- Email: [email protected]
Observation of the growth characteristics of gas hydrate in the quiescent-type formation method using surfactant
Accepted Manuscript Observation of the growth characteristics of gas hydrate in the quiescent-type formation method using surfactant Tatsunori Asaoka, Koya Ikeda PII: DOI: Reference:
S0022-0248(17)30501-8 http://dx.doi.org/10.1016/j.jcrysgro.2017.08.017 CRYS 24268
To appear in:
Journal of Crystal Growth
Received Date: Revised Date: Accepted Date:
4 July 2017 9 August 2017 11 August 2017
Please cite this article as: T. Asaoka, K. Ikeda, Observation of the growth characteristics of gas hydrate in the quiescent-type formation method using surfactant, Journal of Crystal Growth (2017), doi: http://dx.doi.org/10.1016/ j.jcrysgro.2017.08.017
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Observation of the growth characteristics of gas hydrate in the quiescent-type formation method using surfactant Tatsunori Asaokaa,*, Koya Ikedab a
Academic Assembly Institute of Engineering, Shinshu University, 4-17-1 Wakasato, Nagano 380-8553, Japan Department of Mechanical Systems Engineering, Shinshu University, 4-17-1 Wakasato, Nagano 380-8553, Japan
b
*
Corresponding author. E-mail address: [email protected] (T. Asaoka)
Abstract We have observed the growth behavior of gas hydrate with addition of surfactants by several observation methods, and we discuss the mechanism of gas hydrate formation along the wall surface. From observation using colored water, the water moves up through the porous hydrate structure and unreacted water remains in the middle of the porous hydrate. In contrast to previous studies which observed growth from the side, we observed the growth behavior from the top. The hydrate height agrees well with that of water pulled up by capillary action. These results confirm that the driving force for hydrate growth in the vertical direction is capillary action. We discuss the driving force for hydrate growth in the horizontal direction. We suggest that the driving force in the horizontal direction is different because the initial concentration of the surfactant solution affects the growth behavior in the horizontal direction while it does not affect the hydrate height. Keywords:
A1. Growth models, A1. Mass transfer, A1. Heat transfer, A2. Growth from solutions, B1. Gas hydrate
1. Introduction Natural gas hydrate has great potential in natural gas storage and transportation. The storage and transportation costs of natural gas can be decreased by using natural gas hydrate, because natural gas hydrate is stable at higher temperature and lower pressure conditions than liquid natural gas. Gas hydrates are solid-state materials composed of gas and water, which are called clathrate hydrates. There are many types of gas hydrates depending on the guest gas, such as methane hydrate, ethane hydrate, and propane hydrate. In the present work, we used propane hydrate because of its manageability in ambient conditions. It is difficult to continuously form gas hydrates at a high production rate because the formed hydrate covers the surface of water, which inhibits the reaction between the gas and water. Expanding/renewing the gas–liquid interface is necessary to increase the production rate. In industrial fields, splaying [1] and bubbling methods [2] are commonly used, which can be classified as stirred systems. Quiescent systems, in which mechanical agitation or pumping of gas/water is not required, have some advantages. For example, the facility can be simplified and the operation/maintenance costs can be reduced. Various attempts have also been made to enhance the production rate in quiescent systems. Kumano et al. [3] used a fiber layer to increase the gas–liquid interface. Kuhs et al. [4] and Teraoka et al. [5] used fine ice particles. Addition of a surfactant is an effective and convenient method because it can be operated with a simple facility. Addition of surfactant results in continuous formation of gas hydrate without mechanical power, because the gas hydrate grows up along the side wall of the container. A number of researchers have investigated the mechanism of gas hydrate formation [6-12]. They suggest that the driving force for gas hydrate formation should be capillary action, which continuously transports water/surfactant solution to the surface of the gas hydrate. To further investigate the mechanism, Veluswamy et al. [13], Lim et al. [14], Yoslim et al. [15], and Okutani et al. [11] observed the growth of gas hydrates along the wall surface using surfactant solution. However, the gas hydrate covered the observation window and prevented observation of its growth. Hayama et al. [16, 17] observed gas hydrate formation from droplets of surfactant solution and investigated the effect of the surfactant concentration. Lee et al. [18] observed the growth of hydrate around gas bubbles 1
placed in a surfactant solution. Although these types of micro-scale experiments are useful to investigate the mechanism, formation along the wall surface cannot be discussed. In the present study, we used several methods to observe propane hydrate formation, in which colored water was used or the observation angle was varied, and discuss the mechanism of gas hydrate formation along the wall surface.
(a)
(b)
Fig. 1. Experiment apparatus. 1. Constant temperature room. 2. Pressure vessel. 3. Sample container. 4. Sample. 5. Digital camera. 6. Pressure gauge. 7. Thermocouple. 8. Glass windows. 9. Light source. 10. Data logger. 11. Gas reservoir. 12. Constant temperature bath. 13 High pressure cylinder of propane gas. 14. Valves. 15. Relief valve. 16. Vacuum pump. 17. Gas exhaust. 18. Transparent acrylic plate.
2. Experiment apparatus and procedure 2.1. Experiment apparatus Figure 1 shows the experiment apparatus. A stainless-steel pressure vessel with a diameter of 97 mm and a height of 181 mm is placed in a constant temperature room, and a sample container is placed in the pressure vessel. Gas hydrate forms in the sample container by the reaction between the gas and the water/surfactant solution. The pressure vessel shown in Fig. 1(a) has two 15-mm-wide glass windows on the face of the side walls. Formation/crystal growth of the hydrate is observed through the windows. Photographs were periodically taken by a digital camera. Both transmitted and reflected light can be used varying the position of the light source. The temperature and pressure inside the vessel are measured by a thermocouple and a pressure gauge. The thermocouple is T-type and the resolution is 0.1 K. The pressure gauge (GC61, Nagano Keiki Co., Ltd., Tokyo, Japan) is diaphragm type and the resolution is 0.001 MPa. When we observes the sample from the top, the pressure vessel is replaced by that shown in Fig. 1(b), which consists of a transparent plate as the upper wall. The temperature and pressure cannot be measured in this vessel because it is designed to have a wide viewing area. Thus, the pressure vessel is chosen depending on the intended observation direction. The shape/size and material of the sample container can also be changed depending on the aim of the experiment. In this study, we used propane in the experiments because of its manageability close to ambient temperatures and pressures. A gas reservoir was used to keep the pressure in the pressure vessel at the required value. By cooling the reservoir below the liquefied temperature of propane, the propane gas liquefied and liquid propane was contained in the reservoir. Because liquid propane maintains a constant pressure at the vapor–liquid equilibrium, the pressure can be set by controlling the temperature of the reservoir. In the experiments, because we chose a pressure of 0.46 MPa, we kept the reservoir temperature at 0.5 °C, which is the equilibrium temperature at 0.46 MPa. Beforehand the pressure vessel was evacuated to
2
below 0.005 MPa to remove air. Opening the valve between the reservoir and pressure vessel, we can keep the pressure in the pressure vessel at 0.46 MPa. Initial formation of the gas hydrate takes a long time after water is exposed to the gas because of supercooling. To inhibit supercooling, a small hydrate crystal of about 5 g, which acts as nucleus for crystal growth, was placed in the sample container in all of the experiments. A hydrate–water mixture was obtained in the experiments because the sample water never completely converted to propane hydrate. We will call this mixture the mixed product. 2.2. Observation of the formation process Propane hydrate formed along the inner wall of the sample container. A glass container with a diameter of 90 mm and height of 110 mm was used as the sample container. The inner surface of the sample container was coated with a non-ionic hydrophilic surfactant (B-542, LEC, Inc., Tokyo, Japan) as an anti-fog spray. Distilled water (100 g) was supplied to the sample container as the sample. The temperature in the sample container was measured by a thermocouple. Propane gas was injected into the pressure vessel and propane hydrate started to form in the sample container. The temperature of the sample container increased with formation of propane hydrate, although the temperature in the constant temperature room where the pressure vessel was placed was kept at 2 ± 1 °C throughout the experiment. Using colored water, we attempted to observe the movement of water accompanied by propane hydrate formation. At the beginning, propane hydrate formed in the sample container from the uncolored water solution. After the propane hydrate reached the top of the container, we opened the pressure vessel and supplied colored water (50 g) to the sample container. After evacuation of air and injection of propane gas, we restarted observation of the growing propane hydrate and movement of the colored water. Red food coloring containing 85% dextrin and 15% food red No. 102 was used to color the water. Observation was continued until the temperature of the water inside the sample container decreased and reached the temperature of the constant temperature room, which indicates completion of the reaction/formation of propane hydrate. The mixed product was then taken out from the pressure vessel and the surface and cross-section were observed. 2.3. Conversion ratio Mass of the propane hydrate formed in the experiment was estimated by following procedure, and the variation of the mass was investigated. A copper container, which has a good thermal conductivity, with a diameter of 68 mm and height of 133 mm was used as the sample container. The inner surface was coated with a non-ionic hydrophilic surfactant, similar to the experiment in Section 2.2. Distilled water was supplied in the sample container and the mass was varied by the experiments (100, 70, or 50 g). Observation was not performed in this experiment. The temperature of the sample container increased with propane hydrate formation, although the temperature in the constant temperature room was kept at 2 ± 1 °C throughout the experiment. We define time 0 as the time when the temperature of the sample container started to increase and the reaction time as the time from time 0 until the end of the experiment. Several experiments were performed varying the reaction time. After the experiment, the mixed product was taken out of the pressure vessel and the following measurements were performed. We define the conversion ratio f as the ratio of the mass of propane hydrate in the mixed product mh to the theoretical mass of propane hydrate when all of the supplied water is converted to propane hydrate mh0. First, the mass of the mixed product mf was measured. The resolution of the weight scale was 0.1 g. We then charged the mixed product into hot water (>70 °C) filled in a Dewar bottle, and the variation of mass of the bottle was measured. The mass of the water obtained by the dissociation of the mixed product mw can be determined by the mass difference, because the water remains in the hot water while the gas is released into the surrounding. The mass of gas mg was calculated by
mg m f mw
(1)
Given that the hydration number of propane hydrate is 17 [19] and the ratio of the molar weight of propane to water is 44:18, mh can be calculated from mg as follows: 3
mh
350 mg 44
(2)
In addition, mh0 can be calculated by
mh 0
350 mw 306
(3)
From the results, f was calculated using the measured values of mf and mw: f
mh 306 m f mw mh 0 44 mw
(4)
2.4. Effect of the wall material and surfactant concentration The effect of the wall material of the sample container was also investigated. The pressure vessel shown in Fig. 1(b), which has a wide viewing area, was used and we observed the sample from the top. Two types of sample containers were used in this experiment. One was a polypropylene container with a diameter of 79 mm and height of 12 mm, as shown in Fig. 2(a), and the other was the same container covered with an aluminum film, as shown in Fig. 2(b). Sodium dodecyl sulfate (SDS) solution (10 ppm) was used as the sample. The mass of the sample was 100 g and the sample was colored to visualize it clearly. SDS is a popular surfactant and is suitable for this experiment because the physical properties of SDS solution are well known. For example, the critical micelle concentration (CMC) is about 1780 ppm [20] and the surface tension has been reported [21]. We confirmed that the trend of propane hydrate formation was almost same when SDS and the non-ionic hydrophilic surfactant (Section 2.2) were used. We also investigated the effect of the surfactant concentration. A glass container with a diameter of 60 mm and height of 100 mm was used as the sample container. SDS solution (10 or 1000 ppm) was used as the sample. The mass of the sample was 100 g and the sample was colored to visualize it clearly.
(a) Polypropylene surface
(b) Aluminum surface
Fig. 2. Sample containers used for observing the growth behavior on the different wall surfaces. The containers are both made of polypropylene but the container in (b) is covered with an aluminum film.
3. Results 3.1. Observation of the formation process Figure 3 shows the growth behavior of propane hydrate on the glass wall observed from the side. As shown in Fig. 3(a), the window on the opposite side and light reflection from the bottom can be observed in the each photograph. The width of the window is 15 mm. The pictures were taken using transmitted light. Time 0 is defined as the time when gas was supplied to the pressure vessel. Figure 3(b) shows immediately after gas supply. The small propane hydrate crystal can be observed below the water surface, which was placed beforehand as nucleus for crystal growth. Initially, film of propane hydrate grows on the front surface of the sample container (Fig. 3(c) and (d)). As shown in Fig. 3(e), a crack forms in the propane hydrate 4
film, which may be caused by the increase of the volume of the hydrate film. A crack was observed for all of the experiments performed under the same conditions. Colored water was supplied into the sample container at 195 min (Fig. 3(f)). The edge of the propane hydrate crystal becomes slightly smoother in Fig. 3(f) compared with Fig. 3(e) because of dissociation of propane hydrate by the effect of evacuation of air and injection of propane gas. From Fig. 3(g), the colored water moves upward. This indicates that there are gaps in the propane hydrate body or between the propane hydrate body and the wall surface, and water moves through the gaps by capillary action. It should be noted that the surfactant, which covers the inner surface of the sample container, has an important role, because this water movement does not occur without the surfactant.
(a)
(b)
(c)
(d)
(e)
(f)
(g)
Fig. 3. Growth behavior of propane hydrate on the glass wall observed from the side. (a) is the schematic of the view from the window. Time after gas injection are (b) 1 min, (c) 90 min, (d) 164 min, (e) 192 min, (f) 195 min, (g) 200 min.
Figure 4 shows the appearance of the mixed product formed in the above experiment. As shown in Fig. 4(f), the sample for the observation was cut out of the original cylindrically shaped mixed product, and we observed the sample from the directions of view A, B, and C. Figures 4(a) and 4(b) show the surfaces facing the side wall of the container, and Figs. 4(c) and 4(d) show the surfaces exposed to the gas. Figures 4(a) and 4(c) show the sample just after preparation. Figures 4(b) and 4(d) show the sample after slight dissociation, which was kept under ambient conditions for a few minutes. Figure 4(e) shows the cross-section of the sample just after preparation. Although propane hydrate is naturally transparent, as shown in Fig. 3, the propane hydrate that formed in the experiments is white because of diffuse reflection on its porous surface. Additionally, propane hydrate is not colored because it does not capture the coloring material. Because the liquid and the discharged coloring material appear red, we can determine the residual water/surfactant solution by red in the images. In Fig. 4(a), there are many large gaps on the surface. This means that the formed propane hydrate is porous and has large pores around the surface. We can observe large cracks in Fig. 4 (c), which were formed by the volume expansion during hydrate formation, as shown in Fig. 3(e). In Fig. 4(b) and 4(d), there are many small pores around the both surfaces, although there are long gaps only in Fig. 4(b). This means large long gaps exist only around the surface facing the side wall of the container. We suggest that large long gaps were formed by the water flow caused by capillary action. In Fig. 6, the sample is colored in the middle of the propane hydrate layer, while it is white near the surfaces. This means that unreacted water remains in the middle of the mixed product. Because we performed the experiment for a sufficient reaction time, we suggest that reaction of water is inhibited in the middle of the propane hydrate layer.
5
(a) View A. Just after preparation
(b) View A. After slight dissociation
(c) View B. Just after preparation
(d) View B. After slight dissociation
(e) View C. Just after preparation
6
(f) Schematic diagram of the observation directions. Fig. 4. Appearance of hydrate formed in the experiment. View A is the container wall side, view B is the exposed side, and view C is the cross-section of the hydrate.
Conversion ratio f , %
100 80 60 Mass of supplied water 100g 70g 50g
40 20 0
0
20
40 60 80 Reaction time, h
100
Fig. 5. Variation of the conversion ratio with reaction time.
3.2. Conversion ratio Figure 5 shows the relationship between the reaction time and the conversion ratio. The conversion ratio increases with increasing reaction time until about 22 h. After 22 h, the conversion ratio becomes constant at about 90%. It was confirmed that the conversion ratio never reaches 100% even if for a reaction time of more than 90 h. Additionally, from the temperature history, the temperature of the sample increased owing to propane hydrate formation and it decreases to the surrounding temperature after about 22 h. This means that formation of propane hydrate stops at about 22 h and unreacted water remained. As shown in Fig. 5, even when we vary the mass of supplied water, the same trend is observed. Moreover, the same trend has been reported by Okutani et al. [11]. 3.3. Effect of the wall material and surfactant concentration Figure 6 shows the shapes of the propane hydrates formed from 10 ppm SDS solution in containers with different wall surfaces. The time is more than 22 h after gas injection and the images were taken from the top. The maximum heights of propane hydrate are given by white line in the figures. The maximum height is 52 mm using polypropylene, while it is 85 mm using aluminum. The height on the aluminum surface is 1.6 times larger than that on the polypropylene surface. We believe that this is the effect of capillary action between the body of propane hydrate and the side wall of the container. We performed the same experiment several times and confirmed the measured heights were almost the same. Additionally, it should be noted that the heat transfer along with the aluminum film is negligible, because the aluminum film is thin. 7
Moreover the heat flax along with the aluminum film is small, because the formation of hydrate is slow and the temperature distribution on the aluminum film is small.
(a) Polypropylene surface
(b) Aluminum surface
Fig. 6. Propane hydrate formed in containers with different wall surfaces using 10 ppm SDS solution. The time is more than 22 h after gas injection.
Fig. 7. Model of capillary action occurring between porous hydrate body and container wall during the propane hydrate formation process. is contact angle on each surface, h is the height of water pulled up by capillary action, and r is the pore radius.
Figure 7 shows the model of capillary action during the propane hydrate formation process. It shows the capillary action between porous propane hydrate and the container wall. The height of water pulled up by capillary action h is estimated by h
2 cos w rg
(5)
where , , w, r, and g are the surface tension, contact angle on the container surface, density of the water/surfactant solution, radius of the capillary tube, and acceleration by gravity, respectively. h is proportional to cos because the other variables are constant in the experiment. The contact angles of droplets of 10 ppm SDS solution on the polypropylene and aluminum surfaces were measured to be 54° (cos 54° = 0.588) and 30° (cos 30° = 0.866), respectively. From these results, we can estimate that h on the aluminum surface is 1.5 times larger than that on the polypropylene surface, because the cos value is 1.5 times larger for the aluminum surface than for the polypropylene surface. As mentioned above, the height of propane hydrate formed on the aluminum surface is 1.6 times larger than that formed on the aluminum surface, and this ratio agrees well with the ratio of the h values. This shows that formation of propane hydrate occurs because of the supply of water/surfactant solution by capillary action between porous propane hydrate and the container wall. Because water/surfactant solution cannot be supplied to higher than h, propane hydrate cannot grow above this height. By substituting the contact angle on each surface into Eq. (5), we estimated the pore radius, which corresponds to r, to be 0.08 mm for the polypropylene surface and 0.07 mm for the aluminum surface. The surface tension is assumed to be 35 mN/m for the 1780 ppm (at the CMC) SDS solution at 6 °C [21] and the density w is assumed to be approximately 1000 8
kg/m3. Because the estimated pore radius were almost the same (both of them were about 0.1 mm), we consider it is independent of the material of the container surface. Although we attempted to measure the pore radius of the formed propane hydrate, we failed because we could not fix the sample on the scale owing to dissociation of the sample.
(a) 48 min (10 ppm SDS solution)
(b) 82 min (10 ppm SDS solution)
(c) 114 min (10 ppm SDS solution)
(d) 295 min (10 ppm SDS solution)
(e) 6 min (1000 ppm SDS solution)
(f) 40 min (1000 ppm SDS solution)
(g) 88 min (1000 ppm SDS solution)
(h) 241 min (1000 ppm SDS solution) 9
Fig. 8. Growth behavior of propane hydrate observed from the top. The material of the sample container is glass.
Figure 8 show the growth behavior of propane hydrate using 10 and 1000 ppm SDS solution, respectively. A glass container with a diameter of 60 mm and height of 100 mm was used in the experiments. The images were taken from the top. As shown in Fig. 8 (a)-(d), when 10 ppm SDS solution is used, propane hydrate grows almost symmetrically in the vertical direction. Initially, growth in the horizontal direction is not significant. Horizontal growth starts at about 100 min, as shown in Fig. 4(c). In contrast, as shown in Fig. 8(e)-(h), when 1000 ppm SDS solution is used, propane hydrate asymmetrically grows in the vertical direction. In the initial period, a number of long dendritic shaped crystals grow in the horizontal direction. The maximum heights of propane hydrate are almost same for 10 and 1000 ppm SDS solution. This indicates that the driving forces for propane hydrate growth in the vertical and horizontal directions are different. This will be discussed in Section 4. 4. Discussion of crystal growth As mentioned in Section 3, formation of hydrate occurs because of the supply of water/surfactant solution by capillary action between porous hydrate and the container wall. Although this has been already suggested by other researchers, for example Okutani et al. [11], we provide additional evidence in the present work. From observation using colored water, we show the water moves up through the porous hydrate structure and unreacted water remains in the middle of the porous hydrate. As a result, the conversion ratio never reaches 100%. We also show that the hydrate height agrees well with that of water pulled up by capillary action. From these results, we confirmed that the driving force for hydrate growth in the vertical direction is capillary action. We suggest the driving force for hydrate growth in the horizontal direction is different. The initial concentration of the SDS solution affects the growth behavior in the horizontal direction. However, it does not affect the hydrate height. We believe the condensation of SDS solution is important for the hydrate growth in the vertical direction. The concentration of the SDS solution increases with hydrate formation because water is converted to hydrate. The surface tension and contact angle decrease with increasing SDS concentration, but the concentration finally reaches a constant value above the saturated concentration or CMC. Considering micelles do not form at the experimental temperature [20], we expect that the concentration of the SDS solution reaches the saturated concentration at the top of the formed hydrate regardless of its initial concentration. This would be the reason why the initial concentration does not affect the hydrate height. Considering hydrate growth in the horizontal direction, the driving force for water movement is not capillary action because it is strongly affected by the initial concentration of the SDS solution. There are various candidates for the driving force, such as concentration diffusion, micelle formation, and deposition of SDS crystals. If the concentration on the surface exposed to the gas becomes higher than that in the porous hydrate structure because of condensation of the surfactant solution, water should move from the middle to the surface by concentration diffusion. As the initial concentration of surfactant solution increases, the flow rate of water increases and formation of hydrate in the horizontal direction should be facilitated. About the effects of micelle formation, some interesting investigations were reported. Wang et al. [22] used mimic micelles, and Bhattacharjee et al. [23] used co-surfactant (cocamidopropyl betaine) to SDS solution. There are few reports about the effects of deposition of surfactant crystals. Further investigations are required to determine the actual driving force. The two different driving forces, one of which is capillary action and the other is unclear, simultaneously affect the shape of hydrate. When the concentration of the surfactant is low, hydrate symmetrically forms in height. This is mainly because of the effect of capillary action. In contrast, when the concentration is high, hydrate asymmetrically forms. This is because of a different effect, for example, concentration diffusion, micelle formation, and deposition of SDS crystals. Condensation of the surfactant solution is facilitated where hydrate forms, which enhances hydrate growth. Because hydrate growth is locally facilitated, hydrate forms in an asymmetric shape. 5. Conclusion
10
We have observed the growth behavior of propane hydrate with addition of surfactant to investigate the mechanism of gas hydrate formation along the wall surface. From observation using colored water, the water moves up through the porous hydrate structure and unreacted water remains in the middle of the porous hydrate. As a result, the conversion ratio never reaches 100%. Additionally, from observation of the growth behavior from the top, the hydrate height agrees well with that of water pulled up by capillary action. These results confirm that the driving force for hydrate growth in the vertical direction is capillary action. We also discuss the driving force for hydrate growth in the horizontal direction. The driving force for growth in the horizontal direction is different because the initial concentration of the SDS solution affects the growth behavior in the horizontal direction while it does not affect the hydrate height. We suggest concentration diffusion, micelle formation, and deposition of SDS crystals as candidates for the driving force for hydrate growth in the horizontal direction.
References [1] S. Matsuda, H. Tsuda, Y.H. Mori, Hydrate formation using water spraying onto a cooled solid surface in a guest gas, AIChE J. 52(8) (2006) 2978–2987. [2] Y.T. Luo, J.H. Zhu, S.S. Fan, G.J. Chen, Study on the kinetics of hydrate formation in a bubble column, Chem. Eng. Sci. 62 (2007) 1000–1009. [3] H. Kumano, T. Asaoka, K. Sugaya, Y. Mochizuki, M. Okada, Formation characteristics of propane hydrate using fiber layer, Int. J. Air Cond. Refrig. 23(4) (2015) 1550030. [4] W.F. Kuhs, D.K. Staykova, A.N. Salamatin, Formation of methane hydrate from polydisperse ice powders, J. Phys. Chem. B 110(26) (2006) 13283–13295. [5] Y. Teraoka, M. Okada, S. Kaji, Formation of propane hydrate with melting of ice particles, Trans. JSRAE, 24(3) (2007) 235–241 (in Japanese). [6] A. Kumar, G. Bhattacharjee, B.D. Kulkarni, R. Kumar, Role of surfactants in promoting gas hydrate formation, Ind. Eng. Chem. Res. 54(49) (2015), 12217–12232. [7] F. Wang, Z.Z. Jia, S.J. Luo, S.F. Fu, L. Wang, X.S. Shi, C.S. Wang, R.B. Guo, Effects of different anionic surfactants on methane hydrate formation, Chem. Eng. Sci. 137 (2015) 896–903. [8] F. Wang, L. Wang, C.S. Wang, G. Guo, G.Q. Liu, S.J. Luo, R.B. Guo, Direction controlled methane hydrate growth, Cryst. Growth Des. 15 (10) (2015) 5112–5117. [9] M. Mitarai, M. Kishimoto, D. Suh, R. Ohmura, Surfactant effects on the crystal growth of clathrate hydrate at the interface of water and hydrophobic-guest liquid, Cryst. Growth Des. 15 (2) (2015) 812–821. [10] J. Du, H. Li, L. Wang, Effects of ionic surfactants on methane hydrate formation kinetics in a static system, Adv. Powder Technol. 25 (2014) 1227–1233. [11] K. Okutani, Y. Kuwabara, Y.H. Mori, Surfactant effects on hydrate formation in an unstirred gas/liquid system: An experimental study using methane and sodium alkyl sulfates, Chem. Eng. Sci. 63 (2008) 183–194. [12] Y. Zhong, R.E. Rogers, Surfactant effects on gas hydrate formation, Chem. Eng. Sci. 55 (2000) 4175–4187. [13] H.P. Veluswamy, J.Y. Chen, P. Linga, Surfactant effect on the kinetics of mixed hydrogen/propane hydrate formation for hydrogen storage as clathrates, Chem. Eng. Sci. 126 (2015), 488–499. [14] Y.A. Lim, P. Babu, R. Kumar, P. Linga, Morphology of carbon dioxide–hydrogen–cyclopentane hydrates with or without sodium dodecyl sulfate, Cryst. Growth Des. 13 (5) (2013) 2047–2059. [15] J. Yoslim, P. Linga, P. Englezos, Enhanced growth of methane–propane clathrate hydrate crystals with sodium dodecyl sulfate, sodium tetradecyl sulfate, and sodium hexadecyl sulfate surfactants, J. Cryst. Growth 313(1) (2010) 68–80. [16] H. Hayama, M. Mitarai, H. Mori, J. Verrett, P. Servio, R. Ohmura, Surfactant effects on crystal growth dynamics and crystal morphology of methane hydrate formed at gas/liquid interface, Cryst. Growth Des. 16 (10) (2016) 6084–6088. [17] H. Hayama, M. Mitarai, H. Mori, R. Ohmura, Methane hydrate crystal growth at the gas/liquid interface in the presence of sodium dodecyl sulfate, Procedia Eng. 148 (2016) 339–345. [18] S.Y. Lee, H.C. Kim, J.D. Lee, Morphology study of methane–propane clathrate hydrates on the bubble surface in the presence of SDS or PVCap, J. Cryst. Growth 402 (2014) 249–259. [19] E.D. Sloan, C. Koh, Clathrate Hydrates of Natural Gases, third ed., CRC Press, FL, USA, 2007, pp. 86. [20] J.S. Zhang, S. Lee, J.W. Lee, Does SDS micellize under methane hydrate-forming conditions below the normal Krafft point?, J. Colloid Interface Sci. 315(1) (2007) 313–318. 11
[21] K. Watanabe, S. Imai, Y.H. Mori, Surfactant effects on hydrate formation in an unstirred gas/liquid system: An experimental study using HFC-32 and sodium dodecyl sulfate, Chem. Eng. Sci. 60 (2005) 4846–4857. [22] F. Wang, G. Guo, G.Q. Liu, S.J. Luo, R.B. Guo, Effects of surfactant micelles and surfactant-coated nanospheres on methane hydrate growth pattern, Chem. Eng. Sci. 144 (2016), 108–115. [23] G. Bhattacharjee, O.S. Kushwaha, A. Kumar, M.Y. Khan, J. N. Patel, R. Kumar, Ind. Eng. Chem. Res. 56 (13) (2017) 3687–3698.
12
Highlight Observed the growth behavior of gas hydrate with addition of surfactants. Using colored water, the movement of water was observed. Confirmed the driving force for hydrate growth in the vertical direction is capillary action. Discussed the driving force for hydrate growth in the horizontal direction.
13
S0022-0248(17)30501-8 http://dx.doi.org/10.1016/j.jcrysgro.2017.08.017 CRYS 24268
To appear in:
Journal of Crystal Growth
Received Date: Revised Date: Accepted Date:
4 July 2017 9 August 2017 11 August 2017
Please cite this article as: T. Asaoka, K. Ikeda, Observation of the growth characteristics of gas hydrate in the quiescent-type formation method using surfactant, Journal of Crystal Growth (2017), doi: http://dx.doi.org/10.1016/ j.jcrysgro.2017.08.017
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Observation of the growth characteristics of gas hydrate in the quiescent-type formation method using surfactant Tatsunori Asaokaa,*, Koya Ikedab a
Academic Assembly Institute of Engineering, Shinshu University, 4-17-1 Wakasato, Nagano 380-8553, Japan Department of Mechanical Systems Engineering, Shinshu University, 4-17-1 Wakasato, Nagano 380-8553, Japan
b
*
Corresponding author. E-mail address: [email protected] (T. Asaoka)
Abstract We have observed the growth behavior of gas hydrate with addition of surfactants by several observation methods, and we discuss the mechanism of gas hydrate formation along the wall surface. From observation using colored water, the water moves up through the porous hydrate structure and unreacted water remains in the middle of the porous hydrate. In contrast to previous studies which observed growth from the side, we observed the growth behavior from the top. The hydrate height agrees well with that of water pulled up by capillary action. These results confirm that the driving force for hydrate growth in the vertical direction is capillary action. We discuss the driving force for hydrate growth in the horizontal direction. We suggest that the driving force in the horizontal direction is different because the initial concentration of the surfactant solution affects the growth behavior in the horizontal direction while it does not affect the hydrate height. Keywords:
A1. Growth models, A1. Mass transfer, A1. Heat transfer, A2. Growth from solutions, B1. Gas hydrate
1. Introduction Natural gas hydrate has great potential in natural gas storage and transportation. The storage and transportation costs of natural gas can be decreased by using natural gas hydrate, because natural gas hydrate is stable at higher temperature and lower pressure conditions than liquid natural gas. Gas hydrates are solid-state materials composed of gas and water, which are called clathrate hydrates. There are many types of gas hydrates depending on the guest gas, such as methane hydrate, ethane hydrate, and propane hydrate. In the present work, we used propane hydrate because of its manageability in ambient conditions. It is difficult to continuously form gas hydrates at a high production rate because the formed hydrate covers the surface of water, which inhibits the reaction between the gas and water. Expanding/renewing the gas–liquid interface is necessary to increase the production rate. In industrial fields, splaying [1] and bubbling methods [2] are commonly used, which can be classified as stirred systems. Quiescent systems, in which mechanical agitation or pumping of gas/water is not required, have some advantages. For example, the facility can be simplified and the operation/maintenance costs can be reduced. Various attempts have also been made to enhance the production rate in quiescent systems. Kumano et al. [3] used a fiber layer to increase the gas–liquid interface. Kuhs et al. [4] and Teraoka et al. [5] used fine ice particles. Addition of a surfactant is an effective and convenient method because it can be operated with a simple facility. Addition of surfactant results in continuous formation of gas hydrate without mechanical power, because the gas hydrate grows up along the side wall of the container. A number of researchers have investigated the mechanism of gas hydrate formation [6-12]. They suggest that the driving force for gas hydrate formation should be capillary action, which continuously transports water/surfactant solution to the surface of the gas hydrate. To further investigate the mechanism, Veluswamy et al. [13], Lim et al. [14], Yoslim et al. [15], and Okutani et al. [11] observed the growth of gas hydrates along the wall surface using surfactant solution. However, the gas hydrate covered the observation window and prevented observation of its growth. Hayama et al. [16, 17] observed gas hydrate formation from droplets of surfactant solution and investigated the effect of the surfactant concentration. Lee et al. [18] observed the growth of hydrate around gas bubbles 1
placed in a surfactant solution. Although these types of micro-scale experiments are useful to investigate the mechanism, formation along the wall surface cannot be discussed. In the present study, we used several methods to observe propane hydrate formation, in which colored water was used or the observation angle was varied, and discuss the mechanism of gas hydrate formation along the wall surface.
(a)
(b)
Fig. 1. Experiment apparatus. 1. Constant temperature room. 2. Pressure vessel. 3. Sample container. 4. Sample. 5. Digital camera. 6. Pressure gauge. 7. Thermocouple. 8. Glass windows. 9. Light source. 10. Data logger. 11. Gas reservoir. 12. Constant temperature bath. 13 High pressure cylinder of propane gas. 14. Valves. 15. Relief valve. 16. Vacuum pump. 17. Gas exhaust. 18. Transparent acrylic plate.
2. Experiment apparatus and procedure 2.1. Experiment apparatus Figure 1 shows the experiment apparatus. A stainless-steel pressure vessel with a diameter of 97 mm and a height of 181 mm is placed in a constant temperature room, and a sample container is placed in the pressure vessel. Gas hydrate forms in the sample container by the reaction between the gas and the water/surfactant solution. The pressure vessel shown in Fig. 1(a) has two 15-mm-wide glass windows on the face of the side walls. Formation/crystal growth of the hydrate is observed through the windows. Photographs were periodically taken by a digital camera. Both transmitted and reflected light can be used varying the position of the light source. The temperature and pressure inside the vessel are measured by a thermocouple and a pressure gauge. The thermocouple is T-type and the resolution is 0.1 K. The pressure gauge (GC61, Nagano Keiki Co., Ltd., Tokyo, Japan) is diaphragm type and the resolution is 0.001 MPa. When we observes the sample from the top, the pressure vessel is replaced by that shown in Fig. 1(b), which consists of a transparent plate as the upper wall. The temperature and pressure cannot be measured in this vessel because it is designed to have a wide viewing area. Thus, the pressure vessel is chosen depending on the intended observation direction. The shape/size and material of the sample container can also be changed depending on the aim of the experiment. In this study, we used propane in the experiments because of its manageability close to ambient temperatures and pressures. A gas reservoir was used to keep the pressure in the pressure vessel at the required value. By cooling the reservoir below the liquefied temperature of propane, the propane gas liquefied and liquid propane was contained in the reservoir. Because liquid propane maintains a constant pressure at the vapor–liquid equilibrium, the pressure can be set by controlling the temperature of the reservoir. In the experiments, because we chose a pressure of 0.46 MPa, we kept the reservoir temperature at 0.5 °C, which is the equilibrium temperature at 0.46 MPa. Beforehand the pressure vessel was evacuated to
2
below 0.005 MPa to remove air. Opening the valve between the reservoir and pressure vessel, we can keep the pressure in the pressure vessel at 0.46 MPa. Initial formation of the gas hydrate takes a long time after water is exposed to the gas because of supercooling. To inhibit supercooling, a small hydrate crystal of about 5 g, which acts as nucleus for crystal growth, was placed in the sample container in all of the experiments. A hydrate–water mixture was obtained in the experiments because the sample water never completely converted to propane hydrate. We will call this mixture the mixed product. 2.2. Observation of the formation process Propane hydrate formed along the inner wall of the sample container. A glass container with a diameter of 90 mm and height of 110 mm was used as the sample container. The inner surface of the sample container was coated with a non-ionic hydrophilic surfactant (B-542, LEC, Inc., Tokyo, Japan) as an anti-fog spray. Distilled water (100 g) was supplied to the sample container as the sample. The temperature in the sample container was measured by a thermocouple. Propane gas was injected into the pressure vessel and propane hydrate started to form in the sample container. The temperature of the sample container increased with formation of propane hydrate, although the temperature in the constant temperature room where the pressure vessel was placed was kept at 2 ± 1 °C throughout the experiment. Using colored water, we attempted to observe the movement of water accompanied by propane hydrate formation. At the beginning, propane hydrate formed in the sample container from the uncolored water solution. After the propane hydrate reached the top of the container, we opened the pressure vessel and supplied colored water (50 g) to the sample container. After evacuation of air and injection of propane gas, we restarted observation of the growing propane hydrate and movement of the colored water. Red food coloring containing 85% dextrin and 15% food red No. 102 was used to color the water. Observation was continued until the temperature of the water inside the sample container decreased and reached the temperature of the constant temperature room, which indicates completion of the reaction/formation of propane hydrate. The mixed product was then taken out from the pressure vessel and the surface and cross-section were observed. 2.3. Conversion ratio Mass of the propane hydrate formed in the experiment was estimated by following procedure, and the variation of the mass was investigated. A copper container, which has a good thermal conductivity, with a diameter of 68 mm and height of 133 mm was used as the sample container. The inner surface was coated with a non-ionic hydrophilic surfactant, similar to the experiment in Section 2.2. Distilled water was supplied in the sample container and the mass was varied by the experiments (100, 70, or 50 g). Observation was not performed in this experiment. The temperature of the sample container increased with propane hydrate formation, although the temperature in the constant temperature room was kept at 2 ± 1 °C throughout the experiment. We define time 0 as the time when the temperature of the sample container started to increase and the reaction time as the time from time 0 until the end of the experiment. Several experiments were performed varying the reaction time. After the experiment, the mixed product was taken out of the pressure vessel and the following measurements were performed. We define the conversion ratio f as the ratio of the mass of propane hydrate in the mixed product mh to the theoretical mass of propane hydrate when all of the supplied water is converted to propane hydrate mh0. First, the mass of the mixed product mf was measured. The resolution of the weight scale was 0.1 g. We then charged the mixed product into hot water (>70 °C) filled in a Dewar bottle, and the variation of mass of the bottle was measured. The mass of the water obtained by the dissociation of the mixed product mw can be determined by the mass difference, because the water remains in the hot water while the gas is released into the surrounding. The mass of gas mg was calculated by
mg m f mw
(1)
Given that the hydration number of propane hydrate is 17 [19] and the ratio of the molar weight of propane to water is 44:18, mh can be calculated from mg as follows: 3
mh
350 mg 44
(2)
In addition, mh0 can be calculated by
mh 0
350 mw 306
(3)
From the results, f was calculated using the measured values of mf and mw: f
mh 306 m f mw mh 0 44 mw
(4)
2.4. Effect of the wall material and surfactant concentration The effect of the wall material of the sample container was also investigated. The pressure vessel shown in Fig. 1(b), which has a wide viewing area, was used and we observed the sample from the top. Two types of sample containers were used in this experiment. One was a polypropylene container with a diameter of 79 mm and height of 12 mm, as shown in Fig. 2(a), and the other was the same container covered with an aluminum film, as shown in Fig. 2(b). Sodium dodecyl sulfate (SDS) solution (10 ppm) was used as the sample. The mass of the sample was 100 g and the sample was colored to visualize it clearly. SDS is a popular surfactant and is suitable for this experiment because the physical properties of SDS solution are well known. For example, the critical micelle concentration (CMC) is about 1780 ppm [20] and the surface tension has been reported [21]. We confirmed that the trend of propane hydrate formation was almost same when SDS and the non-ionic hydrophilic surfactant (Section 2.2) were used. We also investigated the effect of the surfactant concentration. A glass container with a diameter of 60 mm and height of 100 mm was used as the sample container. SDS solution (10 or 1000 ppm) was used as the sample. The mass of the sample was 100 g and the sample was colored to visualize it clearly.
(a) Polypropylene surface
(b) Aluminum surface
Fig. 2. Sample containers used for observing the growth behavior on the different wall surfaces. The containers are both made of polypropylene but the container in (b) is covered with an aluminum film.
3. Results 3.1. Observation of the formation process Figure 3 shows the growth behavior of propane hydrate on the glass wall observed from the side. As shown in Fig. 3(a), the window on the opposite side and light reflection from the bottom can be observed in the each photograph. The width of the window is 15 mm. The pictures were taken using transmitted light. Time 0 is defined as the time when gas was supplied to the pressure vessel. Figure 3(b) shows immediately after gas supply. The small propane hydrate crystal can be observed below the water surface, which was placed beforehand as nucleus for crystal growth. Initially, film of propane hydrate grows on the front surface of the sample container (Fig. 3(c) and (d)). As shown in Fig. 3(e), a crack forms in the propane hydrate 4
film, which may be caused by the increase of the volume of the hydrate film. A crack was observed for all of the experiments performed under the same conditions. Colored water was supplied into the sample container at 195 min (Fig. 3(f)). The edge of the propane hydrate crystal becomes slightly smoother in Fig. 3(f) compared with Fig. 3(e) because of dissociation of propane hydrate by the effect of evacuation of air and injection of propane gas. From Fig. 3(g), the colored water moves upward. This indicates that there are gaps in the propane hydrate body or between the propane hydrate body and the wall surface, and water moves through the gaps by capillary action. It should be noted that the surfactant, which covers the inner surface of the sample container, has an important role, because this water movement does not occur without the surfactant.
(a)
(b)
(c)
(d)
(e)
(f)
(g)
Fig. 3. Growth behavior of propane hydrate on the glass wall observed from the side. (a) is the schematic of the view from the window. Time after gas injection are (b) 1 min, (c) 90 min, (d) 164 min, (e) 192 min, (f) 195 min, (g) 200 min.
Figure 4 shows the appearance of the mixed product formed in the above experiment. As shown in Fig. 4(f), the sample for the observation was cut out of the original cylindrically shaped mixed product, and we observed the sample from the directions of view A, B, and C. Figures 4(a) and 4(b) show the surfaces facing the side wall of the container, and Figs. 4(c) and 4(d) show the surfaces exposed to the gas. Figures 4(a) and 4(c) show the sample just after preparation. Figures 4(b) and 4(d) show the sample after slight dissociation, which was kept under ambient conditions for a few minutes. Figure 4(e) shows the cross-section of the sample just after preparation. Although propane hydrate is naturally transparent, as shown in Fig. 3, the propane hydrate that formed in the experiments is white because of diffuse reflection on its porous surface. Additionally, propane hydrate is not colored because it does not capture the coloring material. Because the liquid and the discharged coloring material appear red, we can determine the residual water/surfactant solution by red in the images. In Fig. 4(a), there are many large gaps on the surface. This means that the formed propane hydrate is porous and has large pores around the surface. We can observe large cracks in Fig. 4 (c), which were formed by the volume expansion during hydrate formation, as shown in Fig. 3(e). In Fig. 4(b) and 4(d), there are many small pores around the both surfaces, although there are long gaps only in Fig. 4(b). This means large long gaps exist only around the surface facing the side wall of the container. We suggest that large long gaps were formed by the water flow caused by capillary action. In Fig. 6, the sample is colored in the middle of the propane hydrate layer, while it is white near the surfaces. This means that unreacted water remains in the middle of the mixed product. Because we performed the experiment for a sufficient reaction time, we suggest that reaction of water is inhibited in the middle of the propane hydrate layer.
5
(a) View A. Just after preparation
(b) View A. After slight dissociation
(c) View B. Just after preparation
(d) View B. After slight dissociation
(e) View C. Just after preparation
6
(f) Schematic diagram of the observation directions. Fig. 4. Appearance of hydrate formed in the experiment. View A is the container wall side, view B is the exposed side, and view C is the cross-section of the hydrate.
Conversion ratio f , %
100 80 60 Mass of supplied water 100g 70g 50g
40 20 0
0
20
40 60 80 Reaction time, h
100
Fig. 5. Variation of the conversion ratio with reaction time.
3.2. Conversion ratio Figure 5 shows the relationship between the reaction time and the conversion ratio. The conversion ratio increases with increasing reaction time until about 22 h. After 22 h, the conversion ratio becomes constant at about 90%. It was confirmed that the conversion ratio never reaches 100% even if for a reaction time of more than 90 h. Additionally, from the temperature history, the temperature of the sample increased owing to propane hydrate formation and it decreases to the surrounding temperature after about 22 h. This means that formation of propane hydrate stops at about 22 h and unreacted water remained. As shown in Fig. 5, even when we vary the mass of supplied water, the same trend is observed. Moreover, the same trend has been reported by Okutani et al. [11]. 3.3. Effect of the wall material and surfactant concentration Figure 6 shows the shapes of the propane hydrates formed from 10 ppm SDS solution in containers with different wall surfaces. The time is more than 22 h after gas injection and the images were taken from the top. The maximum heights of propane hydrate are given by white line in the figures. The maximum height is 52 mm using polypropylene, while it is 85 mm using aluminum. The height on the aluminum surface is 1.6 times larger than that on the polypropylene surface. We believe that this is the effect of capillary action between the body of propane hydrate and the side wall of the container. We performed the same experiment several times and confirmed the measured heights were almost the same. Additionally, it should be noted that the heat transfer along with the aluminum film is negligible, because the aluminum film is thin. 7
Moreover the heat flax along with the aluminum film is small, because the formation of hydrate is slow and the temperature distribution on the aluminum film is small.
(a) Polypropylene surface
(b) Aluminum surface
Fig. 6. Propane hydrate formed in containers with different wall surfaces using 10 ppm SDS solution. The time is more than 22 h after gas injection.
Fig. 7. Model of capillary action occurring between porous hydrate body and container wall during the propane hydrate formation process. is contact angle on each surface, h is the height of water pulled up by capillary action, and r is the pore radius.
Figure 7 shows the model of capillary action during the propane hydrate formation process. It shows the capillary action between porous propane hydrate and the container wall. The height of water pulled up by capillary action h is estimated by h
2 cos w rg
(5)
where , , w, r, and g are the surface tension, contact angle on the container surface, density of the water/surfactant solution, radius of the capillary tube, and acceleration by gravity, respectively. h is proportional to cos because the other variables are constant in the experiment. The contact angles of droplets of 10 ppm SDS solution on the polypropylene and aluminum surfaces were measured to be 54° (cos 54° = 0.588) and 30° (cos 30° = 0.866), respectively. From these results, we can estimate that h on the aluminum surface is 1.5 times larger than that on the polypropylene surface, because the cos value is 1.5 times larger for the aluminum surface than for the polypropylene surface. As mentioned above, the height of propane hydrate formed on the aluminum surface is 1.6 times larger than that formed on the aluminum surface, and this ratio agrees well with the ratio of the h values. This shows that formation of propane hydrate occurs because of the supply of water/surfactant solution by capillary action between porous propane hydrate and the container wall. Because water/surfactant solution cannot be supplied to higher than h, propane hydrate cannot grow above this height. By substituting the contact angle on each surface into Eq. (5), we estimated the pore radius, which corresponds to r, to be 0.08 mm for the polypropylene surface and 0.07 mm for the aluminum surface. The surface tension is assumed to be 35 mN/m for the 1780 ppm (at the CMC) SDS solution at 6 °C [21] and the density w is assumed to be approximately 1000 8
kg/m3. Because the estimated pore radius were almost the same (both of them were about 0.1 mm), we consider it is independent of the material of the container surface. Although we attempted to measure the pore radius of the formed propane hydrate, we failed because we could not fix the sample on the scale owing to dissociation of the sample.
(a) 48 min (10 ppm SDS solution)
(b) 82 min (10 ppm SDS solution)
(c) 114 min (10 ppm SDS solution)
(d) 295 min (10 ppm SDS solution)
(e) 6 min (1000 ppm SDS solution)
(f) 40 min (1000 ppm SDS solution)
(g) 88 min (1000 ppm SDS solution)
(h) 241 min (1000 ppm SDS solution) 9
Fig. 8. Growth behavior of propane hydrate observed from the top. The material of the sample container is glass.
Figure 8 show the growth behavior of propane hydrate using 10 and 1000 ppm SDS solution, respectively. A glass container with a diameter of 60 mm and height of 100 mm was used in the experiments. The images were taken from the top. As shown in Fig. 8 (a)-(d), when 10 ppm SDS solution is used, propane hydrate grows almost symmetrically in the vertical direction. Initially, growth in the horizontal direction is not significant. Horizontal growth starts at about 100 min, as shown in Fig. 4(c). In contrast, as shown in Fig. 8(e)-(h), when 1000 ppm SDS solution is used, propane hydrate asymmetrically grows in the vertical direction. In the initial period, a number of long dendritic shaped crystals grow in the horizontal direction. The maximum heights of propane hydrate are almost same for 10 and 1000 ppm SDS solution. This indicates that the driving forces for propane hydrate growth in the vertical and horizontal directions are different. This will be discussed in Section 4. 4. Discussion of crystal growth As mentioned in Section 3, formation of hydrate occurs because of the supply of water/surfactant solution by capillary action between porous hydrate and the container wall. Although this has been already suggested by other researchers, for example Okutani et al. [11], we provide additional evidence in the present work. From observation using colored water, we show the water moves up through the porous hydrate structure and unreacted water remains in the middle of the porous hydrate. As a result, the conversion ratio never reaches 100%. We also show that the hydrate height agrees well with that of water pulled up by capillary action. From these results, we confirmed that the driving force for hydrate growth in the vertical direction is capillary action. We suggest the driving force for hydrate growth in the horizontal direction is different. The initial concentration of the SDS solution affects the growth behavior in the horizontal direction. However, it does not affect the hydrate height. We believe the condensation of SDS solution is important for the hydrate growth in the vertical direction. The concentration of the SDS solution increases with hydrate formation because water is converted to hydrate. The surface tension and contact angle decrease with increasing SDS concentration, but the concentration finally reaches a constant value above the saturated concentration or CMC. Considering micelles do not form at the experimental temperature [20], we expect that the concentration of the SDS solution reaches the saturated concentration at the top of the formed hydrate regardless of its initial concentration. This would be the reason why the initial concentration does not affect the hydrate height. Considering hydrate growth in the horizontal direction, the driving force for water movement is not capillary action because it is strongly affected by the initial concentration of the SDS solution. There are various candidates for the driving force, such as concentration diffusion, micelle formation, and deposition of SDS crystals. If the concentration on the surface exposed to the gas becomes higher than that in the porous hydrate structure because of condensation of the surfactant solution, water should move from the middle to the surface by concentration diffusion. As the initial concentration of surfactant solution increases, the flow rate of water increases and formation of hydrate in the horizontal direction should be facilitated. About the effects of micelle formation, some interesting investigations were reported. Wang et al. [22] used mimic micelles, and Bhattacharjee et al. [23] used co-surfactant (cocamidopropyl betaine) to SDS solution. There are few reports about the effects of deposition of surfactant crystals. Further investigations are required to determine the actual driving force. The two different driving forces, one of which is capillary action and the other is unclear, simultaneously affect the shape of hydrate. When the concentration of the surfactant is low, hydrate symmetrically forms in height. This is mainly because of the effect of capillary action. In contrast, when the concentration is high, hydrate asymmetrically forms. This is because of a different effect, for example, concentration diffusion, micelle formation, and deposition of SDS crystals. Condensation of the surfactant solution is facilitated where hydrate forms, which enhances hydrate growth. Because hydrate growth is locally facilitated, hydrate forms in an asymmetric shape. 5. Conclusion
10
We have observed the growth behavior of propane hydrate with addition of surfactant to investigate the mechanism of gas hydrate formation along the wall surface. From observation using colored water, the water moves up through the porous hydrate structure and unreacted water remains in the middle of the porous hydrate. As a result, the conversion ratio never reaches 100%. Additionally, from observation of the growth behavior from the top, the hydrate height agrees well with that of water pulled up by capillary action. These results confirm that the driving force for hydrate growth in the vertical direction is capillary action. We also discuss the driving force for hydrate growth in the horizontal direction. The driving force for growth in the horizontal direction is different because the initial concentration of the SDS solution affects the growth behavior in the horizontal direction while it does not affect the hydrate height. We suggest concentration diffusion, micelle formation, and deposition of SDS crystals as candidates for the driving force for hydrate growth in the horizontal direction.
References [1] S. Matsuda, H. Tsuda, Y.H. Mori, Hydrate formation using water spraying onto a cooled solid surface in a guest gas, AIChE J. 52(8) (2006) 2978–2987. [2] Y.T. Luo, J.H. Zhu, S.S. Fan, G.J. Chen, Study on the kinetics of hydrate formation in a bubble column, Chem. Eng. Sci. 62 (2007) 1000–1009. [3] H. Kumano, T. Asaoka, K. Sugaya, Y. Mochizuki, M. Okada, Formation characteristics of propane hydrate using fiber layer, Int. J. Air Cond. Refrig. 23(4) (2015) 1550030. [4] W.F. Kuhs, D.K. Staykova, A.N. Salamatin, Formation of methane hydrate from polydisperse ice powders, J. Phys. Chem. B 110(26) (2006) 13283–13295. [5] Y. Teraoka, M. Okada, S. Kaji, Formation of propane hydrate with melting of ice particles, Trans. JSRAE, 24(3) (2007) 235–241 (in Japanese). [6] A. Kumar, G. Bhattacharjee, B.D. Kulkarni, R. Kumar, Role of surfactants in promoting gas hydrate formation, Ind. Eng. Chem. Res. 54(49) (2015), 12217–12232. [7] F. Wang, Z.Z. Jia, S.J. Luo, S.F. Fu, L. Wang, X.S. Shi, C.S. Wang, R.B. Guo, Effects of different anionic surfactants on methane hydrate formation, Chem. Eng. Sci. 137 (2015) 896–903. [8] F. Wang, L. Wang, C.S. Wang, G. Guo, G.Q. Liu, S.J. Luo, R.B. Guo, Direction controlled methane hydrate growth, Cryst. Growth Des. 15 (10) (2015) 5112–5117. [9] M. Mitarai, M. Kishimoto, D. Suh, R. Ohmura, Surfactant effects on the crystal growth of clathrate hydrate at the interface of water and hydrophobic-guest liquid, Cryst. Growth Des. 15 (2) (2015) 812–821. [10] J. Du, H. Li, L. Wang, Effects of ionic surfactants on methane hydrate formation kinetics in a static system, Adv. Powder Technol. 25 (2014) 1227–1233. [11] K. Okutani, Y. Kuwabara, Y.H. Mori, Surfactant effects on hydrate formation in an unstirred gas/liquid system: An experimental study using methane and sodium alkyl sulfates, Chem. Eng. Sci. 63 (2008) 183–194. [12] Y. Zhong, R.E. Rogers, Surfactant effects on gas hydrate formation, Chem. Eng. Sci. 55 (2000) 4175–4187. [13] H.P. Veluswamy, J.Y. Chen, P. Linga, Surfactant effect on the kinetics of mixed hydrogen/propane hydrate formation for hydrogen storage as clathrates, Chem. Eng. Sci. 126 (2015), 488–499. [14] Y.A. Lim, P. Babu, R. Kumar, P. Linga, Morphology of carbon dioxide–hydrogen–cyclopentane hydrates with or without sodium dodecyl sulfate, Cryst. Growth Des. 13 (5) (2013) 2047–2059. [15] J. Yoslim, P. Linga, P. Englezos, Enhanced growth of methane–propane clathrate hydrate crystals with sodium dodecyl sulfate, sodium tetradecyl sulfate, and sodium hexadecyl sulfate surfactants, J. Cryst. Growth 313(1) (2010) 68–80. [16] H. Hayama, M. Mitarai, H. Mori, J. Verrett, P. Servio, R. Ohmura, Surfactant effects on crystal growth dynamics and crystal morphology of methane hydrate formed at gas/liquid interface, Cryst. Growth Des. 16 (10) (2016) 6084–6088. [17] H. Hayama, M. Mitarai, H. Mori, R. Ohmura, Methane hydrate crystal growth at the gas/liquid interface in the presence of sodium dodecyl sulfate, Procedia Eng. 148 (2016) 339–345. [18] S.Y. Lee, H.C. Kim, J.D. Lee, Morphology study of methane–propane clathrate hydrates on the bubble surface in the presence of SDS or PVCap, J. Cryst. Growth 402 (2014) 249–259. [19] E.D. Sloan, C. Koh, Clathrate Hydrates of Natural Gases, third ed., CRC Press, FL, USA, 2007, pp. 86. [20] J.S. Zhang, S. Lee, J.W. Lee, Does SDS micellize under methane hydrate-forming conditions below the normal Krafft point?, J. Colloid Interface Sci. 315(1) (2007) 313–318. 11
[21] K. Watanabe, S. Imai, Y.H. Mori, Surfactant effects on hydrate formation in an unstirred gas/liquid system: An experimental study using HFC-32 and sodium dodecyl sulfate, Chem. Eng. Sci. 60 (2005) 4846–4857. [22] F. Wang, G. Guo, G.Q. Liu, S.J. Luo, R.B. Guo, Effects of surfactant micelles and surfactant-coated nanospheres on methane hydrate growth pattern, Chem. Eng. Sci. 144 (2016), 108–115. [23] G. Bhattacharjee, O.S. Kushwaha, A. Kumar, M.Y. Khan, J. N. Patel, R. Kumar, Ind. Eng. Chem. Res. 56 (13) (2017) 3687–3698.
12
Highlight Observed the growth behavior of gas hydrate with addition of surfactants. Using colored water, the movement of water was observed. Confirmed the driving force for hydrate growth in the vertical direction is capillary action. Discussed the driving force for hydrate growth in the horizontal direction.
13