Modeling Low-Salinity Waterflooding in Chalk and ... - ACS Publications

Subscriber access provided by NEW YORK UNIV

Article

Modeling Low Salinity Waterflooding in Chalk and Limestone Reservoirs Changhe Qiao, Russell T. Johns, and Li Li Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.5b02456 • Publication Date (Web): 12 Jan 2016 Downloaded from http://pubs.acs.org on January 16, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1 2

Modeling Low Salinity Waterflooding in Chalk and

3

Limestone Reservoirs

4

Changhe Qiao, Russell Johns, Li Li

5

Pennsylvania State University

6

ABSTRACT

7

Injection of low salinity brines can improve oil recovery (IOR) in carbonate reservoirs by

8

changing the rock wettability from more oil-wet to more water-wet. Existing models use an

9

empirical dependence of wettability on variables including equivalent salinity and ionic strength.

10

We recently developed a process-based model that mechanistically includes the geochemical

11

interactions between crude oil, brine, and the chalk surface that alter rock wettability. Here in

12

this research, we extend the previous model by including mineral dissolution reactions, therefore

13

enabling modeling low salinity flooding in chalks and limestone cores with and without

14

anhydrite, a key factor considered to control the extent of improved oil recovery.

15

We examine the role of mineralogy by including surface complexation, aqueous reactions, and

16

dissolution/precipitation of calcite and anhydrite in the extended model.

17

coupled with the equations of multiphase flow and transport, are solved simultaneously using an

18

in-house simulator, PennSim. Relative permeability functions and residual oil saturation during

19

flooding are adjusted dynamically according to the concentration of oil acids attached to the

These reactions,

ACS Paragon Plus Environment

1

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 41

1

mineral surface. Core flooding experiments from the Stevns Klint (SK) chalk, a limestone with

2

small amount of anhydrite, and a Middle Eastern carbonate with a small volume fraction

3

anhydrite are used to tune the reaction network and predict recovery. Simulation results agree

4

with the effluent concentrations of SO42-, Ca2+ and Mg2+ reported from chromatographic

5

wettability tests and the recoveries under differing compositions in chalk and limestone cores.

6

For the SK chalk without anhydrite, lower Na+ and Cl- concentrations under constant SO42-

7

conditions leads to improved oil recovery (IOR) by as much as 6% OOIP. Lower salinity alone,

8

however, does not lead to IOR in limestones without anhydrite. Instead, anhydrite dissolution

9

provides a natural source of sulfate and increase oil recovery by 5% when injecting diluted

10

formation water. Simulations of 2D five-spot patterns using tuned reaction networks

11

demonstrated that IORs from 5% to 20% OOIP can be obtained after two pore volumes are

12

injected. These IORs depend greatly on the aqueous chemistry of the injected fluid, and sweep.

13

The results highlight the critical importance of understanding the mineralogy and including a

14

mechanistic reaction model in the simulation of low salinity water floods.

15

INTRODUCTION

16

Low salinity water (LSW) is reported in the literature to improve oil recovery (IOR)

17

significantly in sandstone and carbonate core experiments.1-4 Here we stick to the widely-used

18

term `low salinity flooding,’ although both experimental data from Austad et al.1, 5-7 and our

19

model8 suggest that the IOR is not solely caused by low salinity (total dissolved salt). Instead, it

20

is the specific composition of the injection water, in particular the concentrations of the

21

wettability-altering species (SO42-, Ca2+, and Mg2+), that matters. The recoveries in core floods

22

and imbibition tests vary greatly depending on the aqueous chemistry used, the pore volumes

23

injected (PVI), and the carbonate mineralogy. 6, 9


2

Page 3 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1

Improved recoveries were around 5% to 40% OOIP for spontaneous imbibition tests in the

2

Stevns Klint chalk and 5% to 20% OOIP in core flooding experiments in both the Stevns Klint

3

chalk and a Middle Eastern limestone.6, 10, 11 Diluted formation water without sulfate was found

4

to improve oil recovery by 5% OOIP.5 Diluted seawater was reported to improve oil recovery by

5

17% to 19% OOIP in a limestone core containing anhydrite.10 A single well tracer test at field

6

scale also exhibited decreased residual oil saturation by seven saturation unit from injection of

7

diluted seawater. 12

8 9

One of the primary mechanisms proposed for IOR by LSW flooding is wettability alteration from a more oil-wet to a more water-wet state.1,

11

Chromatographic wettability tests with

10

chemically tuned water confirmed wettability alteration by incremental sulfate adsorption onto

11

the rock surface.6, 13 Wettability alteration was also confirmed in contact angle measurements on

12

a flat plane carbonate surface, where smaller contact angles were reported with modified

13

seawater.10 Further, capillary pressure reduction was observed in spontaneous imbibition tests

14

when sulfate-containing seawater was injected compared to sulfate-free water.14

15

relative permeability and decreases in residual oil saturation were demonstrated in coreflooding

16

experiments.11

17

Shifts in

Wettability alteration could be caused by mineral dissolution, surface adsorption/desorption,

18

and surface potential reduction.7, 15, 16 Austad, et al.

1

19

outcrop, the concentrations of potential determining ions (PDIs) such as Mg2+, Ca2+ and SO42-

20

greatly affected the oil recovery in spontaneous imbibition tests. They explained that for mixed

21

wet surfaces, both SO42- and carboxylic groups in the oil adsorb onto the rock surface, altering

22

the surface charge from positive to negative. The negatively-charged solid surface then attracts

23

cations such as Mg2+ and Ca2+ to react with the adsorbed carboxylic group, which releases the

proposed that for the Stevns Klint chalk


3

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 41

1

carboxylic group and causes the surface to become more water-wet. The notion that carboxylic

2

groups leave the chalk surface was inferred from water vapor adsorption isotherms, where SO42-

3

reduced the surface concentrations of adsorbed fatty acids.17 Spontaneous imbibition

4

experiments with varied brine compositions support such chemical mechanisms.6,

5

addition to SO42-, BO32- and PO43- were also found to improve oil recovery.18

7, 9, 13

In

6

Experimental studies have revealed that the brine composition sufficient for enhanced oil

7

recovery in Stevns Klint chalk was not necessarily effective for other carbonate minerals.

8

Instead, diluted formation water and seawater were found to improve oil recovery the most for

9

limestone and dolomites.10, 19 Romanuka, et al.

19

performed spontaneous imbibition tests for

10

Stevns Klint chalk, limestone and dolomite cores and concluded that effective EOR fluids

11

depends significantly on the rock mineralogy. Screening Amott imbibition experiments showed

12

that additional SO42- did not lead to IOR for limestone and dolomite, as well as for the Stevns

13

Klint chalk with diluted formation water. One possible reason for an increase in oil recovery for

14

sulfate-free injection fluids in limestone is the increased sulfate concentration from dissolution of

15

anhydrite in situ.5

16

incrementally in core flooding experiments for limestone.20

In addition, SO42- and Mg2+ were reported to improve oil recovery

17

The above-mentioned experimental evidences have shown significant promise of LSW for

18

IOR however at the same time indicated the complexity of geochemical reactions involved in

19

low salinity flooding in different types of carbonate formations. Currently there is no theory to

20

quantify and explain many of these differences. Various modeling efforts have tried to capture

21

the wettability alteration process. Wettability alteration has been modeled by interpolating

22

capillary pressure, relative permeability and residual oil saturation between two wetting states21,

23

22

using empirical relationships between wettability and water chemistry parameters including


4

Page 5 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1

salinity, ionic strength, or total dissolved salts.21, 23 Contact angles were also used to represent

2

surface wettability.23

3

conditions, they do not take into account complex surface reactions mechanistically and

4

therefore cannot be used to predict wettability alteration under conditions that differ from where

5

the empirical relationships are derived.24 Thus, a mechanistic model is needed to understand the

6

dominant processes and to accurately predicts wettability under varying mineralogical

7

conditions.

8

Qiao, et al.

8

Although these relationships are valid under the given experimental

developed a mechanistic model coupling multiphase flow and transport using a

9

detailed surface and aqueous multi-component reaction network including adsorption/desorption

10

that captures the competitive interactions among oil, brine, and the surface of the Stevns Klint

11

chalk. Natural carbonate reservoirs, however, often contain minerals including anhydrite,

12

dolomite, and quartz. These minerals could potentially affect surface reactions among rock, oil,

13

and water, and therefore influence the oil recovery. For example, Austad, et al.

14

anhydrite can play an important role in wettability alteration.

15

5

reported that

In this paper, we extend the multiphase reactive transport model by Qiao, et al.

8

to include

16

limestone surface complexation and mineral dissolution/precipitation reactions. This extends the

17

model in our in-house simulator PennSim25 as an attempt to provide a unified model for low

18

salinity flooding in mineralogically different carbonate reservoirs. The model was calibrated

19

using core experimental data from the Stevns Klint chalk and limestones with anhydrite. We

20

first introduce the multiphase reactive transport equations, and then describe the wettability

21

alteration model and the mechanisms of altered wettability and oil recovery in different cores.

22

Last, we use the new model to demonstrate the potential of low salinity water flooding at the

23

field pattern scale.


5

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 41

1 2

METHODOLOGY

3

We consider isothermal recovery processes with immiscible oil/water flow. Geochemical

4

reactions occur in the aqueous phase and at the crude oil-brine-mineral interfaces. Desorption of

5

the carboxylic group from the rock surface leads to more water-wet conditions, which further

6

changes the relative permeability and residual oil saturation based on measured curves at two

7

wettability endpoints. The major extension from our previous model8 includes the addition of

8

calcite and anhydrite dissolution/precipitation reactions, surface complexation with the surface

9

site > CO , and additional aqueous complexation such as CaCl and NaSO4 . In this section we

10

describe the governing equations that couple the flow, transport and reaction system, and the

11

method to setup a simulation.

12

Governing Equations.

13

The equation set describes the mass conservation of oil/water and chemical species, Darcy flow,

14

and reaction kinetics and equilibrium. The physical meanings and units of the symbols are in the

15

nomenclature.

16 17

18

Immiscible oil/water flow equations. The mass conservation equations for immiscible oil and water phases are as follows: + ∇ ⋅ = 0,

= , .

(1)

Darcy’s law governs the flow rate of each phase, =

⋅ ∇ − !" ,

= , .

(2)

19

The subscript “” is for the water phase and subscript “” for the oil phase. Capillary pressure

20

relates the pressure of oil and water phases, $%& = % − & .

(3)


6

Page 7 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1

Energy & Fuels

The saturation relation completes the set of equations

% + & = 1.

2 3

(4)

The primary unknowns for the multiphase flow system are % and & .

Reactive transport equations. The mass conservation equation for the primary species ( is /012

/012

.34

.34

)*+ + , -.+ *. 5 + ∇ ⋅ )6+ + , -.+ 6. 5 = 0,

( = 1, … , 8+

(5)

4

where the first term is the accumulation of total moles and the second term is the total molar flux

5

of a primary component p, the chemical building blocks of the reaction systems. The equations

6

are based on the stoichiometric relationship among the species participating in reactions and

7

reaction kinetics. The derivation of the reactive transport equations are given in Qiao, et al. 8.

8

More details of reactive transport equations can also be found in Yeh and Tripathi 26, Lichtner 29,

9

Steefel and Lasaga

27

and Walter, et al.

28

. These equations solve for the concentrations of

10

primary species in the aqueous phase and at the phase interfaces. Secondary species are solved

11

using laws of mass action based on the concentrations of primary species. Reactive transport

12

models have been widely used in subsurface biogeochemistry in the past decades

13

have been relatively new in applications of interests to the oil and gas industry 25, 33-35.

14

29-32

however

Multiphase Reaction Network.

15

Reactions are either kinetically controlled or at equilibrium depending on the typical rates of the

16

specific geochemical reactions. According to the geochemical convention, the reactions in the

17

aqueous phase and surface complexation on the rock and oil surfaces typically occur fast, and are

18

assumed to be at equilibrium 36. In contrast, mineral dissolution and precipitation are relatively

19

slow and modeled with reaction kinetics 37. The reaction network along with the thermodynamic

20

and kinetic constants at 25°C and 110°C are listed in Table 1. The primary species include

21

=A aqueous ions HCO , Cl , SO= , Ca=A , NaA , K A , H A and surface species > < , SCN , Mg


7

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 41

1

CaOH=A , > CO and −COO , where “>” and “−” represent the species at the solid surface and

2

oil surfaces, respectively. The extension from Qiao, et al.

3

(C4-C6), aqueous reactions (A4-A8) and mineral dissolution/precipitation reactions (M1, M2).

4

The equilibrium constants for aqueous and mineral dissolution reactions are from the EQ3/6

5

thermodynamic database.38

6 7 8

8

includes calcite surface reactions

Aqueous complexation reactions in water. The aqueous species includes free ions HCO , Cl ,

= =A =A A A A SO= < , SCN , Mg , Ca , Na , K , H , CO , OH , H= CO and complex ions MgSO< , NaSO< ,

CaSO< , CaClA , MgClA . These aqueous complexes were chosen based on their dominance in the

9

aqueous phase.25 The common practice in geochemistry is to assume all aqueous reactions are

10

fast and at equilibrium with equilibrium constants described by laws of mass action. As an

11

example, the equilibrium constant of reaction (A4) is, DE.,F< =

12

13 14

GHIJK GLMNJO GHILMN

where the activities of the aqueous species (GP ) are calculated by GP = QP RP .

Activity coefficients were determined using an extended Debye-Hückel model.39

High NaA and Cl concentrations reduce surface adsorption in two ways.

First, higher

15

concentrations lead to greater ionic strength and smaller activity coefficients of aqueous species

16

and therefore smaller surface adsorption. Second, Na+ and Cl- form aqueous complexes with

17

=A =A SO= through reactions (A5), (A7), and (A8), which reduces their corresponding < , Mg and Ca

18

free ion concentration available for surface adsorption. Therefore, reducing Na+ and Cl-

19

concentrations can promote surface adsorption, which causes wettability alteration.


8

Page 9 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1

Mineral surface reactions. We considered the surface dissolution/precipitation of calcite and

2

anhydrite (M1 and M2 in Table 1). The mineral reaction rates are written based on the transition

3

state theory (TST) rate laws. 40 For example, calcite dissolution rate is expressed as follows: STUTVWX = YTUTVW X Z K exp ^−

GTUJK GcTVOW _` b GcK d1 − e . Sa DE. GZ K

4

The last term in the parenthesis in the above equation indicates how far away the reaction is from

5

equilibrium. If the reaction is at equilibrium, the rate is zero. The Damkholer number for this

6

reaction is calculated as f`,g`gMWh =

Sg`gMW h , iRg`JK ,j%j`k

7

where i is the flow velocity and CTUJK ,lmlUn is the total Ca=A concentration. The Damknoler

8

number determines if the reaction is modeled using an equilibrium or kinetic rate law. Calcite

9

dissolution provides a source for Ca=A and buffers pH. The calculated Damkholer number for

10

anhydrite dissolution is over 100 in the experimental cases studied later in this paper, indicating

11

much faster dissolution rates than transport. The anhydrite dissolution reaction is therefore

12

modeled at equilibrium. Anhydrite dissolution is an important source or sink for sulfate and

13

calcium.

14

In addition to the mineral dissolution reactions, we also consider the surface

15

adsorption/desorption reactions on calcite surface. Surface complexation models are typically

16 17

used to describe the surface reactions.41-44 A calcite surface contains sites > CaOH and > CO H

that can adsorb aqueous ions and carboxylic groups and form surface species > CaOH=A ,

18

A A 45 > CaSO In this research two models are used < , > CaCO , > CO , > CO Ca and > CO Mg .

19

to describe the surface reactions, the electrical diffusive layer (EDL) and the non-electrical

20

diffusive layer (NEDL) model, the latter of which is also called constant capacitance model.44, 46


9

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 41

1

We used the relatively simple-to-implement NEDL model when the change of surface potential

2

does not have a significant effect. The EDL model is used when there are significant surface

3

potential changes owing to surface adsorption. Thus, the EDL model also affects the surface

4

adsorption equilibrium. In the EDL model, the activity of an aqueous species is corrected

5

through a Boltzmann factor so the equilibrium constant for reaction (C2) is expressed as follows, Dg= =

6 7

26qh r s> RG t RGtv=A uGLMJO N

exp o−

2 A Here s> CaSO < u and s> CaOH= u are the surface concentrations in mol/m . The surface potential

qh is calculated from the Gouy-Chapman theory,

wh = −x8z{ z| }Sa sinh ^−

6qh b. 2Sa

8

In the NEDL model, Ψh is considered constant. The equilibrium constant for reaction (C2) can

9

be written then as Dg= =

s> RG t RGtv=A uGLMJO N

10

while the constant surface potential term is lumped into KC2. A sensitivity analysis in Qiao, et al.

11

47

12 13 14 15 16 17

demonstrated the strong dependency of the oil recovery on the equilibrium constants. For

aqueous reactions, the DE. values are well documented for a large range of temperature. For surface complexation, the DE. values are empirical parameters related to intrinsic thermodynamic

constants.46 The solid surface site densities for >CaOH and >CO H are fixed at 4.1 mol/m2

based on a previous simulation study.41

Oil/water interface reactions. The active species on the oil-water interface include −COOH,

−COO , −COOCaA and −COOMg A . The −NH< oil surface sites are not included here because

18

– COO dominates the interaction between oil and water surfaces.41 Oil surface complexation

19

reactions are described by either NEDL or EDL in a similar way as mineral surface reactions.


10

Page 11 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1

Temperature effects for reactions on the oil surface are considered as not significant48. The

2

surface oil site total concentrations are fixed at 6 µmol/m2.49

3 4

Interfacial interactions and wettability alteration. The calcite-water-oil interface reaction is

represented by reaction (CO1). The species > CaOH=A −COO is the adsorbed carboxylic

5

group on the calcite surface. When the reaction goes from left to right, > CaOH=A −COO

6

desorbs and the system becomes more water wet as is in low salinity waterflooding. If the

7

reaction goes from right to left, the system becomes more oil wet as occurs during the aging of

8

crude oil. This reaction is reversible over the time scale of coreflooding as shown in a recent

9

experimental paper by Hopkins, et al. 50 at the temperature of 50 °C. Reaction (CO1) occurs fast

10

at high temperature while much slower at room temperature1,

11

reversible at high temperature (110°C). For simulation of the chromatographic wettability test at

12

25°C, we assumed irreversibility during the duration of the experiments (around 3 hours). The

13

equilibrium constant of this reaction is not known in the literature so we used this parameter as a

14

tuning parameter. In this research the DgM4 was fixed at -5.9 obtained from reproducing initial

15

wettability of Fathi, et al.

16

6

51-53

. This reaction is assumed

and this value was used for all other simulations. The surface

concentration of > CaOH=A −COO therefore is a wettability indicator.

The higher this

17

concentration, the more oil wet the solid surface is. The wettability indicator is calculated

18

from =

19

− && %& − &&

(6)

where , && and %& are the surface concentrations of > CaOH=A −COO at the current state,

20

the end-point water-wet state and oil-wet state, respectively. Values of && and %& are obtained

21

by speciation calculation at the initial and injection water compositions.

22

For relative permeability values, we use a linear interpolation model as follows:


11

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 41

∗ ∗ ∗ & = 1 − &,&& + &,%& ∗ ∗ ∗ % = 1 − %,&& + %,%&

(7)

1

∗ ∗ ∗ ∗ where %,%& , &,%& , %,&& and &,&& are relative permeability values at the end-point oil-wet

2

(ow) and end-point water-wet states (ww). These end-point states do not have to be at the

3

complete oil-wet or water-wet states, but ideally should be measured at initial reservoir

4

conditions (mixed wet state) and at the most water-wet state possible. The Brooks-Corey model

5

is used to calculate relative permeability at any arbitrary saturation between the end points54, ∗ & = & ∗

∗ % = % 1 − ∗

6

where the normalized water saturation ∗ is calculated by

∗ =

& − &P . 1 − &P − %

7

Here &P is the irreducible saturation and % is the residual oil saturation. Wettability alteration

8

can change contact angle, which reduces the capillary number and then the residual oil saturation.

9

Residual oil saturation actually has been shown to change during low salinity waterflooding, in

10

Nasralla, et al.

11

follows,

11

and shown in Morrow, et al.

55

. Therefore, Sor is a function of wettability as

&& %&

% = 1 − % + % .

12

Figure 1 shows two examples of relative permeability curves measured or calculated at end-

13

point wettability conditions. Relative permeability endpoints at other wettability conditions is

14

determined using Eq. (7).

15

end-point wettability states. Capillary pressure is also important when the capillary number is

16

large. However, most of the coreflooding experiments do not report the relative permeability and

17

capillary pressure curves. In this paper, relative permeability curves at the endpoints were

18

determined by best fit to the oil recoveries, because data was not available. The capillary

Ideally, the relative permeability curves should be measured at the


12

Page 13 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1

pressure is assumed to be zero since the relative permeability curve is tuned, and we prefer not to

2

have more tunable parameters.

3 4

The capillary pressure effects are

specifically studied for

spontaneous imbibition in a previous publication 8. Experimental studies also shows that the &P

change is negligible for high and low salinity water.56 The initial water saturation was assumed

5

to be constant. The water saturation remains high after the wettability alteration and even &P

6

changes the oil recovery will not be affected.

7 8

Figure 1 – Relative permeability curves at the end-point wetting states used in the

9

simulations of the core flooding experiments in (A) Fathi, et al.

10

6

with 100% volume

fraction of calcite and (B) Yousef, et al. 10 with 3% anhydrite. In both cases, is constant

11

when wettability changes. Values of decrease more in (B) than (A). The curves for (A)

12

show significantly less changes in the relative permeability curves compared to (B) because

13

the core in Fathi, et al. 6 is initially more water wet than the core in Yousef, et al. 10.

14

Simulation cases.

Table 2 lists the mineralogical compositions and parameters for the

15

limestone and chalk cores. The Stevns Klint chalk in Fathi, et al.

16

limestone in Strand, et al.

57

6

and the Middle Eastern

contain pure calcite with 100% volume fraction in the solid phase.


13

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

5

10

1

In contrast, the limestone cores from Austad, et al.

2

these experiments, kr curves were not measured. A practical workflow for simulation of core

3

flooding experiments where relative permeability curves are not measured is as follows.

4

contain anhydrite. In

1. Tune kro, krw and Sor at the original wetting state to match oil recovery with injection of

5 6

and Yousef, et al.

Page 14 of 41

formation water where wettability alteration does not occur. 2. Tune Sor at the end-point water-wet case to match ultimate oil recovery.

7

3. Perform speciation calculations given the formation water composition to obtain %& .

8

Namely, solve the nonlinear equation systems that describing the equilibrium condition

9

of all reactions and conservation of the total concentration as in formation water.

10

Perform speciation calculation given the low salinity water that leads to most water

wettability to obtain && . Select either the NEDL or the EDL model so that %& >

11

&& .

12 13

4. Perform simulations using the end-point parameters obtained in step 1, 2, and 3.

14

Reaction equilibrium constants are fixed for simulation of all experiments. We conduct 1-D

15

simulations of core flooding experiments in different carbonate rocks listed in Table 2. The 1-D

16

simulation domains were discretized uniformly into 100 grid blocks, which is the minimum

17

number of grid block to reproduce the tracer breakthrough curves in the chromatographic

18

wettability test. Oil and water viscosities were fixed at 1.0 cp and 0.476 cp, respectively. We also

19

used the reactions and tuned parameters from the 1-D core floods to model 2-D five-spot

20

patterns.

21 22

RESULTS AND DISCUSSION


14

Page 15 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1

Stevns Klint Chalk Outcrop. The Stevns Klint chalk is comprised of 55% (volume) calcite

2

with 45% pore space. The wettability can be altered by injection of ions that causes the oil acids

3

to desorb from the surface, changing the rock to a more water-wet state. Austad, et al.

4

demonstrated this in imbibition tests where a core plug was immersed by aqueous solution

5

containing SO42-. Forced displacement experiments (core floods) also show that chemically

6

tuned seawater injection leads to increased oil recovery in the same cores.6

1

7

Chromatographic Wettability Test. In chromatographic wettability tests, seawater with

8

different compositions (SW1 in Table 3) is injected into a formation water saturated core and a

9

core initially at residual oil saturation (here % = 0.25 ). The purpose of chromatographic

10

wettability tests is to quantify the adsorption of sulfate. Figure 2A shows the breakthrough

11

curves in a reference water-wet core without oil. The diffusion/dispersion coefficients (0.0042

12

ft2/day) were determined by reproducing the breakthrough curves (BTCs) of the nonreactive

13

tracer SCN-. The area between the BTCs for SCN- and SO42- quantifies the amount of adsorbed

14

SO42- as discussed in Zhang, et al. 9, which essentially measures the amount of surface sites

15

available and the water-rock contact surface area.

16

We included all reactions in the simulations except reaction (CO1) because at room

17

temperature the oil surface reactions were assumed not active. Two simulations were performed

18

to reproduce the experimental breakthrough curves

19

were filled with formation water initially. Figure 2B shows the predicted Ca2+ and Mg2+

20

concentrations. Because the injected seawater water contains more Mg2+ and less Ca2+ than the

21

formation water, more Mg2+ adsorbed on the calcite surface replacing adsorbed Ca2+ as the

22

seawater contacted the rock surface, causing the Ca2+ concentration peak near 1.0 PVI. This is

6

with and without residual oil. The cores


15

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1

consistent with experimental observations.

2

injection concentrations.

6, 9

Page 16 of 41

Both concentrations eventually returned to the

3

Figure 2C shows the BTCs for cores at a residual oil saturation of 0.25. The surface area that

4

is in direct contact with the aqueous phase is smaller than that saturated conditions without oil,

5

leading to a smaller difference between the SCN- and SO42- curves. The SCN- and SO42-

6

breakthrough earlier than those in Figure 2A because of the immobile residual oil and less

7

available pore space for flow and less surface area for adsorption. The predicted Mg2+ and Ca2+

8

concentrations in Figure 2D are similar to Figure 2B, except with an earlier and a smaller Ca2+

9

peak. The match of the BTCs with and without oil shows the validity of the model and

10

parameters relevant to the physical processes when no oil reactions are included.

11

12


16

Page 17 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1 2 3

Figure 2 - Breakthrough curves for seawater injection into water-wet cores at room

temperature. is the injection concentration of the corresponding species. (A) SCN- and

4

SO42- breakthrough curves (BTCs) from experiments6 and simulations without oil. (B)

5

Ca2+ and Mg2+ BTCs from simulations without oil.

6

experiments6 and simulations in the presence of residual oil. (D) Ca2+ and Mg2+ BTCs from

7

simulations with oil.

(C) SCN- and SO42- BTCs from

8 9

Forced Displacement. Forced displacement experiments were performed with successive

10

injection of formation water (FW), seawater (SW) and seawater depleted in NaCl (SW0NaCl)6.

11

The initial water saturation is 0.08.

12

Simulation studies are performed in two ways. Simulation A injected FW, SW and SW0NaCl

13

successively in Table 3; in simulation B, NaCl solution was injected with the same total molality

14

of FW, SW and SW0NaCl. As shown in Figure 3A, simulation A closely reproduced the

15

experimental data. The oil recovery for simulation B, however, indicates that without SO42-, Ca2+

16

and Mg2+, oil recovery was the same even though the injection water has increasingly lower

17

salinity. This indicates that for the Stevns Klint chalk, it is the chemical composition of the


17

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 41

1

injected water that matters, not salinity. This is very different from core flooding in limestones,

2

where lower salinity has been shown to improve oil recovery10, 11, 20, 53.

3 4

Figure 3B shows the predicted BTCs of total SO42-, >CaSO4-, >CaOH= −COO and pH for

simulation A. During the FW injection, >CaOH= −COO maintained a high concentration while

5

no sulfate sorption occurred (>CaSO < = 0). During the SW injection, the effluent concentrations

6

and oil recovery did not change much until the seawater reached the outlet at about 3.0 PVI,

7

when the total concentration of SO42- increased along with increasing >CaSO < and decreasing

8 9

>CaOH= −COO . The high affinity of the calcite surface to sulfate, indicated by higher surface concentrations of >CaSO < than >CaOH= −COO , leads to increased oil recovery between 3.0 and

10

4.0 PVI. From 4.0 to 6.0 PVI, SW0NaCl was injected with the same SO42- concentration as in

11

the SW case however without NaCl. With the decrease in ionic strength, SO42- activity increased

12

so that the surface concentration of >CaSO< increased further while the aqueous total SO42-

13

decreased. With continued SO42- injection, the total concentration eventually increased back to

14

the inlet concentration at around 5.0 PVI.

15 16

The increased sulfate adsorption reduces

>CaOH= −COO , which causes the observed change in wettability towards a more water-wet

state and the increased in oil recovery. The predicted SO42- and >CaOH= −COO concentrations

17

are consistent with the chemical mechanism proposed by Austad, et al. 1. The pH of the effluent

18

solution changes between 6.5 and 9.5 and is responsive to the changes in injection water

19

composition. Effluent pH increased during the SW slug and decreased during the SW0NaCl

20

slug. After about 3 PV, the pH change almost mirrored the change in >CaSO4, indicating the

21

dominant control of SO42- sorption on the pH of the system.

22


18

Page 19 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1 2

Figure 3 – (A) Oil recovery from experiments and simulations for forced displacements

3

with successive injection of FW, SW and SW0NaCl. The experimental data is from Fathi,

4

et al. 6. (B) The dimensionless concentration of total SO42-, >CaSO4- and >CaOH2(-COO)

5 6

for simulation A. C0 is 0.024 mol/kg water for SO42-; 4.1 / for surface species,

>CaSO4- and >CaOH2(-COO).

7 8

Limestone without Anhydrite from Strand, et al.

57

. The selected limestone core was

9

composed of 75% calcite and 25% pore space by volume. Thus, most of the reactions are the

10

same as those for the SK chalk. The limestone has lower specific surface area (0.29 m2/g)

11

compared to that of chalk (1.00 m2/g).

12

Chromatographic Wettability Tests. To explore the affinity of SO42- towards the limestone

13

surface, a chromatographic wettability test for a completely water-wet core is simulated. In the

14

test, seawater that contains SCN- and half of SO42- of SW (SW1/2M) was injected (Table 3). A

15

smaller diffusion/dispersion coefficient (0.002 ft2/day) than the SK chalk was obtained by

16

reproducing the BTC of SCN-. As shown in Figure 4A, the predicted SO42- concentration agrees

17

well with the concentration data. The similarity between Figure 4A and Figure 2A indicates that


19

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 41

1

the same equilibrium constants can be used to describe sulfate adsorption in limestone and in SK

2

chalk.

3

Surface Adsorption Test for Ca2+ and Mg2+. A NaCl solution with equal amounts of Ca2+,

4

Mg2+ and SCN- (NaCl-M, Table 3) was injected to test the calcite surface adsorption of Ca2+ and

5

Mg2+ in a similar chromatographic wettability test.57 As shown in Figure 4B, the breakthrough

6

times of Ca2+ and Mg2+ are longer than SCN-, indicating that Ca2+ and Mg2+are adsorbed on the

7

surface. The adsorption reactions are independent from SO42- adsorption and the carboxylic

8

group, as the adsorption is observed without SO42- and oil acids.

Figure 4B shows the

9

comparison of experimental and predicted BTCs for SCN , Ca2+, and Mg2+ at 70°C. The BTCs

10

were matched with a tuned KC5 and KC6 (-2.1 for both reactions) because of the increased

11

temperature using the NEDL model while other reaction parameters are the same with the

12

simulation of Fathi, et al. 6.

13

The two simulations of the chromatographic wettability tests indicated no significant

14

difference between chalk and limestone regarding the surface reactivity behaviors of Ca2+, Mg2+

15

and SO42-. The difference in adsorption is a result of different porosity and specific surface area.

16

That is, we can use the same reaction network listed in Table 1 to describe the surface

17

complexations on the Stevns Klint chalk6 and this Middle Eastern limestone surfaces.57


20

Page 21 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1 2

Figure 4 – Breakthrough curves of chromatographic wettability tests in limestone cores.57

3

A (A) and and A at 70°C. at 25°C. (B) ,

4

Limestone with Anhydrite. Anhydrite dissolution was reported in some low salinity core

5

floods and was regarded as the reason for the cause of the wettability alteration when the

6

injection fluids do not contain sulfate. Here we examine our model results for two cases.

7

Forced Displacement in Limestone Core with Anhydrite from Austad, et al. 5. Diluted

8

formation water was found to improve oil recovery (IOR) in limestone cores.5, 10, 11 Austad, et al.

9

5

observed tertiary IORs of 5% OOIP in limestone rocks with diluted formation water containing

10

no sulfate. Sulfate was detected in the effluent when the core was flooded with deionized water,

11

confirming the presence of anhydrite.

12

anhydrite likely led to the wettability alteration and increased IOR. To test this hypothesis we

13

simulated the forced displacement experiments in Austad, et al.

14

and 100-times-diluted formation water (100 diluted FW) were injected successively.

The increased SO42- concentration from dissolved

5

where formation water (FW)

15

Figure 5A shows that simulation A with anhydrite dissolution reproduced the incremental oil

16

recovery with 100 diluted FW. In simulation B where anhydrite dissolution was removed,

17

Figure 5A indicates that the diluted FW does not lead to IOR. Figure 5B shows the predicted


21

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2

Page 22 of 41

total effluent concentration of Ca2+ and SO42- for simulation A. Upon the injection of 100 diluted FW, Ca2+ concentrations decreased while SO42- concentrations increased. Since the

3

injection water does not contain SO42-, the total SO42- concentration indicates how much

4

anhydrite is dissolved. As shown in Figure 5B, total SO42- did not change much when the

5

injection fluid composition was changed at around 6.0 PVI. This is because the lower Ca2+

6

concentration in the injection water increased anhydrite dissolution while a smaller ionic strength

7

decreased anhydrite dissolution. The net result is that anhydrite dissolution remained nearly the

8

same as indicated by the almost constant total SO42- concentration. Free SO42- concentration,

9

however, increased by more than one order of magnitude because Na+ and Ca2+ concentrations

10

decreased with the 100 diluted FW so that less aqueous complexation of NaSO4- and

11

CaSO4(aq) were formed. Thus, additional free SO42- was available to adsorb onto the solid

12

surface to replace adsorbed oil acids, leading to an increased oil recovery of about 5% OOIP.

13 14

15


22

Page 23 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1

Figure 5 – (A) Oil recovery for forced displacements in cores with anhydrite dissolution

2

(simulation A) and without anhydrite dissolution (simulation B). (B) Breakthrough curves

3

of Ca2+, total Ca2+, SO42- and total SO42- from simulation.

4 5

Forced Displacement in Composite Limestone with Anhydrite in Yousef, et al. 10. Yousef, et

6

al. 10 observed around 20% OOIP incremental oil recovery with successive injection of seawater,

7

2, and 10 diluted seawater into composite limestone cores. Anhydrite dissolution was

8

verified in their experiments using nuclear-magnetic-resonance measurements. The cores were

9

originally oil wet. The NEDL model failed to predict wettability alteration in our first attempt.

10

An EDL model, however, reproduced the data, indicating the importance of including the surface

11

potential changes in simulations with brines of different ionic strength.

12

Figure 6A shows that the model reproduced oil recovery data. Figure 6B shows the predicted

13

breakthrough curves (BTC) of Na+, SO42- and the concentration of >CaOH2(-COO) and >CaSO< .

14

Stepwise decreases in Na+ concentration is because of the slug injection of diluted seawater. The

15

concentration of >CaOH2(-COO) also decreased stepwise, demonstrating the wettability change

16

to more water-wet states. Because the concentration of >CaSO< also decreased, the reason for

17

the decrease in >CaOH2(-COO) is not due to the competition by SO= < adsorption. A lower ionic

18

strength leads to a smaller solid surface potential, and less adsorption of the negatively charged

19

carboxylic group.


23

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 41

1 2

Figure 6 – (A) Oil recovery for forced displacement from core floods (Yousef, et al. 10) and

3

simulation. (B) Effluent concentration of chemical species. The units for Na+, total SO42-

4

and free SO42- concentration are mol/kg water. The units for >CaOH2(-COO)

5

concentration is mol/m2.

6 7

Comparison of Chalk and Limestones. Using the same input parameters for the previously

8

discussed cases, we compare predicted oil recoveries for the different cores with the same

9

injection water composition. The core floods for different mineralogies were modeled with the

10 11

same parameters in NEDL as shown Figs. 3A and 5A, and EDL for Fig. 6A. Figure 7A compares the oil recovery of chalk6 with limestone A (Austad, et al. 5) and 10

12

limestone B (Yousef, et al.

) that contain anhydrite, with the same injection sequence as

13

Yousef, et al. 10. Unlike limestone B, chalk and limestone A show little IOR with the injection of

14

diluted seawater. Figure 7B shows the predicted adsorbed carboxylic acid at the effluent point.

15

All three cores have decreased >CaOH2(-COO) at around 1.0 PVI with seawater injected. With

16

2 and 10 diluted seawater injection, limestone B shows decreased >CaOH2(-COO), indicating

17

a more water wet state. Chalk and limestone A have an increased concentration of >CaOH2(-

18

COO) indicating a more oil wet rock. The increase occured because the sulfate concentration


24

Page 25 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1

decreased as seawater was diluted (even after anhydrite was dissolved). Limestone A and

2

limestone B show significantly different behavior. The different IOR for limestone A and

3

limestone B results from the use of NEDL for limestone A, while EDL is used for limestone B.

4

Figure 7C compares the oil recovery of the three cores with FW and 100 diluted FW

5

injected. Both limestone A and limestone B show increased oil recovery, although the extent of

6

the increase is much smaller in limestone A. The differences in IOR might be caused by shifts in

7

relative permeability curves, which should be measured for each core. In contrast, no IOR is

8

observed for chalk. The >CaOH2(-COO) concentrations in Figure 7D indicate that wettability

9

alteration occurs for limestones but does not occur for chalk, due to the lack of anhydrite

10

dissolution. The different responses of the chalk and limestones to the same injection water

11

indicate the importance of anhydrite dissolution.

12 13

14


25

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 41

1 2

Figure 7 – (A) Oil recovery for chalk, limestone A and limestone B when seawater, 2× and

3

10× diluted seawater were injected. (B) The ratio of the breakthrough concentration of

4

adsorbed carboxylic acids >CaOH2(-COO) with respect to its initial concentration for each

5

case in (A). (C) Oil recovery for chalk, limestone A and limestone B. Formation water

6

(FW) and 100× diluted FW were injected. (D) The ratio of the breakthrough concentration

7

of adsorbed carboxylic acids >CaOH2(-COO) with respect to its initial concentration at the

8

condition of (C). Simulation input parameters are the same as Figure 3A for chalk, Figure

9

5A for limestone A and Figure 6A for limestone B.

10

Extension to the field scale injection. One concern in the core flood using LSW is the large

11

pore volumes injected, for example 30 PVIs in Yousef, et al. 10, to achieve a significant increase

12

in oil recovery. To examine this, we simulated a quarter of a five-spot pattern using the PennSim

13

with the reaction network and parameters that reproduced core flood experiments (Table 1 and

14

2). Two homogeneous models and one heterogeneous model were simulated. The pattern size of

15

300 300 ft2 for all three models was discretized into uniform 2020 grid blocks. No gravity is

16

included.

17

experiments of Austad, et al. 5 and Yousef, et al. 10 respectively. The heterogeneous model used

18

the same porosity, mineral composition and endpoint relative permeability of Yousef, et al.

The first two homogeneous models used the input parameters of core flood

10


26

Page 27 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1

and the permeability field was taken from an upscaling of the 41th layer of the permeability field

2

in SPE10 comparative solution project 58. The original permeability was for 60220 grid blocks

3

and was coarsened into the 2020 grid blocks. The grid coarsening was done by combining

4

311 fine grid blocks into one coarse grid block. The permeability was calculated by using an

5

arithmetic mean. PennSim results for the 2-D five-spot pattern cases matched exactly with

6

simulations using Eclipse for FW injection without wettability alteration.

7

Figure 8A compares the oil recovery curves for the 1-D homogeneous core floods and 2-D

8

homogeneous five-spot patterns. Breakthrough in the 1-D simulations is delayed slightly

9

compared to the 2-D model for both FW and 100 diluted FW injection water. Breakthrough is

10

somewhat quicker in the 2-D homogeneous model owing to poorer sweep. Sweep is excellent in

11

the 2-D models, however, so that ultimate oil recovery quickly approaches the 1-D values. This

12

is expected since the mobility ratio is favorable in these floods (end-point mobility ratios are

13

around 0.5). For both the 1-D and 2-D cases, the injection of 100 diluted FW produces about

14

5% OOIP more oil recovery than injection of FW at 2.0 PVI. This increase is significant and

15

demonstrates that recovery can be enhanced with a realistic volume of water being injected.

16

Figure 8B shows the oil recovery of SW and 10 diluted SW injection for the same 1-D core

17

and 2-D five-spot models. For both injection fluids, the 2-D five-spot patterns exhibits smaller

18

oil recovery at 2.0 PVI than the 1-D core floods. For both 1-D and 2-D cases, injection of 10

19

diluted SW leads to around 20% OOIP more oil recovery at 2.0 PVI than injection of SW alone.

20


27

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 41

1 2

Figure 8 – Oil recoveries for 1-D core floods and 2-D five-spot patterns in a homogeneous

3

limestone reservoir. (A) Using parameters of core flood experiments from Austad, et al. 5.

4

(B) Using parameters of core floods from Yousef, et al. 10.

5 6

Figure 9A shows the permeability field for a heterogeneous 2-D reservoir. The permeability

7

field contains a high permeability channel and therefore injection water breakthroughs much

8

faster.

9

compared to the homogenous case because of poor sweep efficiency, although the benefit of

10

using LSW remains large (6% OOIP). The CPU time for the entire simulation on a MacBook Pro

11

(Intel i5 2.3 GHz, 2011 version) was 307 seconds.

Figure 9B shows a significantly decreased oil recovery and earlier breakthrough

12


28

Page 29 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1 2

Figure 9 – Simulations for 2-D heterogeneous five-spot pattern. (A) Permeability field. (B)

3

Oil recoveries 2-D five-spot patterns using same parameters of Figure 6B.

4

Validation with Previous Model. The reaction network used in this paper is extended from a

5

previous model8. With more reactions, the current model is compared with the previous model

6

for the case of flow in chalk and a series of injection brine compositions including formation

7

water (FW), seawater (SW), SW depleted in NaCl (SW0NaCl) and SW0NaCl with four time

8

spiked sulfate (SW0NaCl4SO4). The brine composition is shown in Table 3. As is shown in

9

Figure 2 for the current model with and without calcite dissolution, the simulations reproduce the

10

decrease of > CaOH= −COO concentration on the rock surface. This indicates that the current

11

model and the previous model predict the same trend of wettability alteration with the tested

12

brine compositions. The effect of calcite dissolution is also tested as calcite dissolution was

13

regarded important during seawater injection processes59. Figure 2 also indicates that calcite

14

dissolution does not affect the trend of wettability alteration.


29

Energy & Fuels

0.70

Surface Fraction of Adsorbed Carboxylic Material

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 41

Previous Model

0.60

Current Model

0.50

Current Model without calcite dissolution

0.40 0.30 0.20 0.10 0.00

1 2

FW

SW

SW0NaCl

SW0NaCl4SO4

Figure 10 – Comparison of the calculated surface fraction of > − for chalk

3

from extended and previous models47. The blue bars represent the values calculated by the

4

previous model. Red bars represent the values calculated by the current model. Green bars

5

represent the values calculated by the current model, but without calcite dissolution.

6

Model limitations and Potential Improvements. This paper gives a promising way forward

7

on understanding the chemical mechanisms for wettability alteration during low salinity

8

waterflooding. The current model does have some limitations that may affect the predictive

9

capability of this model.

10

Four mechanisms of interaction between crude oil, brine and solid surfaces have been

11

reported

in

the

literature,

including

polar,

surface

precipitation,

acid/base

and

12

divalent/multivalent ion binding24. Here we focus on the acid/base mechanism. We did not

13

include ion binding because there is a lack of experimental and thermodynamic data for ion-

14

bridging interaction. Moreover, it is challenging to study those mechanisms independently.


30

Page 31 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1

The use of EDL and NEDL models to reproduce the experiments for limestone A5 and

2

limestone B10 is not desirable. This indicates that the key difference between different forms of

3

CaCO3, limestone and chalk, should be identified. Collection of detailed aqueous geochemical

4

data, including Ca2+, Mg2+, SO42-, and pH at the effluent, as well as measuring zeta potential at

5

the oil and rock surface at high pressure and high temperature, will also allow further constrain

6

the geochemical reaction network.

7 8

CONCLUSIONS

9

In this research, we extended a previous mechanistic wettability alteration model to understand

10

and predict the IOR from low salinity waterflooding for different cases including seawater

11

injection in the Stevns Klint chalk, and diluted formation water and seawater injection in

12

limestone cores containing anhydrite. The extension of the reactions include adding calcite and

13

anhydrite dissolution, the sorption of Ca2+ and Mg2+ on calcite surface, and aqueous

14

complexation. The new model was incorporated into an in-house simulator PennSim, providing

15

a unifying explanation of the recoveries observed under low-salinity waterflooding (LSW)

16

conditions in carbonate reservoirs.

17

The results show that with little tuning the new model can capture the role of mineral reactions

18

in buffering the aqueous composition and could accurately predict wettability alteration and

19

recoveries owing to surface reactions. The main overall conclusions are:

20

•

The 1-D chromatographic wettability tests for the SK chalk and limestone cores were

21

simulated. The predicted effluent concentration matched well with the experimental

22

measurements.

23 24

•

Forced displacement experiments for the SK chalk and limestone cores were also simulated.

The predicted oil recoveries also matched well the experimental


31

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 41

1

measurements. The predicted effluent concentrations were consistent with experimental

2

measurements.

3

•

Simulation of secondary recoveries for the 1-D core flooding models and 2-D five-spot

4

models ranged from 5% to 20% OOIP at 2.0 PVI using a diluted formation water or

5

diluted seawater. More importantly, the 2-D models demonstrated that LSW can yield

6

excellent increases in recovery at realistic pore volumes injected (PVI).

7 8

Additional important conclusions are:

9

•

The same set of reactions can be used for the wettability alteration model.

10

•

The prediction of wettability alteration using the mechanistic model is accurate for the

11 12

demonstrated experimental studies. •

13 14

permeability curves should be measured at two wettability states. •

Non-electrical double layer model cannot capture wettability alteration in the core floods of Yousef, et al. 10. The electrical double layer model was required instead.

15 16

Coreflooding experiments are lacking all data needed to predict recovery. Relative

•

In limestone cores with anhydrite5, diluted water reduces ion strength and aqueous

17

complexation of sulfate with cations. This enhances anhydrite dissolution and

18

availability of free SO42-, thereby improving the oil recovery. The simulation for the

19

same limestone core but without anhydrite does not show IOR.

20 21

•

The IOR at 2.0 PVI for 1-D core floods and 2D five-spot models depends on the rock mineralogy.

22

This is the first mechanistic model for carbonate rocks that considers the coupling of mineral

23

dissolution and surface complexation reactions to predict low salinity waterflooding recoveries


32

Page 33 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1

and wettability changes for both chalk and limestone cores, with and without anhydrite. Other

2

minerals could be easily added as needed in the reaction network.

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

NOMENCLATURE GP

=

activity of species (dimensionless)

A = temperature water1/2/mol1/2)

dependent

parameters

for

the

Debye-Hückel

model

(kg

B = temperature 1/2 1/2 water /mol )

dependent

parameters

for

the

Debye-Hückel

model

(kg

d

Yh f

= = =

_`

=

6+

=

!

=

6

6. }

DE.,

=

=

= =

temperature dependent parameters for the Debye-Hückel model (kg water/mol) the bulk surface area (m2) of mineral species ¡ (m2)

diffusion/dispersion coefficient tensor (ft2/day or m2/s)

activation energy (Btu/lbmol or J/mol)

Farady’s constant (9.648 10< C/mol)

molar rate of the primary species ( (mol/m⋅day)

molar rate of the secondary species ¢ (mol/m⋅day) standard gravitational constant (ft/day2 or m/s2) the ionic strength of water (mol/kg water)

equilibrium constant of reaction £

D

=

permeability (md or m2)

¤

=

reaction rate constant (mol/m2-s)

=

h

*+

=

8+

=

*.

8j%j

=

=

relative permeability of phase ¥

molar density of the primary species ( (mol/m3)

molar density of the secondary species ¢ (mol/m3) the number of the primary species the total number of species


33

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

8hE$

=

¤

=

¦+

=

$¤ S

¤ a

=

= =

the number of secondary reactions

pressure of phase ¥ (psi or Pa)

capillary pressure between phase ¥ and a reference phase (psi or Pa) total molar rate of primary species ( (lbmol/day or mol/s)

universal gas constant (8.31 J/mol-K or 10.73 ft3psi /R-lbmol)

saturation of phase ¥ (dimensionless)

"

=

temperature (R or K)

=

depth (ft or m)

z{

=

§P

=

the charge carried by the species

the dielectric constant of water (55.3, dimensionless)

the permittivity of free space (8.854 104 C ¨ 4 ©ª4 )

z|

=

q%

=

the oil-water interface charge potential (mV)

=

the solid-water interface potential (mV)

w%

qh

=

«¤

=

-P+

=

QP

¤

=

the charge density at the oil surface (C/m2)

molar density of phase ¥ (lbmol/ft3 or mol/m3)

activity coefficient of species (kg-water/mol)

the ( , () entry in the stoichiometry matrix for the reactions in canonical form

=

porosity (dimensionless)

=

density (lb/ft3 or kg/m3)

=

Page 34 of 41

viscosity of phase ¥ (cp)

22

References

23 24 25 26 27 28 29

1. Austad, T.; Madland, M.; Puntervold, T.; Korsnes, R., Seawater in Chalk: An EOR and Compaction Fluid. SPE Journal 2008, 11, (04). 2. Yildiz, H. O.; Morrow, N. R., Effect of Brine Composition on Recovery of Moutray Crude Oil by Waterflooding. Journal of Petroleum science and Engineering 1996, 14, (3), 159168. 3. Lager, A.; Webb, K.; Black, C.; Singleton, M.; Sorbie, K., Low Salinity Oil Recovery-An Experimental Investigation1. Petrophysics 2008, 49, (01).


34

Page 35 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Energy & Fuels

4. Yousef, A.; Al-Saleh, S.; Al-Kaabi, A., Laboratory Investigation of the Impact of Injection-Water Salinity and Ionic Content on Oil Recovery From Carbonate Reservoirs. SPE REE 2011, 14, (05). 5. Austad, T.; Shariatpanahi, S. F.; Strand, S.; Black, C. J. J.; Webb, K. J., Conditions for a Low-Salinity Enhanced Oil Recovery (EOR) Effect in Carbonate Oil Reservoirs. Energy & Fuels 2012, 26, (1), 569-575. 6. Fathi, S. J.; Austad, T.; Strand, S., “Smart Water” as a Wettability Modifier in Chalk: The Effect of Salinity and Ionic Composition. Energy & fuels 2010, 24, (4), 2514-2519. 7. Zhang, P.; Tweheyo, M. T.; Austad, T., Wettability Alteration and Improved Oil Recovery in Chalk: The Effect of Calcium in the Presence of Sulfate. Energy & fuels 2006, 20, (5), 2056-2062. 8. Qiao, C.; Li, L.; Johns, R. T.; Xu, J., A Mechanistic Model for Wettability Alteration by Chemically Tuned Waterflooding in Carbonate Reservoirs. SPE Journal 2015, 20, (04), 767 783. 9. Zhang, P.; Tweheyo, M. T.; Austad, T., Wettability Alteration and Improved Oil Recovery by Spontaneous Imbibition of Seawater into Chalk: Impact of the Potential Determining Ions Ca 2+, Mg 2+, and SO 4 2−. Colloids and Surfaces A: Physicochemical and Engineering Aspects 2007, 301, (1), 199-208. 10. Yousef, A.; Al-Saleh, S.; Al-Kaabi, A., Laboratory Investigation of the Impact of Injection-Water Salinity and Ionic Content on Oil Recovery From Carbonate Reservoirs. SPE REE 2011. 11. Nasralla, R. A.; Sergienko, E.; Masalmeh, S. K.; van der Linde, H. A.; Brussee, N. J.; Mahani, H.; Suijkerbuijk, B.; Alqarshubi, I., Demonstrating the Potential of Low-salinity Waterflood to Improve Oil Recovery in Carbonate Reservoirs by Qualitative Coreflood. In Abu Dhabi International Petroleum Exhibition and Conference, Society of Petroleum Engineers: Abu Dhabi, UAE,10-13 November., 2014. 12. Yousef, A. A.; Liu, J.; Blanchard, G.; Al-Saleh, S.; Al-Zahrani, T.; Al-Zahrani, R.; AlTammar, H.; Al-Mulhim, N. In Smartwater Flooding: Industry’s First Field Test in Carbonate Reservoirs, Proceedings of the SPE Annual Technical Conference and Exhibition, 2012; 2012; pp 8-10. 13. Fathi, S. J.; Austad, T.; Strand, S., Water-Based Enhanced Oil Recovery (EOR) by “Smart Water”: Optimal Ionic Composition for EOR in Carbonates. Energy & Fuels 2011, 25, (11), 5173-5179. 14. Webb, K. J.; Black, C. J. J.; Tjetland, G. In A Laboratory Study Investigating Methods for Improving Oil Recovery in Carbonates, International Petroleum Technology Conference, 2005; International Petroleum Technology Conference: 2005. 15. Hiorth, A.; Cathles, L.; Madland, M., The Impact of Pore Water Chemistry on Carbonate Surface Charge and Oil Wettability. Transport in porous media 2010, 85, (1), 1-21. 16. Tang, G.-Q.; Morrow, N. R., Influence of brine composition and fines migration on crude oil/brine/rock interactions and oil recovery. Journal of Petroleum Science and Engineering 1999, 24, (2), 99-111. 17. Gomari, K. R.; Hamouda, A.; Denoyel, R., Influence of Sulfate Ions on the Interaction Between Fatty Acids and Calcite Surface. Colloids and Surfaces A: Physicochemical and Engineering Aspects 2006, 287, (1), 29-35.


35

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Page 36 of 41

18. Gupta, R.; Smith, G. G.; Hu, L.; Willingham, T.; Lo Cascio, M.; Shyeh, J. J.; Harris, C. R., Enhanced Waterflood for Carbonate Reservoirs - Impact of Injection Water Composition. In Society of Petroleum Engineers: 2011. 19. Romanuka, J.; Hofman, J.; Ligthelm, D. J.; Suijkerbuijk, B.; Marcelis, F.; Oedai, S.; Brussee, N.; van der Linde, H.; Aksulu, H.; Austad, T., Low Salinity EOR in Carbonates. In SPE Improved Oil Recovery Symposium, Society of Petroleum Engineers: Tulsa, Oklahoma, USA, 14-18 April, 2012. 20. Chandrasekhar, S.; Mohanty, K. K. In Wettability Alteration with Brine Composition in High Temperature Carbonate Reservoirs, SPE Annual Technical Conference and Exhibition, New Orleans, Louisiana, USA, 30 September-2 October, 2013; Society of Petroleum Engineers: New Orleans, Louisiana, USA, 2013. 21. Jerauld, G. R.; Webb, K. J.; Lin, C.-Y.; Seccombe, J. C., Modeling Low-Salinity Waterflooding. SPE Reservoir Evaluation & Engineering 2008, 11, (06), 1,000-1,012. 22. Delshad, M.; Najafabadi, N. F.; Anderson, G.; Pope, G. A.; Sepehrnoori, K., Modeling Wettability Alteration by Surfactants in Naturally Fractured Reservoirs. SPE Reservoir Evaluation & Engineering 2009, 12, (03), 361-370. 23. Al-Shalabi, E. W.; Sepehrnoori, K.; Delshad, M.; Pope, G., A Novel Method to Model Low-Salinity-Water Injection in Carbonate Oil Reservoirs. SPE Journal 2015, 20, (05), 1154 1166. 24. Buckley, J.; Liu, Y., Some Mechanisms of Crude Oil/Brine/Solid Interactions. Journal of Petroleum Science and Engineering 1998, 20, (3), 155-160. 25. Qiao, C.; Li, L.; Johns, R. T.; Xu, J., Compositional Modeling of Dissolution-Induced Injectivity Alteration During CO2 Flooding in Carbonate Reservoirs. SPE journal 2015. 26. Yeh, G.-T.; Tripathi, V. S., A Model for Simulating Transport of Reactive Multispecies Components: Model Development and Demonstration. Water Resour. Res 1991, 27, (12), 30753094. 27. Steefel, C. I.; Lasaga, A. C., A Coupled Model for Transport of Multiple Chemical Species and Kinetic Precipitation/Dissolution Reactions with Application to Reactive Flow in Single Phase Hydrothermal Systems. American Journal of Science 1994, 294, (5), 529-592. 28. Walter, A.; Frind, E.; Blowes, D.; Ptacek, C.; Molson, J., Modeling of Multicomponent Reactive Transport in Groundwater: 1. Model Development and Evaluation. Water Resources Research 1994, 30, (11), 3137-3148. 29. Steefel, C. I.; DePaolo, D. J.; Lichtner, P. C., Reactive transport modeling: An essential tool and a new research approach for the Earth Sciences. Earth and Planetary Science Letters 2005, 240, 539-558. 30. MacQuarrie, K. T. B.; Mayer, K. U., Reactive transport modeling in fractured rock: A state-of-the-science review. Earth-Science Reviews 2005, 72, (3-4), 189-227. 31. Bao, C.; Wu, H.; Li, L.; Newcomer, D.; Long, P. E.; Williams, K. H., Uranium Bioreduction Rates across Scales: Biogeochemical Hot Moments and Hot Spots during a Biostimulation Experiment at Rifle, Colorado. Environmental Science & Technology 2014, 48, (17), 10116-10127. 32. Brunet, J. P. L.; Li, L.; Karpyn, Z. T.; Kutchko, B. G.; Strazisar, B.; Bromhal, G., Dynamic Evolution of Cement Composition and Transport Properties under Conditions Relevant to Geological Carbon Sequestration. Energy & Fuels 2013, 27, (8), 4208-4220.


36

Page 37 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Energy & Fuels

33. Surasani, V. K.; Li, L.; Ajo-Franklin, J. B.; Hubbard, C.; Hubbard, S. S.; Wu, Y., Bioclogging and Permeability Alteration by L. mesenteroides in a Sandstone Reservoir: A Reactive Transport Modeling Study. Energy & Fuels 2013, 27, (11), 6538-6551. 34. Vilcáez, J.; Li, L.; Wu, D.; Hubbard, S. S., Reactive Transport Modeling of Induced Selective Plugging by Leuconostoc Mesenteroides in Carbonate Formations. Geomicrobiology Journal 2013, 30, (9), 813-828. 35. Hubbard, C. G.; Cheng, Y.; Engelbrektson, A.; Druhan, J. L.; Li, L.; Ajo-Franklin, J. B.; Coates, J. D.; Conrad, M. E., Isotopic insights into microbial sulfur cycling in oil reservoirs. Frontiers in Microbiology 2014, 5. 36. Morel, F.; Hering, J. G., Principles and Applications of Aquatic Chemistry. Wiley: New York, 1993; p xv, 588. 37. Lasaga, A. C., Chemical kinetics of water-rock interactions. Journal of Geophysical reserach 1984, 89, (B6), 4009-4025. 38. Wolery, T. J., EQ3/6: A Software Package for Geochemical Modeling of Aqueous Systems: Package Overview and Installation Guide (Version 7.0). Lawrence Livermore National Laboratory Livermore, CA: 1992. 39. Helgeson, H. C.; Brown, T. H.; Nigrini, A.; Jones, T. A., Calculation of Mass Transfer in Geochemical Processes Involving Aqueous Solutions. Geochimica et Cosmochimica Acta 1970, 34, (5), 569-592. 40. Lasaga, A. C., Kinetic theory in the earth sciences. Princeton University Press: 2014. 41. Brady, P. V.; Krumhansl, J. L.; Mariner, P. E., Surface Complexation Modeling for Improved Oil Recovery. In SPE Improved Oil Recovery Symposium, Society of Petroleum Engineers: Tulsa, Oklahoma, USA, 14-18 April, 2012. 42. Qiao, C.; Li, L.; Johns, R. T.; Xu, J., A Mechanistic Model for Wettability Alteration by Chemically Tuned Water Flooding in Carbonate Reservoirs. In SPE Annual Technical Conference and Exhibition, Society of Petroleum Engineers: Amsterdam, The Netherlands, 2729 October, 2014. 43. Pokrovsky, O.; Schott, J., Surface Chemistry and Dissolution Kinetics of Divalent Metal Carbonates. Environmental science & technology 2002, 36, (3), 426-432. 44. Goldberg, S., Surface complexation modeling. Reference Module in Earth Systems and Environmental Sciences 2013, 10. 45. Pokrovsky, O. S.; Mielczarski, J. A.; Barres, O.; Schott, J., Surface Speciation Models of Calcite and Dolomite/Aqueous Solution Interfaces and Their Spectroscopic Evaluation. Langmuir 2000, 16, (6), 2677-2688. 46. Langmuir, D.; Hall, P.; Drever, J., Environmental Geochemistry. Prentice Hall, New Jersey: 1997. 47. Qiao, C.; Johns, R.; Li, L.; Xu, J., Modeling Low Salinity Waterflooding in Mineralogically Different Carbonates. In SPE Annual Technical Conference and Exhibition, Society of Petroleum Engineers: Houston,Texas, USA, 28-30 September, 2015. 48. Brady, P. V.; Krumhansl, J. L., A Surface Complexation Model of Oil–Brine–Sandstone Interfaces at 100° C: Low Salinity Waterflooding. Journal of Petroleum Science and Engineering 2012, 81, 171-176. 49. Brady, P. V.; Morrow, N. R.; Fogden, A.; Deniz, V.; Loahardjo, N., Electrostatics and the Low Salinity Effect in Sandstone Reservoirs. Energy & Fuels 2015, 29, (2), 666-677.


37

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

Page 38 of 41

50. Hopkins, P.; Puntervold, T.; Strand, S., Preserving Initial Core Wettability During Core Restoration of Carbonate Cores. In International Symposium of the Society of Core Analysts, St. John's Newfoundland and Labrador, Canada, 2015. 51. Strand, S.; Puntervold, T.; Austad, T., Effect of Temperature on Enhanced Oil Recovery From Mixed-wet Chalk Cores by Spontaneous Imbibition and Forced Displacement Using Seawater. Energy & Fuels 2008, 22, (5), 3222-3225. 52. Zahid, A.; Shapiro, A. A.; Skauge, A. In Experimental Studies of Low Salinity Water Flooding Carbonate: A New Promising Approach, SPE EOR Conference at Oil and Gas West Asia, 2012; Society of Petroleum Engineers: 2012. 53. Mahani, H.; Keya, A. L.; Berg, S.; Bartels, W.-B.; Nasralla, R.; Rossen, W. R., Insights into the Mechanism of Wettability Alteration by Low-Salinity Flooding (LSF) in Carbonates. Energy & Fuels 2015, 29, (3), 1352-1367. 54. Brooks, R. H.; Corey, A., Properties of Porous Media Affecting Fluid Flow. Journal of the Irrigation and Drainage Division 1966, 92, (2), 61-90. 55. Morrow, N. R.; Lim, H. T.; Ward, J. S., Effect of crude-oil-induced wettability changes on oil recovery. SPE Formation Evaluation 1986, 1, (01), 89-103. 56. Tibi G. Sorop, B. V., Shehadeh, K. Masalmeh, Bart M.J.M., Suijkerbuijk, Hilbert A. van der Linde, Hassan Mahani, Niels J. Brussee, Fons A.H.M. Marcelis, and Ab Coorn, Relative Permeability Measurements to Quantify the Low Salinity Flooding Effect at Field Scale. In Abu Dhabi International Petroleum Exhibition and Conference, Abu Dhabi, UAE, 2015. 57. Strand, S.; Austad, T.; Puntervold, T., "Smart Water" for Oil Recovery from Fractured Limestone: A Prelimlinary Study. Energy & fuels 2008. 58. Christie, M.; Blunt, M., Tenth SPE Comparative Solution Project: A Comparison of Upscaling Techniques. In SPE Reservoir Simulation Symposium, Society of Petroleum Engineers: Houston, Texas, 11-14 February, 2001. 59. Andersen, P.; Evje, S.; Madland, M.; Hiorth, A., A geochemical model for interpretation of chalk core flooding experiments. Chemical engineering science 2012, 84, 218-241. 60. Chou, L.; Garrels, R. M.; Wollast, R., Comparative study of the kinetics and mechanisms of dissolution of carbonate minerals. Chemical Geology 1989, 78, 269 - 282. 61. Jeschke, A. A.; Vosbeck, K.; Dreybrodt, W., Surface controlled dissolution rates of gypsum in aqueous solutions exhibit nonlinear dissolution kinetics. Geochimica et Cosmochimica Acta 2001, 65, (1), 27-34. 62. Shariatpanahi, S. F.; Strand, S.; Austad, T., Evaluation of Water-Based Enhanced Oil Recovery (EOR) by Wettability Alteration in a Low-Permeable Fractured Limestone Oil Reservoir. Energy & Fuels 2010, 24, (11), 5997-6008.

36 37


38

Page 39 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1

Energy & Fuels

Table 1: Multiphase reaction network* Number

Reactions

Oil/water interface reactions O1 −Rttv ↔ −Rtt + v A O2 −RttRG A ↔ −Rtt + RG=A O3 −Rtt*!A ↔ −Rtt + *!=A

Solid (Calcite) / water interface reactions C1 > RGtv + v A ↔> RGtv=A C2 > RGtv=A + t RG t RGtv=A + Rt= ↔ > RGRt + v= t C4 > Rt v ↔ > Rt + v A C5 > Rt RGA ↔ > Rt + RG=A C6 > Rt *!A ↔> Rt + *!=A

Calcite-water / Oil-water interface reaction CO1 > RGtv=A −Rtt ↔> RGtv=A + −Rtt

Aqueous-phase reactions A1 v= t ↔ v A + tv A2 vRt ↔ v A + Rt= A3 v= Rt ↔ v A + vRt A4 *! t< ↔ *!=A + t

Modeling Low-Salinity Waterflooding in Chalk and ... - ACS Publications

Recommend Documents