chem. Res. 49 (2010) 8267–8270. [12] G.N. Werezak, Chem. Eng. Prog. Symp. Ser. 65 (1969) 6–18. [13] Y. Zhang, P.G. Debenedetti, R.K. Rrud’homme, B.A. Pethica, J. Phys. Chem. B 108 (2004) 16717–16722. [14] J.S. Zhang, J.W. Lee, J. Chem. Eng. Data 54 (2009) 659–661. [15] M. Nakajima, R. Ohmura, Y.H. Mori, Ind. Eng. Chem. Res. 47 (2008) 8933–8939. [16] S.R. Epton, Trans. Faraday Soc. 44 (1948) 226–230. [17] J.S. Zhang, C. Lo, P. Somasundaran, S. Lu, A. Couzis, J.W. Lee, J. Phys. Chem. C 112 (2008) 12381–12385. [18] C. Lo, J.S. Zhang, P. Somasundaran, A. Couzis, J.W. Lee, Langmuir 24 (2008) 12723–12726. [19] G.P. Funkhomer, M.P. Arevalo, D.T. Glatzhofer, E.A. O’Reart, Langmuir 11 (1995) 1443–1447. [20] J. Drzymala, Z. Sadowski, L. Holysz, E. Chibowski, J. Colloid Interface Sci. 220 (1999) 229–234.