rXXXX American Chemical Society A dx.doi.org/10.1021/es201261y | Environ. Sci. Technol. XXXX, XXX, 000–000

ARTICLE

pubs.acs.org/est

Effect of Organic Matters on CO2 Hydrate Formation in Ulleung BasinSediment SuspensionsRheo B. Lamorena, Daeseung Kyung, and Woojin Lee*

Department of Civil and Environmental Engineering, Korea Advanced Institute of Science and Technology, 291 Daehak-ro,Yuseong-gu, Daejeon 305-701, Korea

bS Supporting Information

’ INTRODUCTION

Considerable efforts have been invested in the development ofCO2 sequestration technologies to offer key solutions to climatechange due to anthropogenic activities. A potential promisingtechnological option to mitigate the rising concentration ofCO2 is to sequester CO2 in a specific geologic formation withinthe gas hydrate stability zone (GHSZ) where gas hydratehas been naturally formed and safely contained for a long time.The gas hydrate (e.g., CH4 hydrate), a potential alternativeenergy source in the near future, is a clathrate consisting of watermolecules trapping and containing a suitable guest molecule suchas methane under high pressure (0.1�80 MPa) and low tem-perature (268�300 K).1�4 GHSZ is generally located beneathsea bottom or in permafrost area where its environmental con-ditions are proper for the hydrate formation. It has been wellestablished that hydrate formations in GHSZ were influenced bytemperature, pressure, salinity, guest molecule type, and solidsurface area.1,5�8 Infill history of gas hydrate deposits in UlleungBasin (UB) at East Sea of Korea shows that the GHSZ in UB hasbeen accumulated by organic matter-rich sediments and organiccarbon content in its pore water has increased.9,10 Becausecatalytic roles of organic matters in geochemical reactions areknown,11,12 the new findings have attracted researchers’ atten-tion and triggered their curiosity about whether the sedimentorganic matters (SOMs) abundant in the hydrate deposits affectthe formation of gas hydrate and its stability under subseasediment environments.13,14

Microorganisms such as methanogenic and sulfur-reducingbacteria have been found at gas venting ecosystems and gashydrate deposits along continental margins.15 A plethora ofbiosurfactants and SOMs have been measured in the marinesediment environments as a result of microbial activities. Ithas been suggested that microorganisms in seabed sedimentsproduce biosurfactants (e.g., surfactin and rhamnolipids)through anaerobic microbial metabolism enhancing CH4 hydrateformation in marine porous media.16,17 SOMs are complexesand/or heterogeneous mixtures of diverse organic moleculesfrom vascular plant materials and animal remains in marinesediment environments having diverse characteristics due todifferent functional groups originated from various environmentsand their chemical configurations.18,19 Chemical structures ofSOMs found at gas hydrate deposits have been identified tobe kerogen, n-alkanes (C15—C37), fatty acids (C12—C34),n-alkanols, and isoprenoids.20 Sediments from the Gulf ofMexico are composed of intact proteinaceous materials, lipid-like compounds, and iron-bearing soil minerals, while a kerogen-type has been found in sediment samples from the NankaiTrough.20,21 The presence of SOMs in the hydrate depositshas been considered as a simple observation rather than a

Received: April 13, 2011Accepted: June 8, 2011Revised: June 7, 2011

ABSTRACT: Marine sediment core samples collected from agas hydrate deposit site (Ulleung Basin (UB), East Sea, Korea)were explored to identify the role of sediment organic matters(SOMs) on the formation of CO2 hydrate. Two distinct CO2

hydrate formation regimes (favorable (e40 min) and unfavor-able (>250 min)) were observed from the hydrate formationtests. CO2 hydrate induction time in UB sediment suspensionswas approximately seven times faster than that in UB sedimentsuspensions without SOMs (baked UB), showing a directinfluence of SOMs. Spectrometric and spectroscopic analysesconfirmed the existence of different types of SOMs includingnonhumic and humic substances in UB sediment samples. Wefound SOMs with aromatic ring structures in all sediment extracts and SOMs with amine and amide groups and lignin in alkalineextracts. SOMs were extracted fromUB sediment core samples (1 g each). Measured CO2 hydrate induction times were different inbaked UB sediment suspensions with different extracts of UB sediments. The experimental results demonstrated that SOMs canplay a significant role to accelerate the formation of CO2 hydrate in UB sediment suspensions, suggesting that the gas hydratedeposit site at UB may be a proper place for CO2 sequestration as a form of CO2 hydrate.

B dx.doi.org/10.1021/es201261y |Environ. Sci. Technol. XXXX, XXX, 000–000

Environmental Science & Technology ARTICLE

significant evidence to verify the enhanced gas hydrate formationand its safe containment observed at the deposits. Many differenttypes of SOMs having different chemical structures and physicalconformationsmay exist in the hydrate deposit sites, which seems topotentially influence water cage structures and nucleation sites forthe formation of gas hydrates. No significant research has beenconducted to verify the effect of SOMs on gas hydrate formationand its stability in marine sediment environments to date.

CO2 injected into the GHSZ can be contained as gas, liquid,and solid phases of CO2 depending on the temperature andpressure of subsea sediment environment. Unlike methanehydrate forming at the hydrate deposits naturally, CO2 hydratehaving a similar crystal structure of methane hydrate can beconsidered as a manmade solid waste. To successfully and legallysequester CO2 by following international environmental con-ventions and protocols,22,23 gaseous and liquid CO2 should notseep out from the GHSZ during CO2 sequestration and trans-form to CO2 hydrate in short time. Researchers have providedbasic knowledge on CO2 hydrate formation kinetics and sig-nificant geochemical factors (e.g., pH, mineral type, concentra-tions of chemical components) affecting CO2 hydrate formationto meet the goals of successful CO2 sequestration.

24�26 How-ever, no significant study has been achieved to identify the effectof SOMs on CO2 hydrate formation to date. The essential inves-tigation may provide verification on an accelerated gas hydrateformation and safe containment by SOMs in the GHSZ.

In this study, we characterized an enhanced CO2 hydrateformation in marine sediment suspensions obtained from a gashydrate deposit site in UB and investigate the effect of SOMs onCO2 hydrate formation in the suspensions. We additionallycharacterized the SOMs to identify which biochemical signaturesaffect the enhanced CO2 hydrate formation in UB sedimentsuspensions.

’EXPERIMENTAL SECTION

Materials. Details on the sources of chemicals and UB sedi-ment samples and their characterization methods were describedin the Supporting Information (SI).Experimental Procedure. Experimental setup and proce-

dures for CO2 hydrate formation tests have been previouslyreported in detail 25,27 and briefly described in SI. Batch experi-ments were conducted to check if SOM was a significant con-tributor to a favorable hydrate formation regime. Two distinctregimes (favorable (e40 min) and unfavorable (>250 min))were observed throughout the experiments. UB sediments werebaked in a muffle furnace at 550 �C for 2 h to remove SOMs fromtheir surfaces (hereafter designated as baked UB sediments). Theexperiments were conducted using UB sediment suspensions byfollowing the experimental procedures. CO2 hydrate inductiontimes in baked UB sediment suspensions were also tested ascontrols and compared to those in UB sediment suspensions.Geochemical characteristics of SOMs in UB sediments were

identified by spectroscopic and spectrometric analyses. SOMswere extracted from UB sediments by two different extractionmethods, that is, solvent extractions (P-extract and NP-extract)and alkaline extraction (H-extract). The solvent extracts weredried by a rotary evaporator and set aside for the analyseshowever alkaline-extracts were used without any further treat-ment. Details on SOM extraction methods and sample prepara-tion procedures for the analyses were described in SI.

To investigate which organic fraction in the marine sedimentplayed a significant role in CO2 hydrate formation, extractsprepared above were added to baked UB sediment suspensionsfor hydrate formation tests. All extracts described above wereused for the tests. We followed the same experimental procedureabove for the formation of CO2 hydrate with the extracts. No pHadjustment was conducted to keep the suspension pH constant.Samples and controls were prepared in duplicate as describedpreviously.25

Analytical Procedure. Organic matter content in UB sedi-ments was measured by mass loss on ignition (MLOI) method,dissolved organic carbon (DOC) analyzer, and carbon, hydro-gen, nitrogen, and sulfur (CHNS) analyzer. Solvent and alkalineextracts were characterized by spectroscopic and spectrometrictechniques (see SI). P- and NP-extracts were characterized toidentify key chemical properties of SOMs by Fourier transforminfrared (FT/IR), proton nuclear magnetic resonance (1H-NMR), and ultra performance liquid chromatography-electro-spray ionization mass spectrometry (UPLC-ESI-MS). H-extractwas characterized by FT/IR and UPLC-ESI-MS. All details fromthe sample preparation to analytical conditions and procedureswere described in SI.

’RESULTS AND DISCUSSION

Characteristics of UB andUB Sediments.UB is a deep bowl-shaped back-arc basin in the southwestern part of the East Sea,Korea. The general sedimentation pattern suggested that south-ern UB was filled mostly by mass-transport deposits, whilenorthern basin was dominantly comprised of hemipelagic sedi-ments and distal turbidites. No major rivers drain into the basinalong the East Coast of Korean Peninsula. Significant gashydrate-bearing reservoirs have been found at the depth ofsea between 1800 and 2100 and 150 m below the seafloor.28�31

UB sediment core samples were collected from one samplingsite (WGS84; 130�55.55070E and 36�35.90050N) in the basin.

Table 1. Mineralogical and Organic Contents of UBSediments

sediment composition

mineral components content (%)

quartz 17.3%

muscovite 15.1%

albite 13.2%

opal 7.7%

illite 7.1%

hornblende 6.9%

orthoclase 4.7%

chlorite 4.5%

sodium chloride 3.5%

pyrite 3.4%

microcline 3.0%

kaolinite 17.3%

organic components content (% or ppm)

organic content 10%

organic carbon 3.4%

dissolved organic carbon 13 ppm

C dx.doi.org/10.1021/es201261y |Environ. Sci. Technol. XXXX, XXX, 000–000

Environmental Science & Technology ARTICLE

The average water depth of the site was 2055 m and sedimentcore porosity was 73�85%. The temperature of bottom waterwas between 1 and 2 �C and its salinity was ∼34 ppt.32

Identification of inorganic and organic compositions in the UBsediment was conducted by spectroscopic and spectrometricanalyses. The mineralogical content of the sediment analyzed byXRD is tabulated in Table 1. The sediment included relativelyhigh organic content (10%) in which dissolved organic carboncontent was 13 ppm and organic carbon content was 3.4%. Thesesedimentary components are expected to significantly affect thehydrate formation kinetics and equilibrium along with marineenvironmental conditions such as seawater temperature andpressure.CO2 Hydrate Formation in UB and Baked UB Sediment

Suspensions.CO2 hydrate formed in UB sediment suspensions

in 40 min, while it did not form in baked UB sediment sus-pensions during experimental runs. This indicates that UBsediment suspensions had much favorable hydrate formationconditions compared to baked UB sediment suspensions. It hasbeen reported that surfaces of clay mineral mixtures acceleratedthe formation of CO2 hydrates by providing nucleation sites andchanging the hydrate formation conditions more favorable.27,33,34

XRD analysis on the UB sediment showed that its major portionwas composed of phyllosilicate clay minerals. This indicatesthat surfaces of the clay mineral mixtures in UB sediment mayalso play a key role for the fast formation of CO2 hydrate. XRDdiffractograms of UB and baked UB sediments were compared inSI Figure S2. No significant difference was found out exceptdecreases in peak intensities of baked UB sediment, indicating nostructural deformation of UB sediment and no significant loss in

Figure 1. 1H-NMR spectra of P-extract (a) andNP-extract (b). Labeled peaks are extraction solvents and solvents for NMRmeasurements. FTIR-ATRspectra of NP-extract, P-extract, and H-extract (c).

D dx.doi.org/10.1021/es201261y |Environ. Sci. Technol. XXXX, XXX, 000–000

Environmental Science & Technology ARTICLE

its original composition during a combustion process. The resultimplies that the surfaces of clay mineral mixtures may not be amain factor resulting in the significantly different hydrate induc-tion times inUB and bakedUB sediment suspensions. The bakedUB sediment, however, did not include organic carbon, while UBsediment had relatively high organic carbon content (3.4%).Therefore, SOMs removed during the combustion of UB sedi-ment may be a potential promoter for the fast hydrate formationin the UB sediment suspension. SOMs usually have high waterholding capacity,35 which can affect a solvation-driven behavioraround water and CO2 hydrate forming molecules. The removalof SOMs on the surface of UB sediment could remarkably reduceits hydrate formation capacity by removing effective nucleationsites. It is likely that the removal has triggered the disruption ofhydrogen bonding network shared by water and CO2 moleculesand SOMs appropriate for a rapid clathrate nucleation. Thecharacteristics of SOMs in UB sediments and their effects onCO2 hydrate formation were identified in the next section.Characteristics of Extractable SOMs from UB Sediments.

Aliphatic and aromatic chemical components of UB SOMs werecharacterized by complementary analytical techniques with dif-ferent type of extractants. 1H-NMR spectra for solvent-extractedSOMs from UB sediments are shown in Figure 1. Both P-extract(Figure 1(a)) and NP-extract (Figure 1(b)) showed a number ofaliphatic and aromatic signals with notable intensities. Peaksbetween 0.4 and 2.0 ppm arose from hydrogens of aliphatic

moieties. Saturated aliphatic carbons containing oxygen (H—C—O)resulted from aliphatic alcohols, ethers, and esters were detectedin the chemical shift range of 2.1�3.6 ppm. NMR signalsbetween 6.7 and 7.5 ppm indicating the characteristics ofaromatic rings (Ar—H) were weak in both spectra. Additionalinformation for different functional groups of SOMs, provided byFT/IR analyses, is shown in Figure 1(c). Only P-extract showedO—H stretch of carboxylic group at 3362 cm�1 indicating theinclusion of oxidized constituents. In the range of 3100 and2800 cm�1, all extracts showed C—H bands reflecting thestretchingmotions ofmethyl (CH3) andmethylene (CH2) groupsof aliphatic chains. The peaks around∼1453 and∼1410 cm�1 areattributed to vibrations of CH3 and CH2 of aliphatic chains. Thespectra exhibited a shoulder near ∼1600 cm�1, preferentiallyascribed to CdC stretching or COO� group.Weak bands shownat ∼1734 and ∼1711 cm�1 are attributed to conjugated CdOstretch of carboxyl groups mainly originated from fatty esters,which indicates the presence of lipid compounds in UB SOMs.36

Occurrence of aromatic groups was also evident at severalregions in the range of 690�1350 cm�1 (e.g., C—H out-of-plane deformation bands between 690 and 1000 cm�1, 37,38).Results of FT/IR analyses help determine chemical compo-

nents of H-extract which can extract a humic fraction of UBSOMs. Figure 1(c) shows the functional groups correspondingto pattern of bands usually associated with humic substanceswith covalent linkages. A peak at 3341 cm�1 represents O—H

Table 2. List of Calculated Mass Values, Elemental Compositions, Confirmation of Compounds by Chemspider and Pubchem,And Number of Isomers for Peaks at Different Retention Times in a Chromatogram of H-Extract

R.T. (min) measured mass (m/z) calculated mass (m/z) formula [M] classification no. of isomers

2.1 367.1339 367.1347 C13H22N2O10 Ya 29

367.1322 C17H22N2O5S Y 100

367.1356 C14H26N2O5S2 Y 10

2.3�2.4 344.9813 344.9797 C5H4N4O14 NFb

344.9837 C10H4N2O12 NF

344.9805 C6H8N4O9S2 NF

344.9812 C14H4N2O7S NF

344.9846 C11H8N2O7S2 Y 1

344.9787 C18H4N2O2S2 NF

3.1�3.2 329.1328 329.1343 C14H20N2O7 Y 79

329.1303 C9H20N4O9 NF

329.1318 C18H20N2O2S Y 2199

329.1352 C15H24N2O2S2 Y 45

329.1312 C10H24N4O4S2 NF

3.4 409.1600 409.1605 C19H24N2O8 Y 67

409.158 C23H24N2O3S Y 100

409.1639 C16H28N2O8S Y 3

409.1614 C20H28N2O3S2 Y 100

409.1574 C15H28N4O5S2 Y 67

425.1341 425.1343 C22H20N2O7 Y 100

425.1303 C17H20N4O9 Y 7

425.1362 C10H24N4O14 NF

425.1312 C18H24N4O4S2 Y 236

425.1377 C19H24N2O7S Y 105

425.1352 C23H24N2O2S2 Y 273

425.1318 C26H20N2O2S Y 125

425.1337 C14H24N4O9S Y 37aCompound has a match in ChemSpider or PubChem. bCompound has no match in ChemSpider or PubChem.

E dx.doi.org/10.1021/es201261y |Environ. Sci. Technol. XXXX, XXX, 000–000

Environmental Science & Technology ARTICLE

or N—H stretch; twin peaks at 2928 and 2857 cm�1 correspondto aliphatic components; peaks at 1650 cm�1 and 1546 cm�1

correspond to aromatic components of H-extract where a peak at1650 cm�1 is attributed to CdO stretch of amides, C—Nstretch, or aromatic CdC stretch and 1546 cm�1 repre-sents aromatic CdC stretch or its stretch to other C—Nand amino functionalities; peaks at 1444 and 1380 cm�1 showdeformation of phenolic and aliphatic groups; finally, a peak at1226 cm�1 corresponds to C—O stretch of ethers or carboxylgroups.37�39 Bands at 1025 and 3341 cm�1 could be attributedto polysaccharides or polysaccharide-like components. Peaks at1546 and 1650 cm�1 in H-extract spectrum additionally giveevidence that proteinaceous compounds with amide groups areincluded in UB SOMs.UPLC chromatograms for different extracts and positive

ESI/MS spectrum for P-extract are shown in SI Figure S3(a)and (b), respectively. Molecular structures of UB SOMs in theextracts can be identified by ESI/MS analysis. Approximately71% of chemical compounds from the extracts were suggested asmatching compounds by ChemSpider and PubChem databasesystems40,41 and the remaining 29% were considered as un-known. Potential candidates of chemical compounds with highmolecular weight (C5—C26) at each peak in the chromatogramof H-extract and number of isomers for each candidate arepresented in Table 2 (SI Table S3 and S4 for P- and NP-extract,respectively). By plotting van Krevelen diagram (H/C ratio vsO/C ratio) using generated molecular formulas of UB SOMs, wecan identify the functionality of proposed chemical compoundsin UB SOMs of various extracts and determine their functionalitydistributions.42 The diagram for H-extract shown in Figure 2(a)is categorized by six distinct regions; lipids (H:C = 1.5�2.0,O:C = 0�0.3), proteins (H:C = 1.5�2.2, O:C = 0.3�0.67),lignins (H:C = 0.7�1.5, O:C = 0.1�0.67), carbohydrates(H:C = 1.5�2.4, O:C = 0.67�1.2), unsaturated hydrocarbons(H:C = 0.7�1.5, O:C = 0�0.1), and condensed aromaticstructures (H:C = 0.2�0.7, O:C = 0�0.67).43 The potentialchemical compounds in H-extract covered all six regions in thevan Krevelen diagram and a major clustering was observed inlignin region. More separated peaks were observed in thechromatogram of P-extract indicating more various candidatecompounds of UB SOMs. SI Figure S4(a) shows that thecandidate compounds in P-extract were mostly clustered inlipid region of van Krevelen diagram and remainings were evenlylocated in unsaturated hydrocarbon and protein regions. Nocandidate compounds were observed in carbohydrates and con-densed aromatic structure regions. Compared to H- and P- extractcases, there were not many candidate compounds suggested forUB SOMs of NP-extract. Therefore only several points appearedin the diagram (SI Figure S5(a)).No candidate compounds ofNP-extract were observed in carbohydrate and condensed aromaticstructure regions, which is consistent to the result of P-extract.Other candidate compounds were sparsely distributed in otherfunctionality regions. The molecular formulas shown in Table 2provide the information of possible multiple bonds and rings intheir chemical structures. Double bond equivalent of a possiblecandidate chemical compound (DBE for CxHyNzOnSs: = x �y/2 + z/2 + 1) was calculated and subsequently normalizedby the number of carbon atoms (DBE/C) to investigate itsaromaticity.44,45 The distribution of DBE/C for H-extract withrespect to atomic H/C ratio in Figure 2(b) shows that half of thecandidate compounds inH-extract of UB SOMs are located aboveDBE/C = 0.5 and half of them are located above 0.7. Because a

chemical compound of whichDBE/C value is greater than 0.5 and0.7 shows a strong tendency to have an aromatic and condensedaromatic structure respectively,46 the candidate compounds inH-extract are likely to exhibit strong aromatic compound char-acteristics. As shown in SI Figure S4(b) and S5(b), most ofcandidate chemical compounds in P- and NP-extracts are locatedbelow DBE/C = 0.5. Therefore, compared to the candidate com-pounds in H-extract, those in P- and NP-extracts are less likely toshow the aromatic characteristics.The results suggest that a majority of candidate compounds

of UB SOMs are composed of lipids (38.5%), lignins (30.8%),proteins (17.3%), and unsaturated hydrocarbons (13.4%). Waterand CO2 molecules may form clathrate clusters being influencedby the aliphatic and aromatic moieties of UB SOMs underhydrate forming conditions. A similar behavior of biochemicalcompounds (natural and synthetic proteins) and their effect onfast ice nucleation have been investigated.47�49 The biochemicalice nucleator theory could be applied to CO2 hydrate formationin UB sediment suspensions and help explain the fast hydrateformation in the suspensions with UB SOMs. The biochemicalcompounds containing hydroxy, ether-oxygen, or amine groupsare likely to help form associations of water and CO2 moleculesand cluster the associations to ice-like patterns.48 The formationof associations with UB SOMs may conform to appropriate

Figure 2. A van Krevelen diagram of H-extract (a). Distinct regions ofbiomolecular compound classes: lipids, proteins, carbohydrates, unsa-turated hydrocarbons, lignins, and condensed aromatic structures. Agraphical representation of DBE/C demonstrating the aromaticity ofH-extract (b).

F dx.doi.org/10.1021/es201261y |Environ. Sci. Technol. XXXX, XXX, 000–000

Environmental Science & Technology ARTICLE

configurations in ice nucleation sites.50 They may be reorientedwith water molecules forming an ideal hydrogen bondingnetwork in a hydrate-like (hexagonal) structure which is largeenough to be potential sites for the rapid nucleation of CO2

hydrate.CO2 Hydrate Formation in Baked UB Sediment Suspen-

sions with UB SOM Extracts. CO2 hydrate induction times inbaked UB sediment suspensions with different type of extractsare shown in Figure 3. The induction times in baked UBsediment suspensions with solvent extracts were different, thatis, CO2 hydrate formed in the suspensions with NP-extract in143.0 ( 42.4 min, while it did not form in the suspensions withP-extract during the experiment. Because major chemical con-stituents of P-extract are composed of lipids and unsaturatedhydrocarbons, their chemical structures could not establishsubstantial affinities 51,52 to watermolecules, disrupt water clusterstructures by weakening hydrogen bonding interactions, andfinally result in no CO2 hydrate formation in the suspensionswith P-extract. CO2 hydrate induction times in the suspensionswith H-extract (150( 24.0 min) were similar to those with NP-extract showing a favorable hydrate formation. This suggests thatNP- and H-extracts contain better CO2 hydrate nucleatorsplaying a significant role for triggering fast CO2 hydrate nuclea-tion than lipids and unsaturated hydrocarbons in P-extract. FT/IR and ESI/MS data demonstrated that a major fraction of UBSOMs in H- and NP-extracts were lignins and organic com-pounds with amide and amine groups. This indicates that themajor compounds in H- and NP-extracts may be the hydratenucleators to significantly promote CO2 hydrate formation.H-extract also contained humic substances as shown in thespectroscopic analyses. Hydrate formation tests were conductedin bakedUB suspensions with commercial humic acid (100 ppm)to identify the effect of humic substances on CO2 hydrateinduction time. The hydrate induction time in the suspensionswith commercial humic acid (89.5 ( 26.2 min) was approxi-mately twice faster than those with H- and NP-extracts showingmuch favorable CO2 hydrate formation. Humic substances have

been well-known as supramolecules combined with low molec-ular weight organic compounds including carboxyl and phenolichydroxyl groups.19,53 The supramolecules can form associationswith water molecules and chemical species on soil mineralsurface affecting the arrangement of structured or ordered waterclusters, which leads to the enhanced CO2 hydrate formation inbaked UB sediment suspensions. Hydrate induction times in thesuspensions with a mixture of three extracts (50.2 ( 5.1 min)were very similar to those in UB sediment suspensions andapproximately three times faster than those in baked UB sedi-ment suspensions with H- and NP-extracts, respectively. Thisindicates that UB SOMs in the extracts significantly enhancedCO2 hydrate formation and affected its induction time. Some ofthem (lignins, humic substances, and compounds with amide andamine groups) can accelerate the hydrate induction time, whileothers (lipids and unsaturated hydrocarbons) can slow down.The results suggest that the presence of high content of SOMs inUB sediments is the potential reason why such huge methanehydrate deposits can be developed and safely contained at UBsite in the East Sea of Korea. The site could be also a potentialplace where huge amount of CO2 can be sequestered as a form ofCO2 hydrate in the near future.Environmental Implications for CO2 Sequestration. The

research has demonstrated the characteristics and role of SOMsin CO2 hydrate formation in UB sediment suspensions. Weobserved the enhanced CO2 hydrate formation in the presenceof inorganic (sediment minerals) and organic components(SOMs) of UB sediments. CO2 hydrate induction time in UBsediment suspensions were influenced by UB SOMs providingsignificant implications on the stability of UB as gas hydratereservoir and its suitability as CO2 hydrate storage, althoughnatural sea sediment system in UB with low water contentallows more interactions of water and CO2 molecules withaggregated structures of SOMs on the sediment surface leadingto much faster CO2 hydrate formation. One of main researchthrusts in the field of current energy and environmental studiesin developed and developing countries is to decrease theemission of greenhouse gases by storing their huge amount inspecific geological formations and deep sea sediment systems.Where to store and how to sequester has been one of the mostimportant issues. The discovery of methane hydrate at UB site,East Sea of Korea could provide an appropriate solution foreach side of energy and environmental issues. By understandingthe characteristics of hydrate formation at the site and evaluat-ing its value as an alternative energy source, we can get closer tothe feasible solution for the impending energy and environ-mental problems.54�56 Two potential sites in southern UB areconsidered as suitable CO2 sequestration sites, because theycan offer appropriate CO2 hydrate forming conditions con-tributing to long-term stability of the sites. Moreover, UB mayprovide suitable sites to achieve methane swapping for CO2 inmethane hydrate by injecting CO2 to its deposit for bothsequestration and energy mining purposes. Current researcheson the application of gas hydrate swapping mechanism havenot considered the heterogeneity of hydrate forming systemscomposed of sediment surfaces and SOMs.57,58 Althoughexperimental results obtained from this study do not cover afull understanding of successful methane hydrate swapping andCO2 sequestration, they provide basic knowledge to determinesuitable CO2 sequestration sites and assess their stability anddeliver insight for the viability of methane swapping mechanismas a CO2 sequestration option.

Figure 3. Hydrate induction times observed in UB and baked UBsediment suspensions with different extracts. NoCO2 hydrate formationwas observed in baked UB sediment suspension and the suspension withP-extract.

G dx.doi.org/10.1021/es201261y |Environ. Sci. Technol. XXXX, XXX, 000–000

Environmental Science & Technology ARTICLE

’ASSOCIATED CONTENT

bS Supporting Information. Detailed schematic diagram ofexperimental setup, extraction procedures, and instrumentationdescriptions. This material is available free of charge via theInternet at http://pubs.acs.org.

’AUTHOR INFORMATION

Corresponding Author*Phone: +82-42-350-3624; fax: +82-42-350-3610; e-mail: woojin_lee@kaist.ac.kr.

’ACKNOWLEDGMENT

We thank the three anonymous reviewers for their criticalreviews to enhance the quality of this manuscript. This work wasfully supported by a research grant from the Expertise ResearchProgram of National Research Foundation (NRF-2009-0086664).

’REFERENCES

(1) Sloan, E. D., Clathrate Hydrates of Natural Gases, 2nd ed.; MarcelDekker: New York, 1998.(2) House, K. Z.; Schrag, D. P.; Harvey, C. F.; Lackner, K. S.

Permanent carbon dioxide storage in deep-sea sediments. Proc. Natl.Acad. Sci. 2006, 103, 12291–12295.(3) Novel approaches to carbon management: separation, capture,

sequestration, and converstion to useful products. InWorkshop Report toCommittee on Novel Approaches to the Management of Greenhouse Gasesfrom Energy Systems; National Research Council: Washington, DC,2003.(4) Milkov, A. V.; Sassen, R. Thickness of the gas hydrate stability

zone, Gulf of Mexico continental slope. Mar. Pet. Geol. 2000, 17,981–991.(5) Buffet, B. A. Clathrate hydrates.Annu. Rev. Earth Planet Sci. 2000,

28, 477–507.(6) Clarke, M. A.; Majumbar, A.; Bishnoi, P. R. Experimental

investigation of carbon dioxide hydrate formation conditions in thepresence of KNO3, MgSO4, and CuSO4. J. Chem. Eng. Data 2004, 49,1436–1439.(7) Park, S.-H.; Garrison, S. Do montmorillonite surfaces promote

methane hydrate formation? Monte carlo and molecular dynamicssimulations. J. Phys. Chem. B 2003, 107, 2281–2290.(8) Seo, Y.; Lee, H.; Uchida, T. Methane and carbon dioxide hydrate

phase behavior in small porous silica gels: Three-phase equilibriumdetermination and thermodynamic modeling. Langmuir 2002, 18,9164–9170.(9) Lee, G. H.; Kim, B. Infill history of the Ulleung Basin, East Sea

(Sea of Japan) and implications on source rocks and hydrocarbons.Mar.Pet. Geol. 2002, 19, 829–845.(10) Lim, D.; Choi, J.; Xu, Z.; Kim, M.; Choi, D.; Jung, H.; Lee, P.

Methane-derived authigenic carbonates from the Ulleung basin sedi-ments, East Sea of Korea. Cont. Shelf Res. 2009, 29, 1588–1596.(11) Lenczewski, M.; Jardine, P.; McKay, L.; Layton, A. Natural

attenuation of trichloroethylene in fractured shale bedrock. J. Contam.Hydrol. 2003, 64, 151–168.(12) Fukushima, M.; Shigematsu, S.; and Nagao, S. Degradation of

pentachlorophenol in a contaminated soil suspension using hybridcatalysts prepared via urea-formaldehyde polycondensation betweeniron (III)-tetrakis (p-hydroxyphenyl) porphyrin and humic acid.(13) Fukushima, M.; Tanabe, Y.; Morimoto, K.; Tatsumi, K. Role

of humic acid fraction with higher aromaticity in enhancing the activityof a biomimetic catalyst, tetra(p-sulfonatophenyl)porphineiron (III).Biomacromolecules 2007, 8, 386–391.

(14) Nurmi, J. T.; Tratnyek, P. G. Electrochemical properties ofnatural organic matter (NOM), fractions of NOM, and model biogeo-chemical electron shuttles. Environ. Sci. Technol. 2002, 36, 617–624.

(15) Joye, S. B.; Boetius, A.; Orcutt, B. N.; Montoya, J. P.; Schulz,H. N.; Erickson, M. J.; Lugo, S. K. The anaerobic oxidation of methaneand sulfate reduction in sediments from Gulf of Mexico cold seeps.Chem. Geol. 2004, 205, 219–238.

(16) Rogers, R. E.; Kothapalli, C.; Lee, M. S.;Woolsey, J. R. Catalysisof gas hydrates by biosurfactants in seawater-saturated sand/clay. Can. J.Chem. Eng. 2003, 81, 973–980.

(17) Rogers, R. E.; Zhang, G.; Dearman, J.; Woods, C. Investigationsinto surfactant/gas hydrate relationship. J. Pet. Sci. Eng. 2007, 56, 82–88.

(18) Stevenson, F. J. Humus Chemistry: Genesis, Compositions, Reac-tions, 2nd ed.; John Wiley& Sons Ltd: New York, 1994.

(19) Sutton, R.; Sposito, G. Molecular structure in soil humicsubstances:The new view. Environ. Sci. Technol. 2005, 39, 9009–9015.

(20) Saito, H.; Suzuki, N. Terrestrial organic matter controlling gashydrate formation in the Nankai Trough accretionary prism, offshoreShikoku, Japan. J. Geochem. Exp. 2007, 95, 88–100.

(21) Zhang, C. L.; Pancost, R. D.; Sassen, R.; Qian, Y.; Macko, S. A.Archael lipid biomarkers and isotopic evidence of anaerobic methaneoxidation associated with gas hydrates in the Gulf of Mexico. Org.Geochem. 2003, 34, 827–836.

(22) Wilson, E. J.; Friedmann, S. J.; Pollak, M. F. Research fordeployment: Incorporating risk, regulation, and liability for carboncapture and sequestration. Environ. Sci. Technol. 2007, 41, 5945–5952.

(23) Reiner, D. M.; Herzog, H. J. Developing a set of regulatoryanalogs for carbon sequestration. Energy 2004, 29, 1561–1570.

(24) Kang, S.-P.; Lee, J.-W. Kinetic behaviors of CO2 hydrates inporous media and effect of kinetic promoter on the formation kinetics.Chem. Eng. Sci. 2010, 65, 1840–1845.

(25) Lamorena, R. B.; Lee, W. Effect of pH on carbon dioxidehydrate formation in mixed soil mineral suspensions. Environ. Sci.Technol. 2009, 43, 5908–5914.

(26) Zhang, J.; Lee, J.W. Enhanced kinetics of CO2 hydrate formationunder static conditions. Ind. Eng. Chem. Res. 2009, 48, 5934–5942.

(27) Lamorena, R. B.; Lee, W. Formation of carbon dioxide hydratein soil and soil mineral suspensions with electrolyte. Environ. Sci. Technol.2008, 42, 2753–2759.

(28) Lee, G. H.; Kim, H. J.; Han, S. J.; Kim, D. C. Seismicstratigraphy of the deep Ulleung Basin in the East Sea (Japan Sea)back-arc basin. Mar. Pet. Geol. 2001, 18, 615–634.

(29) Ryu, B. J.; Riedel, M.; Kim, J. H.; Hyndman, R. D.; Lee, Y. J.;Chung, B. H.; Kim, I. S. Gas hydrates in the western deep-water UlleungBasin, East Sea of Korea. Mar. Pet. Geo. 2009, 26, 1483–1498.

(30) Bahk, J.-J.; Kim, J.-H.; Kong, G.-S.; Lee, H.; Park, Y.; Park, K. P.Occurrence of near-seafloor gas hydrates and associated cold vents in theUlleung Basin, East Sea. Geosciences J. 2009, 13, 371–385.

(31) Park, K. P.; Bahk, J.-J.; Kwon, Y.; Kim, G. Y.; Riedel, M.;Holland, M.; Schultheiss, P.; and Rose, K. U.S. Department of Energy,Office of Fossil Energy, National Energy Technology Laboratory.UBGH-1 Scientific Party, 2008. Korea national program expeditionconfirms rich gas hydrate deposits in the Ulleung Basin, East Sea.Methane Hydr. Newsl. 2008, 6�9.

(32) Kim, Y. G.; Kim, K. Intermediate waters in the East/Japan Sea.J. Oceanogr. 1999, 55, 123–132.

(33) Park, S. H.; Sposito, G. Do montmorillonite surfaces promotemethane hydrate formation? Monte carlo and molecular dynamicssimulations. J. Phys.Chem. B 2003, 107, 2281–2290.

(34) Cha, M.; Ouar, H.; Wildeman, T. R.; Sloan, E. D. A third-surface effect on hydrate formation. J. Phys. Chem. 1988, 92, 6492–6494.

(35) Sparks, D. L. Environmental Soil Chemistry, 2nd ed.; AcademicPress, Elsevier Science: California, 2003.

(36) Nielsen, S. S. Food Analysis Laboratory Manual. KluwerAcademic/Plenum Publishers:New York, 2003.

(37) Pavia, D. L.; Lampman, G. M.; Kriz, G. S.; and Vyvyan, J. R.Introduction to Spectroscopy, 4th ed.; Brooks/Cole Cengage Learning:California, 2009.

H dx.doi.org/10.1021/es201261y |Environ. Sci. Technol. XXXX, XXX, 000–000

Environmental Science & Technology ARTICLE

(38) Coates, J. Interpretation of infrared spectra, a practical ap-proach. In Encyclopedia of Analytical Chemistry; Meyers, R.A., Ed.; JohnWiley&Sons Ltd.: Chichester, 2000(39) Chen, J.; Gu, B.; LeBoeuf, E. J.; Pan, H.; Dai, S. Spectroscopic

characterization of the structural and functional properties of naturalorganic matter fractions. Chemosphere 2002, 48, 59–68.(40) ChemSpider. www.chemspider.com (accessed January 2011).(41) The PubChem Project. pubchem.ncbi.nlm.nih.gov/ (accessed

January 2011).(42) Kim, S.; Kramer, R. W.; Hatcher, P. G. Graphical method

for analysis of ultrahigh-resolution broadband mass spectra ofnatural organic matter, the van krevelan diagram. Anal. Chem. 2003, 75,5336–5344.(43) Hockaday, W. C.; Purcell, J. M.; Marshal, A.; Baldock, J. A.;

Hatcher, P. G. Electrospray and photoionization mass spectrometry forthe characterization of organic matter in natural waters: A qualitativeassessment. Limnol. Oceanogr. Meth. 2009, 7, 81–95.(44) Reemtsma, T. Determination of molecular formulas of natural

organic matter molecules by (ultra-) high-resolution mass spectroscopy:Status and needs. J. Chromatogr., A 2009, 1216, 3687–3701.(45) Sleighter, R. L.; Hatcher, P. G. The application of electrospray

ionization coupled to ultrahigh resolution mass spectrometry for themolecular characterization of natural organic matter. J. Mass Spectrom.2007, 42, 559–574.(46) Koch, B. P.; Dittmar, T. Frommass to structure: an aromaticity

index for high-resolution mass data of natural organic matter. RapidCommun. Mass Spectrom. 2006, 20, 926–932.(47) Franks, F. Nucleation of ice and its management in ecosystems.

Philos. Trans. R. Soc., A 2003, 361, 557–574.(48) Head, R. B. Ice nucleation by some cyclic compounds. J. Phys.

Chem. Solids 1962, 23, 1371–1378.(49) Matsumoto, M.; Saito, S.; Ohmine, I. Molecular dynamics

simulation of the ice nucleation and growth process leading to waterfreezing. Nature 2002, 416, 409–413.(50) Brush, R. A.; Griffith, M.; Mlynarz, A. Characterization and

quantification of intrinsic ice nucleators in winter rye (secale cereale)leaves. Plant Physiol. 1994, 104, 725–735.(51) Murphy, D. J. Plant Lipids: Biology, Utilization, And Manipula-

tion; Blackwell Publishing Ltd.: Cornwall, 2005; pp 103�105.(52) Graham Solomons, T. W. Fundamentals of Organic Chemistry,

5th ed.; Wiley: New York, 1997.(53) Wershaw, R. L. Evaluation of Conceptual Models of Natural

Organic Matter (Humus) from a Consideration of the Chemical andBiochemical Processes of Humifications; U.S. Geological Survey: Reston,VA, 2004.(54) Huh, C.; Kang, S. G.; Hong, S.; Choi, J. S.; Baek, J. H.; Lee, C. S.;

Park, Y. C.; Lee, J. S. CO2 storage in marine geological structure: Areview of latest progress and its application in Korea. Energy Proc. 2009,1, 3993–4000.(55) Shin, K.; Park, Y.; Cha, M.; Park, K. P.; Huh, D. G.; Lee, J.; Kim,

S. J.; Lee, H. Swapping phenomena occurring in deep-sea gas hydrates.Energy Fuel 2008, 22, 3160–3163.(56) Englezos, P.; Lee, J. D. Gas hydrates: A cleaner source of energy

and opportunity for innovative technologies. Korean J. Chem. Eng. 2005,22, 671–681.(57) Ersland, G.; Husebo, J.; Graue, A.; Baldwin, B. A.; Howard, J.;

Stevens, J. Measuring gas hydrate formation and exchange with CO2 inBentheim sandstone using MRI tomography. Chem. Eng. Sci. 2010,158, 25–31.(58) Zhou, X.; Fan, S.; Liang, D.; Du, J. Determination of appro-

priate condition on replacing methane from hydrate with carbondioxide. Energy Convers. Manage. 2008, 49, 2124–2129.