ARMA 11-540 Stresses around a Production Well in Gas Hydrate-Bearing Formation

Sadegh Badakhshan Raz

Harold Vance Department of Petroleum Engineering, Texas A&M University, College Station, TX, USA Ahmad Ghassemi Harold Vance Department of Petroleum Engineering, Texas A&M University, College Station, TX, USA
Copyright 2011 ARMA, American Rock Mechanics Association
th This paper was prepared for presentation at the 45 US Rock Mechanics / Geomechanics Symposium held in San Francisco, CA, June 2629, 2011.

This paper was selected for presentation at the symposium by an ARMA Technical Program Committee based on a technical and critical review of the paper by a minimum of two technical reviewers. The material, as presented, does not necessarily reflect any position of ARMA, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of ARMA is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgement of where and by whom the paper was presented.

ABSTRACT: This paper presents an analytical solution for the temperature, pore pressure, strain and stress distributions around a wellbore in a gas hydrate reservoir during gas production by depressurization. The problem of decomposition of gas hydrate in porous rock is treated as a Stefan type moving boundary problem. The pressure distribution as well as temperature distribution is obtained with solution of gas diffusivity equations and conductive-convective heat transfer equations. These are then used in a displacement function in terms of the pore pressure and temperature distributions to find the resulting stress distributions in the rock. A number of simulations are presented to highlight the impact of production on the pore pressure and stress. The results of this semi-analytical study can be used to benchmark more complex numerical methods for the wellbore stability analysis in gas hydrate formations.

1. INTRODUCTION
Gas hydrates are metastable crystalline materials consisting of one or more type of gas molecules inside a molecular cage made out of hydrogen bonded water molecules. The gas hydrates form because of existence of hydrogen bonds between water molecules as well as Vander Waals forces between water and gas molecules [1]. The stability of gas hydrates compounds relates to parameters like temperature, pressure, gas composition and even external electrical and magnetic fields [2, 3]. Gas hydrate prospects on earth are the biggest unlocked recourses of energy in the world. The estimated amount of organic carbon in the form of gas hydrate in earth is 10000 giga tones which is equal to 105 to 3109 Tcf of gas reserves [4, 5]. The significance of gas hydrate resources will be recognized when we consider that the total amount of non hydrate gas reservoirs in the world is just 13000 Tcf [5]. A proposed method for gas production from gas hydrate reservoirs is depressurization. For production from gas hydrate reservoirs by depressurization, a production model that could predict the behavior of gas hydrate reservoir during production is absolutely essential. The knowledge of stress and strain distribution around the wellbore during production is needed to assess the problems like wellbore stability and potential for sand production.

Different researchers addressed the problem of stress and strain distribution around wellbore with different approaches. Freij-Ayoub et al [6] used FLAC to numerically calculate stress and strain distribution around the wellbore induced drilling through gas hydrate bearing strata. Rutqvist el [8] used TOUGH+HYDRATE to numerically simulate pressure and temperature distribution around the wellbore induced by different thermal and mechanical conditions during gas hydrate dissociation. Then, FLAC3D was used to calculate stress distribution around the wellbore. Kimoto et al. [8], on the other hand, treated hydrate bearing reservoir as a chemo thermomechanical material and used an elasto viscoplastic model to address plastic deformations in the soil during gas hydrate production. However, the selection of appropriate model to simulate the condition of stress and strain around the wellbore is greatly affected by the geology of reservoir as well as the condition of production from the reservoir. Waite et al pointed out that gas hydrate accumulations in coarse grain sands are more prone to plastic deformation and sand production during production period than fine grain hydrate sediments [9]. The selection between a poroelastic or poroplastic model is related to the gas production rate and whether the gas hydrates bear loads in the reservoir or merely fill the voids in the pore space. In this paper, it is assumed that the reservoir remains elastic during gas production period. In addition, the intrinsic

permeability of rock remains the same despite the change in gas effective permeability in non decomposed and decomposed zones. It is assumed that the fluid flow is single phase flow i.e. gas flow, and water remains stagnant in the reservoir [3]. A poroelstic model is used as a semi-analytical method to calculate the induced total stress in gas hydrate reservoir during gas production.

(1) Where n is equals 1 for decomposed gas hydrate layer zone and is 2 for non decomposed zone. Also P is the pore pressure and k is permeability to gas, is gas viscosity, s is water saturation, is hydrate saturation, phi is porosity in both decomposed and non-decomposed layer. In addition, (1-s)m m2 = (1-) m The above equation is non linear and can be linearized with respect to P in order to be solved analytically. For the linearization of gas diffusivity equation, we consider following approximations [3]. (2) (3) Where Pe is the reservoir initial pressure in MPa and PD, MPa, is equilibrium pressure at the interface between decomposed and non decomposed layers. After linearization, we will obtain the following equations: (4)

2. MATHEMATICAL MODELS
For calculation of induced stress and strain in gas hydrate reservoir during production, we need to know the pressure and temperature distribution in the reservoir during decomposition of hydrate layer. The area around a wellbore can be viewed as two parts, one corresponding to the decomposed region and the other the non-decomposed gas hydrate layers. The schematic of these layers is shown in figure 1.

Fig. 1. Different zones during gas production, rw is wellbore radius, R is the radius of gas hydrate decomposing front and re is reservoir radius

Where: (5)

During gas production, the diameter of decomposed layer, R, will grow with time. The movement of the boundary between the two areas introduces a physical problem with free moving boundary condition which is called Stefan problem named after Joef Stefan, the Slovene physicist who studied the problems of ice formation around 1890[10]. Stafans problems include at least two differential equations with their own boundary conditions which are related together through Stefans condition. The mathematical model of gas hydrate production as well as the resulted strain and stress field around the wellbore are discussed in following sections. In the following section, we elaborate on the method introduced by Makogon et al [3] to calculate temperature and pressure distribution during gas production from gas hydrate bearing reservoir.

(6) The boundary conditions for diffusivity equations are: P(rw , t) = PW P2(r, 0) = P2 (, t) =Pe P1(R(t), t) = P2 (R(t), t) =PD (7) (8) (9)

Where rw is wellbore radius in meter, PW is wellbore pressure, R(t) is the radius of the interface between the decomposed and un-decomposed layer. The solutions of linearized gas diffusivity equation are [3, 11]: (10) (11) The functions and coefficients in equations 10 and 11 are [3]:

2.1. PRESSURE DISTRIBUSION

The governing equation for pressure distribution around the wellbore during gas hydrate production is gas diffusivity equation in polar coordinates [3, 11]:

(12) (13) (14) (15) (16) Where in equations 9 to 13 we have: Q: production rate of methane gas per unit length of well 0: density of methane gas at atmospheric pressure P0 and temperature T0. (0.706 kg/m3) P0 : atmospheric pressure (0.101 MPa) h: is unit length of pay zone( 1 m) : constant which determines movement velocity of dissociation front (m2/s) Te :reservoir temperature at initial time (K)

and obtain: (18) In addition to conditions mentioned in equations (7)(9), the following boundary conditions for Equation (18): T1(r, 0) = T2(r, 0) = T2 (, t) = Te T1(R (t), t) = T2 (R (t), t) =TD (19) (20)

The solutions of above conductive-convective heat transfer equations are [3]:

(21)

(22) With (23) (24) (25) (26) (27)

2.2. TEMPERATURE DISTRIBUSION

To consider the effect of temperature changes on induced stress in a gas hydrate reservoir, heat transfer equation need to be considered. The governing equation for heat transfer in the reservoir is considered to be conductive-convective heat transfer equation which could be written as [3, 11]:

(17) Where in this equation: n=1 and 2 for decomposed and non decomposed layer respectfully T1: Temperature in decomposed layer (K) T2: Temperature in non-decomposed layer (K) an: thermal conductivity of zones 1 and 2 (m2/s) cv: volume heat capacity of gas (J/K.kg) c1: heat capacity of zone 1 (J/K.kg) c2: heat capacity of zone 2 (J/K.kg) : adiabatic coefficient of gas (K/Pa) : throttling coefficient of gas (K/Pa)

(28) In the pressure and temperature solutions, the temperature at the gas hydrate decomposition front, TD, the pressure at the interface, PD, and the boundary velocity constant, , are unknown. They should be solved in order to find the pressure and temperature solutions around the wellbore. Noting that the TD satisfies Equations (21) and (22), we replace T2 by TD in Equations (22) and obtain:

Note that the Joule-Thompson cooling process is considered in the equations. In order to solve the problem analytically, we assume conduction heat transfer is negligible in gas hydrate reservoir in comparison to convection heat transfer. So we let: an =0

(29) Furthermore, if TD is substituted into Equation 21, we get TD = TD which is not useful. Similarly, if PD is put instead P1 and P2 in Equations 10 and 11, it results in PD= PD. Therefore, we need to use the Stans condition to have a second equation to solve for TD, PD and . If mass and energy conservation are considered at the interface of decomposed and intact zones, we get following equation [3]:

Where (30) Where z: compressibility of gas 3: density of hydrate As it is mentioned, R is the radius of decomposing front in meter which could be calculated according to following equation. (31) Where, t is time in second. Now, if equations 10 and 11 are plugged into equation 30, after derivations and simplification, we will get following equation: is poroelastic stress coefficient and is equal to: (36) : Biots coefficient : Poissons ratio G: Shear modulus, MPa K: Bulk modulus, MPa : The thermal expansion coefficient of rock, K-1 Also P and T are changes in pressure and temperatures from their initial state and are given by: P = Pn-Pi T = Tn -Ti (32) The third equation is an imperial correlation between equilibrium temperature and pressure of methane hydrate formation is [3, 11]. (33) Where a = 0.0342 K-1 b = 0.0005 K-2 c = 6.4804 For finding the unknowns, we need to solve the system of nonlinear equations of 29, 32 and 33. After finding TD, PD and , we could plug them to the equations 10 to 16 and 21 and 22 to have pressure and temperature solutions around the wellbore. And the tangential stress is: (37) (37) (38)

The strains can then be used in the constitutive equations to calculate the stresses. As it is shown in figure 3, the maximum temperature change during production is 4.3 K. The amount of induced stress in reservoir due to this temperature change is small in comparison with induced stress by pressure change and is neglected. In this way, the assumption of formation homogeneity can be maintained. Ignoring the temperature terms, equations 36 and 37 are used to find the stresses:

(36)

2.3. INDUCED STRESS DISTRIBUSION

On the basis of linear coupled thermo-poroelasticity theory, the strain distribution around the wellbore caused by changes in pore pressure and temperature can be calculated by the following expression (e.g., Tao and Ghassemi [12, 13])

The sign convention is tension positive. Now, by substituting the pressure functions into equations 37 and 38 the induced stress around the wellbore can be calculated. Since there is no analytical way to calculate the integral, they need to be calculated numerically.

P+ (34) Also the tangential strain is given (35)

3. RESULTS
In this section the results for the calculation of pressure and temperature distribution as well as related induced stress and strain in the gas hydrate reservoir are

presented. The main assumption in this study is that the gas production rate is constant. Other assumption is that mechanical behavior of reservoir remains in elastic region. For coefficients and mentioned parameters in the paper, the hypothetical case of gas hydrate reservoir is considered with the parameters same as those in Ji et al. [11] and Freji-Ayoub [9]. These parameters are mentioned in table 1.
Table1. The amount of parameters used in this study Parameter : Biots coefficient : The thermal expansion coefficient of rock, K-1 : hydrate saturation, % : throttling coefficient of gas (K/Pa) : adiabatic coefficient of gas (K/Pa) : gas viscosity, Pa.s : Poissons ratio : water saturation, % c1: heat capacity of zone 1 (J/K.kg) c2: heat capacity of zone 2 (J/K.kg) cv: volume heat capacity of gas (J/K.kg) 0: density of methane gas at P0 and T0, kg/m3 3:density of hydrate, kg/m3 w: density of water, kg/m3 G: Shear modulus, MPa k1: gas permeability in zone 1, md K2: gas permeability in zone 2, md K: Bulk modulus, MPa P0 : atmospheric pressure, MPa T0 : atmospheric temperature, m: porosity, % z: compressibility of gas Amount 0.79 1.2 10-5 0.15 8 10-7 3.2 10-6 1.5 10-5 0.2 0.15 2400.2 1030.2 3000 0.706 0.91 103 1 103 6000 5.2 0.4 8000 0.101 273.15 0.19 0.88 Ref. 8 9 3 3 3 3 8 3 3 3 3 3

calculations, the initial reservoir pressure, Pi, and initial reservoir temperature, Ti, are considered to be 15 MPa and 287 K, respectively. Also, the permeability in decomposed layer is k1= 5.210-15 m2 (5.2 md) and in the non decomposed hydrate layer is k2= 0.410-15 m2 (0.4 md). Other gas properties like viscosity and permeability are assumed to be constant. This seems to be reasonable assumption for low temperature and fairly low pressure gas hydrate reservoirs. A Mathematica code was written to do all the necessary calculation. 3.1 The EFFECT OF PRODUCTION TIME Consider the constant gas production rate of 0.04 kg/second, production times are considered to be 120, 365, 100 and 730, 1460 and 2920 days (0.33, 1, 2, 4 and 8 years, respectively). The effect of production time on pressure and temperature around the wellbore are shown in Figures 2 and 3. As it is shown in Figure 2, there are two different zones of pressure in the gas hydrate on either sides of gas hydrate decomposing front. The position of the decomposition front coincides with a bump in pressure and temperature graphs. The bump is caused by considerable permeability difference on either sides of the front. The lower permeability in the nondecomposed gas hydrate layer corresponds to less pressure drop at constant production rate. In contrast, the considerable higher permeability in the decomposed layer caused sharper pressure drop.

284.5 284 283.5

Temperature, K

3 3 8 3 3 8 3 3 3 3

283 282.5 282

Time increase

281.5 281 280.5 0 5 10 15 20 25 30 35

Radius, m

Fig. 2. The effect of production time on pressure distribution around the wellbore Figure 3 shows the temperature distribution in the reservoir in different production times. Like pressure graphs in figure 2, there is a temperature bump at the location of decomposition front because of this fact that two different partial differential equations stated at equation 18 represents the physics of problem at either sides of the front. In the decomposed layer near the wellbore, there is more pressure drops and therefore

In this study, the effect of production time and gas production rate on pressure, temperature, strain and stress distribution in reservoir were investigated. The wellbore radius, rw, is assumed to be 0.13 m. For all the

Tangential tensile stress, MPa

higher velocity which causes higher temperature decrease due to higher convective heat transfer and stronger cooling Joule-Thompson effect.

2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0 50 100 150 r/rw 200 250 Time increase

13 12.5 12

Pressure, MPa

11.5 11 10.5 10 9.5 9 8.5 8 0 5 10 15 20 25 30 35

Time increase

Fig.4. The effect of production time on tangential strain around the wellbore field,

Radius, m

Fig.3. The effect of production time on temperature distribution around the wellbore Also like the effect of production time on pressure distribution, the overall temperature in the reservoir decreases with increase in production time.
0.9 0.8 Radial tensile stress, MPa 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 50 100 r/rw 150 200 250 Time increase

3.2 THE EFFECT OF PRODUCTION RATE

In constant production time of 365 days, the effect of gas production rate on induced strain and stress distribution in the reservoir during production period is studied. Different production rates of 0.04, 0.06, 0.08, 0.1 and 0.12 Kg of gas per second are considered. Figure 5, shows the effect of production rate on induced radial stress in the reservoir. The type of radial strain is tensional strain. As it is shown in figure 5, increase in production rate causes the increase of induced radial stress due to increase in induced pressure and temperature in reservoir. The effect of increase in production rate on induced radial stress is shown in figure 5. The type stress here is compressional stress. The induced radial stress is zero in the wellbore and then increases to a maximum near the wellbore. The induced radial stress vanishes with increase in radius.

Radial tensile stress, MPa

Fig.3. The effect of production time on normal strain around the wellbore field, The effect of production time on radial stress is shown in Figure 3. The type of stress here is compressional stress. As it is shown in the Figure 3, the radial stress is zero at the wellbore radius and then increases to a maximum near the wellbore. The induced radial compressive stress increases with increase in time due to increase in induced pressure and temperature. Also, the effect of production time on tangential stress is shown in Figure 4. The type of induced tangential stress here is compressive stress. The maximum of induced tangential stress happens in the wellbore wall and then decreases with increase in radius. Like induced radial stress, tangential stress increase with increase in production time.

1.2 1

Rate increase

0.8

0.6 0.4

0.2 0 0 5 10 15 20 r/rw 25 30 35 40

Fig.5. The effect of gas production rate on radial stress field, around the wellbore Finally as it is shown in figure 6, the increase in production rate causes increase in induced tangential stress in the reservoir.

3.5 3 Radial tensile stress, MPa 2.5 2 Rate increase

2.

Makogon Yuri F., Stephen A. Holditch, Gas solubility in water, kinetics and morphology of gas hydrate, Annual report for Arco Exploration and production technology, Texas A&M University,1998 Makogon Yuri F., Hydrates 1997,Pennwell books, USA of Hydrocarbons,

3. 4. 5.
0 2 4 6 8 10 r/rw 12 14 16 18 20

1.5 1

http://oceanexplorer.noaa.gov/explorations/deepeast01/ background/fire/media/carb_dist.html Editorial, An introduction to natural gas hydrate/clathrate: The major organic carbon reserve on earth, Journal of Petroleum Science and Engineering, Vol. 56, PP 1-8, 2007

0.5

Fig.6. The effect of gas production rate on tangential strain field, around the wellbore As it is observed from the radial and tangential stress results, the highest amount of stresses are induced around the wellbore. The tangential stresses are maximum in the wellbore periphery and radial stress reaches the maximum value near the wellbore because there is the maximum pressure drop around the wellbore. Consequently, the area near the wellbore is the most prone zone in the reservoir to meet the failure criteria during gas production.

6.

Reem Freij-Ayoub, Chee Tan, Ben Clennell, Bahman Tohidi, Jinhai Yang, A wellbore stability model for hydrate bearing sediments, Journal of Petroleum Science and Engineering 57 (2007) 209220 Jonny Rutqvist and George J. Moridis, Numerical Studies on the Geomechanical Stability of Hydrate-Bearing Sediments, Offshore Technology Conference, 2007, Houston, U.S.A. Sayuri Kimoto, Fusao Oka, Tomohiko Fushita, A chemothermomechanically coupled analysis of ground deformation induced by gas hydrate dissociation, International Journal of Mechanical Sciences, 52,2010, 365376 W. F. Waite, J. C. Santamarina, D. D. Cortes, B. Dugan, D. N. Espinoza, J. Germaine, J. Jang, J. W. Jung, T. J. Kneafsey, H. Shin, K. Soga, W. J. Winters, T.-S. Yun, Physical properties of hydrate-bearing sediments, Reviews of Geophysics, Vol. 47,2009

7.

8.

CONCLUSION
1. The pressure and temperature distribution in gas hydrate bearing reservoir during gas production are calculated analytically 2. The results of pressure and temperature distribution are coupled to a linear thermoporoelstic model to calculate induced strain and stress distribution in the reservoir during gas production. 3. Increase in production time causes an increase in induced stress and strain in the reservoir 4. Increase in production rate causes an increase in induced stress and strain in the reservoir 5. The amount of tangential strain and radial stress is zero in the wellbore periphery which increases to a maximum near the wellbore and again decreases with increase in radius. 6. The amount of radial strain and tangential stress is maximum in wellbore and decrease with increase in radius.

9.

10. L.I. Rubinstein, The Stefan Problem, American Mathematical Society,1971

11. Chaung Ji, Goodarz Ahmadi, Duane Smith,

Constant rate natural gas production from a well in a hydrate reservoir, Energy Conversion & Management, Vol. 44, 2003, PP2403-2423
12. A. Ghassemi, Q. Tao, A. Diek, Influence of coupled chemo-poro-thermoelastic processes on pore pressure and stress distributions around a wellbore in swelling shale, Journal of Petroleum Science and Engineering 67 (2009) 5764 13. Qingfeng Tao, Wellbore Stability in water sensitive shales, Master of Science Thesis, University of North Dakota, 2006 14. Detournay, E. and Cheng, A.H.-D., Fundamentals of poroelasticity,Chapter 5 in Comprehensive Rock Engineering: Principles, Practice and Projects, Vol. II, Analysis and Design Method, ed. C. Fairhurst, Pergamon Press, pp. 113-171, 1993

REFERENCES
1. Sloan E. D., C.A. Koh, Clathrate Hydrates of Natural Gas, Third Edition, CRC Press, 2009