Journal of Colloid and Interface Science 412 (2013) 1–6
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Adsorption of Gemini surfactants onto clathrate hydrates O. Salako a, C. Lo a,b, A. Couzis a, P. Somasundaran a, J.W. Lee a,c,⇑ a
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 c Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon 305-701, Republic of Korea b
a r t i c l e
i n f o
Article history: Received 8 July 2013 Accepted 3 September 2013 Available online 17 September 2013 Keywords: Adsorption Gemini surfactants Clathrate hydrates
a b s t r a c t This work addresses the adsorption of two Gemini surfactants at the cyclopentane (CP) hydrate–water interface. The Gemini surfactants investigated here are Dowfax C6L and Dowfax 2A1 that have two anionic head groups and one hydrophobic tail group. The adsorption of these surfactants was quantified using adsorption isotherms and the adsorption isotherms were determined using liquid–liquid titrations. Even if the Gemini surfactant adsorption isotherms show multi-layer adsorption, they possess the first Langmuir layer with the second adsorption layer only evident in the 2A1 adsorption isotherm. Zeta potentials of CP hydrate particles in the surfactant solution of various concentrations of Dowfax C6L and Dowfax 2A1 were measured to further explain their adsorption behavior at the CP hydrate–water interface. Zeta potentials of alumina particles as a model particle system in different concentrations of sodium dodecyl sulfate (SDS), Dowfax C6L and Dowfax 2A1 were also measured to confirm the configuration of all the surfactants at the interface. The determination of the isotherms and zeta-potentials provides an understanding framework for the adsorption behavior of the two Gemini surfactants at the hydrate–water interface. Ó 2013 Elsevier Inc. All rights reserved.
1. Introduction Clathrate hydrates are a non-stoichiometric, ice-like crystalline compound. They are formed when host water molecules provide three-dimensional cages for trapping guest low molecular weight molecules such as methane, ethane, carbon dioxide, and cyclopentane (CP) [1]. Since the discovery of large deposits of natural gas hydrates in oceans and permafrost areas, there has been a spike in studies of clathrate hydrate [1]. It has been estimated that the amount of natural gas trapped in the natural gas hydrate deposits is approximately 1.5 1016 m3 [2]. This amount makes natural gas hydrates a potentially viable energy resource if properly recovered. Clathrate hydrates can be also used a gas storage medium because natural gas can be stored in the clathrate hydrate form; up to 170 volumes of natural gas per volume of hydrates [1,3]. Other potential applications of clathrate hydrates are for CO2 separation from flue gases and desalination of seawater [4]. There are negative aspects to clathrate hydrates as they can form inside oil and gas pipelines, which cause the nuisance of continuous oil/gas production.
⇑ Corresponding author. Address: Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology (KAIST), 291 Daehak-ro, Yuseong-gu, Daejeon 305-701, Republic of Korea. Fax: +82 42 350 3910. E-mail address: [email protected] (J.W. Lee). 0021-9797/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcis.2013.09.007
Understanding the kinetics of clathrate hydrate formation is essential for both the utilization of the positive aspects of clathrate hydrate technology and the risk management of its negative aspect of hydrate blockage in the pipeline. One major obstacle to applying hydrate technology for gas storage is that clathrate hydrate formation is very slow because of the formation of a thin layer of hydrate at the gas/water interface which effectively blocks the diffusion of the guest molecule into the water phase; hence, ending the process of hydrate formation [5]. The kinetics of hydrate formation can be accelerated through mechanical agitation or addition of surfactants to the reaction system [3,6–8]. The high energy cost of stirring in a large reactor system makes the method of agitation less economically favorable when compared to addition of surfactants that is very effective for promoting hydrate formation even at low surfactant concentrations. Several studies [2] have shown that sodium dodecyl sulfate (SDS) is a very effective promoter of the kinetics of hydrate formation. Adsorption of surfactants at the hydrate–water interface was proposed by Zhang et al. [9] to be the reason for surfactant’s ability to accelerate hydrate formation. In a previous work, Lo et al. [10] studied the adsorption of SDS on cyclopentane (CP) hydrate–water interface and proposed a two-step adsorption mechanism of SDS at the CP hydrate–water interface. Salako et al. [11] investigated the effect of salt on the adsorption of SDS on the CP hydrate–water interface and proposed a pseudo-monolayer adsorption mechanism at the Langmuir adsorption range. It was shown that Gemini
2
O. Salako et al. / Journal of Colloid and Interface Science 412 (2013) 1–6
surfactants could also be good promoters of hydrate formation kinetics [12–16]. However, the adsorption of Gemini surfactants has never been investigated at hydrate–water interface and understanding their adsorption behaviors at the interface is very important to accelerate the formation kinetics. Gemini surfactant, also known as dimeric surfactants [17,18], are a relatively new class of surfactants. Gemini surfactants are composed of two identical conventional surfactants joined at the head groups or very close to the head groups by a spacer group [19–22]. Gemini surfactants have attracted lots of attention because their critical micelle concentrations (CMC) are much lower than the CMC of conventional single head, single-tail surfactant [18,23]. Also, Gemini surfactants are more efficient in reducing surface tension than their corresponding conventional surfactants [24,25]. In this work, we will study the adsorption of two Dowfax Gemini surfactants at the CP hydrate–water interface by determining adsorption isotherms and zeta-potentials. The two Gemini surfactants studied in this work are Dowfax 2A1 and Dowfax C6L. These surfactants are asymmetric Gemini surfactants [22] because they have two head groups and one hydrophobic tail. The isotherms will be determined by liquid–liquid titrations. We will also investigate the change in the surface charge of CP hydrate particles in different Gemini surfactant concentrations in order to have a better understanding of the adsorption isotherms and the configuration of the surfactants at the CP hydrate–water interface. Since very little is known about the surface properties of clathrate hydrate in the literature, we will use alumina particles as a model system to understand the orientation of Dowfax 2A1 and Dowfax C6L at the CP hydrate–water interface.
2.3. Adsorption isotherms A 10 g of Gemini surfactant solutions were charged into 25 mL vials after which the vials were quickly transferred into a chiller at 275 K for about 12 h. 10 g of CP hydrate slurry was quickly transferred into the vial, tightly sealed and then returned into the chiller one after the other. The vials were left in the chiller for a week to allow surfactant adsorption to reach equilibrium. The vials were periodically shaken to accelerate the adsorption of surfactants. At the end of 1 week, several milliliters of the Gemini surfactant solution were extracted from the lower part of the vials with a syringe. The sample extraction usually lasts for less than 1 min and it is done inside the chiller to avoid CP hydrate melting. The Gemini surfactant solutions were analyzed by taking the difference between the concentration of the surfactant solution before and after the addition of CP hydrate particles. This difference gives the amount of Gemini surfactant adsorbed per gram of CP hydrate particles. 2.4. Liquid–liquid titrations One milliliter of both initial Gemini surfactant solution and extract was pipetted into different 20 mL test-tubes followed by the addition of 2.5 mL methylene blue solution (0.003 wt.% Methylene Blue, 1.2 wt.% H2SO4 and 5 wt.% Na2SO4) and 2.5 mL chloroform. The test-tube was vigorously shaken till the contents of the testtube split into two phases. The upper part of the content in the test-tube is clear and the lower part of the test-tube is blue. Hyamine solution (titrant) was added to the test-tube in drops till both the top and the bottom phase have the same shade of blue color. This signifies the endpoint. The triplicate measurements done with this procedure have an error within 2%.
2. Experimental section 2.1. Materials and methods Anionic Gemini surfactants of Dowfax 2A1 (>98% surfactant solution) and Dowfax C6L (>97% surfactant solution) were obtained from Dow Chemicals. Sodium dodecyl sulfate (SDS) and cyclopentane (CP) of 99% purity were purchased from Sigma–Aldrich. Methylene blue with indicator purity was purchased from Sigma–Aldrich. The deionized (D.I.) water used for the experiment has a resistivity of 18 MX cm 1. The sodium sulfate and sulfuric acid, with purity of 99% and 96% respectively, were purchased from Sigma–Aldrich. 100 nm AG alumina powder was purchased from Ocean State Abrasive.
2.2. Preparation of CP slurry A 300 mL of CP and water mixture (10 wt.% CP) was charged into a 1 l bottle; after which the bottle was tightly sealed, vigorously shaken and then transferred into a freezer set at 263 K. After the ice formation is noticed at the interface of CP and water, the bottle was shaken at ambient conditions to melt the ice. As the ice begins to melt, CP is quickly enclathrated. The appearance of whitish particles confirms the beginning of CP hydrate formation. The bottle was quickly transferred into the chiller for at least a week, during which the bottle was shaken at least five times per day to accelerate enclathration and ensure complete formation. Calorimetric measurements showed that the concentration of CP hydrate in the CP hydrate slurry is 51 wt.%. We used the CP hydrate slurry for this experiment in order to minimize variations in CP hydrate particle surface area during adsorption isotherm and zeta potential experiments and to avoid moisture condensation on hydrates.
2.5. Zeta potential measurements A 10 g of surfactant solutions were added to 25 mL vials and the vials were transferred into a chiller at 275 ± 0.2 K for about 12 h after which 1 g of CP hydrate was quickly added to the vial. The vials was then tightly sealed and quickly transferred into the chiller for 1 week to allow surfactant adsorption to reach equilibrium. For the zeta potential reading, the CP hydrate particles from the chilled vials were quickly transferred into a chilled folded capillary cell with a chilled pipette. The capillary cell was then loaded into the zeta potential machine (Zetasizer Nano ZS, Malvern Instrument) that was preset to a temperature of 277 K; after which the zeta potential measurements were recorded. For the zeta potential of alumina particles, 0.05 g of alumina particles was added to vials containing 10 g of surfactant solution. Each vial was well shaken and stored under ambient conditions for a day to allow adsorption to completely reach equilibrium. The zeta potential measurements for alumina particles were taken at 298 K. 3. Results and discussion Gemini surfactants used for this study are Dowfax C6L and Dowfax 2A1 whose two head groups are negatively charged (Fig. 1) with one hydrophobic tail. They have been shown to increase the rate and amount of hydrate enclathration and some Gemini surfactants could be as good as sodium dodecyl sulfate (SDS) in promoting hydrate enclathration [12]. Zhang et al. [9] proposed that the adsorption of SDS at the hydrate–water interface is the explanation behind SDS’s ability to accelerate hydrate enclathration. Therefore, it is important for us to study the adsorption behavior of Gemini surfactants at hydrate–water interface.
O. Salako et al. / Journal of Colloid and Interface Science 412 (2013) 1–6
Fig. 1. General structural formula for Dowfax C6L and Dowfax 2A1.
The mechanism of SDS adsorption at the CP hydrate–water interface was firstly proposed by Lo et al. [10]. Salako et al. [11] also proposed a mechanism for the adsorption of SDS at the CP hydrate–water interface in the presence of NaCl. This work focuses on investigating the adsorption behavior of the two anionic Gemini surfactants at CP hydrate–water interface. We will also use alumina as a model system for CP hydrate particles to explain critical parts of the adsorption isotherm and to understand the orientation of these Gemini surfactants at the CP hydrate–water interface. 3.1. Adsorption isotherms Fig. 2 shows the adsorption isotherms of Dowfax 2A1, Dowfax C6L and SDS from our previous study [11]. The isotherms are used to quantify the equilibrium amount of the surfactants adsorbed at the CP hydrate–water interface. Monolayer coverage seems to occur at approximately 0.010 mM/g and 0.015 mM/g for Dowfax 2A1 and Dowfax C6L, respectively. Both Dowfax 2AI and Dowfax C6L show a short first monolayer like a previous work on SDS adsorption and tend to form so-called pseudo-monolayer [11]. Dowfax 2A1 shows a second plateau at approximately 0.03 mM/g but the second plateau is not evident in the DowfaxC6L adsorption isotherm. Both adsorption isotherms do not show a final plateau in the range of concentration investigated in this work and are types of multi-layer adsorption. The adsorption amount of Dowfax C6L is consistently higher than that of Dowfax 2A1 surfactant. The first monolayer at 0.01 mM/g of CP hydrate in Fig. 2 corresponds to the pseudo-monolayer level for Dowfax 2A1 at the CP hydrate–water interface. The pseudo-monolayer level for Dowfax 2A1 is the same as that of SDS in our previous study [10,11] and the equilibrium concentration of Dowfax 2A1 (x-axis) at which the monolayer occurs also falls in the same range as that of SDS. The second plateau in Fig. 2, at approximately 0.03 mM Dowfax 2A1/
0.12
Dowfax 2A1 Dowfax C6L SDS
Surfactant Adsorbed (mM/g)
0.11 0.10 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00 -0.01
0
1
2
3
4
5
Equilibrium Concentration (mM) Fig. 2. Adsorption isotherms of Dowfax 2A1, Dowfax C6L and SDS [11].
3
g of CP hydrate, is higher than that of SDS at approximately 0.018 mM SDS/g of CP hydrate [10,11]. It is very interesting to see that both Dowfax 2A1 and SDS with the same hydrophobic tail length but different number of hydrophilic head group (Dowfax 2A1 with two head groups and SDS with one head group) form a pseudo-monolayer at the same monolayer coverage and at the same range equilibrium surfactant concentration. We conjectured that two head groups of each Dowfax surfactant would adsorb at the CP hydrate interface. But our adsorption isotherms reveal that this is not the case because if the two head groups in each surfactant adsorb at the interface, the isotherm amount should be half of that of SDS for a given number of adsorption sites. As alluded to in Lo et al. [10] and Salako et al. [11], adsorption of anionic surfactant at the CP hydrate–water interface is possible via hydrogen bonding between the head group of the surfactant monomers and the pendant hydrogen atoms of CP hydrate particles. For both heads of the Gemini surfactant to adsorb at the CP hydrate–water interface, the distance between two adsorption sites must approximately equal to the distance of separation between the two head groups in one surfactant molecule. If the distance between the adsorption sites is shorter than the distance between the head groups, then the two head groups can only adsorb if the spacer group between the surfactant hydrophilic head groups is flexible. The spacer group in both Dowfax 2A1 and Dowfax C6L is a diphenyl oxide and it is not flexible. At this moment, no information is available about the distance between the adsorption sites at the CP hydrate–water interface but our results imply that it is greater than the length of the spacer group. This is because both SDS and Dowfax 2A1 show the same monolayer level at the same equilibrium concentration range, which only means that one of the head groups is adsorbed at the CP hydrate–water interface. The monolayer level of Dowfax C6L is at approximately 0.015 mM/g CP hydrate and it is a little higher than that of Dowfax 2A1 and SDS. Thus, judging from this monolayer level of the Dowfax C6L, one of its head groups is adsorbed at the interface. In our previous study [11], we postulated that the hydrophobic tail of SDS blocks some adsorption sites at the pseudo-monolayer level and these sites are exposed at the onset of the bilayer formation when the hydrophobic tails of adsorbed SDS begin to interact with the hydrophobic tails of SDS in solution. The Dowfax C6L adsorption isotherm supports our postulate. Dowfax C6L has a hydrophobic tail with a chain length of six carbons. Both Dowfax 2A1 and SDS have a hydrophobic tail that has a chain length of 12 carbons. The hydrophobic tail of Dowfax C6L is not long enough to block as many adsorption sites as the hydrophobic tails of Dowfax 2A1 and SDS, therefore more Dowfax C6L molecules will adsorb at the pseudo-monolayer level when compared to SDS and Dowfax 2A1 molecules. This clearly shows the importance of surfactant hydrophobic tail in the formation pseudo-monolayer at the CP hydrate–water interface. Unlike the SDS isotherm in our previous study [10,11], which levels off at approximately 0.018–0.02 mM/g of CP hydrate in Fig. 2, both Dowfax 2A1 and Dowfax C6L isotherms keep rising without finally leveling off in the range of concentrations tested in our studies although the Dowfax 2A1 adsorption isotherm shows a second plateau at approximately 0.03 mM/g of CP hydrate. According to Yueying et al. [25], if it is assumed that the monolayer is the center of aggregation, then the aggregation number of surfactants will be the ratio of the second surface coverage to the first surface coverage. If the aggregation number is approximately two, then the entire surface coverage can be called a bilayer. The aggregation number for our SDS experiment is approximately two; therefore, we concluded that the second surface coverage of SDS signifies bilayer coverage. The ratio of the second to the first surface coverage in the case of Dowfax 2A1 is approximately 3. This
O. Salako et al. / Journal of Colloid and Interface Science 412 (2013) 1–6
signifies the formation of hemimicelles at the second layer and beyond for Dowfax 2A1 at the CP hydrate–water interface. The lack of evidence of a second plateau of surface coverage for Dowfax C6L reflects the importance of the hydrophobic tail on stabilizing the adsorption process. After the formation of the pseudomonolayer, the hydrophobic tails of adsorbed surfactant will begin to interact with the hydrophobic tails of surfactant in solution. For Dowfax C6L, the hydrophobic tail length is not long enough to prevent considerable interaction between the head group of adsorbed Dowfax C6L and those of Dowfax C6L in solution at the onset of the second layer formation. The second layer for Dowfax C6L is unstable and it is reflected in the adsorption isotherm as there is no evidence of a second plateau in Fig. 2. 3.2. Zeta potentials at the CP hydrate–water interface
Zeta Potential of CP Hydrate Particles (mV)
4
-40 -45 -50 -55 -60 -65 -70 -75 -80 -85 0
1
2
3
4
Dowfax C6L Concentration (mM) Fig. 4. Zeta potential of CP hydrate particles at different Dowfax C6L concentrations.
3.3. Adsorption of SDS, Dowfax 2A1 and Dowfax C6L on alumina particles Very little is known in the literature about the surface property of CP hydrates, so we intended to understand the adsorption of SDS, Dowfax 2A1 and Dowfax C6L by using alumina particles as our model system. For this experiment, we used zeta potential measurements to study the change in the surface charge of alumina particles at different concentrations of SDS, Dowfax 2A1 and Dowfax C6L surfactants. Figs. 5–7 are plots of zeta potential of alumina particles in different surfactant concentrations. All the zeta potential plots show similar trends. All plots have the first and second plateaus. The zeta potential values of the first plateaus are 17.5 mV, 31 mV and 31 mV for SDS, Dowfax 2A1 and Dowfax C6L respectively. These first step values give us a clearer picture of the configuration of these surfactants at the alumina–water interface. These values reinforce our earlier assertion that both Dowfax 2A1 and Dowfax C6L surfactants adsorb at the alumina–water interface with one of their hydrophilic head groups. This will be the case because, just like the adsorption isotherms of SDS, Dowfax 2A1 and
15
-30
Zeta Potential of Alumina Particles (mV)
Zeta Potential of CP Hydrate Particles (mV)
Zeta potential measurements also support our understanding of these adsorption isotherms. The zeta potential measurement of CP hydrate particles in varying concentrations of Dowfax 2A1 (Fig. 3) shows two clear plateaus; the first ranging from approximately 0.6–1.25 mM Dowfax 2A1 and the second plateau is seen from approximately 2.2 mM Dowfax 2A1. Fig. 4 shows the zeta potential measurement of CP hydrate particles at different Dowfax C6L solution concentrations with no clear first plateau and a clear second plateau which is seen from 2.2 to 4.0 mM Dowfax C6L. The zeta potential curves of both Dowfax 2A1 and Dowfax C6L from Figs. 3 and 4 respectively level off at approximately 77.5 mV at a concentration of 2.2 mM. Since the adsorption isotherms for both Dowfax 2A1 and Dowfax C6L keep increasing, charge shielding by forming aggregates must be responsible for the leveling off of these zeta potential plots. The outermost surfactant head and tails of the adsorbed layer shield the innermost surfactant head groups. The other important feature of the zeta potential plots is the clear first plateau for Dowfax 2A1 similar to the SDS zeta potential measurement in our previous study. However, the first plateau in the Dowfax C6L zeta potential plot is not clear corresponding to the short plateau of the isotherm in Fig. 2. This also emphasizes the importance of the hydrophobic tail in stabilizing the pseudo-monolayer before transitioning to the formation of the second layer. In the case of Dowfax C6L, there is no distinction between the formation of the pseudo-monolayer and the onset of the second layer formation because of the short hydrophobic tail length.
-40
-50
-60
-70
-80 0
1
2
3
4
Dowfax 2A1 Concentration (mM)
10 5 0 -5 -10 -15 -20 -25 -30 0
2
4
6
8
SDS Concentration (mM) Fig. 3. Zeta potential of CP hydrate particles at different Dowfax 2A1 concentrations.
Fig. 5. Zeta potential of alumina particles at different SDS concentrations.
O. Salako et al. / Journal of Colloid and Interface Science 412 (2013) 1–6
5
Zeta Potential of Alumina Particles
10
0
-10
-20
-30
-40 0
2
4
6
8
Dowfax 2A1 Concentration (mM)
Zeta Potential of Alumina Particles (mV)
Fig. 6. Zeta potential of alumina particles at different Dowfax 2A1 concentrations.
10
0 Fig. 8. Proposed adsorption schematics for Dowfax 2A1 and Dowfax C6L at the CP hydrate–water interface.
-10
3.4. Proposed adsorption scheme for Gemini surfactant
-20
-30
-40
-50 0
2
4
6
8
Dowfax C6L Concentration (mM) Fig. 7. Zeta potential of alumina Particles at different Dowfax C6L concentration.
Dowfax C6L, the first step for these surfactants all appear in the same concentration range but the zeta potential value at the first step for SDS with one negatively charged head is almost half that of Dowfax C6L and Dowfax 2A1. The zeta potential values for the second step on the zeta potential plots are 25 mV, 40 mV, and 42 mV for SDS, Dowfax 2A1 and Dowfax C6L, respectively above the surfactant concentration of 4.2 mM. The zeta potential of Dowfax 2A1 and Dowfax C6L at the second step cannot be exactly twice as high as that of SDS because of the charge shielding. Also, the zeta potential of Dowfax C6L is slightly higher than that of Dowfax 2A1. This is consistent with Fig. 2 which shows that the amount of Dowfax C6L adsorbed at the CP hydrate–water interface is slightly higher than that of Dowfax 2A1. The zeta potential plot of Dowfax C6L in the case of alumina shows a clear first plateau when compared to the zeta potential plot of Dowfax C6L adsorption at the CP hydrate–water interface. This could mean that the pseudo-monolayer of Dowfax C6L at the alumina–water interface is more densely packed than at the CP hydrate–water interface. The packed pseudo-monolayer will increase hydrophobic interaction between the tails of the adsorbed Dowfax C6L molecules and this will increase the stability of the pseudo-monolayer at the alumina–water interface when compared to the adsorption at the CP hydrate–water interface.
From the results of the adsorption isotherms and zeta potential measurements as described above, we propose an adsorption scheme for Dowfax 2A1 and Dowfax C6L. Fig. 8 shows the schematics of the proposed adsorption scheme for both. At low surfactant concentration, depicted by (1) in Fig. 8, the surfactants adsorb at the CP hydrate–water interface with their head group forming hydrogen bonds with the pendant hydrogen of water cages; while the hydrophobic tail of the surfactant is in a ‘‘lay down’’ configuration because of the hydrophobic interactions between the guest molecule (CP) and the hydrophobic tail of the surfactant. The pseudo-monolayer coverage is depicted by (2) in Fig. 8. At this coverage, more surfactants are adsorbed at the CP hydrate–water interface momentarily covering the entire surface of the CP hydrate particle. The hydrophobic tails of the surfactant at this coverage block some of the vacant adsorption sites at the CP hydrate–water interface. The hydrophobic tails of the adsorbed surfactants begin to interact with the hydrophobic tails of the surfactant in solution. The intermediate surfactant configuration, at the CP hydrate–water interface, between the first and the second plateau is depicted by (3) in Fig. 8 while the surface coverage at the second plateau is depicted by (4). The adsorption isotherm of Dowfax 2A1 shows the second plateau at surface coverage that is three times as large as the first plateau. Since there is no final leveling off of the isotherms within the surfactant concentration range investigated in this work, we conjecture that the surfactant aggregate at the CP hydrate–water interface shown in (5) in Fig. 8 will keep adding surfactant till the surfactant aggregation number of each respective surfactant is reached. At this point, we expect the adsorption isotherms to reach a final plateau. 4. Conclusions This study has investigated the adsorption of two Gemini surfactants, Dowfax 2A1 and Dowfax C6L, at the CP hydrate–water
6
O. Salako et al. / Journal of Colloid and Interface Science 412 (2013) 1–6
interface. The adsorption was analyzed using adsorption isotherms. Zeta potential measurements of CP hydrate particles in the presence of Dowfax 2A1 and Dowfax C6L were used to explain the shape of the adsorption isotherms. We also used alumina particles as a model system for CP hydrate particles to further explore the configuration of the Gemini surfactants used in this study and SDS at the CP hydrate–water interface. Our adsorption isotherms show that more Dowfax C6L adsorbs at the CP hydrate–water interface compared to Dowfax 2A1. The adsorption isotherm of Dowfax C6L only has one clear plateau while that of Dowfax 2A1 has two plateaus. Both adsorption isotherms did not show a final plateau but were a multilayer surface coverage at the CP hydrate–water interface. The zeta potential of alumina particles, used as a model, in the presence of both Gemini surfactants is approximately twice as high as that of SDS at the monolayer coverage in the same surfactant concentration range. This shows that Gemini surfactants used in this study adsorb with only one of their two anionic head groups. Acknowledgments The authors are grateful for the financial support from the Advanced Biomass R&D Center (ABC) as a Global Frontier Project funded by the Ministry of Science, ICT and Future Planning and from ACS–PRF (under a Grant Number of 51991-ND9). This research was also made possible in part by a Grant from BP/GoMRI. The authors also appreciate initial discussions with Ms. Franchette Viloria and Mr. Mohammad Mahmud at CCNY. References [1] E.D. Sloan, C. Koh, Clathrate Hydrates of Natural Gas, third ed., CRC Press, Boca Raton, FL, 2008 .
[2] Y.F. Makogon, S.A. Holditch, T.Y. Makogon, J. Pet. Sci. Eng. 56 (2007) 14–31. [3] Y. Zhong, R.E. Roger, Chem. Eng. Sci. 55 (2000) 4175–4187. [4] C.A. Koh, E.D. Sloan, A.K. Sum, D.T. Wu, Annu. Rev. Chem. Biomol. Eng. 2 (2011) 237–257. [5] E.D. Sloan Jr., Clathrate Hydrates of Natural Gas, second ed., Marcel Dekker, 1998. [6] P. Englezos, N. Kalogerakis, P.D. Dholabhai, P.R. Bishnoi, Chem. Eng. Sci. 42 (1987) 2659–2666. [7] W.X. Wang, C.L. Bray, D.J. Adams, A.I. Cooper, J. Am. Chem. Soc. 230 (2008) 11608–11609. [8] 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. [9] J.S. Zhang, S.Y. Lee, J.W. Lee, Ind. Eng. Chem. Res. 46 (2007) 6353–6359. [10] C. Lo, J.S. Zhang, A. Couzis, P. Somasundaran, J.W. Lee, J. Phys. Chem. C 114 (2010) 13385–13389. [11] O. Salako, C. Lo, J.S. Zhang, A. Couzis, P. Somasundaran, J.W. Lee, J. Colloid Interface Sci. 386 (2012) 333–337. [12] C. Lo, The role of surface active agents on hydrate formation, PhD Thesis, The City College of New York, New York, New York, 2011. [13] Y.A. Kwon, J.M. Park, K.E. Jeong, C.U. Kim, T.W. Kim, H.J. Chae, S.Y. Jeong, J.H. Yim, Y.K. Park, J.D. Lee, J. Ind. Eng. Chem. 17 (2011) 120–124. [14] B.M. Lee, H.C. Kang, J.M. Park, J.H. Yoon, US 6,392,064 B2, 2002. [15] B.M. Lee, H.C. Kang, J.M. Park, KR 0578716 B1, 2006. [16] R. Zana, Curr. Opin. Colloid Interface Sci. 1 (1996) 566–571. [17] M.J. Rosen, J. Chemtech. (1993) 30–33. [18] R. Zana, J. Colloid Interface Sci. 248 (2002) 203–220. [19] S.K. Hait, S.P. Moulik, Curr. Sci. 82 (2002). No. 9. [20] F.M. Menger, C.A. Littau, J. Am. Chem. Soc. 115 (1993) 10083–10090. [21] S. Manne, T.E. Schaffer, Q. Huo, P.K. Hansma, D.E. Morse, G.D. Stucky, I.A. Aksay, Langmuir 13 (1997) 6382–6387. [22] Y.I. Rabinovich, J.R. Kanicky, S. Pandey, H. Oskarsson, K. Holmberg, B.M. Moudgil, D.O. Shah, J. Colloid Interface Sci. 288 (2005) 583–590. [23] F. Devinsky, I. Lacko, F. Bittererova, L. Tomeckova, J. Colloid Interface Sci. 114 (1986) 314–322. [24] Y.P. Zhu, A. Masuyama, Y. Kobata, Y. Nakatsuji, M. Okahara, M.J. Rosen, J. Colloidal Interface Sci. 158 (1993) 40–45. [25] Y. Gao, J. Du, T. Gu, J. Chem. Soc., Faraday Trans. 1 (83) (1987) 2671–2679.
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Adsorption of Gemini surfactants onto clathrate hydrates O. Salako a, C. Lo a,b, A. Couzis a, P. Somasundaran a, J.W. Lee a,c,⇑ a
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 c Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon 305-701, Republic of Korea b
a r t i c l e
i n f o
Article history: Received 8 July 2013 Accepted 3 September 2013 Available online 17 September 2013 Keywords: Adsorption Gemini surfactants Clathrate hydrates
a b s t r a c t This work addresses the adsorption of two Gemini surfactants at the cyclopentane (CP) hydrate–water interface. The Gemini surfactants investigated here are Dowfax C6L and Dowfax 2A1 that have two anionic head groups and one hydrophobic tail group. The adsorption of these surfactants was quantified using adsorption isotherms and the adsorption isotherms were determined using liquid–liquid titrations. Even if the Gemini surfactant adsorption isotherms show multi-layer adsorption, they possess the first Langmuir layer with the second adsorption layer only evident in the 2A1 adsorption isotherm. Zeta potentials of CP hydrate particles in the surfactant solution of various concentrations of Dowfax C6L and Dowfax 2A1 were measured to further explain their adsorption behavior at the CP hydrate–water interface. Zeta potentials of alumina particles as a model particle system in different concentrations of sodium dodecyl sulfate (SDS), Dowfax C6L and Dowfax 2A1 were also measured to confirm the configuration of all the surfactants at the interface. The determination of the isotherms and zeta-potentials provides an understanding framework for the adsorption behavior of the two Gemini surfactants at the hydrate–water interface. Ó 2013 Elsevier Inc. All rights reserved.
1. Introduction Clathrate hydrates are a non-stoichiometric, ice-like crystalline compound. They are formed when host water molecules provide three-dimensional cages for trapping guest low molecular weight molecules such as methane, ethane, carbon dioxide, and cyclopentane (CP) [1]. Since the discovery of large deposits of natural gas hydrates in oceans and permafrost areas, there has been a spike in studies of clathrate hydrate [1]. It has been estimated that the amount of natural gas trapped in the natural gas hydrate deposits is approximately 1.5 1016 m3 [2]. This amount makes natural gas hydrates a potentially viable energy resource if properly recovered. Clathrate hydrates can be also used a gas storage medium because natural gas can be stored in the clathrate hydrate form; up to 170 volumes of natural gas per volume of hydrates [1,3]. Other potential applications of clathrate hydrates are for CO2 separation from flue gases and desalination of seawater [4]. There are negative aspects to clathrate hydrates as they can form inside oil and gas pipelines, which cause the nuisance of continuous oil/gas production.
⇑ Corresponding author. Address: Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology (KAIST), 291 Daehak-ro, Yuseong-gu, Daejeon 305-701, Republic of Korea. Fax: +82 42 350 3910. E-mail address: [email protected] (J.W. Lee). 0021-9797/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcis.2013.09.007
Understanding the kinetics of clathrate hydrate formation is essential for both the utilization of the positive aspects of clathrate hydrate technology and the risk management of its negative aspect of hydrate blockage in the pipeline. One major obstacle to applying hydrate technology for gas storage is that clathrate hydrate formation is very slow because of the formation of a thin layer of hydrate at the gas/water interface which effectively blocks the diffusion of the guest molecule into the water phase; hence, ending the process of hydrate formation [5]. The kinetics of hydrate formation can be accelerated through mechanical agitation or addition of surfactants to the reaction system [3,6–8]. The high energy cost of stirring in a large reactor system makes the method of agitation less economically favorable when compared to addition of surfactants that is very effective for promoting hydrate formation even at low surfactant concentrations. Several studies [2] have shown that sodium dodecyl sulfate (SDS) is a very effective promoter of the kinetics of hydrate formation. Adsorption of surfactants at the hydrate–water interface was proposed by Zhang et al. [9] to be the reason for surfactant’s ability to accelerate hydrate formation. In a previous work, Lo et al. [10] studied the adsorption of SDS on cyclopentane (CP) hydrate–water interface and proposed a two-step adsorption mechanism of SDS at the CP hydrate–water interface. Salako et al. [11] investigated the effect of salt on the adsorption of SDS on the CP hydrate–water interface and proposed a pseudo-monolayer adsorption mechanism at the Langmuir adsorption range. It was shown that Gemini
2
O. Salako et al. / Journal of Colloid and Interface Science 412 (2013) 1–6
surfactants could also be good promoters of hydrate formation kinetics [12–16]. However, the adsorption of Gemini surfactants has never been investigated at hydrate–water interface and understanding their adsorption behaviors at the interface is very important to accelerate the formation kinetics. Gemini surfactant, also known as dimeric surfactants [17,18], are a relatively new class of surfactants. Gemini surfactants are composed of two identical conventional surfactants joined at the head groups or very close to the head groups by a spacer group [19–22]. Gemini surfactants have attracted lots of attention because their critical micelle concentrations (CMC) are much lower than the CMC of conventional single head, single-tail surfactant [18,23]. Also, Gemini surfactants are more efficient in reducing surface tension than their corresponding conventional surfactants [24,25]. In this work, we will study the adsorption of two Dowfax Gemini surfactants at the CP hydrate–water interface by determining adsorption isotherms and zeta-potentials. The two Gemini surfactants studied in this work are Dowfax 2A1 and Dowfax C6L. These surfactants are asymmetric Gemini surfactants [22] because they have two head groups and one hydrophobic tail. The isotherms will be determined by liquid–liquid titrations. We will also investigate the change in the surface charge of CP hydrate particles in different Gemini surfactant concentrations in order to have a better understanding of the adsorption isotherms and the configuration of the surfactants at the CP hydrate–water interface. Since very little is known about the surface properties of clathrate hydrate in the literature, we will use alumina particles as a model system to understand the orientation of Dowfax 2A1 and Dowfax C6L at the CP hydrate–water interface.
2.3. Adsorption isotherms A 10 g of Gemini surfactant solutions were charged into 25 mL vials after which the vials were quickly transferred into a chiller at 275 K for about 12 h. 10 g of CP hydrate slurry was quickly transferred into the vial, tightly sealed and then returned into the chiller one after the other. The vials were left in the chiller for a week to allow surfactant adsorption to reach equilibrium. The vials were periodically shaken to accelerate the adsorption of surfactants. At the end of 1 week, several milliliters of the Gemini surfactant solution were extracted from the lower part of the vials with a syringe. The sample extraction usually lasts for less than 1 min and it is done inside the chiller to avoid CP hydrate melting. The Gemini surfactant solutions were analyzed by taking the difference between the concentration of the surfactant solution before and after the addition of CP hydrate particles. This difference gives the amount of Gemini surfactant adsorbed per gram of CP hydrate particles. 2.4. Liquid–liquid titrations One milliliter of both initial Gemini surfactant solution and extract was pipetted into different 20 mL test-tubes followed by the addition of 2.5 mL methylene blue solution (0.003 wt.% Methylene Blue, 1.2 wt.% H2SO4 and 5 wt.% Na2SO4) and 2.5 mL chloroform. The test-tube was vigorously shaken till the contents of the testtube split into two phases. The upper part of the content in the test-tube is clear and the lower part of the test-tube is blue. Hyamine solution (titrant) was added to the test-tube in drops till both the top and the bottom phase have the same shade of blue color. This signifies the endpoint. The triplicate measurements done with this procedure have an error within 2%.
2. Experimental section 2.1. Materials and methods Anionic Gemini surfactants of Dowfax 2A1 (>98% surfactant solution) and Dowfax C6L (>97% surfactant solution) were obtained from Dow Chemicals. Sodium dodecyl sulfate (SDS) and cyclopentane (CP) of 99% purity were purchased from Sigma–Aldrich. Methylene blue with indicator purity was purchased from Sigma–Aldrich. The deionized (D.I.) water used for the experiment has a resistivity of 18 MX cm 1. The sodium sulfate and sulfuric acid, with purity of 99% and 96% respectively, were purchased from Sigma–Aldrich. 100 nm AG alumina powder was purchased from Ocean State Abrasive.
2.2. Preparation of CP slurry A 300 mL of CP and water mixture (10 wt.% CP) was charged into a 1 l bottle; after which the bottle was tightly sealed, vigorously shaken and then transferred into a freezer set at 263 K. After the ice formation is noticed at the interface of CP and water, the bottle was shaken at ambient conditions to melt the ice. As the ice begins to melt, CP is quickly enclathrated. The appearance of whitish particles confirms the beginning of CP hydrate formation. The bottle was quickly transferred into the chiller for at least a week, during which the bottle was shaken at least five times per day to accelerate enclathration and ensure complete formation. Calorimetric measurements showed that the concentration of CP hydrate in the CP hydrate slurry is 51 wt.%. We used the CP hydrate slurry for this experiment in order to minimize variations in CP hydrate particle surface area during adsorption isotherm and zeta potential experiments and to avoid moisture condensation on hydrates.
2.5. Zeta potential measurements A 10 g of surfactant solutions were added to 25 mL vials and the vials were transferred into a chiller at 275 ± 0.2 K for about 12 h after which 1 g of CP hydrate was quickly added to the vial. The vials was then tightly sealed and quickly transferred into the chiller for 1 week to allow surfactant adsorption to reach equilibrium. For the zeta potential reading, the CP hydrate particles from the chilled vials were quickly transferred into a chilled folded capillary cell with a chilled pipette. The capillary cell was then loaded into the zeta potential machine (Zetasizer Nano ZS, Malvern Instrument) that was preset to a temperature of 277 K; after which the zeta potential measurements were recorded. For the zeta potential of alumina particles, 0.05 g of alumina particles was added to vials containing 10 g of surfactant solution. Each vial was well shaken and stored under ambient conditions for a day to allow adsorption to completely reach equilibrium. The zeta potential measurements for alumina particles were taken at 298 K. 3. Results and discussion Gemini surfactants used for this study are Dowfax C6L and Dowfax 2A1 whose two head groups are negatively charged (Fig. 1) with one hydrophobic tail. They have been shown to increase the rate and amount of hydrate enclathration and some Gemini surfactants could be as good as sodium dodecyl sulfate (SDS) in promoting hydrate enclathration [12]. Zhang et al. [9] proposed that the adsorption of SDS at the hydrate–water interface is the explanation behind SDS’s ability to accelerate hydrate enclathration. Therefore, it is important for us to study the adsorption behavior of Gemini surfactants at hydrate–water interface.
O. Salako et al. / Journal of Colloid and Interface Science 412 (2013) 1–6
Fig. 1. General structural formula for Dowfax C6L and Dowfax 2A1.
The mechanism of SDS adsorption at the CP hydrate–water interface was firstly proposed by Lo et al. [10]. Salako et al. [11] also proposed a mechanism for the adsorption of SDS at the CP hydrate–water interface in the presence of NaCl. This work focuses on investigating the adsorption behavior of the two anionic Gemini surfactants at CP hydrate–water interface. We will also use alumina as a model system for CP hydrate particles to explain critical parts of the adsorption isotherm and to understand the orientation of these Gemini surfactants at the CP hydrate–water interface. 3.1. Adsorption isotherms Fig. 2 shows the adsorption isotherms of Dowfax 2A1, Dowfax C6L and SDS from our previous study [11]. The isotherms are used to quantify the equilibrium amount of the surfactants adsorbed at the CP hydrate–water interface. Monolayer coverage seems to occur at approximately 0.010 mM/g and 0.015 mM/g for Dowfax 2A1 and Dowfax C6L, respectively. Both Dowfax 2AI and Dowfax C6L show a short first monolayer like a previous work on SDS adsorption and tend to form so-called pseudo-monolayer [11]. Dowfax 2A1 shows a second plateau at approximately 0.03 mM/g but the second plateau is not evident in the DowfaxC6L adsorption isotherm. Both adsorption isotherms do not show a final plateau in the range of concentration investigated in this work and are types of multi-layer adsorption. The adsorption amount of Dowfax C6L is consistently higher than that of Dowfax 2A1 surfactant. The first monolayer at 0.01 mM/g of CP hydrate in Fig. 2 corresponds to the pseudo-monolayer level for Dowfax 2A1 at the CP hydrate–water interface. The pseudo-monolayer level for Dowfax 2A1 is the same as that of SDS in our previous study [10,11] and the equilibrium concentration of Dowfax 2A1 (x-axis) at which the monolayer occurs also falls in the same range as that of SDS. The second plateau in Fig. 2, at approximately 0.03 mM Dowfax 2A1/
0.12
Dowfax 2A1 Dowfax C6L SDS
Surfactant Adsorbed (mM/g)
0.11 0.10 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00 -0.01
0
1
2
3
4
5
Equilibrium Concentration (mM) Fig. 2. Adsorption isotherms of Dowfax 2A1, Dowfax C6L and SDS [11].
3
g of CP hydrate, is higher than that of SDS at approximately 0.018 mM SDS/g of CP hydrate [10,11]. It is very interesting to see that both Dowfax 2A1 and SDS with the same hydrophobic tail length but different number of hydrophilic head group (Dowfax 2A1 with two head groups and SDS with one head group) form a pseudo-monolayer at the same monolayer coverage and at the same range equilibrium surfactant concentration. We conjectured that two head groups of each Dowfax surfactant would adsorb at the CP hydrate interface. But our adsorption isotherms reveal that this is not the case because if the two head groups in each surfactant adsorb at the interface, the isotherm amount should be half of that of SDS for a given number of adsorption sites. As alluded to in Lo et al. [10] and Salako et al. [11], adsorption of anionic surfactant at the CP hydrate–water interface is possible via hydrogen bonding between the head group of the surfactant monomers and the pendant hydrogen atoms of CP hydrate particles. For both heads of the Gemini surfactant to adsorb at the CP hydrate–water interface, the distance between two adsorption sites must approximately equal to the distance of separation between the two head groups in one surfactant molecule. If the distance between the adsorption sites is shorter than the distance between the head groups, then the two head groups can only adsorb if the spacer group between the surfactant hydrophilic head groups is flexible. The spacer group in both Dowfax 2A1 and Dowfax C6L is a diphenyl oxide and it is not flexible. At this moment, no information is available about the distance between the adsorption sites at the CP hydrate–water interface but our results imply that it is greater than the length of the spacer group. This is because both SDS and Dowfax 2A1 show the same monolayer level at the same equilibrium concentration range, which only means that one of the head groups is adsorbed at the CP hydrate–water interface. The monolayer level of Dowfax C6L is at approximately 0.015 mM/g CP hydrate and it is a little higher than that of Dowfax 2A1 and SDS. Thus, judging from this monolayer level of the Dowfax C6L, one of its head groups is adsorbed at the interface. In our previous study [11], we postulated that the hydrophobic tail of SDS blocks some adsorption sites at the pseudo-monolayer level and these sites are exposed at the onset of the bilayer formation when the hydrophobic tails of adsorbed SDS begin to interact with the hydrophobic tails of SDS in solution. The Dowfax C6L adsorption isotherm supports our postulate. Dowfax C6L has a hydrophobic tail with a chain length of six carbons. Both Dowfax 2A1 and SDS have a hydrophobic tail that has a chain length of 12 carbons. The hydrophobic tail of Dowfax C6L is not long enough to block as many adsorption sites as the hydrophobic tails of Dowfax 2A1 and SDS, therefore more Dowfax C6L molecules will adsorb at the pseudo-monolayer level when compared to SDS and Dowfax 2A1 molecules. This clearly shows the importance of surfactant hydrophobic tail in the formation pseudo-monolayer at the CP hydrate–water interface. Unlike the SDS isotherm in our previous study [10,11], which levels off at approximately 0.018–0.02 mM/g of CP hydrate in Fig. 2, both Dowfax 2A1 and Dowfax C6L isotherms keep rising without finally leveling off in the range of concentrations tested in our studies although the Dowfax 2A1 adsorption isotherm shows a second plateau at approximately 0.03 mM/g of CP hydrate. According to Yueying et al. [25], if it is assumed that the monolayer is the center of aggregation, then the aggregation number of surfactants will be the ratio of the second surface coverage to the first surface coverage. If the aggregation number is approximately two, then the entire surface coverage can be called a bilayer. The aggregation number for our SDS experiment is approximately two; therefore, we concluded that the second surface coverage of SDS signifies bilayer coverage. The ratio of the second to the first surface coverage in the case of Dowfax 2A1 is approximately 3. This
O. Salako et al. / Journal of Colloid and Interface Science 412 (2013) 1–6
signifies the formation of hemimicelles at the second layer and beyond for Dowfax 2A1 at the CP hydrate–water interface. The lack of evidence of a second plateau of surface coverage for Dowfax C6L reflects the importance of the hydrophobic tail on stabilizing the adsorption process. After the formation of the pseudomonolayer, the hydrophobic tails of adsorbed surfactant will begin to interact with the hydrophobic tails of surfactant in solution. For Dowfax C6L, the hydrophobic tail length is not long enough to prevent considerable interaction between the head group of adsorbed Dowfax C6L and those of Dowfax C6L in solution at the onset of the second layer formation. The second layer for Dowfax C6L is unstable and it is reflected in the adsorption isotherm as there is no evidence of a second plateau in Fig. 2. 3.2. Zeta potentials at the CP hydrate–water interface
Zeta Potential of CP Hydrate Particles (mV)
4
-40 -45 -50 -55 -60 -65 -70 -75 -80 -85 0
1
2
3
4
Dowfax C6L Concentration (mM) Fig. 4. Zeta potential of CP hydrate particles at different Dowfax C6L concentrations.
3.3. Adsorption of SDS, Dowfax 2A1 and Dowfax C6L on alumina particles Very little is known in the literature about the surface property of CP hydrates, so we intended to understand the adsorption of SDS, Dowfax 2A1 and Dowfax C6L by using alumina particles as our model system. For this experiment, we used zeta potential measurements to study the change in the surface charge of alumina particles at different concentrations of SDS, Dowfax 2A1 and Dowfax C6L surfactants. Figs. 5–7 are plots of zeta potential of alumina particles in different surfactant concentrations. All the zeta potential plots show similar trends. All plots have the first and second plateaus. The zeta potential values of the first plateaus are 17.5 mV, 31 mV and 31 mV for SDS, Dowfax 2A1 and Dowfax C6L respectively. These first step values give us a clearer picture of the configuration of these surfactants at the alumina–water interface. These values reinforce our earlier assertion that both Dowfax 2A1 and Dowfax C6L surfactants adsorb at the alumina–water interface with one of their hydrophilic head groups. This will be the case because, just like the adsorption isotherms of SDS, Dowfax 2A1 and
15
-30
Zeta Potential of Alumina Particles (mV)
Zeta Potential of CP Hydrate Particles (mV)
Zeta potential measurements also support our understanding of these adsorption isotherms. The zeta potential measurement of CP hydrate particles in varying concentrations of Dowfax 2A1 (Fig. 3) shows two clear plateaus; the first ranging from approximately 0.6–1.25 mM Dowfax 2A1 and the second plateau is seen from approximately 2.2 mM Dowfax 2A1. Fig. 4 shows the zeta potential measurement of CP hydrate particles at different Dowfax C6L solution concentrations with no clear first plateau and a clear second plateau which is seen from 2.2 to 4.0 mM Dowfax C6L. The zeta potential curves of both Dowfax 2A1 and Dowfax C6L from Figs. 3 and 4 respectively level off at approximately 77.5 mV at a concentration of 2.2 mM. Since the adsorption isotherms for both Dowfax 2A1 and Dowfax C6L keep increasing, charge shielding by forming aggregates must be responsible for the leveling off of these zeta potential plots. The outermost surfactant head and tails of the adsorbed layer shield the innermost surfactant head groups. The other important feature of the zeta potential plots is the clear first plateau for Dowfax 2A1 similar to the SDS zeta potential measurement in our previous study. However, the first plateau in the Dowfax C6L zeta potential plot is not clear corresponding to the short plateau of the isotherm in Fig. 2. This also emphasizes the importance of the hydrophobic tail in stabilizing the pseudo-monolayer before transitioning to the formation of the second layer. In the case of Dowfax C6L, there is no distinction between the formation of the pseudo-monolayer and the onset of the second layer formation because of the short hydrophobic tail length.
-40
-50
-60
-70
-80 0
1
2
3
4
Dowfax 2A1 Concentration (mM)
10 5 0 -5 -10 -15 -20 -25 -30 0
2
4
6
8
SDS Concentration (mM) Fig. 3. Zeta potential of CP hydrate particles at different Dowfax 2A1 concentrations.
Fig. 5. Zeta potential of alumina particles at different SDS concentrations.
O. Salako et al. / Journal of Colloid and Interface Science 412 (2013) 1–6
5
Zeta Potential of Alumina Particles
10
0
-10
-20
-30
-40 0
2
4
6
8
Dowfax 2A1 Concentration (mM)
Zeta Potential of Alumina Particles (mV)
Fig. 6. Zeta potential of alumina particles at different Dowfax 2A1 concentrations.
10
0 Fig. 8. Proposed adsorption schematics for Dowfax 2A1 and Dowfax C6L at the CP hydrate–water interface.
-10
3.4. Proposed adsorption scheme for Gemini surfactant
-20
-30
-40
-50 0
2
4
6
8
Dowfax C6L Concentration (mM) Fig. 7. Zeta potential of alumina Particles at different Dowfax C6L concentration.
Dowfax C6L, the first step for these surfactants all appear in the same concentration range but the zeta potential value at the first step for SDS with one negatively charged head is almost half that of Dowfax C6L and Dowfax 2A1. The zeta potential values for the second step on the zeta potential plots are 25 mV, 40 mV, and 42 mV for SDS, Dowfax 2A1 and Dowfax C6L, respectively above the surfactant concentration of 4.2 mM. The zeta potential of Dowfax 2A1 and Dowfax C6L at the second step cannot be exactly twice as high as that of SDS because of the charge shielding. Also, the zeta potential of Dowfax C6L is slightly higher than that of Dowfax 2A1. This is consistent with Fig. 2 which shows that the amount of Dowfax C6L adsorbed at the CP hydrate–water interface is slightly higher than that of Dowfax 2A1. The zeta potential plot of Dowfax C6L in the case of alumina shows a clear first plateau when compared to the zeta potential plot of Dowfax C6L adsorption at the CP hydrate–water interface. This could mean that the pseudo-monolayer of Dowfax C6L at the alumina–water interface is more densely packed than at the CP hydrate–water interface. The packed pseudo-monolayer will increase hydrophobic interaction between the tails of the adsorbed Dowfax C6L molecules and this will increase the stability of the pseudo-monolayer at the alumina–water interface when compared to the adsorption at the CP hydrate–water interface.
From the results of the adsorption isotherms and zeta potential measurements as described above, we propose an adsorption scheme for Dowfax 2A1 and Dowfax C6L. Fig. 8 shows the schematics of the proposed adsorption scheme for both. At low surfactant concentration, depicted by (1) in Fig. 8, the surfactants adsorb at the CP hydrate–water interface with their head group forming hydrogen bonds with the pendant hydrogen of water cages; while the hydrophobic tail of the surfactant is in a ‘‘lay down’’ configuration because of the hydrophobic interactions between the guest molecule (CP) and the hydrophobic tail of the surfactant. The pseudo-monolayer coverage is depicted by (2) in Fig. 8. At this coverage, more surfactants are adsorbed at the CP hydrate–water interface momentarily covering the entire surface of the CP hydrate particle. The hydrophobic tails of the surfactant at this coverage block some of the vacant adsorption sites at the CP hydrate–water interface. The hydrophobic tails of the adsorbed surfactants begin to interact with the hydrophobic tails of the surfactant in solution. The intermediate surfactant configuration, at the CP hydrate–water interface, between the first and the second plateau is depicted by (3) in Fig. 8 while the surface coverage at the second plateau is depicted by (4). The adsorption isotherm of Dowfax 2A1 shows the second plateau at surface coverage that is three times as large as the first plateau. Since there is no final leveling off of the isotherms within the surfactant concentration range investigated in this work, we conjecture that the surfactant aggregate at the CP hydrate–water interface shown in (5) in Fig. 8 will keep adding surfactant till the surfactant aggregation number of each respective surfactant is reached. At this point, we expect the adsorption isotherms to reach a final plateau. 4. Conclusions This study has investigated the adsorption of two Gemini surfactants, Dowfax 2A1 and Dowfax C6L, at the CP hydrate–water
6
O. Salako et al. / Journal of Colloid and Interface Science 412 (2013) 1–6
interface. The adsorption was analyzed using adsorption isotherms. Zeta potential measurements of CP hydrate particles in the presence of Dowfax 2A1 and Dowfax C6L were used to explain the shape of the adsorption isotherms. We also used alumina particles as a model system for CP hydrate particles to further explore the configuration of the Gemini surfactants used in this study and SDS at the CP hydrate–water interface. Our adsorption isotherms show that more Dowfax C6L adsorbs at the CP hydrate–water interface compared to Dowfax 2A1. The adsorption isotherm of Dowfax C6L only has one clear plateau while that of Dowfax 2A1 has two plateaus. Both adsorption isotherms did not show a final plateau but were a multilayer surface coverage at the CP hydrate–water interface. The zeta potential of alumina particles, used as a model, in the presence of both Gemini surfactants is approximately twice as high as that of SDS at the monolayer coverage in the same surfactant concentration range. This shows that Gemini surfactants used in this study adsorb with only one of their two anionic head groups. Acknowledgments The authors are grateful for the financial support from the Advanced Biomass R&D Center (ABC) as a Global Frontier Project funded by the Ministry of Science, ICT and Future Planning and from ACS–PRF (under a Grant Number of 51991-ND9). This research was also made possible in part by a Grant from BP/GoMRI. The authors also appreciate initial discussions with Ms. Franchette Viloria and Mr. Mohammad Mahmud at CCNY. References [1] E.D. Sloan, C. Koh, Clathrate Hydrates of Natural Gas, third ed., CRC Press, Boca Raton, FL, 2008 .
[2] Y.F. Makogon, S.A. Holditch, T.Y. Makogon, J. Pet. Sci. Eng. 56 (2007) 14–31. [3] Y. Zhong, R.E. Roger, Chem. Eng. Sci. 55 (2000) 4175–4187. [4] C.A. Koh, E.D. Sloan, A.K. Sum, D.T. Wu, Annu. Rev. Chem. Biomol. Eng. 2 (2011) 237–257. [5] E.D. Sloan Jr., Clathrate Hydrates of Natural Gas, second ed., Marcel Dekker, 1998. [6] P. Englezos, N. Kalogerakis, P.D. Dholabhai, P.R. Bishnoi, Chem. Eng. Sci. 42 (1987) 2659–2666. [7] W.X. Wang, C.L. Bray, D.J. Adams, A.I. Cooper, J. Am. Chem. Soc. 230 (2008) 11608–11609. [8] 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. [9] J.S. Zhang, S.Y. Lee, J.W. Lee, Ind. Eng. Chem. Res. 46 (2007) 6353–6359. [10] C. Lo, J.S. Zhang, A. Couzis, P. Somasundaran, J.W. Lee, J. Phys. Chem. C 114 (2010) 13385–13389. [11] O. Salako, C. Lo, J.S. Zhang, A. Couzis, P. Somasundaran, J.W. Lee, J. Colloid Interface Sci. 386 (2012) 333–337. [12] C. Lo, The role of surface active agents on hydrate formation, PhD Thesis, The City College of New York, New York, New York, 2011. [13] Y.A. Kwon, J.M. Park, K.E. Jeong, C.U. Kim, T.W. Kim, H.J. Chae, S.Y. Jeong, J.H. Yim, Y.K. Park, J.D. Lee, J. Ind. Eng. Chem. 17 (2011) 120–124. [14] B.M. Lee, H.C. Kang, J.M. Park, J.H. Yoon, US 6,392,064 B2, 2002. [15] B.M. Lee, H.C. Kang, J.M. Park, KR 0578716 B1, 2006. [16] R. Zana, Curr. Opin. Colloid Interface Sci. 1 (1996) 566–571. [17] M.J. Rosen, J. Chemtech. (1993) 30–33. [18] R. Zana, J. Colloid Interface Sci. 248 (2002) 203–220. [19] S.K. Hait, S.P. Moulik, Curr. Sci. 82 (2002). No. 9. [20] F.M. Menger, C.A. Littau, J. Am. Chem. Soc. 115 (1993) 10083–10090. [21] S. Manne, T.E. Schaffer, Q. Huo, P.K. Hansma, D.E. Morse, G.D. Stucky, I.A. Aksay, Langmuir 13 (1997) 6382–6387. [22] Y.I. Rabinovich, J.R. Kanicky, S. Pandey, H. Oskarsson, K. Holmberg, B.M. Moudgil, D.O. Shah, J. Colloid Interface Sci. 288 (2005) 583–590. [23] F. Devinsky, I. Lacko, F. Bittererova, L. Tomeckova, J. Colloid Interface Sci. 114 (1986) 314–322. [24] Y.P. Zhu, A. Masuyama, Y. Kobata, Y. Nakatsuji, M. Okahara, M.J. Rosen, J. Colloidal Interface Sci. 158 (1993) 40–45. [25] Y. Gao, J. Du, T. Gu, J. Chem. Soc., Faraday Trans. 1 (83) (1987) 2671–2679.
