Investigation on Gas Storage in Methane Hydrate

Journal of Natural Gas Chemistry 13(2004)107-112CIENCE PRESSInvestigation on Gas Storage in Methane HydrateZhigao Sun", Rongsheng Ma, Shuanshi Fan2, Kaihua Guo2, Ruzhu WangiI. School of Envirommental Science and Engineering, Yangzhou University, Yangzhou 225009, China;2. Guangzhou Institute of Energy Conversion, Guangzhou 570010, Ch3. Institute of Refrigeration and Cryogenics, Shanghai Jiaotong University, Shanghai 200090, ChinaManuscript received April 05, 2004; revised April 23, 2004Abstract: The effect of additives(anionic surfactant sodium dodecyl sulfate(SDS), nonionic surfactantalkyl polysaccharide glycoside(APG), and liquid hydrocarbon cyclopentane(CP)on hydrate inductiontime and formation rate, and storage capacity was studied in this work. Micelle surfactant solutions werefound to reduce hydrate induction time, increase methane hydrate formation rate and improve methanetorage capacity in hydrates. In the presence of surfactant, hydrate could form quickly in a quiescentsystem and the energy costs of hydrate formation were reduced. The critical micelle concentrations of SDSand APG water solutions were found to be 300x 10-6 and 500x10-6 for methane hydrate formation systemrespectively. The effect of anionic surfactant(SDS)on methane storage in hydrates is more pronouncedcompared to a nonionic surfactant(APG). CP also reduced hydrate induction time and improved hydrateformation rate, but could not improve methane storage in hydrates.Key words: methane hydrate, surfactant, cyclopentane, gas storage1. IntroductionGas hydrates have drawn much attention today as notonly a new natural energy resource but also a newNatural gas is a kind of clean fuel. As the main means for natural gas storage and transportcomponent of natural gas is methane whose boilingNatural gas storage in hydrates has been investi-point is about"C under 0.1 MPa, it is difficult gated because hydrates store large quantities of nat-to store and transport. There are two main methods ural gas 2. Gudmundsson et al. 3) showed that hyfor natural gas storage and transport today. One is drate could be stored at -15oC under atmosphericliquefied natural gas (LNG), and the other is natural pressure for 15 days, retaining almost all the gasgas storage and transport by pipeline, and both the Their study [4] also showed that a substantial costmethods are costly. In order to enlarge naturalsaving(24%)for the transport of natural gas in hyconsumers in China, economic method for natural gas drates compared to LNG from the northern Northstorage and transport should be foundSea to Central Europe. The hydrate formation withNatural gas hydrates are crystalline inclusion additives for the purpose of natural gas storage andcompounds composed of water and natural gas. There transport has been reported recently[2,5-8. How-are three known structures(structure I, structure II ever, industrial applications of hydrate storage pro-and structure H) in which water molecules arrange cesses have been hindered by some problems, such asthemselves around guest molecules, depending prin- slow formation rates, unreacted interstitial water as acipally on the molecular size of guest molecules [1, 2. large percentage of the hydrate mass, the reliabilityCorrespondingauthor.E-mail:szgyzu@163.com;Tel:0514-7863852Supported by Chinese Jiangsu Province Education Committee Progra中国煤化工 Science Foundation(50176051)YHaCNMHGZhigao Sun et al. /Journal of Natural Gas Chemistry Vol. 13 No 220(4of hydrate storage capacity and economy of process K Jacket coolant. of the cell is circulating cooling wath enough ethylene glycol to depress theThe main purpose of this work is to investigate freezing point. The coolant is circulated from a refrig.the effect of anionic surfactant, nonionic surfactant erating bath capable of maintaining the bath temper-and liquid hydrocarbon on mcthane hydrate forma- ature within +0.01 K of the set point to a low tempertion, particularly to investigate their ability of reduc- ature capability of 258 K. The cell is enclosed with in-ing induction time, increasing the formation rate of sulation. Two platinum resistance thermometers weregas hydrate, improving hydrate storage capacity and used to measure the experimental temperature, withreducing energy costs of hydrate formationan accuracy of +0. 01 K One extends into the bottomof the cell, the other extends into the gas phase at the2. Experimental apparatusp. The pressure of the cell was measured using a 10MPa gauge with an accuracy of +0. 25% of full scale.Hydrate formation experiments were carried outA constant pressure regulator can maintain conng a high-pressure system shown in Figure 1. Gas stant pressure of the cell. a mass gas flowmeter washydrates form in a cylindrical high-pressure stainless used to measure gas added to the cell during hydratesteel cell with available volume of about 1000 cm. formation. The flowmeter has a capacity of 0-1 stana stainless flange, which has appropriate ports for dard liter per minute with an accuracy of +2% of fullaccess to the interior, is used to seal the cell on the scale. There is a data-logger to record parameters intop. The cell is designed to operate at pressurethe process of hydrate formation as the function ofto 20 MPa and temperature in the range of 253-323 timePressure regulatorThermometeracuum pumpCoolantThermometerGas supplyComputertic diagram3. Experimental procedurehen evacuated by a vacuum pump. Approximately00 cmwater solution was charged into the cell forOne anionic surfactant(SDS), one nonionic sur- each experiment. Methane was then injected into thefactant(AG)and one liquid hydrocarbon(CP)wereell up to about 1.0 MPa. The system was cooled tochosen as the hydrate promoters in this work. The 274.05- 277. 55 K with a pressure below the methanepressure of the cell was kept constant during each hydrate formation pressure, Pressure of the cell wasexperimental run. A typical procedure was as folal pressure over a 3ows: the cell was first rinsed with distilled water, andH中国煤化工rbon gas into the cell.CNMHGJournal of Natural Gas Chenistry Vol. 13 No, 2 2004109The beginning of hydrate formation was judged by the iment in a stirring system is about 5 min. The induc-gas consumption, which was shown by a flowmeter. tion time of water -SDS-methane and water-SDS-CPThe hydrate formation was considered to be at an methane experiments in a quiescent system is 1-1.5end when the methane consumption speed was less h and 10-20 min, respectively. The experimental re-than 0.04x10-3m/min. Hydrate formation as well sults showed that additives reduce the induction times temperature and gas mass flow were recorded and greatly in a quiescent system. The explanation for CPdisplayed on the data acquisition systemreducing the induction time of hydrate formation wasThe test materials used in this work were given in that there was a significant shift in hydrate formationTable 1. The chemicals were not further purified be. pressures to lower values in the system of methane andfore usage. Surfactants were weighed on an electronic CP compared to the system of methane [10balance with a readability of +0. 1 mg. Distilled wa-ter was used in all experiments. Water and CP were 4.2. Effect of surfactant on hydrate formationweighed on an electronic balance with a readability of rate and storage capacity士001gMethane storage capacity in hydrates could beTable 1. Test materialscalculated by the consumption of methane in theComponent Purity(‰)hydrate formation process experiments [11].ItwasMethaneshown in Figure 2 that the formation rate of methaneCP98.4Meilong Chenical Cohydrate and methane storage capacity in hydrates>98 Guangzhou Chenical Reagent Cowere very small in a stirring pure water system. Sur-98 Guangzhou Chernical Reagent CoDistilledfactant SDS was then added to improve methane fomation in a quiescent system. Methane hydrate for-mation rate and gas storage capacity were improvedgreatly in the presence of surfactant SDS compared to4. Experimental results and discussionthe hydrate formation in a stirring pure water system4.1. Effect of additives on hydrate formationnduction timeNucleation of a hydrate crystal requires an ex-cess energy to create a nucleus surface. Since a therbe related to the difference in the chemical potentialof the hydrate components in the liquid and hydratephases at the existing thermodynamic conditions, isnecessary to overcome a nucleation barrier, hydratenucleated from a solution which is cooled below theequilibrium temperature. Induction time (9 of hydrate formation is mainly dependent on the time of Figure 2. Surfactant increases hydrate formation ratehydrate nucleation(1)300x10-6 SDS, (2) Pure water 400 rpm; Reaction condiIn the initial work, the ways of reducing induction tions: p= 5.76 MPa, T=277.55 Ktime of hydrate formation were first studied. Underthe conditions of 5.76 MPa and 277. 55 K, the expere above experiments showed that hydrate foriments of pure water-methane(with 400 rpm stirring mation could be carried out in a quiescent system invelocity and without stirring), water-SDS-methane the presence of a surfactant. Hydrate formation in aand water-SDS-CP-methane in a quiescent system quiescent system can reduce processing costs as stirwere carried out to observe the effect of stirring and ring is not needed. The following experiments for gasadditives on hydrate formation induction time. The storage in this work were carried out in the presenceinduction time of pure water-methane hydrate experi- of a surfactant in a quiescent systemment in a quiescent system is more than 28 h, and theH中国煤化工 nism of gas hyinduction time of pure water-methane hydrate exper- drateform at the in-CNMHG110higao Sun et al. / Journal of Natural Gas Chemistry Vol. 13 No. 22004terface between water and gas in a pure water sys- result of hydrate formation. Figure 3(a)showed thattem as the solubility of gas in water is small. Hy- the hydrate formation rate and storage capacity weredrates covered on the gas-water interface block fur- very low over a period of about 20 hours without addi-ther conversion of water to hydrate. In the presence of tives in a stirred system(stirring velocity is about 400surfactant, there are three conditions that contribute rpm). There was no apparent adsorption or symmet-to hydrate formation. First, solubility of hydrocar- rical packing on the metal cell surface Hydrate cov.bon gas is improved greatly in the presence of sur- ered on the gas-water interface. Figure 3(b)showedfactant [ 12. Second, there is a layer of water(10 that the hydrate formation rate and storage capacity100 nm thick) between water and metal cell surface, were large over a period of about 8 h in the pres-whose structure is described as an"ice-like"molecular ence of 300x10-6 SDS in a quiescent system.Theconfiguration 13 Surfactant displaces this tightly dark center shows the bottom of the cell. There wereheld water on metal. Surfactant micelle could form apparent adsorption and symmetrical packing on thewith sufficient surfactant adsorption on the metal sur- metal surface. a possible explanation for the promo-face and solubilize the hydrocarbon gas in contact tion effect on hydrate formation and storage capacitywith the configured water. The " ice-like"water is was that the agglomerating hydrate particles movedfavored to hydrate formation. Third, the metal sur- radially to be adsorbed on the cell walls in a quies-face dissipates the latent heat of hydrate formation cent system with SDS presence. The adsorption ofquickly by conduction. Hydrate could form both at hydrate on the walls prevented the hydrate from hinvapor-water interface and at subsurface of bulk water, dering further conversion of water to hydrate in theand at a high rate in the presence of a surfactant.Figure 3 was a photograph of the experimentalFigure 3. Photograph of experimental result of hydrate formationL Hydrate formation in pure water in a stirred system,(b) Hydrate formation in SDS solutions in a quiescent system4.3. Critical micelle concentration of surfac- surfactant in solutions that gave the highest methanetants in hydrate formation systemstorage capacity in hydrates is the critical micelle con-The solubility of methane is small, but it can beFigure 4 showed the relation of gas storage caimproved by adding surfactant [11. Micelle forms in pacity in hydrates and surfactant concentration. Gassolutions and its solubility attains a maximal value storage capacity in hydrates first increased sharply,phen surfactant concentration reaches some value. and then dropped a little with the increase ofThis concentration is defined as the critical micelle factant concentration. This showed that there wasconcentration. If surfactant concentration in solution the largest storage capacity when the concentrationsexceeds the critical micelle concentration, solubility of surfactant solutions for SDS and APG were aboutof gas cannot be further improved [11]. In this work, 300- and 500x10-b, respectively. It also showedour main aim is to improve the gas storage capacity中国煤化工 ntration of SDs wain hydrates, So we define that the concentration ofHnd that of APg wasCNMHGJournal of Natural Gas Chemistry Vol. 13 No. 22004111500x10-0 for gas storage in hydrates. The follow-The effect of SDs or APG on methane hydrg experiments of methane storage were done under formation under the conditions of 4.34 MPa andcritical micelle concentration water solution274.05 K was shown in Figure 5. The hydrate formation rate in aqueous APG solution was almost equalto that in aqueous SDS solution during hydrate for-mation(0 215 seconds) after hydrate induction timeBut the time of hydrate formation in aqueous APGsolutions at hydrate growth stage was shorter tharthat in aqueous SDS solutions. The effect of SDSmethane storage in hydrates was more pronouncedcompared to APG20Figure 4. Effect of surfactant concentration on gas(1) APG at275.15K,(2) SDS at276.75K;4.4. Methane storage in hydratesSixteen methane storage experiments in hydrateswere carried out in the presence of additives(SDSFigure 5. Comparison of the effect of different sur-factants on methane hydrate formationAPG, or CP)in a quiescent system. The experimen- (1)300x 10-6 SDS, (2)500x10-6 APGtal results were tabulated in Table 2Reaction conditions: p= 4.34 MPa, T=274.05 KTable 2. Experimental results of methane storage in hydratesAdditonT/KP/MPaExperimental time cost(minStorage capacity(V/v42512345678927755326APG274.0370SDS+CP274.056.18123456329277.55Solution concentration: DPG=500x 10-6, SDS=300x10-6, CP=1.0%Figure 6 compared the effect of SDS and SDS+ CP storage capacity, and higher hydrate formation rateon hydrate formation. It showed that use of these ad- under the same expcrimental pressure. It also showedditives brought about lower temperature, larger gas that中国煤化工 nation rate.butCNMHGZhigao Sun et al. Journal of Natural Gas Chemistry Vol. 13 No. 2 2004ad no effect on the gas storage capacity in hydrates. also increase hydrate formation rate and reduce hy-That is to say, CP shortened the time costs of hy- drate formation induction time, but could not im-drate formation. CP can form structure II hydrates prove nethane storage in the hydrates. The tests9. The system of SDS+CP+methane formed a mix- also showed that methane hydrate could be formedture of structure I and structure II hydrates in this in a quiescent system in the presence of SDS or APGwork as only 1.0%CP was involved in the experimen-tal system.References[1 Sloan E D Clathrate Hydrates of Natural Gases. 2nd. New York: Marcel Dekker Inc. 1998. 402 Khokhar AA, Gudmundsson J S, Sloan E D. Fluid(2)Phase Equilibria, 1998, (150-151):383[3] Gudmundsson J S, Borrehaug A. Frozen Hydrate forTransport of Natural Gas. In: Guillon O. 2u Inter-national Conference on Nature Gas Hydrate. Francune26,1996.415[4 Gudmundsson J S, Borrehaug A Petroleum Review996,50:2325] Saito Y, Kawasaki T, Okui T, Kondo T, Hiraoka R.In: Guillon O. Methane Storage in Hydrate Phasewith Soluble Guests. 2"d International Conference orFigure 6. Comparison of the effect of SDs andature Gas Hydrate. France: Toulouse, June 2-6SDS+CP on methane hydrate formation1996.459(U)300x 10-6 sDs at 274.05K, (2)300x 10-6 SDS at 277.55 (6 Karaaslan U, Uluneye E, Parlaktuna M Journal ofK,(3)300×10-6sDs+10%CPat27405K,(4)300×10-6Petroleum Science and Engineering, 2002, 35(1-2)SDS+1. 0%CP at 277.55 K7] Zhong Y, Rogers R E. Chemical Enginecring Science,2000,55(19):41755. Conclusions8 Guo Y K, Fan SS, Guo K H, Chen Y Storage Capac-ity of Methane in Hydrate Using Calcium Hypochlo-Anionic surfactant SDs, nonionic surfactantrite as Additive. In: Mori. 4th International ConferAPG, and liquid hydrocarbon CP were used as theence on Gas Hydrates. Yokohama, 2002. 1040nethane hydrate formation promoters. The tests9 Kashchiev D, Firoozabadi A. Journal of Crystalshowed that the critical Inicelle concentrations ofGrowth,2003,250(34):499SDS and APG water solutions were 300x10-6 and [10] Sun Z G, Fan SS, Guo K H et al. Journal of Chemical500x10-0 in hydrate formation system, respectively. [11] Yevi G Y, Rogers R E. Journal of Energy ResourcesIt also revealed by the experiments that surfactantsTechnology, 1996, 118(9)SDS and APG reduced the hydrate induction time, [12] MacKerell A D Jr. Journal of Physical Chemistry,improved the hydrate formation rate and gas stor995,99(7):1846age capacity. The effect of APG on hydrate forma- [13] Wanless E J, Ducker W A. Journal of Physical Chem-tion is less pronounced compared to SDS. CP couldastro,1996,100(8):3207中国煤化工CNMHG