Evaluation of Gas Production from Marine Hydrate

energies Article

Evaluation of Gas Production from Marine Hydrate Deposits at the GMGS2-Site 8, Pearl River Mouth Basin, South China Sea Yi Wang 1,2,3 , Jing-Chun Feng 1,2,3,4 , Xiao-Sen Li 1,2,3, *, Yu Zhang 1,2,3 and Gang Li 1,2,3 1

2 3 4

*

Key Laboratory of Gas Hydrate, Guangzhou Institute of Energy Conversion, Chinese Academy of Sciences, Guangzhou 510640, China; [email protected] (Y.W.); [email protected] (J.-C.F.); [email protected] (Y.Z.); [email protected] (G.L.) Guangzhou Center for Gas Hydrate Research, Chinese Academy of Sciences, Guangzhou 510640, China Guangdong Key Laboratory of New and Renewable Energy Research and Development, Guangzhou 510640, China University of Chinese Academy of Sciences, Beijing 100083, China Correspondence: [email protected]; Tel.: +86-020-8705-7037

Academic Editor: Richard Coffin Received: 22 November 2015; Accepted: 23 February 2016; Published: 21 March 2016

Abstract: Natural gas hydrate accumulations were confirmed in the Dongsha Area of the South China Sea by the Guangzhou Marine Geological Survey 2 (GMGS2) scientific drilling expedition in 2013. The drilling sites of GMGS2-01, -04, -05, -07, -08, -09, -11, -12, and -16 verified the existence of a hydrate-bearing layer. In this work gas production behavior was evaluated at GMGS2-8 by numerical simulation. The hydrate reservoir in the GMGS2-8 was characterized by dual hydrate layers and a massive hydrate layer. A single vertical well was considered as the well configuration, and depressurization was employed as the dissociation method. Analyses of gas production sensitivity to the production pressure, the thermal conductivity, and the intrinsic permeability were investigated as well. Simulation results indicated that the total gas production from the reference case is approximately 7.3 ˆ 107 ST m3 in 30 years. The average gas production rate in 30 years is 6.7 ˆ 103 ST m3 /day, which is much higher than the previous study in the Shenhu Area of the South China Sea performed by the GMGS-1. Moreover, the maximum gas production rate (9.5 ˆ 103 ST m3 /day) has the same order of magnitude of the first offshore methane hydrate production test in the Nankai Trough. When production pressure decreases from 4.5 to 3.4 MPa, the volume of gas production increases by 20.5%, and when production pressure decreases from 3.4 to 2.3 MPa, the volume of gas production increases by 13.6%. Production behaviors are not sensitive to the thermal conductivity. In the initial 10 years, the higher permeability leads to a larger rate of gas production, however, the final volume of gas production in the case with the lowest permeability is the highest. Keywords: natural gas hydrate; Guangzhou marine geological survey 2 (GMGS2); Pearl River Mouth Basin; massive hydrate layer; dual hydrate layer; depressurization

1. Introduction Natural gas hydrates are an ice-crystalline, naturally occurring compounds composed of water and gas molecules (methane CH4 is the most common ingredient). Natural gas hydrates form under certain thermodynamically favorable conditions of high pressure and relatively low temperature. Under these conditions, gas molecules are trapped in a cage constituted by hydrogen-bonded water molecules. When dissociated under standard conditions, one volume of solid natural gas hydrate will release approximately 160 volumes of natural gas [1], indicating that methane hydrates have a

Energies 2016, 9, 222; doi:10.3390/en9030222

www.mdpi.com/journal/energies

Energies 2016, 9, 222

Energies 2016, 9, 222

2 of 22

2 of 22

high energy density. In addition, technology can bein applied in the of thestorage hydrogen energy density. In addition, hydratehydrate technology can be applied the process ofprocess the hydrogen [2], storage [2], sea water desalination [3,4], gas separation [5], and capture of carbon dioxide [6]. sea water desalination [3,4], gas separation [5], and capture of carbon dioxide [6]. The came from The first first confirmation confirmation of of naturally-occurring naturally-occurring gas gas hydrates hydrates came from the the Siberia Siberia and and the the Black Black Sea. By the mid-1990s, it was estimated that the volume of the energy stored in gas hydrates Sea. By the mid-1990s, it was estimated that the volume of the energy stored in gas hydrates could could exceed that of of all all the the world’s world’s coal, coal,oil, oil,and andconventional conventionalnatural naturalgas gascombined combined[7,8]. [7,8].Nowadays, Nowadays, it exceed that it is is known that gashydrate hydrateisisa apotential potentialenergy energyresource, resource,which which occurs occurs naturally naturally in in the known that gas the permafrost permafrost regions and in the ocean sediments of continental margins [7,8]. Therefore, developing regions and in the ocean sediments of continental margins [7,8]. Therefore, developing methods methods for for methane production from hydrate reservoirs and investigating their production behavior are attracting methane production from hydrate reservoirs and investigating their production behavior are considerable attention from governments and scientists over theall world. attracting considerable attention from governments andall scientists over the world. Recently, an increasing number of deep ocean drilling expeditions Recently, an increasing number of deep ocean drilling expeditions have have been been carried carried out out to to determine deposit sand gather thethe geological factors controlling the determine the thelocation locationofofthe themarine marinehydrate hydrate deposit sand gather geological factors controlling occurrence of gas hydrates. The premier drilling projects the occurrence of gas hydrates. The premier drilling projectsare arethe theOcean OceanDrilling DrillingProgram Program (ODP) (ODP) [9] [9] and the Integrated Ocean Drilling Program (IODP) [10]. The Department of Energy (DOE) of the USA and the Integrated Ocean Drilling Program (IODP) [10]. The Department of Energy (DOE) of the USA and and some some oil oil companies companies sponsored sponsored the the Joint Joint Industry Industry Project Project to to address address the the challenges challenges of of gas gas hydrate hydrate exploration and exploitation in the Northern Gulf of Mexico [11]. In Asia, information exploration and exploitation in the Northern Gulf of Mexico [11]. In Asia, information about about the the occurrence of gas hydrate deposits in the sea area of the Indian Peninsula and the Andaman convergent occurrence of gas hydrate deposits in the sea area of the Indian Peninsula and the Andaman margin was documented the project of National Hydrate Program 01 [12]. convergent margin was by documented byIndian the project of Gas Indian National Gas Expedition Hydrate Program Hydrate samples and the corresponding geologic data have been acquired in the Ulleung Basin located Expedition 01 [12]. Hydrate samples and the corresponding geologic data have been acquired in the in the east sea of Koreain bythe theeast Korea Hydrate Expedition in Hydrate 2007 andExpedition 2010, respectively In Ulleung Basin located sea Gas of Korea by the Korea Gas in 2007[13,14]. and 2010, China, hydrate resources appear to be abundant. In 2007, the Guangzhou Marine Geological Survey respectively [13,14]. In China, hydrate resources appear to be abundant. In 2007, the Guangzhou (GMGS) successfully completed thesuccessfully GMGS1 hydrate exploration program in theexploration Shenhu Area of the Marine Geological Survey (GMGS) completed the GMGS1 hydrate program South China Sea [15,16]. Subsequently, the second Chinese marine hydrate expedition GMGS2 was in the Shenhu Area of the South China Sea [15,16]. Subsequently, the second Chinese marine hydrate performed in the Dongsha Area of the South China Sea in the summer of 2013. expedition GMGS2 was performed in the Dongsha Area of the South China Sea in the summer of 2013. Figure Figure 11 shows shows the the location location map map of of the the GMGS2 GMGS2 scientific scientific drilling drilling expedition expedition in in the the Dongsha Dongsha Area. The Dongsha Area explored by GMGS2 is situated at the east of the Pearl River Mouth Area. The Dongsha Area explored by GMGS2 is situated at the east of the Pearl River Mouth Basin Basin in the South South China ChinaSea Sea[17]. [17].An An initial Logging While Drilling phase employed by GMGS2, in the initial Logging While Drilling phase waswas employed by GMGS2, and and then a coring and sampling phase were used. During the GMGS2expedition, a total of 13 sites then a coring and sampling phase were used. During the GMGS2expedition, a total of 13 sites (shown (shown in2), Figure 2),includes which includes 10 logging-while-drilling (LWD) and three downhole line in Figure which 10 logging-while-drilling (LWD) and three downhole wire linewire logging logging (DWL) pilotwere holes,investigated, were investigated, the existence of hydrate-bearing layers (DWL) pilot holes, amongamong which which the existence of hydrate-bearing layers was was confirmed at nine sites (sites GMGS2-01, -04, -05, -07, -08, -09, -11, -12, and -16). The water depth confirmed at nine sites (sites GMGS2-01, -04, -05, -07, -08, -09, -11, -12, and -16). The water depth of of drilling sites ranged from667 667toto1747 1747m. m.As Asa aresult, result,the theDongsha Dongshaarea areawas was concluded concluded to to be be one one thethe drilling sites ranged from of of the the most most complex complex and and the the richest richest marine marine gas gas hydrate hydrate accumulations accumulations that that exist. exist.

Figure 1.1.Map of of northeastern partpart of South ChinaChina Sea. The rectangle shows theshows locationthe of the drilling Figure Map northeastern of South Sea. The rectangle location ofarea. the drilling area.

Energies 2016, 9, 222

3 of 22

Energies 2016, 9, 222

3 of 22

Figure 2. Bathymetric map of gas hydrate drilling area and sites.

Figure 2. Bathymetric map of gas hydrate drilling area and sites. At Site GMGS2-08, double gas hydrate intervals with saturation of 45%–55% have been detected.

At GMGS2-08, double gas hydrate intervalsatwith of 45%–55% havewith beenvein, detected. AsSite shown in Figure 3, gas hydrates were identified this saturation site in fine grained sediments nodular, and laminated In identified the shallowat part (9.0–23.0 a roughlysediments 14 m-thick layer As shown in Figure 3, gas morphologies. hydrates were this site inmbsf), fine grained with vein, of gas hydrates, widelymorphologies. distributed in the form of nodules veins, occurs inmbsf), a rangeaofroughly 9.0 to 23.014 mbsf. nodular, and laminated In the shallowand part (9.0–23.0 m-thick In the middle part (58.0–63.0 mbsf), a 5 m-thick authigenic carbonates deposit was found just above layer of gas hydrates, widely distributed in the form of nodules and veins, occurs in a range of 9.0 to the top of the lower hydrate-bearing sediments. In the deep part (63.0–98.0 mbsf), the interval is 23.0 mbsf. In the middle part (58.0–63.0 mbsf), a 5 m-thick authigenic carbonates deposit was found just dominated by silty clay with some partially dispersed muddy carbonates. A 35-m-thick gas hydrate above the top of the lower hydrate-bearing sediments. In the deep part (63.0–98.0 mbsf), the interval is layer, tightly distributed with a massive form, was found in a range of 63.0–98.0 mbsf. Meanwhile, a dominated by silty clayofwith partially carbonates. A 35-m-thick gas hydrate 20 cm-thick section puresome gas hydrate wasdispersed recovered muddy in this interval [17]. Ideally, the best natural layer,gas tightly distributed with a massive form, was found in a range of 63.0–98.0 mbsf. Meanwhile, a hydrate (NGH) is found in sands, in particular turbidite sands bounded by impermeable shales. 20 cm-thick section of pure gas hydrate was recovered in this interval [17]. Ideally, the best natural gas This is because the industry already has the tools and the knowledge to produce natural gas from beds like(NGH) it [18]. is found in sands, in particular turbidite sands bounded by impermeable shales. This hydrate production teststhe in tools the field to assess the feasibility of exploiting is becauseObviously, the industry already has and are the essential knowledge to produce natural gas from beds like methane hydrate deposits. However, until now, just a few short-term field tests have been conducted it [18]. because of the huge cost [19]. Therefore, numerical studies and laboratory tests are underway to Obviously, production tests in the field are essential to assess the feasibility of exploiting methane advance our understanding of the technologies to exploit this potential energy resource [15,19–23]. hydrate deposits. However, until now, just a few short-term field tests have been conducted because After the GMGS1hydrate exploration program in the Shenhu Area of the South China Sea, a series of of thenumerical huge cost [19]. ofTherefore, numerical studies laboratory tests are underway to advance studies gas production from the hydrateand reservoir in the Shenhu Area were carried out our understanding of the technologies to exploit this potential energy resource [15,19–23]. After the [24–27]. The application of the vertical well [28], the single horizontal well [25], and the dual horizontal GMGS1hydrate exploration program the Shenhu of the South Sea, a series of numerical well [15] technologies in the hydrateinreservoir at theArea Shenhu Area wereChina investigated. Evaluations of theofgas production from hydrate reservoirreservoir by depressurization [29], thermal stimulation huff The studies gas production from the hydrate in the Shenhu Area were carried[30], out the [24–27]. and puff [26], and conjunction with warm [31] were application ofmethod the vertical welldepressurization [28], the singlein horizontal well [25], andbrine the stimulation dual horizontal well [15] also studied. However, up to now, there was little literature about the evaluation of gas production technologies in the hydrate reservoir at the Shenhu Area were investigated. Evaluations of the gas from the hydrate reservoir in the Dongsha Area. production from hydrate reservoir by depressurization [29], thermal stimulation [30], the huff and puff In this work, the main objective was to evaluate the production potential of the gas hydrate method [26], and depressurization in conjunction with warm brine stimulation [31] were also studied. accumulation at the Site GMGS2-8 of the Dongsha Area in the Pearl River Mouth Basin. This is the However, up tothe now, there was little literature thebeen evaluation of gasnumerical production from hydrate first time gas production potential in thisabout area has studied using tools. Allthe of the reservoir in the Dongsha Area. parameter selection is based on the logging data and exploration data of the reservoir. On the other In thisthe work, the main objective was the production potential of the gas hydrate hand, TOUGH-hydrate code is one of to theevaluate most popular tools for evaluating the gas production in hydrateat reservoirs, and this code verified by experiments our previous [32–34]. accumulation the Site GMGS2-8 of has thebeen Dongsha Area in the PearlinRiver Mouth work Basin. This is the Therefore, theproduction results of this paper should be area valuable the studied future hydrate this area. first time the gas potential in this hasfor been usingexploitation numerical in tools. All of the The single vertical well is considered as the well configuration. Depressurization is employed as the other parameter selection is based on the logging data and exploration data of the reservoir. On the dissociation method. Analyses of the sensitivity of gas production to various dissociation parameters hand, the TOUGH-hydrate code is one of the most popular tools for evaluating the gas production were carried out as well. in hydrate reservoirs, and this code has been verified by experiments in our previous work [32–34]. Therefore, the results of this paper should be valuable for the future hydrate exploitation in this area. The single vertical well is considered as the well configuration. Depressurization is employed as the dissociation method. Analyses of the sensitivity of gas production to various dissociation parameters were carried out as well.

Energies 2016, 9, 222 Energies 2016, 9, 222

4 of 22 4 of 22

Figure 3. Gas hydrate morphologies in core samples recovered from GMGS2 drilling sites, South

Figure 3. Gas hydrate morphologies in core samples recovered from GMGS2 drilling sites, South China China Sea. 1 and 2 (Site GMGS2-08F) massive; 3 and 4 (Site GMGS2-08E) laminated; 5 (Site GMGS2-08E) Sea. and 1 and 2 (Site GMGS2-08F) massive; 3 and 4 (Site GMGS2-08E) laminated; 5 disseminated. (Site GMGS2-08E) and 6 (Site GMGS2-08C) nodular; 7 (Site GMGS2-08E) vein; 8 (Site GMGS2-16D) 6 (Site GMGS2-08C) nodular; 7 (Site GMGS2-08E) vein; 8 (Site GMGS2-16D) disseminated.

2. System Description and Production Strategy

2. System Description and Production Strategy 2.1. System Description and Geometry Figure 4 shows and a sketch of this system, and provides some basic information on the thickness of 2.1. System Description Geometry the various layers. All the data in Figure4 are based on the best available data from Site GMGS2-8 of Figure 4 shows a sketch of this system, and provides some basic information on the thickness of the Dongsha Area [16]. The water depth at the GMGS2-8 site is 798 m. Preliminary available the various layers. All the inare Figure 4 are based on layers the best available data from GMGS2-8 information indicates thatdata there two hydrate-bearing at the GMGS2-8 site. TheSite upper gas

of the Dongsha Area [16]. The water depth at the GMGS2-8 site is 798 m. Preliminary available information indicates that there are two hydrate-bearing layers at the GMGS2-8 site. The upper gas

Energies 2016, 9, 222

5 of 22

hydrate layer2016, (Hydrate Layer I, HBL-I) is 9 m below the seafloor, and the thickness of this layer is 14 m. Energies 9, 222 5 of 22 According to the drilling sample results, hydrate occurs within fine-grained sediments in this layer, hydrate layer (Hydrate Layer I, HBL-I) is from 9 m below and the of this layersaturation is 14 m. and the saturation of the hydrate ranges 10%the to seafloor, 14%. Thus, in thickness this workhydrate in According to the drilling sample results, hydrate occurs within fine-grained sediments in this layer, this layer was selected as 12%. In the middle part (58–63 mbsf), a 5 m-thick authigenic carbonates and the saturation of the hydrate ranges from 10% to 14%. Thus, in this workhydrate saturation in deposited was found just above the top of the lower hydrate-bearing sediment. The permeability of this layer was selected as 12%. In the middle part (58–63 mbsf), a 5 m-thick authigenic carbonates this carbonates layer is very low. The lower gas hydrate layer (Hydrate Layer II, HBL-II) ranges from deposited was found just above the top of the lower hydrate-bearing sediment. The permeability of 63 to 98 mbsf with alayer thickness 35 The m, and the of the hydrate in this layer ranges rangesfrom from 30% this carbonates is veryof low. lower gassaturation hydrate layer (Hydrate Layer II, HBL-II) to 50%. in this work the hydrate saturation inofthis was selected as 40%. Meanwhile, 63Therefore, to 98 mbsf with a thickness of 35 m, and the saturation the layer hydrate in this layer ranges from 30% the analysis pressureincoring located at thesaturation depth ofin74–77 mbsf indicates thisMeanwhile, interval is rich to 50%.of Therefore, this work the hydrate this layer was selected that as 40%. the analysis of pressureAcoring located atsection the depth 74–77 mbsf indicates that this interval richinterval. of of massive gas hydrates. 20-cm-thick of of pure gas hydrate was recovered inisthis massive gas hydrates. A 20-cm-thick section of pure gas hydrate was recovered in this interval. Thus, Thus, in the cylindrical system used in the simulations (Figure 4), there is a massive hydrate interval the cylindrical system used in the simulations (Figure 4), there is a massive hydrate interval ranges rangesinfrom 74 to 77 mbsf, which is characterized by high porosity, high intrinsic permeability, and from 74 to 77 mbsf, which is characterized by high porosity, high intrinsic permeability, and high high hydrate saturation. The upper hydrate layer is overlain by an overburden with a thickness of 9 m, hydrate saturation. The upper hydrate layer is overlain by an overburden with a thickness of 9 m, and the lower hydrate layer is underlain with a thickness 22Meanwhile, m. Meanwhile, and the lower hydrate layer is underlainby byan an underburden underburden with a thickness of 22ofm. there is a 35m-thick mud between the upper and the lower hydrate layers. The cylindrical system there is a 35m-thick mud between the upper and the lower hydrate layers. The cylindrical system extends over an area withwith a radius of of 250 extends over an area a radius 250m. m.

Figure 4. Description thecylindrical cylindrical system in in thethe simulations. Figure 4. Description ofofthe systemused used simulations.

Table 1 shows the geological and system properties based on the exploration results and

Table 1 shows the geological and system properties based on the exploration results and the literature. the literature. Table 1. Geophysical properties and simulation models in marine hydrate reservoir at Site GMGS2-8

Table 1. Geophysical and simulation models in marine hydrate reservoir at Site GMGS2-8 of the Pearl Riverproperties Mouth Basin. of the Pearl River Mouth Basin. Parameter Layer thicknesses and porosities Parameter Hydrate saturation of HBL Layer thicknesses Initial pressure atand topporosities of HBL (PT) Initial temperature at top of HBL(TT) Hydrate saturation of HBL Initial temperature at base of HBL (TB) Initial pressure at top of HBL (PT ) Water depth Initial permeability temperature of at sediment top of HBL(T T ) layer Intrinsic of each Initial temperature at base of HBL (TB )

Value As in Figure 4 As in Value Figure 4 As in MPa Figure 4 8.11 277.60 K 4 As in Figure 281.53 K 8.11 MPa 798 m K4 As in277.60 Figure 281.53 K

Water depth

798 m

Intrinsic permeability of sediment of each layer

As in Figure 4

Energies 2016, 9, 222

6 of 22

Table 1. Cont. Parameter

Value

Capillary pressure model [35]

Pcap = ´P01 [(Sˆ)´1{λ ´1] 1´λ Sˆ = (SA ´ SirA )/(SmxA ´ SirA )

P01 [23]

105 Pa

λ

0.45

Dry thermal conductivity (kΘRD ) (all formations) [23]

1.0 W/m/K

Wet thermal conductivity (kΘRW ) (all formations) [23]

3.1 W/m/K

Composite thermal conductivity model

kΘC = kΘRD + (SA 1/2 + SH 1/2 ) (kΘRW ´kΘRD ) + ϕSI kΘI

Relative permeabilityModel [20]

krA = (SA ˆ )n krG = (SG ˆ )nG SA ˆ = (SA ´ SirA )/(1 ´ SirA ) SG ˆ = (SG ´ SirG )/(1 ´ SirA )

n [23]

3.572

nG [23]

3.572

SirG [23]

0.05

SirA [23]

0.30

Sea floor temperature T0 [23]

277.15 K

Geothermal gradient G [23]

0.045 K/m

Water salinity (mass fraction)

3.50%

The intrinsic permeabilities of the overburden, the HBL, and the underburden, which are a function of the sediment structure only (not of the fluid and hydrate), are estimated according to the limited measuring data [17]. The drilling data show that the zone above the lower hydrate-bearing layer is highly impermeable carbonate. Thus, the intrinsic permeability of the carbonate layer is chosen as 7.5 ˆ 10´20 m2 . The intrinsic permeability of the massive hydrate layer is chosen as 7.5 ˆ 10´12 m2 [36]. 2.2. Method of Production and Well Design The technologies for commercial production of natural gas from hydrate are still developing. The common methods for hydrate dissociation are: (1) the depressurization method, in which the hydrate reservoir pressure is reduced below the equilibrium decomposition pressure to decompose the hydrate [37,38]; (2) the thermal stimulation method, in which the hydrate reservoirs are heated above the equilibrium decomposition temperature to decompose the hydrate [21,39]; (3) the chemical injection method, in which chemicals (such as methanol or ethylene glycol) are injected into the reservoir to change the equilibrium hydrate decomposition conditions and induce hydrate dissociation [40,41]; and (4) the CO2 replacement method, in which CO2 is injected into the hydrate reservoirs to replace the methane gas [42,43]. A series of field production tests from the practical hydrate reservoirs have confirmed the availability of these methods, such as the test at the Mackenzie Delta (Northwest Territories, Canada) by thermal stimulation and depressurization methods [19], and the offshore test at the Nankai Trough, Japan by the depressurization method during 12–18 March 2013 [44]. Depressurization has been regarded as an effective method because of its economic and technical effectiveness [45]. Thus, in this work, depressurization is applied for the hydrate dissociation, and the effects of the hydrate dissociation by depressurization are analyzed. The carbonate layer and the low intrinsic permeability of marine clays (muds) precluded the use of horizontal wells. Thus, a vertical well model was used exclusively in this study (Figure 4). The well design is quite simple and involves two perforated intervals that cover the entire upper and lower hydrate-bearing intervals. The well radius in this work is rw = 0.1m. The numerical representation of a constant Pw involves treating the well as an internal boundary. In the case of a vertical well, this boundary is placed in the grid block above the uppermost cell in the well.

Energies 2016, 9, 222

7 of 22

Energies 2016, 9, 222

3. The Numerical Models and Simulation Approach

7 of 22

3. The Numerical Models and Simulation Approach

3.1. The Numerical Simulation Codes

3.1. The Numerical Simulation In this study, the parallel Codes version of the numerical code TOUGH+HYDRATE, which consists of an equilibrium andthe a kinetic numerical simulation.which Recently, bothofthe In this study, parallel model, version was of theemployed numericalfor code TOUGH+HYDRATE, consists an equilibrium a kinetic model, simulation. Recently, etboth the equilibrium and theand kinetic model havewas beenemployed validatedfor bynumerical Li, et al. [32–34] and Moridis, al. [45,46] equilibrium and the kinetic model have been validated by Li, et al. [32–34] and Moridis, et al. [45,46] using the experimental data. using the experimental data.

3.2. Domain Discretization 3.2. Domain Discretization

Figure 5 shows the mesh of the simulation system. The cylindrical domain of the single vertical Figurewas 5 shows the meshinto of the system. The cylindrical domain of thedissociation single vertical well problem discretized 72 simulation ˆ 71 = 5112 gridblocks in (r,z). The hydrate was well problem was discretized into 72 × 71 = 5112 gridblocks in (r,z). The hydrate dissociation was treated as an equilibrium reaction [47] and the effect of the salinity on hydrate dissociation was treated as an equilibrium reaction [47] and the effect of the salinity on hydrate dissociation was considered, which resulting in a system of 5112 equations. Discretization along the radial direction considered, which resulting in a system of 5112 equations. Discretization along the radial direction increases from rw to rmax , which was non-uniform. Discretization along the z-axis was uniform (with increases from rw to rmax, which was non-uniform. Discretization along the z-axis was uniform (with ∆z = 1.0 m) within the upper and lower hydrate-bearing layers (Figure 5). The discretization was Δz = 1.0 m) within the upper and lower hydrate-bearing layers (Figure 5). The discretization was nonnon-uniform theoverburden, overburden, center mud layer, underburden. uniform ininthe center mud layer, andand underburden.

Figure 5. Domain discretizationof ofthe thehydrate hydrate reservoir Figure 5. Domain discretization reservoirin inthe thesimulations. simulations.

3.3. Boundary and Initial Conditions

3.3. Boundary and Initial Conditions

The pressure and temperature conditions in the uppermost overburden and the lowermost The pressure and temperature conditions in the uppermost overburden and the lowermost underburden are set as constant, which are corresponding to the inactive boundaries. The initial underburden are of setthe as constant, which are corresponding to the inactive initial temperatures top and bottom boundaries are determined on the boundaries. basis of the The seafloor temperatures of the bottom gradient boundaries are determined on thepressure basis of at thethe seafloor temperature temperature andtop the and geothermal (shown in Table 1). Initial top of the HBL-I and (P the geothermal gradient (shown in Table 1). Initial pressure at the top of the HBL-I (P 8.11 MPa) T =domains T = 8.11 MPa) is determined by a hydrostatic distribution. Then, the P- and T- profiles in the is determined by athrough hydrostatic distribution. Then,the theboundary P- and T-temperature, profiles in the aregradient, initialized are initialized a short simulation using the domains hydrostatic through short simulation the Afterwards, boundary temperature, the hydrostatic and the adjusted and athe adjusted brine using density. the other initial parametersgradient, shown in Table 1 are simulated another short simulation. Unless a perturbation occurs from the outside, any conditions brine density.by Afterwards, the other initial parameters shown in Table 1 are simulated by another the system remain 6 shows the outside, distribution the initial hydrate saturation, shortinsimulation. Unlessunchangeable. a perturbationFigure occurs from the anyofconditions in the system remain initial pressure, and initial temperature in this simulation. As shown in Figure 6, there are twoand unchangeable. Figure 6 shows the distribution of the initial hydrate saturation, initial pressure, hydrate-bearing layers the Site GMGS2-8. The in hydrate in the upper gas hydrate layer is at initial temperature in thisat simulation. As shown Figuresaturation 6, there are two hydrate-bearing layers 12%,GMGS2-8. and the saturation of thesaturation hydrate inin most of gas the lower gaslayer hydrate layerand is 40%. However, of the Site The hydrate the area upper hydrate is 12%, the saturation there is a massive hydrate interval in the lower gas hydrate layer, and the hydrate saturation of this the hydrate in most area of the lower gas hydrate layer is 40%. However, there is a massive hydrate hydrate interval is 55%. The pressure and temperature distributions accord with geologic conditions.

interval in the lower gas hydrate layer, and the hydrate saturation of this hydrate interval is 55%. The pressure and temperature distributions accord with geologic conditions.

Energies 2016, 9, 222 Energies 2016, 9, 222

8 of 22 8 of 22

Figure 6. Spatial distributions of hydrate saturation, pressure, and temperatureat t = 0 day. Figure 6. Spatial distributions of hydrate saturation, pressure, and temperatureat t = 0 day.

3.4. Simulation Process and Approach 3.4. Simulation Process and Approach Because of significant uncertainties about the values of the important parameters, conditions Because of significant about valuessystem of the behavior, importantour parameters, and and variables that affect gasuncertainties production and thethe overall approachconditions involved two variables production and the systemcase, behavior, approach involved twolikely steps. steps. Thethat firstaffect stepgas involved the study ofoverall a reference whichour was based on the most The first step involved the study of a reference case, which was based on the most likely (expected) (expected) values of all the critical parameters and properties. The second step involved extensive values of all the critical parameters properties.ofThe involved sensitivity sensitivity analyses, which covered and the spectrum the second possiblestep variations inextensive the parameter and analyses, which covered the spectrum of the possible variations in the parameter and property values, property values, and thus defined the envelope of the expected system performance during longand thus definedHowever, the envelope of the performance production. term production. the 2D (r-z)expected model insystem this work is hard to during capturelong-term all the characteristics However, 2D (r-z) model this the work is hard toconditions capture allare theassumed characteristics of the flow dynamics. of the flowthe dynamics. In this in work, geological as rotational symmetry. If In this work, the geological conditions are assumed as rotational symmetry. If the comprehensive the comprehensive geological data for the hydrate deposit can be obtained, a 3D model can be geological data for the hydrate deposit potential can be obtained, a 3D model can be established to evaluate the established to evaluate the production in the future. production potential in the future.

Energies Energies2016, 2016,9,9,222 222

99ofof22 22

3.5. Evaluation Method 3.5. Evaluation Method The production potential of the hydrate deposits in the GMGS2-8 site can be evaluated by two The production potential of the hydrate deposits in the GMGS2-8 site can be evaluated by two production criteria: an absolute criterion and a relative criterion. To satisfy the absolute criterion, production criteria: an absolute criterion and a relative criterion. To satisfy the absolute criterion, production potential (i.e., QP, VP and QR, VR values) needs to be attained over the duration of the production potential (i.e., QP , VP and QR , VR values) needs to be attained over the duration of the study. The relative criterion is satisfied when the water/gas ratio RWG = MW/VP is low, indicating low study. The relative criterion is satisfied when the water/gas ratio RWG = MW /VP is low, indicating low production cost because of low energy requirements for the water lift and water disposal problems. production cost because of low energy requirements for the water lift and water disposal problems. 4. Production 4. Production from from GMGS2-Site GMGS2-Site 88 in in the the Pearl Pearl River River Mouth Mouth Basin: Basin: The The Reference Reference Case Case In this thiswork, work,the theproduction productionpressure pressureofof the well is set MPa, which is the same as that in In the well is set as as 4.54.5 MPa, which is the same as that in the the field test of the Nankai Trough [44]. field test of the Nankai Trough [44]. 4.1. Gas Gas and and Water Water Production Production 4.1. Figure 77 shows shows the the evaluations evaluations of of the the cumulative cumulative volume volume of ofproduced producedgas gas(V (Vpp),), cumulative cumulative Figure volumeof ofdissociated dissociatedgas gas(V(V R ), volumetric flow rate of produced gas (Q pand ), and volumetric flow volume ), volumetric flow rate of produced gas (Q ), volumetric flow raterate of p R of dissociated gas (Q R ) overtime. As shown in Figure 7, the change of the volumetric flow rate of gas dissociated gas (QR ) overtime. As shown in Figure 7, the change of the volumetric flow rate of gas (Q p (Qp and R) can be divided stages 1–S In S the S1 (0–1.8 years), the Qp and QR firstly reach and QR ) Q can be divided into into fourfour stages (S1 –S(S4 ). In4).the 1 (0–1.8 years), the Q p and QR firstly reach to 3/day and 3 /day to the maximum in 0.11 year (40 days). The maximum values p and are9560 9560mm the maximum in 0.11 year (40 days). The maximum values forfor thethe Q pQand QRQRare and 10,600m m33/day, /day, respectively. Afterwards, in in the the remaining remainingtime timeof ofthe theSS11,, the the value valueof ofthe theQQpp keeps keeps at at 10,600 respectively. Afterwards, 3 3 3 3 approximately9500 9500mm/day, /day, and andthe theQQRR decrease decreaseto to9500 9500mm /day. /day. The The Q QRR is is higher higher than than the theQ QPP in in the the approximately S 1 , because some of the dissociated gas remains in the hydrate reservoir. In the S 2 (1.8–4.6 years), the S1 , because some of the dissociated gas remains in the hydrate reservoir. In the S2 (1.8–4.6 years), the 3/day in a same rate. The decrease of QR and Qp Qpp and and Q QRR rapidly decrease to to approximately approximately7500 7500mm3 /day Q rapidly decrease in a same rate. The decrease of QR and Q p is caused caused by by the the reduction reduction of of the the hydrate hydrate saturation saturation near near the the vertical vertical well. well. In In the the SS33 (4.6–11.5 years), is (4.6–11.5 years), 3/day, indicating that the rate of 3 the curves of Q p and Q R fluctuate in a range from 7400 to 7800 m the curves of Q p and QR fluctuate in a range from 7400 to 7800 m /day, indicating that the rate of hydrate dissociation dissociation in in this this stage stage remains remains stable. stable. It It could could be be attributed attributed to to the the fact fact that that the the water water with with hydrate a higher temperature in the underburden flows to the HBL-II, and the sensible heat of the water is a higher temperature in the underburden flows to the HBL-II, and the sensible heat of the water is applied for hydrate dissociation in this stage. This hypothesis will be discussed in the next figures of applied for hydrate dissociation in this stage. This hypothesis will be discussed in the next figures of the temperature temperatureand andhydrate hydratedistributions. distributions.In Inthe thelast laststage stageSS4 4(11.5–30 (11.5–30years), years),the theQQppand andQ QRR gradually gradually the 3/day, which is caused by the low permeability of the system. 3 decrease from 7500 to 4700 m decrease from 7500 to 4700 m /day, which is caused by the low permeability of the system.

Figure 7. 7. Reference Referencecase: case:evolution evolutionofofcumulative cumulativevolume volume produced ), cumulative volume Figure of of produced gasgas (V p(V ), pcumulative volume of dissociated gas (V volumetric flow rate produced gas (Q p ),gas and(Q volumetric flow rate offlow dissociated of dissociated gas R), volumetric flowof rate of produced p), and volumetric rate of R ),(V gas (QR ) overtime. dissociated gas (QR) overtime.

Energies 2016, 9, 222

10 of 22

Energies 2016, 9, 222 released and produced volumes of gas in this figure are consistent 10 with of 22 the The cumulative observations made on the relative magnitude of Q p and QR , and their evolution overtime. The total The cumulative released and produced volumes of gas in this figure are consistent with the gas production is approximately 7.3 ˆ 107 ST m3 in 30 years. Thus, the average gas production rate in observations made on the relative magnitude of Qp and QR, and their evolution overtime. The total 30 years 6.7 ˆ 103isST m3 /day, which than Thus, the previous study the Shenhu 7 ST m3higher gas is production approximately 7.3 ×is 10much in 30 years. the average gas in production rateArea in of 3 /day) [48]. the South China Sea by GMGS-1 (211 ST m 30 years is 6.7 × 103 ST m3/day, which is much higher than the previous study in the Shenhu Area of Moreover, the maximum gas production rate has the South China Sea by GMGS-1 (211 ST m3/day) [48].the same order of magnitude of the first offshore 3 /day) [49]. Moreover, the maximum gas rate hasJapan the same ofm magnitude of This the first methane hydrate production test in theproduction Nankai Though, (2 ˆorder 104 ST work is 4 ST m3/day) [49]. This offshore methane hydrate production test in the Nankai Though, Japan (2 × 10 based on the local conditions at GMGS2-Site 8 with limited resources. The actual gas production from work is based local conditions GMGS2-Site 8 with limited resources. The actual gas hydrate deposits in on thethe Dongsha area may at obtain more economically feasible. production from hydrate deposits in the Dongsha area may obtain more economically Figure 8 shows the evolution of the cumulative mass of the produced water (MW )feasible. and gas-to-water Figure 8 shows the evolution of the cumulative mass of the produced water (MW) and gas-toratios (RGW = VP /MW ) overtime. As shown in Figure 8, MW increases continuously, and the rate of water ratios (RGW = VP/MW) overtime. As shown in Figure 8, MW increases continuously, and the rate water production increases over time. The final MW is 6.75 ˆ 101010 kg, which leads to an average of water production increases over time. The final MW is 6.75 × 10 kg, which leads to an average 6 waterwater production raterate of 6.16 ˆ 10 also presents a continuous decreasing trend. After 6 kg/day. R production of 6.16 × 10kg/day. RGW GW also presents a continuous decreasing trend. After t = 30t=days, RGW is smaller uneconomical[50]. [50].The The large amount of water production 30 days, RGW is smallerthan than5,5,which which is is uneconomical large amount of water production is mainly caused by the of this TheThe thickness of the overburden is just 9 meters, is mainly caused bythin the overburden thin overburden of system. this system. thickness of the overburden is just 9 meters, and difference the pressure difference between seawater and wellbore is over 3.5 MPa. seawater Therefore,easily and the pressure between seawater and wellbore is over 3.5 MPa. Therefore, flows through flowsseawater througheasily the overburden to the theoverburden wellbore. to the wellbore.

Figure 8. Reference case: evolution thecumulative cumulative mass water (M(M w) and gas to water Figure 8. Reference case: evolution ofofthe massof ofproduced produced water w ) and gas to water ratio (R GW) overtime. ratio (RGW ) overtime.

4.2. Spatial Distributions of P

4.2. Spatial Distributions of P

Figure 9 shows the evolution of pore pressure distribution in the hydrate reservoir over time. As

Figure 9 shows the pore pressure distribution in thethan hydrate reservoir time. As shown in Figure 9, theevolution carbonateof layer has initially lower permeability the other layers,over resulting shown in Figure 9, the carbonate layer has initially lower permeability than the other layers, resulting in an obvious pressure interrupt between the HBL-II and the middle mud layer. Meanwhile, the 3min anlayer obvious pressure interrupt between the HBL-II and the middle mud layer. Meanwhile, of massive hydrate in the HBL-II has initially higher permeability, resulting in a larger pressure the drop of over a larger hydrate distance. in As the shown, the largest pressurehigher drops in the domain occur near the 3m-layer massive HBL-II has initially permeability, resulting inwell a larger and drop correspond the distance. hydrate dissociation As time advances, region effected by near pressure over a to larger As shown,region. the largest pressure dropsthe in the domain occur depressurization increases continuously in the HBL-II. As also shown in Figure 9, the pressure the well and correspond to the hydrate dissociation region. As time advances, the region effected increases in the vicinity of the borehole afterin10the years. The reason forshown this behavior may 9, bethe thatpressure the by depressurization increases continuously HBL-II. As also in Figure dissociation gas migrates to the vicinity of the borehole because of the pressure gradient. The high increases in the vicinity of the borehole after 10 years. The reason for this behavior may be that the gas saturation causes the increase of the relative permeability in this region before 10 years, which dissociation gas migrates to the vicinity of the borehole because of the pressure gradient. The high gas leads to a low pressure area. Afterwards, the decrease of the gas saturation in the vicinity of the saturation causes of the relative permeability incausing this region before 10 years, which borehole leadsthe to increase the decrease of the relative permeability, the increase of pressure in theleads to a low pressure Afterwards, of the gas saturation in have the vicinity of thewhich borehole vicinity of the area. borehole. However,the the decrease pressure distributions in the HBL-I little change, leadsindicates to the decrease of the permeability, causing increase of pressure in the vicinity of the influence byrelative depressurization in the HBL-I hasthe little change over time. the borehole. However, the pressure distributions in the HBL-I have little change, which indicates the influence by depressurization in the HBL-I has little change over time.

Energies 2016, 9, 222 Energies 2016, 9, 222 Energies 2016, 9, 222

11 of 22 11 of 22 11 of 22

Figure9.9.Reference Reference case: case: evolution evolution of Figure of the the pore pore pressure pressure (P) (P) distribution distribution overtime. overtime. Figure 9. Reference case: evolution of the pore pressure (P) distribution overtime.

4.3. Spatial Distributions of T 4.3. 4.3.Spatial SpatialDistributions DistributionsofofTT Figure 10 shows the evolution of the T-distribution in the entire hydrate reservoir over time. As Figure evolution of ofthe theT-distribution T-distributioninin entire hydrate reservoir Figure10 10shows shows the the evolution thethe entire hydrate reservoir overover time.time. As shown in this figure, the area in the HBL-I near the well is cooling down, which is due to the As shown thisfigure, figure,the thearea areaininthe theHBL-I HBL-Inear nearthe thewell wellisiscooling coolingdown, down,which whichisisdue due to to the the shown in inthis endothermic reaction of hydrate dissociation and the penetration of cooler water from the overburden. endothermic andthe thepenetration penetrationofofcooler cooler water from overburden. endothermicreaction reactionof of hydrate hydrate dissociation dissociation and water from thethe overburden.

Figure 10. Reference case: evolution of the Temperature (T) distribution overtime. Figure10. 10.Reference Referencecase: case:evolution evolution of of the the Temperature Temperature (T) (T) distribution distribution overtime. Figure overtime.

Energies 2016, 9, 222 Energies 2016, 9, 222

12 of 22 12 of 22

Thecooling coolingarea areaexpands expandsfrom fromt t==1–5 1–5years. years.Afterwards, Afterwards,ititremains remainsunchanged, unchanged,indicating indicatingthat that The thecorresponding correspondingarea areafor forhydrate hydratedissociation dissociationalso alsodoes doesnot notexpend expendfrom fromt t==5–30 5–30years, years,or orthere thereisis the an equilibrium between the hydrate dissociation (which may expend) and the heat transport from an equilibrium between the hydrate dissociation (which may expend) and the heat transport from theenvironment. environment. On On the the well is continuously warmed up, the the other other hand, hand,the thearea areaininthe theHBL-II HBL-IInear near the well is continuously warmed andand thisthis area gradually expands from warmer water water from fromthe the up, area gradually expands fromt t==1–30 1–30years. years.This Thisisis because because the the warmer underburden enters the system. The low temperature region (box with black dots line) appears in the underburden enters the system. The low temperature region (box with black dots line) appears in the figureofoftemperature temperaturedistribution distributionatatt t==22years, years,which whichisiscaused causedby bythe theendothermic endothermicnature natureofofthe the figure hydrate dissociation. The low temperature region disappears in the next figure, because hydrate hydrate dissociation. The low temperature region disappears in the next figure, because hydrate in thisin this region is exhausted thewater warm water flows from the underburden into The the dissociation HBL-II. The region is exhausted and theand warm flows from the underburden into the HBL-II. of hydrate increases the of permeability theenhances HBL-II, and enhancesof the of the ofdissociation hydrate increases the permeability the HBL-II,ofand the mobility themobility warm water warm water from the underburden. Therefore, the warm area and the corresponding area for hydrate from the underburden. Therefore, the warm area and the corresponding area for hydrate dissociation dissociation expend continually. expend continually. 4.4. 4.4.Spatial SpatialDistributions DistributionsofofSSHH The Theevolution evolutionofofSSHHdistributions distributionsin inFigure Figure11 11shows showsthe theprogressive progressivedestruction destructionofofthe thehydrate hydrate through throughdissociation. dissociation.As Asseen seenininFigure Figure11, 11,the thehydrate hydratedissociation dissociationarea areaininthe theHBL-I HBL-Iexpands expandsslowly slowly over overtime. time.The Thereason reasonfor forthe theslow slowrate rateof ofhydrate hydratedissociation dissociationin inthis thislayer layerisisthat that(1) (1)the thepressure pressure gradient dissociation in HBL-I is lower than than that in HBL-II; (2) the (2) lowthe permeability causes gradientfor forhydrate hydrate dissociation in HBL-I is lower that in HBL-II; low permeability the limitation of the effect region byregion pressure (3) the cool(3) water fromwater the overburden enters this causes the limitation of the effect bydrop; pressure drop; the cool from the overburden layer. thechange S H distribution HBL-I has already in the P-distribution entersThe thischange layer. of The of the SHof distribution of HBL-Ibeen hasidentified already been identified in theof PFigure 9 and T-distribution Figure 10. distribution of Figure 9 andofT-distribution of Figure 10.

Figure 11. Reference case: evolution of the hydrate saturation (S H ) distribution overtime. Figure 11. Reference case: evolution of the hydrate saturation (SH) distribution overtime.

On Onthe theother otherhand, hand,the thechange changeofofSS distributionin inthe theHBL-II HBL-IIcan canbe bedivided dividedinto intotwo twostages. stages. HHdistribution Stage Stage11isisshown shownininthe thefigures figuresofoft t==11year yearand and22years. years.The The3-m 3-mmassive massivehydrate hydratelayer layerisisdissociated dissociated rapidly, rapidly,and andthe thedissociation dissociationregion regionininthe thethin thinlayer layerreaches reachestotoxx==50 50mmininthe thefirst firstyear. year.At Atthe thesame same time, time,the theother otherhydrate hydratedissociation dissociationregions regionsininthe theHBL-II HBL-IIexpand expandvery veryslowly. slowly.The Thedissociation dissociationregion region ininthe theother otherlayers layersininthe theHBL-II HBL-IIonly onlyreaches reachestotoxx==10 10m mininthe thefirst firstyear. year.ItItmay maybe bebecause becausethat thatthe the higher higherpermeability permeabilityin inthe themassive massivehydrate hydratelayer layerleads leadsto toaahigher higherrate rateof ofmass masstransfer, transfer,causing causingaa higher hydrate dissociation rate in this layer. The rapid hydrate dissociation rate causes the lower temperature region in Figure 10. Stage 2 is shown in the figures of t = 5–30 years. In this stage, the

Energies 2016, 9, 222

13 of 22

Energies 9, 222 dissociation higher2016, hydrate

of 22 rate in this layer. The rapid hydrate dissociation rate causes the 13 lower temperature region in Figure 10. Stage 2 is shown in the figures of t = 5–30 years. In this stage, the hydrate in the layer below the massive hydrate layer is dissociated quickly, and the maximum region hydrate in the layer below the massive hydrate layer is dissociated quickly, and the maximum region for hydrate dissociation reaches x = 190 m. This is due to the fact that the warm water from the for hydrate dissociation reaches x = 190 m. This is due to the fact that the warm water from the underburden enters the lower layer of the HLB-II. However, the hydrate in the layer above the underburden enters the lower layer of the HLB-II. However, the hydrate in the layer above the massive massive hydrate layer is dissociated slowly. The region for hydrate dissociation above the massive hydrate layer is dissociated slowly. The region for hydrate dissociation above the massive hydrate hydrate layer has almost no change from t = 5–30 years, which indicates that the warm water has little layer has almost no change from t = 5–30 years, which indicates that the warm water has little or no or no effect on the dissociation of hydrate in this region, because the warm water can directly flow effect on the dissociation of hydrate in this region, because the warm water can directly flow into the into the wellbore through the dissociated massive hydrate layer. wellbore through the dissociated massive hydrate layer.

5. 5. Sensitivity SensitivityAnalysis Analysisof ofProduction Productionfrom fromGMGS2-Site GMGS2-Site88in inPearl PearlRiver RiverMouth MouthBasin Basin In In this this work, work, we we investigated investigated the the sensitivity sensitivity of of gas gas production production to to the the following following conditions conditions and and parameters: the production pressure P W, the thermal conductivity kΘ, and the intrinsic permeability k parameters: the production pressure PW , the thermal conductivity kΘ , and the intrinsic permeability k of of the the reservoir. reservoir. 5.1. Sensitivity Sensitivity to to the the Production Production Pressure PW W Figures 13 show show the thedependence dependenceofofVVP P, ,VVRR,, M MWW, R RGW GW on W,, Figures 12 and 13 on the the production production pressure PW respectively. As seen seen in in Figure Figure 12, 12, the the V VPP and respectively. As and V VRRincrease increasewith withthe thedecrease decreaseof ofthe theproduction production pressure pressure PW Thefinial finialcumulative cumulativevolumes volumesof of gas gas production production for for the the cases cases with with the PW 2.3MPa, MPa, 3.4 3.4 MPa, MPa, W. .The W ==2.3 and are 1.0 1.0 ˆ× 10 1088 ST m m33, 8.8 ˆ × 1077 ST m3,, and 7.3 ˆ × 107 ST m3, respectively. Thus, and 4.5 MPa are Thus, the corresponding average average gas gasproduction productionrates ratesare are9.1 9.1ˆ×10 103 3ST STmm3 ,3,8.0 8.0ˆ× 10 1033 ST ST m m33, and and 6.7 6.7 ˆ × 1033 ST ST m33,, respectively. Obviously, the hydrate dissociation rate, which is Obviously, the the decrease decreaseof ofthe thePPWWcan canenhance enhance the hydrate dissociation rate, which caused byby thethe larger is caused largerpressure pressuredriving drivingforce forcefor forhydrate hydratedissociation dissociation(difference (difference between between the the pore pressure and the hydrate dissociation pressure) and the large pressure gradient for the movement of the W decreases the warm water from the underburden to the HBL-II. When PW decreases from from 4.5 4.5 to to 3.4 3.4 MPa, the volume of gas production increases increases by by 20.5%, 20.5%, and and when when PPW W decreases decreases from from 3.4 to to 2.3 2.3 MPa, MPa, the the volume volume of gas production increases by 13.6%. Additionally, Additionally, the the lowest lowest PPW W is is 2.3 2.3 MPa, MPa, which which is is lower lower than the quadruple point point of methane methane hydrate hydrate (2.60 (2.60 MPa MPa [51]). However, However, the ice blocking blocking does does not happen happen in this case.

Figure12. 12.Sensitivity Sensitivityto tothe theproduction productionpressure pressure(P (PW):): evolution evolution of of V VP and VR overtime. Figure P and V R overtime. W

Energies 2016, 9, 222 Energies2016, 2016,9,9,222 222 Energies

14 of 22

14ofof22 22 14

Figure13. 13.Sensitivity Sensitivityto tothe theproduction productionpressure pressure(P (PW):):evolution evolution ofM MW andRRGW GWovertime. overtime. Figure Figure 13. Sensitivity to the production pressure (PWW): evolutionof of MWWand and RGW overtime.

Asseen seenin inFigure Figure13, 13,the thecumulative cumulativemass massof ofwater waterproduction productionincreases increaseswith withthe thedecline declineof ofthe the As As seenfinial in Figure 13, the cumulative mass of water production increases with the decline of the P W . The cumulative masses of water production for the cases with the P W = 2.3 MPa, PW. The finial cumulative masses of water production for the cases with the PW = 2.3 MPa, PW 3.4 . The finial of water production for the with the PW = 2.3The MPa,GW3.4 MPa, MPa, andcumulative 4.5MPa MPaare aremasses 1.08××10 10 Kg,8.85 8.85 101010Kg, Kg,and and 6.75cases 101010Kg, Kg,respectively. respectively. inthe the 1111Kg, 3.4 MPa, and 4.5 1.08 ××10 6.75 ××10 The RRGW in 11 Kg, 8.85 ˆ 1010 Kg, and 6.75 ˆ 1010 Kg, respectively. The R and 4.5 MPa are 1.08 ˆ 10 in the case GW case with a lower production pressure before t = 3.7 years (1350 days) is higher. However, after case with a lower production pressure before t = 3.7 years (1350 days) is higher. However, after with lower production before t = 3.7production years (1350 days) isisis higher. t the =the 3.7 years, 3.7 years, theRRGW GWin inpressure thecase casewith withaalower lower production pressure lower.However, Thefinal finalRRafter GWfor for cases tt==a3.7 years, the the pressure lower. The GW cases thewith RGWthe inPthe with a lower production pressure is lower. The final R for the cases with the with the PWW==case 2.3 MPa, 3.4 MPa, and 4.5 MPa are 0.92, 0.99, and 1.09, respectively. The final R GW for GW 2.3 MPa, 3.4 MPa, and 4.5 MPa are 0.92, 0.99, and 1.09, respectively. The final RGW for PWthese = 2.3cases MPa,are 3.4 and 4.5 MPa are 0.92, 0.99, and 1.09, respectively. The final R for these cases these cases are MPa, close, because the gas production and water production both increase with the GW close, because the gas production and water production both increase with the decrease ofthe thePP aredecrease close, because the of WW . . gas production and water production both increase with the decrease of the PW .

Figure 14.Sensitivity Sensitivitytoto the production pressure (PWW):):evolution evolution andP, distribution at 30years. years. at Figure 14. tothe the production pressure (P of S distribution H , and T at W ): evolution Figure 14. Sensitivity production pressure (P ofofP,P,SSHH, ,and TTdistribution tt==30 t = 30 years.

Figure14 14shows showsthe theevolution evolutionof ofP, P,SSHH, ,and andTTdistributions distributionsat attt==30 30years yearsfor forthe thecases caseswith withthe the Figure P W = 2.3 MPa, 3.4 MPa, and 4.5 MPa, respectively. As seen in this figure, the regions effected by the PWFigure = 2.3 MPa, 3.4 MPa, 4.5 MPa, seen in thisatfigure, regions effected the 14 shows the and evolution of respectively. P, S H , and T As distributions t = 30the years for the cases by with the depressurization, the hydrate dissociated area, as well as the low-temperature region in the HBL-I the hydrate area, as well as the region effected in the HBL-I PWdepressurization, = 2.3 MPa, 3.4 MPa, and 4.5dissociated MPa, respectively. As seen in low-temperature this figure, the regions by the depressurization, the hydrate dissociated area, as well as the low-temperature region in the HBL-I and

Energies 2016, 9, 222 Energies 2016, 9, 222 Energies 2016, 9, 222

15 of 22 15 of 22 15 of 22

high-temperature region in the in HBL-IIHBL-II enlarge with the decline of theofproduction pressure. The reason and enlarge with the decline decline the production production pressure. The and high-temperature high-temperature region region in the the HBL-II enlarge with the of the pressure. The is that the larger pressure drop can supply the higher pressure driving force for hydrate dissociation, reason can supply supply the the higher higher pressure pressure driving driving force force for for hydrate hydrate reason is is that that the the larger larger pressure pressure drop drop can and the lower P can enhance the flow rates of the water in the overburden and underburden dissociation, and the lower P W can enhance flow rates of the water in the overburden andthe W dissociation, and the lower PW can enhance the flow rates of the water in the overburden into and underburden into the hydrate bearing layers. In addition, the temperature distribution in the case hydrate bearing layers. In addition, the temperature distribution in the case with the P = 2.3 MPa underburden into the hydrate bearing layers. In addition, the temperature distribution inW the case with the P Wthe = 2.3 MPa indicates that the temperatures of the HBL do not decrease to the freezing indicates that temperatures of the HBL do not decrease to the freezing point, even when the P with the PW = 2.3 MPa indicates that the temperatures of the HBL do not decrease to the freezingW is point, even the is quadruple point. point. lower than thewhen quadruple point, even when the P PWW point. is lower lower than than the the quadruple Sensitivity the Thermal Conductivityk 5.2.5.2. Sensitivity toto the 5.2. Sensitivity to theThermal ThermalConductivity ConductivitykkΘΘΘ Figures and show the dependence ofVV V VRRR,, M MW W, RGW on the thermal conductivity Figures 1515 and 1616 show the dependence onthe thethermal thermal conductivity Figures 15 and 16 show the dependenceofof W,, RGW conductivity kkΘΘkΘ PP,P,,VV GWon (describing fully water-saturated sediments), respectively. As seen in Figures 15 and 16, a change of kΘ (describing fully water-saturated sediments), 16 aa change (describing fully water-saturated sediments),respectively. respectively.As Asseen seen in in Figures Figures 15 and 16, change of k Θ from 2.1 to 4.1 W/mK leads to almost no change for the VP, VR, MW, and RGW. This result indicates kΘ from 4.1 W/mK leads to almost change theVV W,,and . This result indicates from 2.1 to2.1 4.1toW/mK leads to almost nono change forfor the andRRGW . This result indicates P ,P,VVRR, MW GW that the production behaviors are not sensitive for the thermal conductivity k Θ.. Thus, Thus, the effect ofthe thethe that the production behaviors are not sensitive for the thermal conductivity k Θ the effect of that the production behaviors are not sensitive for thermal conductivity kΘ . Thus, the effect of heat conduction during hydrate dissociation in this case is not obvious. heat conduction duringhydrate hydratedissociation dissociation in in this this case heat conduction during case is is not notobvious. obvious.

Figure 15. conductivity (k (kΘ Θ PP and Figure Sensitivity thethermal thermalconductivity Θ): ): evolution of and VVRR overtime. overtime. Figure 15.15. Sensitivity totothe ): evolution evolutionof ofVVV P and V R overtime.

Figure 16. 16. Sensitivity Sensitivity to to the the thermal thermal conductivity overtime. Figure conductivity (k (kΘΘ):): evolution evolution of of M MWW and and RRGW GW overtime. Figure 16. Sensitivity to the thermal conductivity (kΘ ): evolution of MW and RGW overtime.

Energies 2016, 9, 222 Energies 2016, 9, 222

16 of 22 16 of 22

Figure 1717 shows SHH, ,and andTTdistributions distributions 30 years forcases the cases with Figure showsthe theevolution evolution of of P, P, S at tat= t30=years for the with the 2016, 9, 222 16 of 22 thekEnergies kΘΘ==2.1 2.1 W/mK, 3.1 W/mK, and 4.5 W/mK, respectively. As seen in this figure, the P and W/mK, 3.1 W/mK, and 4.5 W/mK, respectively. As seen in this figure, the P and SHS H distributions for thethe cases with the kΘkΘ= =2.1 W/mK, and4.5 4.5W/mK W/mKare arealmost almost identical. distributions with the W/mK, 3.1 3.1 W/mK, W/mK, and identical. Figure 17for showscases the evolution of P, S2.1 H, and T distributions at t = 30 years for the cases with the However, the T distributions for these cases have a slight difference: the high-temperature region However, the T distributions for these cases have a slight difference: the high-temperature region kΘ = 2.1 W/mK, 3.1 W/mK, and 4.5 W/mK, respectively. As seen in this figure, the P and SHinin thethe HBL-II with thethe lower kΘkwith isislarger with the higher kΘΘ..4.5 However, thisalmost difference has HBL-II with lower Θ larger than that with3.1 the higherand However, difference has nono distributions for the cases the kΘthan = 2.1that W/mK, W/mK, W/mK this are identical. effect production behavior. effect on on thethe production behavior. However, the T distributions for these cases have a slight difference: the high-temperature region in the HBL-II with the lower kΘ is larger than that with the higher kΘ. However, this difference has no effect on the production behavior.

Figure Sensitivity to conductivity (kΘ): evolution of P, SH, and t = 30 years.at Figure 17.17.Sensitivity tothe thethermal thermal conductivity (kΘ ): evolution of P,T distribution S H , and T at distribution t = 30 years.

Figure 17. Sensitivity to the thermal conductivity (kΘ): evolution of P, SH, and T distribution at t = 30 years. 5.3. Sensitivity to the Intrinsic Permeability k

5.3. 5.3. Sensitivity Intrinsic Permeability kk Figuresto18the 19 show the dependence of VP, VR, MW, RGWon the intrinsic permeability k, Sensitivity toand the Intrinsic Permeability respectively. In the sensitivity cases, the k of the 3 m-massive hydrate layer (7.5 × 1012 m2) and the Figures 18 18 and 19 19 show the2 dependence ofofVV , RGW on the intrinsic permeability P ,P,VVRR,, M Figures show dependence MW W, RGW intrinsic k, k, 20 mthe carbonate layer and (7.5 × 10 ) keep no change, and the k of on thethe other layerspermeability changes from 12 2m2 ) and 12 respectively. In the sensitivity cases, the k of the 3 m-massive hydrate layer (7.5 ˆ 10 respectively. In mD) the sensitivity cases, of the 3 m-massive layer (7.5 × 10 m ) and the the 15 m2the 2 (750 mD), 7.5 × 1014 m2 (75 into 7.5 × 10 (7.5kmD) and 7.5 × 1013 mhydrate respectively. 20 2 20 2 carbonate layerlayer (7.5 ˆ 10 × m no change, and theand k of the the other layers changes 7.5 ˆ from 1014 m2 carbonate (7.5 10 ) keep m ) keep no change, k of the other layersfrom changes 15 2 13 2 14 2 15 2 13 2 (75 mD) intom7.5(75 ˆ mD) 10 into m (7.5 and(7.5 7.5mD) ˆ 10andm respectively. 7.5 × 10 7.5 ×mD) 10 m 7.5 (750 × 10 mD), m (750 mD), respectively.

Figure 18. Sensitivity to the intrinsic permeability (k): evolution of VP and VR overtime. Figure 18. Sensitivity to the intrinsic permeability (k): evolution of VP and VR overtime.

Figure 18. Sensitivity to the intrinsic permeability (k): evolution of VP and VR overtime.

Energies 2016, 9, 222 Energies 2016, 9, 222

17 of 22 17 of 22

Energies 2016, 9, 222

17 of 22

Figure 19. Sensitivity to the intrinsic permeability (k): evolution of MW and RGW overtime.

Figure 19. Sensitivity to the intrinsic permeability (k): evolution of MW and RGW overtime. Figure 19. Sensitivity to the intrinsic permeability (k): evolution of MW and RGW overtime. As seen in Figure 18, the changes of the VP and VR are evident. In the initial 10 years, the higher seen ainlarger Figurerate 18,ofthe changes of theHowever, V and VR are evident. In the initial 10 years,decrease, the higher k k As leads rates of gas In production As to seen in Figure 18, gas the production. changes of the PVP and the VR are evident. the initialgradually 10 years, the higher leads to a larger rate of gas production. However, the rates of gas production gradually decrease, and the decrease of the gas production rate in the highest k case is the quickest. Finally, the final VPand k leads to a larger rate of gas production. However, the rates of gas production gradually decrease, and V R in the case with the k = 7.5 mD are the highest. As seen in Figure 19, the M W in the case theand decrease of the gas production rate in rate the highest k case is the quickest. Finally,Finally, the final andVPVR Pwith the decrease of the gas production in the highest k case is the quickest. theVfinal lower k with is smaller, and the Rare GW with lower k As is higher. On the aspect ofMthe water production, the in the case the k = 7.5 mD the highest. seen in Figure 19, the in the case with lower W and VR in the case with the k = 7.5 mD are the highest. As seen in Figure 19, the MW in the case withk is production in R the lower case has better commercial value smaller, and the withklower k isahigher. On the aspect ofand the potential. water production, the production in GW and the RGW with lower k is higher. On the aspect of the water production, the lower k is smaller, theproduction lower k case a better commercial value and potential. in has the lower k case has a better commercial value and potential.

Figure 20. Sensitivity to the intrinsic permeability (k): evolution of P, SH, and T distribution at t = 30 years.

Figure 20 shows the evolution of P,permeability SH, and(k): T distributions atH,tand = 30 years for T the with the Figure Sensitivity totothe intrinsic permeability evolution of P, S T distribution atdistribution tcases = 30 years. Figure 20.20. Sensitivity the intrinsic (k): evolution of P, S H , and at k t==7.5 mD, 75 mD, and 750 mD, respectively. As seen in the figure of P distributions, the regions 30 years. Figure by 20 shows the evolution of , and Tenlarge distributions at tdecline = 30 years fork,the cases the influenced the depressurization in P, theSHHBL-II with the of the and thatwith in the kHBL-I = 7.5 mD, 75 mD, respectively. Ask.seen in theinfigure of P distributions, the regions is similar in and the 750 casesmD, with the different As seen the figure of SH distributions, the Figure 20 shows the evolution of P, S H , and T distributions at t = 30 years for the cases with the dissociatedby hydrate area in HBL-I enlarges with the increase of the the decline k. However, HBL-II, influenced the depressurization in the HBL-II enlarge with of theink,the and that inthe the k = 7.5 mD, 75 mD, and 750 mD, respectively. As seen in the figure of P distributions, the regions HBL-I is similar in the cases with the different k. As seen in the figure of SH distributions, the influenced by the depressurization inenlarges the HBL-II enlarge with the of the k, and thatHBL-II, in the HBL-I dissociated hydrate area in HBL-I with the increase ofdecline the k. However, in the the

Energies 2016, 9, 222

18 of 22

is similar in the cases with the different k. As seen in the figure of S H distributions, the dissociated hydrate area in HBL-I enlarges with the increase of the k. However, in the HBL-II, the regions of hydrate dissociation with different k are totally different. In the case with the k = 7.5 mD, the hydrate in the massive hydrate layer is dissociated rapidly, which is on account of the high intrinsic permeability and porosity of this layer. The dissociated region in this layer reaches about r = 180 m, which is the largest among the three cases. However, the hydrate in the other layer of the HBL-II is dissociated slowly. In the case with the k = 750 mD, the dissociated region in the massive hydrate layer only reaches about r = 120 m, but the dissociated region in the layer above the massive hydrate is the largest in the three cases. As seen in the figure of T distributions, the low temperature region in the HBL-I and the high temperature region in the HBL-II enlarge with the increase of k. This is because the reservoir with the higher k can enhance the mass of water from the over- and under-burden. Therefore, in the case with k = 7.5 mD, the hydrates in the HBL-II are mainly dissociated under the effect of depressurization. However, in the case with k = 750 mD, the hydrates in the HBL-II are mainly dissociated under the effect of warm water flow. 6. Conclusions In this work, the production potential of the gas hydrate accumulation at Site GMGS2-8 of the Dongsha Area in the Pearl River Mouth Basin has been firstly evaluated by the TOUGH-Hydrate code. A single vertical well is considered as the well configuration, and depressurization is employed as the dissociation method. Analyses of gas production sensitivity to the production pressure PW , the thermal conductivity kΘ , and the intrinsic permeability k of the reservoir are investigated as well. The following conclusions can be obtained: 1.

2.

3. 4.

The total gas production is approximately 7.3 ˆ 107 ST m3 in 30 years. Thus, the average gas production rate in the entire 30 years is 6.7 ˆ 103 ST m3 /day, which is much higher than the previous study in the Shenhu Area of the South China Sea by the GMGS-1. Moreover, the maximum gas production rate (9.5 ˆ 103 ST m3 /day) has the same order of magnitude of the first offshore methane hydrate production test in the Nankai Though. The decrease of the PW can enhance the hydrate dissociation rate. When PW decreases from 4.5 to 3.4 MPa, the volume of gas production increases by 20.5%, and when PW decreases from 3.4 to 2.3 MPa, the volume of gas production increases by 13.6%. Furthermore, the lowest PW is 2.3 MPa, which is lower than the quadruple point of methane hydrate. However, the ice blocking does not happen in this case. Production behaviors are not sensitive to the thermal conductivity kΘ . In the initial 10 years, the higher k leads to a larger rate of gas production, but the final VP and VR in the case with the lowest k are the highest.

Acknowledgments: This work is supported by National Science Fund for Distinguished Young Scholars of China (51225603), National Natural Science Foundation of China (51406210 and 51476174), and Key Arrangement Programs of the Chinese Academy of Sciences (KGZD-EW-301-2) which are gratefully acknowledged. Author Contributions: Yi Wang and Jing-Chun Feng conceived and designed the models; Yi Wang and Jing-Chun Feng performed the experiments; Yi Wang, Gang Li, and Yu Zhang analyzed the data; Xiao-Sen Li contributed analysis tools; Yi Wang wrote the paper. Conflicts of Interest: The authors declare no conflict of interest.

Notation G H k keff krA

thermal gradient within the sea (˝ C/m) depth of the sea water (m) intrinsic permeability (m2 ) effective permeability (m2 ) aqueous relative permeability (m2 )

Energies 2016, 9, 222

krG kΘC kΘRD kΘRW kΘI VP VR MW P PB P0 PW PW0 Q Qavg Qinj r RGW S t T T0 TW TB TT W x,y,z SH SG XS ∆Hc ∆PW ∆x ∆y ∆z ϕ η λ

gas relative permeability (m2 ) thermal conductivity (W/m/K) thermal conductivity of dry porous medium (W/m/K) thermal conductivity of fully saturated porous medium (W/m/K) thermal conductivity of ice (W/m/K) cumulative volume of produced gas (m3 ) cumulative volume of released gas (m3 ) cumulative mass of produced water (kg) pressure (MPa) initial pressure at base of HBL (MPa) atmosphere pressure (MPa) working pressure at the well (MPa) initial pressure at the well (MPa) injected heat (J) average gas production rate (ST m3 /day/m of well) heat injection rate (W/m of well) radius (m) the gas to water production ratio (ST m3 of CH4 / m3 of H2 O) phase saturation time (days) temperature (˝ C) the temperature of the sea floor (˝ C) injected warm water temperature (˝ C) initial temperature at the base of HBL (˝ C) initial temperature at the top of HBL (˝ C) pump work (J) cartesian coordinates (m) hydrate saturation gas saturation the mass fraction of salt in the aqueous phase combustion enthalpy of produced methane (J) driving force of depressurization, PW0 —PW (MPa) discretization along the x-axis (m) discretization along the y-axis (m) discretization along the z-axis (m) porosity energy ratio van Genuchten exponent—Table 1

Subscripts and Superscripts 0 A B cap G H

denotes initial state aqueous phase base of HBL Capillary gas phase solid hydrate phase

19 of 22

Energies 2016, 9, 222

20 of 22

References 1. 2. 3.

4. 5. 6. 7. 8. 9.

10. 11. 12. 13. 14.

15.

16.

17.

18. 19.

20. 21. 22.

Sloan, E.D. Fundamental principles and applications of natural gas hydrates. Nature 2003, 426, 353–359. [CrossRef] [PubMed] Veluswamy, H.P.; Kumar, R.; Linga, P. Hydrogen storage in clathrate hydrates: Current state of the art and future directions. Appl. Energy 2014, 122, 112–132. [CrossRef] Babu, P.; Kumar, R.; Linga, P. Unusual behavior of propane as a co-guest during hydrate formation in silica sand: Potential application to seawater desalination and carbon dioxide capture. Chem. Eng. Sci. 2014, 117, 342–351. [CrossRef] Osegovic, J.P.; Blake-Collins, B.; Slattery, I.M.; Max, M.D. Accelerated Hydrate Formation and Dissociation. U.S. Patent 8,334,418, 18 December 2012. Kumar, A.; Kumar, R. Role of Metallic Packing and Kinetic Promoter in Designing a Hydrate-Based Gas Separation Process. Energy Fuels 2015, 29, 4463–4471. [CrossRef] Yang, S.H.B.; Babu, P.; Chua, S.F.S.; Linga, P. Carbon dioxide hydrate kinetics in porous media with and without salts. Appl. Energy 2016, 162, 1131–1140. [CrossRef] Koh, C.A.; Sum, A.K.; Sloan, E.D. Gas hydrates: Unlocking the energy from icy cages. J. Appl. Phys. 2009, 106. [CrossRef] Sloan, E.D.; Koh, C.A. Clathrate Hydrates of Natural Gases; Marcel Dekker Inc.: New York, NY, USA, 2008. Bohrmann, G.; Kuhs, W.F.; Klapp, S.A.; Techmer, K.S.; Klein, H.; Murshed, M.M.; Abegg, F. Appearance and preservation of natural gas hydrate from Hydrate Ridge sampled during ODP Leg 204 drilling. Mar. Geol. 2007, 244, 1–14. [CrossRef] Pohlman, J.W.; Kaneko, M.; Heuer, V.B.; Coffin, R.B.; Whiticar, M. Methane sources and production in the northern Cascadia margin gas hydrate system. Earth Planet. Sci. Lett. 2009, 287, 504–512. [CrossRef] Ruppel, C.; Boswell, R.; Jones, E. Scientific results from Gulf of Mexico Gas Hydrates joint Industry Project Leg 1 drilling: Introduction and overview. Mar. Petrol. Geol. 2008, 25, 819–829. [CrossRef] Sain, K.; Rajesh, V.; Satyavani, N.; Subbarao, K.V.; Subrahmanyam, C. Gas-hydrate stability thickness map along the Indian continental margin. Mar. Petrol. Geol. 2011, 28, 1779–1786. [CrossRef] Moridis, G.J.; Reagan, M.I.; Kim, S.J.; Seol, Y.; Zhang, K. Evaluation of the Gas Production Potential of Marine Hydrate Deposits in the Ulleung Basin of the Korean East Sea. Spe J. 2009, 14, 759–781. [CrossRef] Kwon, T.H.; Lee, K.R.; Cho, G.C.; Lee, J.Y. Geotechnical properties of deep oceanic sediments recovered from the hydrate occurrence regions in the Ulleung Basin, East Sea, offshore Korea. Mar. Petrol. Geol. 2011, 28, 1870–1883. [CrossRef] Feng, J.C.; Li, X.S.; Li, G.; Li, B.; Chen, Z.Y.; Wang, Y. Numerical Investigation of Hydrate Dissociation Performance in the South China Sea with Different Horizontal Well Configurations. Energies 2014, 7, 4813–4834. [CrossRef] Ning, F.; Zhang, K.; Wu, N.; Zhang, L.; Li, G.; Jiang, G.; Yu, Y.; Liu, L.; Qin, Y. Invasion of drilling mud into gas-hydrate-bearing sediments. Part I: Effect of drilling mud properties. Geophys. J. Int. 2013, 193, 1370–1384. [CrossRef] Liu, C.L.; Meng, Q.G.; He, X.L.; Li, C.F.; Ye, Y.G.; Zhang, G.X.; Liang, J.Q. Characterization of natural gas hydrate recovered from Pearl River Mouth basin in South China Sea. Mar. Petrol. Geol. 2015, 61, 14–21. [CrossRef] Max, M.D.; Johnson, A.H. Hydrate petroleum system approach to natural gas hydrate exploration. Petrol. Geosci. 2014, 20, 187–199. [CrossRef] Moridis, G.J.; Collett, T.S.; Pooladi-Darvish, M.; Hancock, S.; Santamarina, C.; Boswell, R.; Kneafsey, T.; Rutqvist, J.; Kowalsky, M.B.; Reagan, M.T.; et al. Challenges, Uncertainties, and Issues Facing Gas Production From Gas-Hydrate Deposits. Spe Reserv. Eval. Eng. 2011, 14, 76–112. [CrossRef] Li, X.-S.; Wang, Y.; Li, G.; Zhang, Y. Experimental Investigations into Gas Production Behaviors from Methane Hydrate with Different Methods in a Cubic Hydrate Simulator. Energy Fuels 2011, 26, 1124–1134. [CrossRef] Wang, Y.; Li, X.-S.; Li, G.; Zhang, Y.; Li, B.; Feng, J.-C. A three-dimensional study on methane hydrate decomposition with different methods using five-spot well. Appl. Energy 2013, 112, 83–92. [CrossRef] Chong, Z.R.; Yang, S.H.B.; Babu, P.; Linga, P.; Li, X.-S. Review of natural gas hydrates as an energy resource: Prospects and challenges. Appl. Energy 2016, 162, 1633–1652. [CrossRef]

Energies 2016, 9, 222

23.

24.

25. 26.

27.

28. 29.

30.

31. 32.

33. 34.

35.

36.

37.

38. 39.

40.

41.

21 of 22

Li, X.S.; Yang, B.; Li, G.; Li, B. Numerical Simulation of Gas Production from Natural Gas Hydrate Using a Single Horizontal Well by Depressurization in Qilian Mountain Permafrost. Ind. Eng. Chem. Res. 2012, 51, 4424–4432. [CrossRef] Chen, D.X.; Wu, S.G.; Dong, D.D.; Mi, L.J.; Fu, S.Y.; Shi, H.S. Focused fluid flow in the Baiyun Sag, northern South China Sea: Implications for the source of gas in hydrate reservoirs. Chin. J. Oceanol. Limnol. 2013, 31, 178–189. [CrossRef] Li, G.; Li, X.S.; Chen, Q.; Chen, Z.Y. Numerical Simulation of Gas Production from Gas Hydrate Zone in Shenhu Area, South China Sea. Acta Chim. Sin. 2010, 68, 1083–1092. Li, G.; Moridis, G.J.; Zhang, K.; Li, X.S. The use of huff and puff method in a single horizontal well in gas production from marine gas hydrate deposits in the Shenhu Area of South China Sea. J. Petrol. Sci. Eng. 2011, 77, 49–68. [CrossRef] Su, Z.; Cao, Y.C.; Wu, N.Y.; Chen, D.F.; Yang, S.X.; Wang, H.B. Numerical investigation on methane hydrate accumulation in Shenhu Area, northern continental slope of South China Sea. Mar. Petrol. Geol. 2012, 38, 158–165. [CrossRef] Su, Z.; Moridis, G.J.; Zhang, K.N.; Wu, N.Y. A huff-and-puff production of gas hydrate deposits in Shenhu area of South China Sea through a vertical well. J. Petrol. Sci. Eng. 2012, 86–87, 54–61. [CrossRef] Li, G.; Li, X.S.; Zhang, K.N.; Moridis, G.J. Numerical simulation of gas production from hydrate accumulations using a single horizontal well in Shenhu Area, South China Sea. Chin. J. Geophys. 2011, 54, 2325–2337. Su, Z.; Cao, Y.C.; Wu, N.Y.; He, Y. Numerical Analysis on Gas Production Efficiency from Hydrate Deposits by Thermal Stimulation: Application to the Shenhu Area, South China Sea. Energies 2011, 4, 294–313. [CrossRef] Feng, J.C.; Li, G.; Li, X.S.; Li, B.; Chen, Z.Y. Evolution of Hydrate Dissociation by Warm Brine Stimulation Combined Depressurization in the South China Sea. Energies 2013, 6, 5402–5425. [CrossRef] Li, B.; Li, G.; Li, X.S.; Li, Q.P.; Yang, B.; Zhang, Y.; Chen, Z.Y. Gas Production from Methane Hydrate in a Pilot-Scale Hydrate Simulator Using the Huff and Puff Method by Experimental and Numerical Studies. Energy Fuels 2012, 26, 7183–7194. [CrossRef] Li, B.; Li, X.S.; Li, G. Kinetic studies of methane hydrate formation in porous media based on experiments in a pilot-scale hydrate simulator and a new model. Chem. Eng. Sci. 2014, 105, 220–230. [CrossRef] Li, G.; Li, B.; Li, X.S.; Zhang, Y.; Wang, Y. Experimental and Numerical Studies on Gas Production from Methane Hydrate in Porous Media by Depressurization in Pilot-Scale Hydrate Simulator. Energy Fuels 2012, 26, 6300–6310. [CrossRef] Kneafsey, T.J.; Moridis, G.J. X-Ray computed tomography examination and comparison of gas hydrate dissociation in NGHP-01 expedition (India) and Mount Elbert (Alaska) sediment cores: Experimental observations and numerical modeling. Mar. Petrol. Geol. 2014, 58, 526–539. [CrossRef] Moridis, G.J.; Kim, J.; Reagan, M.T.; Kim, S.J. Feasibility of gas production from a gas hydrate accumulation at the UBGH2–6 site of the Ulleung basin in the Korean East Sea. J. Petrol. Sci. Eng. 2013, 108, 180–210. [CrossRef] Li, X.-S.; Yang, B.; Zhang, Y.; Li, G.; Duan, L.-P.; Wang, Y.; Chen, Z.-Y.; Huang, N.-S.; Wu, H.-J. Experimental investigation into gas production from methane hydrate in sediment by depressurization in a novel pilot-scale hydrate simulator. Appl.Energy 2012, 93, 722–732. [CrossRef] Wang, Y.; Feng, J.C.; Li, X.S.; Zhang, Y.; Li, G. Large Scale Experimental Evaluation to Methane Hydrate Dissociation below Quadruple Point in Sandy Sediment. Applied Energy. 2016, 162, 372–381. [CrossRef] Wang, Y.; Feng, J.C.; Li, X.S.; Zhang, Y. Analytic Modeling and Large-scale Experimental Study of Mass and Heat Transfer during Hydrate Dissociation in Sediment with Different Dissociation Methods. Energy 2015, 90, 1931–1948. [CrossRef] Li, X.S.; Wan, L.H.; Li, G.; Li, Q.P.; Chen, Z.Y.; Yan, K.F. Experimental Investigation into the Production Behavior of Methane Hydrate in Porous Sediment with Hot Brine Stimulation. Ind. Eng. Chem. Res. 2008, 47, 9696–9702. [CrossRef] Li, G.; Li, X.S.; Tang, L.G.; Zhang, Y. Experimental investigation of production behavior of methane hydrate under ethylene glycol injection in unconsolidated sediment. Energy Fuels 2007, 21, 3388–3393. [CrossRef]

Energies 2016, 9, 222

42.

43. 44. 45.

46. 47. 48.

49. 50. 51.

22 of 22

Bai, D.S.; Zhang, X.R.; Chen, G.J.; Wang, W.C. Replacement mechanism of methane hydrate with carbon dioxide from microsecond molecular dynamics simulations. Energy Environ. Sci. 2012, 5, 7033–7041. [CrossRef] Komatsu, H.; Ota, M.; Smith, R.L.; Inomata, H. Review of CO2 -CH4 clathrate hydrate replacement reaction laboratory studies—Properties and kinetics. J. Taiwan Inst. Chem. E 2013, 44, 517–537. [CrossRef] Konno, Y.; Jin, Y.; Shinjou, K.; Nagao, J. Experimental evaluation of the gas recovery factor of methane hydrate in sandy sediment. RSC Adv. 2014, 4, 51666–51675. [CrossRef] Moridis, G.J.; Collett, T.S.; Boswell, R.; Kurihara, M.; Reagan, M.T.; Koh, C.; Sloan, E.D. Toward Production From Gas Hydrates: Current Status, Assessment of Resources, and Simulation-Based Evaluation of Technology and Potential. Spe Reserv. Eval. Eng. 2009, 12, 745–771. [CrossRef] Birkedal, K.A.; Freeman, C.M.; Moridis, G.J.; Graue, A. Numerical Predictions of Experimentally Observed Methane Hydrate Dissociation and Reformation in Sandstone. Energy Fuels 2014, 28, 5573–5586. [CrossRef] Kowalsky, M.B.; Moridis, G.J. Comparison of kinetic and equilibrium reaction models in simulating gas hydrate behavior in porous media. Energy Convers. Manag. 2007, 48, 1850–1863. [CrossRef] Su, Z.; He, Y.; Wu, N.Y.; Zhang, K.N.; Moridis, G.J. Evaluation on gas production potential from laminar hydrate deposits in Shenhu area of south China sea through depressurization using vertical wells. J. Petrol. Sci. Eng. 2012, 86–98. [CrossRef] Yamamoto, K. Japan Completes First Off shore Methane Hydrate Production Test—Methane Successfully Produced from Deepwater Hydrate Layers. Fire Ice 2013, 13, 1–2. Li, G.; Moridis, G.J.; Zhang, K.N.; Li, X.S. Evaluation of Gas Production Potential from Marine Gas Hydrate Deposits in Shenhu Area of South China Sea. Energy Fuels 2010, 24, 6018–6033. [CrossRef] Li, X.S.; Zhang, Y.; Li, G.; Chen, Z.Y.; Yan, K.F.; Li, Q.P. Gas hydrate equilibrium dissociation conditions in porous media using two thermodynamic approaches. J. Chem. Thermodyn. 2008, 40, 1464–1474. [CrossRef] © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).