Authors Accepted Manuscript

An easy and efficient way to evaluate mechanical

properties of gas hydrate-bearing sediments: The
direct shear test

Zhichao Liu, Houzhen Wei, Li Peng, Changfu Wei,

Fulong Ning
www.elsevier.com/locate/petrol

PII: S0920-4105(16)30456-9
DOI: http://dx.doi.org/10.1016/j.petrol.2016.09.040
Reference: PETROL3647
To appear in: Journal of Petroleum Science and Engineering
Received date: 29 May 2015
Revised date: 11 November 2015
Accepted date: 21 September 2016
Cite this article as: Zhichao Liu, Houzhen Wei, Li Peng, Changfu Wei and
Fulong Ning, An easy and efficient way to evaluate mechanical properties of gas
hydrate-bearing sediments: The direct shear test, Journal of Petroleum Science
and Engineering, http://dx.doi.org/10.1016/j.petrol.2016.09.040
This is a PDF file of an unedited manuscript that has been accepted for
publication. As a service to our customers we are providing this early version of
the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting galley proof before it is published in its final citable form.
Please note that during the production process errors may be discovered which
could affect the content, and all legal disclaimers that apply to the journal pertain.
 An easy and efficient way to evaluate mechanical properties of gas

hydrate-bearing sediments: The direct shear test

Zhichao Liua, Houzhen Weib*, Li Penga, Changfu Weib, Fulong Ninga*
a
Faculty of Engineering, China University of Geoscience, Wuhan 430074, China

b
State Key Laboratory of Geomechanics and Geotechnical Engineering, Institute of Rock and Soil Mechanics, Chinese Academy

of Sciences, Wuhan 430071, China

nflzx@cug.edu.cn

weihouzhen@163.com
*
Corresponding authors

Abstract

Understanding the mechanical behaviours of gas hydrate-bearing sediments (GHBS) is important for

their associated applications in wellbore stability, stratum deformation during exploitation, geological

disaster prevention, and the risk assessment of the exchange of CH4 with CO2 in hydrate reservoirs and

CO2 sequestration in oceans. However, triaxial tests on mechanical properties of GHBS are taxing and

time-consuming. Here, we presented an easy and efficient way to evaluate these by using a

self-developed direct shear apparatus. Then a series of direct shear tests on GHBS represented by CO2

hydrate-bearing silt were performed to investigate their mechanical behaviours and strength indices by

changing the axial pressure, CO2 hydrate saturation, shear rate and hydrate synthesis temperature. Our

results indicate that CO2 hydrate significantly strengthens specimens by cementing silt grains. In

addition, when hydrate saturation increases, the cohesions are enhanced from 0.09 MPa to 2.39 MPa,

and the internal friction angles increase and decrease at the range from 28.6 to 43.3 under the

experimental conditions. These findings have direct implications for evaluating the stability and safety

of natural gas hydrate reservoirs, CO2 replacement to extract CH4 and CO2 sequestration.

Keywords: gas hydrate; CO2; direct shear; sediment; mechanical property

1. Introduction
Natural gas hydrate in nature is an unconventional energy resource that is mainly found in

onshore permafrost and offshore regions (Kvenvolden, 1988; Sloan, 1997; Mahajan et al, 2007) which

1
can satisfy future energy demands for 1000 years (Ghiassian and Grozic, 2013). However, it is difficult

to achieve safe and effective gas production from hydrate reservoirs due to its complicated geological

distribution and sensitive depositional environment (Hyodo et al., 2013b), which cause a big challenge

to accurately understand the mechanical behaviours of gas hydrate-bearing sediments (GHBS).

Compared with conventional recovery methods (Demirbas, 2010), the exchange of CH4 with CO2

in hydrate reservoirs (CH4nH2O+CO2CO2nH2O+CH4) is a creative idea that has been proposed in

recent years (Ohgaki et al., 1994, 1996; Ota et al., 2005; Nakazono et al., 2008; Ors and Sinayuc, 2014)

because the replacement reaction is a spontaneous process under proper conditions and CO2 can fill the

void of released CH4 which could mitigate the problems of wellbore stability, stratum deformation,

geohazards and greenhouse effect. Thus, it is important to study the mechanical properties of CO2

hydrate-bearing sediments and compare them with those of natural GHBS especially CH 4

hydrate-bearing sediments (Ning et al., 2012, 2015; Mekala et al., 2014; Yasin et al., 2015).

Currently, the mechanical properties of gas hydrate-bearing sediments are primarily studied in

laboratory experiments because it is difficult and costly to perform in situ mechanical tests. But there

are some mechanical experiments on in situ hydrate samples from natural gas hydrate deposits. Winters

et al. (2000, 2007) conducted mechanical comparison experiments using permafrost hydrate samples

from the Mallik area and artificial hydrate-bearing sediments. And then mechanical experiments were

performed to some in situ hydrate samples from oceanic hydrate deposits such as the eastern Nankai

Trough (Masui et al., 2007, 2008a, 2008b; Yoneda et al., 2013a, 2013b, 2015a, 2015b) and the

Krishnae Godavari Basin (Priest et al., 2014). These experiments indicated that artificial hydrate

samples exhibited mechanical behaviours similar to those of in situ samples.

Therefore, the more common approach to investigate mechanical properties of GHBS is to test

artificial hydrate-bearing sediments by laboratory mechanical experiments. Yun et al. (2007)

investigated the mechanical properties of artificial sand, silt and clay samples containing

tetrahydrofuran (THF) hydrate. Hyodo et al. (2005, 2013b, 2014a, 2014b) performed a series of

comparison experiments to investigate the effects of temperature, pore pressure, effective confining

pressure and hydrate saturation on the mechanical properties of artificial sand samples containing CH4

and CO2 hydrates. Song et al. (2010), Li et al. (2011, 2013) and Liu et al. (2013, 2014) used a triaxial

apparatus to study the effect of confining pressure on hydrate strength in artificial sediment samples

2
consisting of clay, hydrate and ice grains and compared the strengths of CO2, CH4 and CO2-CH4 hybrid

hydrate-bearing sediments after incomplete CO2-CH4 replacement.

In the above-mentioned studies, most experiments were performed by triaxial tests, which were

complex, taxing and time-consuming. To simplify the experimental methods, here we designed a

simple direct shear apparatus and performed a series of direct shear tests on gas-hydrate-bearing silt

specimens to investigate their mechanical behaviours and strength indices by changing the axial

pressure, CO2 hydrate saturation, shear rate and hydrate synthesis temperature. The results are

comparable to those of natural and artificial samples from triaxial tests, which prove the direct shear

test is an easy and efficient way to study the mechanical properties of GHBS. In addition, the strength

indices obtained from this method, such as the shear strength, cohesion and internal friction angle,

could be used to evaluate the practical failure of GHBS in the future.

2. Experimental method
2.1. Apparatus

The direct shear apparatus consists of six parts, shear boxes, a horizontal displacement unit, an

axial pressure-loading unit, a temperature controller, a gas supply unit, and a data acquisition system.

Figure1a and b illustrates the apparatus that we used in the direct shear experiments. The shear boxes

can hold cylindrical specimens that are 61.8 mm in diameter and 20 mm thick and is divided into three

sections, a top shear box, a bottom shear box and a pulley support base (Figure 1a and c). The top and

bottom shear boxes hold the specimens for the shear tests. The horizontal displacement unit provide

shear force by using a constant flow pump and cut-off valves. The axial pressure-loading unit applies

an axial pressure on the specimens using an axial pressure piston that is driven by N2 gas. The

temperature controller can regulate the temperature from -20 to 150 C, with an accuracy of 0.5 C

via an air bath. The gas supply unit has two functions, providing N2 gas for applying the axial pressure

and supplying CO2 gas for hydrate synthesis. The data acquisition system collects and digitizes signals

from the sensors and displays them on the screen in real-time.

Our apparatus can prepare hydrate specimens in shear boxes under stable P-T conditions

compared with conventional direct shear apparatus. In addition, the specimens are thinner than those

used in typical hydrate triaxial apparatus, which allows for the synthesis of high-quality hydrate

specimens because gas can easily permeate into the specimens.

3
1. Pressure pad 2. Ventilation gaskets 3. Artificial hydrate-bearing silt specimens 4. Top shear box 5.

Bottom shear box 6. Pulley support base 7. Axial pressure-loading piston 8. Displacement sensor 9.

Temperature sensor 10. Horizontal displacement piston 11. Vacuum pump 12. Constant flow pump 13.

CO2 gas cylinder 14. Nitrogen gas cylinder 15. Pressure gauge 16. Incubator 17. Pressure chamber 18.

Cut-off valve 19. Data acquisition system

(a)

(b) (c)

Figure 1 The direct shear apparatus for hydrate-bearing sediments. (a) a schematic diagram. (b)

the main parts in the incubator. (c) the shear boxes

2.2. Materials and tools

The experimental materials used in this work include silt, distilled water, CO2 gas with a purity of

99.9% and N2 gas with a purity of 99.9%. The silt collected from the alluvial plain of the Yellow River

is a kind of flood deposited soils. The silts were firstly pre-treated by drying to remove moisture and

sieving to remove other impurities. The basic characteristics of the silt are shown in Figure 2. The tools

4
used for preparing the specimens include a standard cutting ring (61.8 mm in diameter, 20 mm in

height, and 43.0 g in weight), an electronic balance (maximum weight of 2100 g and accuracy of 0.01

g), a homemade pressure box for compressing the silt specimen, and hydraulic jacks with a maximum

load of 16 Tons.

100
3
Gravity 2.68 g/cm
90 Liquild limit 29%
Plastic limit 19%
80
Cumulative frequency/%

70
60
50
40
30
20
10
0
1E-3 0.01 0.1 1
Grain size/mm

Figure 2 Basic characteristics of the silt specimens without hydrates

2.3. Procedures

The detailed test processes and P-T conditions during the hydrate synthesis (Figure 3) are as

follows:

(1) Specimen preparation: The masses of pre-treated silt and water are determined according to

the required hydrate saturation (see appendix for detail calculation). When the silt and water are stirred

thoroughly, the mixtures are placed into a homemade pressure box that is matched with a standard

cutting ring. Next, the mixtures are pressed into remoulded cylindrical specimens (61.8 mm in diameter,

20 mm thick, and restricted by the homemade pressure box and standard cutting ring) at 20 MPa for 5

minutes using a hydraulic jack.

(2) Specimen installation: Specimens are placed into the shear box through the standard cutting

ring, and the ventilation gaskets are placed on the top and bottom surfaces of the specimens. Then, the

pressure chamber is connected to the gas system. Simultaneously, the acquisition system records

real-time data.

(3) CO2 gas injection: Although the air compounds (except CO2) do not form hydrates under our

test conditions, the pressure chamber is vacuumed for 5 minutes before injecting CO2 gas to ensure that

the CO2 hydrate is pure. Next, CO2 gas is injected into the chamber at a pre-set pressure (3.6 MPa)

5
using a pressure-regulating valve, which was maintained for 24 hours to allow the CO2 gas to fully

penetrate the specimens in advance.

(4) CO2 hydrate synthesis: The temperature gradually decreases from about 15 C to a pre-set

value (3 C) while the pressure chamber is always connected to the CO2 gas cylinder through the

pressure-regulating valve. As shown in Figure 3, the CO2 hydrates can stabilize under the temperature

and pressure. This condition is maintained for 48 hours until the pressure stops decreasing to ensure

that all of the water is consumed.

(5) Direct shear tests: After hydrate synthesis, axial loading is applied to the hydrate specimens

and the constant flow pump is set to the corresponding speed. Then, direct shear tests are performed

and measured, and the shear operation is ended when the shear displacement reaches 15 mm

(approximately 25% of the specimen diameter).

Figure 3 Pressure and temperature conditions during preparation of the hydrate specimens (the

CO2 hydrate phase diagram was obtained from Pahlavanzadeh et al., 2012)

2.4. Experimental design

Test designs are shown in Table 1: CO2 hydrate specimens with different hydrate saturations are

prepared based on initial water contents, and different axial pressures are applied to obtain shear

stress-displacement curves. In addition, the effects of experimental conditions (including the shear rate

and hydrate synthesis temperature) on the mechanical properties of the specimens are investigated.

6
 Table1 The shear test designs for the CO2 hydrate-bearing silt specimens

Gas Saturation of
Test Temperature Shear rate
pressure CO2 hydrate Axial pressure /MPa
sequence t/C /mm/min
p/MPa Sh/%

1 15* 0.1* 0 0.31,0.47,0.62,0.78 3

2 3 3.6 21.38 0.31,0.47,0.62,0.78 3

3 3 3.6 48.85 0.31,0.47,0.62,0.78 3

4 3 3.6 65.29 0.31,0.47,0.62,0.78 3

5 3 3.6 48.85 0.62 1,9

6 1,2 3.6 48.85 0.62 3

*For hydrate-free specimens, the average atmospheric temperature and pressure are 15 C and 0.1 MPa, respectively.

3. Results and discussions

3.1. Effect of axial pressure

Figure 4 shows the shear stress-displacement curves of the specimens with hydrates (48.85%

saturation) under different axial pressures. All the curves can be divided into four stages. In the first

stage, it shows elastic behaviour, the shear strengths linearly increase with the increase of shear

displacement. The second stage is an elastoplastic deformation and reaches a peak point as the slope

decreases. In the third stage, it shows soften behaviour, the shear stress rapidly decrease to a constant

value as the specimen is broken by coupling with hydrate cementation failure (see section 3.2) and

hydrate's own failure (Wu et al., 2015) and stress has to readjust. The last stage shows stress retention,

the residual stress changes at a nearly constant rate until the shear movement ends at approximately 15

mm. The end stress of stage II and IV are designated as peak and residual strength, respectively. By

combining the characteristics of the four stages, the curves reveal an apparent strength maximum and a

brittle failure tendency that is similar to the failure mode of the in situ hydrate samples from the eastern

Nankai Trough (Yonedaet al., 2015a, 2015b).

7
 2.0 Axial pressure 0.31MPa
Axial pressure 0.47MPa
Axial pressure 0.62MPa
Axial pressure 0.78MPa
1.5

Shear stress/MPa
1.0

0.5

I II III IV
0.0
0 2 4 6 8 10 12 14 16
Shear displacement/mm

Figure 4 The shear stress-displacement of the specimens with 48.85% hydrate saturation under

various axial pressures

According to Figure 4, the initial slopes in the first stage, which represent the difficulties of

specimen displacement, become steeper and the peak strengths enhanced as the axial pressure increases.

This tendency is similar to that of hydrate-bearing turbidite sediments in the eastern Nankai Trough and

shown to be a function of the confining pressure (Yoneda et al., 2015b). In addition, the residual

strengths also increased as the axial pressure increases. Therefore, the high axial pressure enhanced the

resistance of the specimens throughout the shear process because it resulted in dense specimens that

could resist displacement and have high friction force during the relative movement on the shear plane.

3.2. Effect of hydrate saturation

Figure 5 shows the shear stress-displacement curves of the specimens with different hydrate

saturations under the same axial pressure (at 0.62 MPa). The increase in shear strength with gas hydrate

saturation is consistent with previous results based on measurements of artificial and natural

hydrate-bearing sediments by triaxial tests (Masui et al., 2007; Yun et al., 2007; Miyazaki et al., 2010;

Yoneda et al., 2015a). The curves of the specimens with high hydrate saturations have the same brittle

failure tendency, as shown in Figure 4, and the curves of the specimens without hydrates or with low

hydrate saturation exhibit ductile failures, which are observed in triaxial tests of natural pressure core

samples (Yoneda et al., 2015a). Based on the four curves shown in Figure 5: (1) high hydrate

saturations result in high peak strengths and an especially pronounced increase when the hydrate

saturation is greater than 50%, which agrees with the results reported by Yun et al. (2007). (2) In

addition, the specimens without hydrates do not exhibit an obviously decreasing shear stress, and (3)

8
the brittle failure tendency of the specimens with the hydrates becomes more obvious as the hydrate

saturation increases.

3.0
Hydrate saturation 0%
Hydrate saturation 21.38%
2.5 Hydrate saturation 48.85%
Hydrate saturation 65.29%

Shear strength/MPa
2.0

1.5

1.0

0.5

0.0
0 2 4 6 8 10 12 14 16
Shear displacement/mm

Figure 5 The shear stress-displacement of the specimens with different hydrate saturations under

an axial pressure of 0.62 MPa

According to the hydrate growth pattern in gas-saturation sediments (Waite et al., 2004; Yun et al.,

2007), the hydrate particles are more likely to fill pore spaces and cement or coat silt grains for

increasing hydrate contents. Pore-scale morphology using X-ray computed microtomography also

showed that CO2 hydrates formed in an excess gas condition had a cementation effect on grains (Ta et

al., 2015). Therefore, when hydrate saturation is low, the hydrate particles have little or no effect on the

cementing interactions between the silt grains. However, hydrate particles in the specimens with high

hydrate saturation grow together and cement themselves and silt grains, which make it difficult to

destroy the specimens. In this case, the peak strengths are mostly contributed to cementation force and

are partly contributed to interface friction force. With increasing displacement, the effect of

cementation force on the shear plane gradually decrease and only the effect of friction force should be

considered. Therefore, higher hydrate saturations correspond to larger cementation effects by the

hydrates and larger differences between the peak strengths and the residual strengths which show

obvious brittle failures when the hydrate saturation is greater than 48.85% (Figure 5). In addition, the

residual strengths are enhanced with increasing hydrate saturation. This may be attributed to the entire

structure of the specimens with high hydrate saturations, which has a greater integrity and a lower

tendency to crumble due to dendritic or tree-branch geometry of hydrates growing into pore spaces (Ta

et al., 2015). These traits result in the following displacement after failure when compared with the low

saturation specimens. For the specimens without hydrates, no decrease in stress is observed because the

9
hydrate cementation does not enhance the shear stress of the specimens. The whole mechanism can be

illustrated in Figure 6.

Figure 6 The mechanisms of silt and hydrate movement during the shear process (Modified from

Waite et al., 2004, Yun et al., 2007 and Ta et al., 2015)

3.3. Effect of shear rate

Figure 7 shows the shear stress-displacement of the specimens at different shear rates. The

increase of shear rate results in enhanced peak strengths because more force is required to balance the

corresponding resistance during shear motion if the specimen structure need to be destroyed quickly,

which has been confirmed in ice, frozen soils and CH4 hydrate-bearing sediments (Hawkes and Mellor,

1972; Baker, 1979; Song et al., 2010; Liu et al., 2013). However, an irregular tendency exists for

residual strengths. This phenomenon may be explained by two contradiction effects caused by high

shear rate: (1) the increase effect of squeezing among grains with the increase of shear rate, which

increases the friction force on the shear plane (Liu et al., 2013), and (2) the decrease effect of less time

for grains rearranging to make specimen dense, which decrease their corresponding residual strengths.

Therefore, the specimens may have random residual strengths after failure.

10
 2.5
Shear rate of 1mm/min
Shear rate of 3mm/min
2.0 Shear rate of 9mm/min

Shear strength/MPa
1.5

1.0

0.5

0.0
0 2 4 6 8 10 12 14 16
Shear displacement/mm

Figure 7 The shear stress-displacement of specimens with a hydrate saturation of 48.85% at

various shear rates

3.4. Effect of hydrate synthesis temperature

Figure 8 shows the shear stress-displacement curves of the specimens at different synthesis

temperatures. It is consistent with the results of prior studies in which lower temperatures resulted in

higher peak strengths (Song et al., 2010; Yu et al., 2011; Hyodo et al., 2013b) because hydrates at low

temperatures are believed to exhibit better thermal stability and greater intermolecular forces (Hyodo et

al., 2014a). In contrast, the tendency of the residual strengths appears different from that of the peak

strengths. We speculate that hydrate may form quickly under high super-cooling degrees and is

non-uniformly distributed in the specimen. Therefore, the residual strength is lower than that of the

specimens at high synthesis temperatures because the structure and cementation of the specimens at

low temperature may be broken completely after the specimen is damaged.

2.5
Synthesis temperature 1 C
Synthesis temperature 2 C
2.0 Synthesis temperature 3 C
Shear stress/MPa

1.5

1.0

0.5

0.0
0 2 4 6 8 10 12 14 16
Shear displacement/mm

Figure 8 The shear stress-displacement of the specimens with a hydrate saturation of 48.85% at

various synthesis temperatures

11
3.5. Mechanical parameters

3.5.1. Peak strength

Except for the axial pressure of 0.62 MPa shown in Figure 5, the specimens with different hydrate

saturations are tested using axial pressures of 0.34 MPa, 0.47 MPa and 0.78 MPa and compared with

previously reported results from field samples (Masui et al., 2007; Santamarina et al., 2015; Yoneda et

al., 2015a, 2015b) and artificial samples containing CH4 and CO2 hydrate (Grozic and Ghiassian, 2010;

Miyazaki et al., 2010; Ordonez and Grozic, 2011; Hyodo et al., 2013a, 2014a) obtained by triaxial

compression tests (Figure 9). To remove the influences of other factors, we used the differences

between the strengths of each of the hydrate-bearing and hydrate-free sediment to analyse the changes

in the peak strengths of the specimens. Although our results are slightly lower than those of field

samples, which are either frozen by liquid nitrogen or sand or silt-sand, they are comparable with those

from field or other artificial samples.

This study-CO2-silt
Hyodo et al. -CO2-sand [2014]
Ordonez et al. -CO2-sand [2011]
6 Hyodo et al. -CH4-sand [2013a]
Miyazaki et al. -CH4-sand [2010]
Grozic et al. -CH4-sand [2010]
5
Difference of peak strength/MPa

Yoneda et al. -field-silty sand [2015a]

Yoneda et al. -field-silt [2015b]
Santamarina et al. -field-silty sand [2015]
4 Masui et al. -field-sand [2007]

2 Empirical equation for synthetic

methane hydrate-bearing sand
Miyazaki et.al [2010]
1
DS=1.91e Sh

0
0 10 20 30 40 50 60 70 80
Hydrate saturation/%

Figure 9 The relationship between the hydrate saturation and difference in the peak strength (The

solid marks represent data acquired in this study and the other artificial samples containing CH4 and CO2 hydrate, and

the hollow marks represent data acquired from marine-recovered field samples.)

12
 Similar to the behaviour shown in Figure 5, the differences in the peak strengths of the specimens

significantly increased as the hydrate saturation increased (Figure 9). The increased rates of the peak

strength remains nearly unchanged when the hydrate saturation is less than 48.85%, and a nearly

exponential increase in the peak strength is observed when the hydrate saturation increases to 65.29%,

as reported by Miyazaki et al. (2010). An empirical equation for synthetic methane hydrate-bearing

sand is DS=1.91e-3*Sh1.80, where DS represents the difference in the strength between hydrate-bearing

and hydrate-free sediments, as proposed by Miyazaki et al. (2010). Here, the trend of the peak strength

difference for CO2 hydrate-bearing silt with high hydrate saturation generally agrees with that of the

empirical equation. Therefore, the hydrate saturation combined with the hydrate distribution in the

sediments is an essential factor used to evaluate wellbore stability and sediment deformation, especially

when the hydrate saturation decreases from high to low levels due to global warming and human

activities.

3.5.2. Cohesion and internal friction angle

According to the Mohr-Coulomb criterion, the strength envelope curves of the specimens can be

expressed as follows:

=tan+c (1)

where is the peak strength of a direct shear test, is the axial pressure of the direct shear test, is the

internal friction angle, and c is the cohesion.

According to Equation (1), the cohesion and internal friction angle of the specimens with different

hydrate saturations are acquired in Table 2, and the relationships between cohesion, internal friction

angle and hydrate saturation in this study and previous studies of artificial samples containing CH4,

CO2 and THF hydrate are plotted in Figures 10 and 11.

Table2 The Mohr-Coulomb formulas of the specimens with different hydrate saturations

Hydrate Mohr-Coulomb formula Cohesion c/MPa Internal friction

saturation Sh/% angle /

0 =0.72+0.09 0.09 35.9

21.38 =0.94+0.42 0.42 43.3

48.85 =0.69+1.29 1.29 34.8

65.29 =0.55+2.39 2.39 28.6

13
 5 This study-CO2-silt
Ghiassian et al.-CH4-sand [2013]
Zhang et al.-CH4-sand [2012]
4 Ordonez et al.-CO2-sand [2011]
Yun et al.-THF-sand,silt,clay [2007]
Masui et al.-CH4-sand [2005]
fitting formula of this study:

Cohesion/MPa
3 Masui et al.-field-sand [2006]
2.02
c=0.09+4.97e-4*Sh

2
fitting curve of Waite et al.
-CH4-sand [2009]
1
fitting curve of Miyazaki et al.
-CH4-sand [2010]

0
0 10 20 30 40 50 60 70 80 90 100
Hydrate saturation/%

Figure 10 The relationship between cohesion and hydrate saturation

This study-CO2-silt
Ghiassian et al.-CH4-sand [2013]
Zhang et al.-CH4-sand [2012]
60 Ordonez et al.-CO2-sand [2011]
Yun et al.-THF-sand,silt,clay [2007]
Masui et al.-CH4-sand [2005] fitting curve of Waite et al.
50
Masui et al.-field-sand [2006] -CH4-sand [2009]
Internal friction angle/ C
O

40

30
fitting curve of Miyazaki et al.
-CH4-sand [2010]
20
fitting formula of this study:
=0.35Sh+35.9(Sh<23.86%)
10 =-0.38Sh+53.2(Sh23.86%)

0
0 10 20 30 40 50 60 70 80 90 100
Hydrate saturation/%

Figure 11 The relationship between the internal friction angle and hydrate saturation

The cohesions in this study rapidly increase by following an exponential tendency as the hydrate

saturation increase (Figure 10). This tendency coincides with that observe in previous studies (Masui et

al., 2005, 2006; Yun et al., 2007; Waite et al., 2009; Miyazaki et al., 2010; Ordonez and Grozic, 2011;

Zhang et al., 2012; Ghiassian and Grozic, 2013), which illustrates that hydrate saturation has a large

effect on the cohesion and that the mechanism of this effect is similar to that responsible for the effect

of hydrates on the peak strength, which was discussed in section 3.2. As shown in Figure 10, the

mathematical expression of the fitting curve for the relationship between cohesion and hydrate

14
saturation is c=0.09+4.97e-4*Sh2.02, where R2=0.98. The internal friction angles shown in Figure 11

firstly increased from 35.9 to 43.3 and subsequently decreased to 34.8 and 28.6. Although Waite et

al. (2009) and Miyazaki et al. (2010) found that the internal friction angles were nearly independent of

hydrate saturation, Zhang et al. (2012) found that the internal friction angles decreased as the hydrate

saturation increased. Here, our data remain within a reasonable range compared with previous studies

(Masui et al., 2005, 2006; Yun et al., 2007; Waite et al., 2009; Miyazaki et al., 2010; Ordonez and

Grozic; 2011, Zhang et al., 2012; Ghiassian and Grozic, 2013). The fitting curve between the internal

friction angle and hydrate saturation can be simplified as =0.35Sh+35.9 when the saturation is less

than 23.86% and as =-0.38Sh+53.2 when the saturation is greater than 23.86% (Figure 11).

The reversal of the internal friction angle may be attributed to the grain movement on a shear

plane, as shown in Figure 6. When the silt specimens do not contain hydrate, the silt grains normally

roll and slip over one another on a shear plane as there is water lubrication between them. Thus, the

internal friction angle is normal. When small amounts of hydrate (e.g., 21.38% saturation) are present

in the silt specimens, small hydrate grains replacing the water occupy the spaces among the silt grains

or partly coat them. When relative motion occurs on the shear plane, both the large silt grains and the

small hydrate grains roll and slip on the shear plane, which makes the shear process difficult. Thus, the

internal friction angle increases, and when hydrate saturation reaches a critical point (e.g., 48.85%

saturation), the mode of hydrate action in sediment partially transforms from filling pores to cementing

or coating the silt grains. Then, the shear plane becomes smoother because gas hydrate is an ice-like

crystalline material. Therefore, the internal friction angle decreases and remains slightly lower than that

of the silt specimens without hydrate. When the hydrate saturation of the silt specimens increases to a

high level (e.g., 65.29% saturation), the mode of hydrate action in the silt specimens is mostly

transformed to cement and even fully coat the silt grains. Thus, the internal friction angle is smallest

because the shear plane is the smoothest. In conclusion, as the hydrates fill the pores, the internal

friction angle increases with increasing hydrate saturation. When the mode of hydrate action hydrates

becomes the cementation or coating of the grains, the internal friction angles decreases with increase of

hydrate saturation.

Finally, when considering the hydrate saturation, the Equation (1) can be modified as a function of

the hydrate saturation as follows:

=tan[(Sh)]+c(Sh) (7)

15
where (Sh)=0.35Sh+35.9, (Sh<=23.86%), (Sh)=-0.38Sh+53.2, (Sh>23.86%), and

c(Sh)=0.09+4.97e-4*Sh2.02. If the stratum stress and the hydrate saturation are known, the shear strength

of the hydrate-bearing sediment can be approximately evaluated using Equation (2). Furthermore, the

relationship between the cohesion, internal friction angle and the hydrate saturation can provide

mechanical parameters for future numerical simulations.

4. Conclusions and prospects

In this work, we designed a direct shear apparatus for evaluating the mechanical properties of

GHBS and tested the stress-displacement curves under the conditions of different axial pressures,

hydrate saturations, shear rates and synthesis temperatures. The strength indices such as shear strength,

cohesion and internal friction angle, which are main parameters for the risk assessment of wellbore

stability, sediment deformation during gas production, geological disaster and the safe exchange of CH4

with CO2 in hydrate reservoirs, are also investigated by taking CO2 hydrate as an example. The results

show the self-developed direct shear apparatus can successfully form the artificial hydrate specimens to

simulate the natural GHBS, and the mechanical behaviours and strength indices obtained from this

apparatus are comparable to the previous results obtained from the taxing and time-consuming triaxial

measurements on natural and lab-synthesized GHBS. It is proved that this method is easy and efficient

to simulate and evaluate the mechanical properties of GHBS.

However, the results obtained from the direct shear apparatus may represent a minimum value

relative to homogeneous hydrate-bearing sediment samples due to the edge effect and sample

heterogeneity (Yoneda et al., 2015a). This study only considers CO2 hydrate-bearing silt for the

application of CO2 replacement in NGH-bearing sediments and does not fully represent actual

NGH-bearing sediments. In future studies, specimens with skeleton materials consisting predominantly

of clay and sand, and other hydrate synthesis gases, such as CH4 and C2H6, should be tested and

compared with the specimens presented here. And more detailed hydrate behaviours at the pore scale

before and after shear tests would be analyzed in future study.

16
Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant No.

51274177, 51239010 and 41572295) and was partly supported by the Program for New Century

Excellent Talents in University (NCET-13-1013), Fok Ying Tong Education Foundation (132019) and

the Fundamental Research Funds for the Central Universities, China University of Geosciences

(Wuhan) (CUG120112).

Appendix
1. Hydrate synthesis and determination of its saturation

Three methods are typically used to form hydrate-bearing sediment specimens in the laboratory

(Waite et al., 2004; Ghiassion et al., 2013). The first method involves mixing water and gas in advance

to form pure hydrate grains or hydrate grains containing ice under suitable temperature and pressure

conditions. Then, the grains of the hydrate or hydrate-containing ice are mixed with soil grains at a

high pressure and the remoulded mixtures are used for tests at low temperatures. In the second method,

soil specimens are prepared with water. Next, the soil specimens are placed in the closed chamber of

the test apparatus and gas is injected to pressurize the chamber. When the temperature decreases into

the left zone of the hydrate phase-equilibrium boundary, hydrates form in the specimens. The third

method is to dissolve former gas in water and cycles through specimens under hydrate synthesis

conditions. When hydrates form in the specimen pores, the specimen can be used for the following tests.

In our tests, silt specimens partially saturated by water are adopted to form CO2 hydrate-bearing silt

specimens via the second method.

Hydrate saturation can be estimated from the volume of pore water (Vw) available in the

specimens during hydrate synthesis (Ghiassian and Grozic, 2013). The following assumptions are made

in this study: 1) all water has been converted to hydrate and, 2) the gas is held in only 13% of the

hydrate (by volume) while the remaining 87% of the hydrate is occupied by water (Kvenvolden, 1993).

Vh=Vw/0.87 (A-1)

where Vh and Vw are the volumes of the formed hydrate and initial water, respectively.

17
 The following relationship between the hydrate saturation (Sh) and the initial water saturation (Sw)

is obtained.

Sh=Vh/V=Vw/0.87V=Sw/0.87 (A-2)

where Sh and Sw are the saturations of the formed hydrate and initial water, respectively, and V is the

volume of the pores in the specimen, which remains constant.

The initial water saturation (Sw) can be expressed using the following soil mechanics equations

(Punmia et al., 2005):

=


= 1 (A-3)


=
{ 1 +

where w is the initial water content, e is the void ratio of the specimens, s is the density of the silt

grains, w is the density of water, d is the density of the dried specimens, and is the bulk density of

the specimens.

Replacing e and d in Equation (A-3) yields the following:

= (A-4)
[(1 + ) ]

Finally, by inserting Equation (A-4) into Equation (A-2), the following expression for hydrate

saturation (Sh) is obtained:

= (A-5)
0.87 [(1 + ) ]

The initial water content (w) is pre-determined according to the experimental requirement, and

each silt specimen mixed with the water in the standard cutting ring (60 cm3) contains 102.6 g of the

mixture. Therefore, because the density of the silt grains (s=2.68 g/cm3), the bulk density of the

specimens (=102.6/60=1.71 g/cm3), the density of water (w=1.00 g/cm3) and the initial water content

(w is pre-determined) are clear, the hydrate saturation of the specimens can be calculated according to

Equation (A-5). In our tests, the initial water content varies from 4.4% to 18% (less than the liquid limit

29% and plastic limit 19% to ensure specimen without water loss during compression moulding and

not being out of shape after compression moulding), and the porosity of the silt specimens is

36.6-59.0%, which is similar to that of the in situ hydrate samples (Lee et al., 2013).

18
References
Baker, T. H. W., 1979. Strain rate effect on the compressive strength of frozen sand. Engineering

Geology 13(1), 223-231.

Demirbas, A., 2010. Methane hydrates as potential energy resource: Part 2-Methane production

processes from gas hydrates. Energy Conversion and Management 51(7), 1562-1571.

Ghiassian H, Grozic J. L. H., 2013. Strength behavior of methane hydrate bearing sand in undrained

triaxial testing. Marine and Petroleum Geology 43, 310-319.

Grozic, J. L. H., Ghiassian, H., 2010. Undrained shear strength of methane hydrate-bearing sand:

preliminary laboratory results. In Proceedings 6th Canadian permafrost conference and 63rd

Canadian geotechnical conference, Calgary, 459-466.

Hawkes, I., Mellor, M., 1972. Deformation and fracture of ice under uniaxial stress. Glaciology 11(61),

103-131.

Hyodo, M., Nakata, Y., Yoshimoto, N., Ebinuma, T., 2005. Basic research on the mechanical behavior

of methane hydrate-sediments mixture. Soils and foundations 45(1), 75-85.

Hyodo, M., Yoneda, J., Yoshimoto, N., Nakata, Y., 2013a. Mechanical and dissociation properties of

methane hydrate-bearing sand in deep seabed. Soils and foundations 53(2), 299-314.

Hyodo, M., Li, Y., Yoneda, J., Nakata, Y., Yoshimoto, N., Nishimura, A., Song, Y., 2013b. Mechanical

behavior of gas-saturated methane hydrate-bearing sediments. Journal of Geophysical Research:

Solid Earth 118(10), 5185-5194.

Hyodo, M., Li, Y., Yoneda, J., Nakata, Y., Yoshimoto, N., Kajiyama, S., Nishimura, A., Song, Y., 2014a.

A comparative analysis of the mechanical behavior of carbon dioxide and methane

hydrate-bearing sediments. American Mineralogist 99(1), 178-183.

Hyodo, M., Li, Y. Yoneda, J., Nakata, Y., Yoshimoto, N., Nishimura, A., 2014b. Effects of Dissociation

on the Shear Strength and Deformation Behavior of Methane Hydrate-Bearing Sediments. Marine

and Petroleum Geology 51,52-62.

Kvenvolden, K. A., 1988. Methane hydrate-a major reservoir of carbon in the shallow geosphere?

Chemical Geology 71(1), 41-51.

Kvenvolden, K. A., 1993. A primer on gas hydrate. In: Howell, D.G., et al. (Eds.), The Future of

Energy Gases. US Geological Survey Professional, 1570, pp. 279-291.

19
Lee, J. Y., Kim, G. Y., Kang, N. K., Yi, B. Y., Jung, J. W., Im, J. H., Son, B. K., Bahk, J. J., Chun, J. H.,

Ryu, B. J., Kim, D. S., 2013. Physical properties of sediments from the Ulleung Basin, East Sea:

results from second Ulleung Basin gas hydrate drilling expedition, East Sea (Korea). Marine and

Petroleum Geology 47, 43-55.

Li, Y., Song, Y., Liu, W., Yu, F., Wang, R., 2013. A new strength criterion and constitutive model of gas

hydrate-bearing sediments under high confining pressures. Journal of Petroleum Science and

Engineering 109, 45-50.

Li, Y., Song, Y., Yu, F., Liu, W., Wang, R., 2011. Effect of confining pressure on mechanical behavior

of methane hydrate-bearing sediments. Petroleum Exploration and Development 38(5), 637-640.

Liu, W., Zhao, J., Luo, Y., Song, Y., Li, Y., Yang, M., Zhang, Y., Liu, Y., Wang, D., 2013. Experimental

measurements of mechanical properties of carbon dioxide hydrate-bearing sediments. Marine and

Petroleum Geology 46, 201-209.

Liu, W., Shen, Z., Li, Y., Song, Y., Lu, Y., Zhu, Y., 2014. Analysis of mechanical properties of hybrid

hydrate sediments after CO2-CH4 incomplete replacement. In: The 8th International Conference

on Gas Hydrates (ICGH 2014), Beijing, China.

Mahajan, D., Taylor, C. E., Mansoori, G. A., 2007. An introduction to natural gas hydrate/clathrate: The

major organic carbon reserve of the Earth. Journal of Petroleum Science and Engineering. 56(1-3).

1-8.

Masui, A., Haneda, H., Ogata, Y., Aoki, K., 2005. The effect of saturation degree of methane hydrate on

the shear strength of synthetic methane hydrate sediments. In: Fifth International Conference on

Gas Hydrates, pp. 657-663, Tapir Acad., Trondheim,Norway.

Masui, A., Haneda, H., Ogata, Y., Aoki, K., 2006. Triaxial compression test on submarine sediment

containing methane hydrate in deep sea off the coast of Japan. In: the 41st Annual Conference,

Japanese Geotechnical Society, Kagoshima, Japan.

Masui, A., Haneda, H., Ogata, Y., Aoki, K., 2007. Mechanical properties of sandy sediment containing

marine gas hydrates in deep sea offshore Japan. In: The 7th (2007) International Offshore and

Polar Engineering Conference, Lisbon, Portugal.

Masui, A., Miyazaki, K., Haneda, H., Ogata, Y., and Aoki, K., 2008a. Mechanical properties of natural

gas hydrate bearing sediments retrieved from eastern Nankai trough. In: The 39th (2008) Offshore

Technology Conference, Huston, America.

20
Masui, A., Miyazaki, K., Haneda, H., Ogata, Y., Aoki, K., 2008b. Mechanical characteristics of natural

and artificial gas hydrate bearing sediments. In: The 6th International Conference on Gas Hydrates

(ICGH 2008), Vancouver, British Columbia, Canada.

Mekala, P., Busch, M., Mech, D., Patel, R. S., Sangwai, J. S., 2014. Effect of silica sand size on the

formation kinetics of CO2 hydrate in porous media in the presence of pure water and seawater

relevant for CO2 sequestration. Journal of Petroleum Science and Engineering 122, 1-9.

Miyazaki, K., Yamaguchi, T., Sakamoto, Y., Tenma, N., Ogata, Y., Aoki, K., 2010. Effect of confining

pressure on mechanical properties of sediment containing synthetic methane hydrate. Journal of

Mining and Metallurgical Institute of Japan 126, 408-417.

Nakazono, M., Jiang, Y., Tanabashi, Y., 2008. Study on the use possibility of carbon dioxide hydrate in

methane hydrate dissolution. In: The Fifth (2008) China-Japan Joint Seminar for the Graduate

Students in Civil Engineering, Shanghai, China.

Ning, F., Yu, Y., Kjelstrup, S., Vlugt, T. J., Glavatskiy, K., 2012. Mechanical properties of clathrate

hydrates: status and perspectives. Energy & Environmental Science 5(5), 6779-6795.

Ning, F., Glavatskiy, K., Ji, Z., Kjelstrup, S., Vlugt, T. H., 2015. Compressibility, thermal expansion

coefficient and heat capacity of CH4 and CO2 hydrate mixtures using molecular dynamics

simulations. Physical Chemistry Chemical Physics 17(4), 2869-2883.

Ohgaki, K., Takano, K., Moritoki, M., 1994. Exploitation of CH4 hydrates under the Nankai Trough in

combination with CO2 storage. Kagaku Kogaku Ronbunshu 20, 121-123.

Ohgaki, K., Takano, K., Sangawa, H., Matsubara, T., Nakano, S., 1996. Methane exploitation by

carbon dioxide from gas hydrates-phase equilibria for CO2-CH4 mixed hydrate system. Journal of

chemical engineering of Japan 29(3), 478-483.

Ordonez, C., Grozic, J. L., 2011. Strength and compressional wave velocity variation in carbon dioxide

hydrate bearing Ottawa sand. In: Pan-Am CGS Geotechnical Conference, Toronto, Ontario,

Canada.

Ors, O., Sinayuc, C., 2014. An experimental study on the CO2-CH4 swap process between gaseous CO2

and CH4 hydrate in porous media. Journal of Petroleum Science and Engineering 119, 156-162.

Ota, M., Abe, Y., Watanabe, M., Smith Jr, R. L., Inomata, H., 2005. Methane recovery from methane

hydrate using pressurised CO2. Fluid Phase Equilibria 228, 553-559.

21
Pahlavanzadeh, H., Kamran-Pirzaman, A., Mohammadi, A. H., 2012. Thermodynamic modeling of

pressure-temperature phase diagrams of binary clathrate hydrates of methane, carbon dioxide or

nitrogen+tetrahydrofuran, 1, 4-dioxane or acetone. Fluid Phase Equilibria 320, 32-37.

Priest, J. A., Clayton, C. R., Rees, E. V., 2014. Potential impact of gas hydrate and its dissociation on

the strength of host sediment in the Krishna-Godavari Basin. Marine and Petroleum Geology 58,

187-198.

Punmia, B. C., Jain, A. K., 2005. Soil mechanics and foundations. Firewall Media. pp. 50-52.

Santamarina, J. C., Dai, S., Terzariol, M., Jang, J., Waite, W. F., Winters, W. J., Nagao, J., Yoneda, J.,

Konno, Y., Fujii, T., Suzuki, K., 2015. Hydro-bio-geomechanical properties of hydrate-bearing

sediments from Nankai Trough. Marine and Petroleum Geology 66(2), 434-450. doi:

10.1016/j.marpetgeo.2015.02.033.

Sloan, E. D., 1997. Clathrate Hydrates of Natural Gases. Marcel Dekker, New York. pp. 1-2.

Song, Y., Yu, F., Li, Y., Liu, W., Zhao, J., 2010. Mechanical property of artificial methane hydrate under

triaxial compression. Journal of Natural Gas Chemistry 19(3), 246-250.

Ta, X. H., Yun T. S., Muhunthan B., and Kwon T.-H. (2015), Observations of pore-scale growth

patterns of carbon dioxide hydrate using X-ray computed microtomography, Geochem. Geophys.

Geosyst., 16, 912-924, doi:10.1002/2014GC005675.

Waite, W. F., Santamarina, J. C., Cortes, D. D., Dugan, B., Espinoza, D. N., Germaine, J., Jang, J., Jung,

J. W., Kneafsey, T. J., Shin, H., Soga, K., Winters, W. J., Yun, T. S., 2009. Physical properties of

hydrate-bearing sediments. Reviews of Geophysics 47(4), 465-484. doi: 10.1029/2008RG000279.

Waite, W. F., Winters, W. J., Mason, D. H., 2004. Methane hydrate formation in partially

water-saturated Ottawa sand. American Mineralogist 89(8-9), 1202-1207.

Winters, W. J., Waite, W. F., Mason, D. H., Gilbert, L. Y., Pecher, I. A., 2007. Methane gas hydrate

effect on sediment acoustic and strength properties. Journal of Petroleum Science and Engineering

56(1), 127-135.

Winters, W. J., Dallimore, S. R., Collett, T. S., Jenner, K. A., Katsube, J. T., Cranston, R. E., Wright, J.

F., Nixon, F. M., Uchida, T., 2000. Relation between Gas Hydrate and Physical Properties at the

Mallik2L-38 Research Well in the Mackenzie Delta. Annals of the New York Academy of

Sciences 912(1), 94-100.

22
Wu. J., Ning. F., Trinh. T. T., Kjelstrup. S., Vlugt. T. J. H., He. J., Skallerud. B. H., Zhang. Z, 2015.

Mechanical instability of monocrystalline and polycrystalline methane hydrates. Nature

communications 6, 8743. DOI: 10.1038/ncomms9743.

Yasin, N. H. M., Maeda, T., Hu, A., Yu, C. P., Wood, T. K., 2015. CO2 sequestration by methanogens in

activated sludge for methane production. Applied Energy 142, 426-434.

Yoneda, J., Hyodo, M., Yoshimoto, N., Nakata, Y., Kato, A., 2013a. Development of high-pressure

low-temperature plane strain testing apparatus for methane hydrate-bearing sand. Soils Found

53(5), 774-783. doi: 10.1016/j.sandf.2013.08.014.

Yoneda, J., Masui, A., Tenma, N., Nagao, J., 2013b. Triaxial testing system for pressure core analysis

using image processing technique. Rev. Sci. Instrum. 84, 114503, doi: 10.1063/1.4831799.

Yoneda, J., Masui, A., Konno, Y., Jin, Y., Egawa, K., Kidab, M., Ito, T., Nagao, J., Tenma, N., 2015a.

Mechanical behavior of hydrate-bearing pressure-core sediments visualized under triaxial

compression. Marine and Petroleum Geology 66(2), 451-459. doi:

10.1016/j.marpetgeo.2015.02.028.

Yoneda, J., Masui, A., Konno, Y., Jin, Y., Egawa, K., Kidab, M., Ito, T., Nagao, J., Tenma, N., 2015b.

Mechanical properties of hydrate-bearing turbidite reservoir in the first gas production test site of

the Eastern Nankai Trough. Marine and Petroleum Geology 66(2), 471-486. doi:

10.1016/j.marpetgeo.2015.02.029.

Yu, F., Song, Y., Liu, W., Li, Y., Lam, W., 2011. Analyses of stress strain behavior and constitutive

model of artificial methane hydrate. Journal of Petroleum Science and Engineering 77, 183-188.

Yun, T. S., Santamarina, J. C., Ruppel, C., 2007. Mechanical properties of sand, silt, and clay

containing tetrahydrofuran hydrate. Journal of Geophysical Research: Solid Earth (1978- 2012)

112(B4).

Zhang, X., Lu, X., Zhang, L., Wang, S., Li, Q., 2012. Experimental study on mechanical properties of

methane-hydrate-bearing sediments. Acta Methanica Sinica 28(5), 1356-1366.

23
Highlights:

A direct shear apparatus was developed for gas hydrate-bearing sediments.

Shear strength was measured under various testing conditions.

The shear strength, cohesion and internal friction angle were comparable with previous

studies.

24