0 Bewertungen0% fanden dieses Dokument nützlich (0 Abstimmungen)

5 Ansichten12 Seitenjhk

Jul 29, 2017

© © All Rights Reserved

PDF, TXT oder online auf Scribd lesen

jhk

© All Rights Reserved

Als PDF, TXT **herunterladen** oder online auf Scribd lesen

0 Bewertungen0% fanden dieses Dokument nützlich (0 Abstimmungen)

5 Ansichten12 Seitenjhk

© All Rights Reserved

Als PDF, TXT **herunterladen** oder online auf Scribd lesen

Sie sind auf Seite 1von 12

journal homepage: www.elsevier.com/locate/jngse

alteration

Saurabh Naik, Zhenjiang You*, Pavel Bedrikovetsky

Australian School of Petroleum, The University of Adelaide, Adelaide, SA 5005, Australia

a r t i c l e i n f o a b s t r a c t

Article history: In unconventional water-wet gas reservoirs with extremely low permeability, water blockage due to the

Received 4 February 2015 capillary end effect near the fracture face and in the wellbore vicinity is a severe obstacle to gas pro-

Received in revised form duction. Wettability alteration of rock surface is discussed as an effective way to reduce water blockage

21 May 2015

and enhance gas production. It can also be applied to accelerating dewatering and preventing drilling

Accepted 9 June 2015

and fracturing uid leak-off. The current paper develops analytical models for steady-state two-phase

Available online 25 June 2015

linear and radial ows, accounting for the contact-angle-dependent capillary pressure and relative

permeability. The model has been validated by comparison with laboratory test. Applications of the

Keywords:

Wettability alteration

proposed model to porous media with constant and piecewise-constant contact angles are presented.

Production rate The modelling shows that the competition between the viscous and capillary effects on productivity

Unconventional gas reservoirs results in an optimal contact angle for the maximum well productivity indexes for water and gas. The

Capillary end effect productivity index increases with the capillary-viscous ratio and the altered zone length.

Two phase ow 2015 Elsevier B.V. All rights reserved.

Contact angle

bility modication of rock surface towards less water-wet results in

Natural gas from the unconventional reservoirs is one of the key decrease of capillary pressure and mitigation of the water-blockage

energy resources that can meet the current world's energy and effect.

environmental objectives (Economides and Wood, 2009). However, Altering wettability to more water-wet increases the relative

the unconventional reservoirs have extremely low well produc- permeability for gas (Anderson, 1987), which is a positive factor for

tivity. Therefore, hydraulic fracturing is commonly used to stimu- gas production. However, it is accompanied by the increase of water

late wells in coal seam, shale and tight gas reservoirs (Wang et al., blocking effect which restricts gas ow from reservoir matrix to-

2012). For reservoirs with very low permeability, fracturing and wards well or fractures (Xie et al., 2009). These two competitive

drilling uid leak-off along with capillary end effect leads to an factors suggest the existence of the optimal contact angle (wetta-

increase of water saturation in the near wellbore or fracture-wall bility), corresponding to the maximum ow rate during gas pro-

regions (Xie et al., 2009; Bahrami et al., 2011; Odumabo et al., duction. Yet, the existence of the optimal contact angle for gas

2014). Fluid invasion into the reservoir is greatly affected by the production has not been reported in the literature.

contact angle (Li et al., 2014). Two competitive factors also suggest the existence of the con-

The increased water saturation in the fracture or wellbore tact angle that maximises water rate with gas-water production.

neighbourhood throttles gas ow from reservoir to the well. This Fast dewatering in coal seams or shale gas reservoirs is required at

water entrapment is referred to as water blockage. A few studies the beginning of the reservoir exploitation, in order to facilitate

have been conducted on water block clean-up (Mahadevan et al., rapid decrease of the reservoir pressure which intensies gas

2007, 2009) and the long-term effects on production (Bahrami desorption from the matrix. Numerous experimental studies on

et al., 2011). One of the most effective methods to reduce water wettability alteration have been performed. The contact angle of

rock surfaces can be altered articially through injection of sur-

factants (Adibhatla et al., 2006; Bryant et al., 2006; Kumar et al.,

* Corresponding author. 2008; Chen and Mohanty, 2014), high temperature steam

E-mail address: zyou@asp.adelaide.edu.au (Z. You).

http://dx.doi.org/10.1016/j.jngse.2015.06.016

1875-5100/ 2015 Elsevier B.V. All rights reserved.

1574 S. Naik et al. / Journal of Natural Gas Science and Engineering 26 (2015) 1573e1584

s swc

(Nasralla et al., 2013) and water with different pH (Morsy et al., Ss (1)

2014). Signicant change of contact angle in sandstones and car- 1 swc sgr

bonates with variations of salinity and pH has been reported with

regards to low salinity waterooding (Fogden et al., 2011; Morrow where s is the water saturation, and swc and sgr are the connate

and Buckley, 2011). Presently, intensive research is ongoing for water and irreducible gas saturations, respectively.

wettability alteration using nanoparticles (Roustaei et al., 2012; The contact angle at rock surface q is measured between the

Arab and Pourafshary, 2013; Hendraningrat et al., 2013; Assef wateregas interface and the wateresolid interface. The rock is

et al., 2014). To some extent, surface contact angle can be water-wet (hydrophilic) if q < p/2 and is hydrophobic if q > p/2.

controlled by varying concentrations and compositions of injected Therefore, saturation of the wetting phase Sw can be expressed as:

agents. It inspires the investigation of contact angle as a continuous 8 p

parameter, controllable by surface treatment. Although wettability >

< Ss; q<

2

alteration has been extensively studied in sandstone and carbonate Sw S; q (2)

>

: 1 Ss; p

reservoirs, research on unconventional rocks such as coals and q>

2

shales is still very limited. There are a few studies on shale (Morsy

et al., 2014) and tight sandstone (Kathel and Mohanty, 2013) oil Unconventional reservoirs exhibit a variety of mineralogy and

reservoirs. However, comprehensive research on wettability alter- surface types such as organic matter (e.g. kerogen), aluminosilicate

ation in unconventional gas reservoirs is unavailable in the clays and silica sands; each mineral has its own wettability. This

literature. leads to the necessity of using fractional wettability to describe the

Rate enhancement by wettability alteration has been dis- rock surface.

cussed by Kathel and Mohanty (2013). Piecewise wettability In the present work, the LeveretteCassie equation (O'Carroll

alteration may lead to saturation shock on the boundary of et al., 2005) is applied to describe a porous medium which has

altered and unaltered zones similarly to those on the boundary of three fractions of surface area (f1, f2, f3) with different contact an-

rocks with piecewise permeability (Barenblatt et al., 1990; gles. The major fraction f1 has the contact angle which is considered

Bedrikovetsky, 1993; Van Duijn et al., 1995). The saturation alterable from 0 to p by some wettability altering agent. A small

shock may highly affect well productivity index. However, fraction f2 is considered permanent water wet with contact angle

steady-state solutions for two-phase linear ows have been 0 and another small fraction f3 is permanently hydrocarbon wet

presented for uniform rock wettability only (Barenblatt et al., with contact angle p. The overall capillary pressure of porous me-

1990; Chaouche et al., 1993; Van Lingen et al., 1996; Huang and dium with fractional wettability is:

Honarpour, 1998). Analytical models accounting for wettability

2s

variation along the core or near to fracture face and wellbore are Pc s; q s Js; q;

not available in the literature. k

In the current study, the effects of contact angle alteration on f

well productivity for both water and gas are investigated using

Js; q f1 cosqCSw s; q a f2 CSs a f3 C1 Ss a

1 1 1

the analytical model. The formulae for steady-state saturation

distribution in heterogeneous rocks with piecewise-constant (3)

contact angle and for well index are derived. It is found that

there does exist an optimal contact angle to maximise the water where s is the interfacial tension, k is the rock permeability, f is the

production rate; another value of the contact angle maximises porosity, f1, f2, f3 are the major fraction and two minor fractions of

the production rate for gas. Calculations using the analytical rock surface respectively, f1 f2 f3 1; C and a are empirical

model show that the wettability alterations cause increase in model parameters.

productivity index 2.0e4.7 times for gas and around 2.0 times for The relative permeabilities of water (krw) and gas (krg) are

water. expressed as functions of water saturation as (Chen et al., 2013):

The structure of the paper is as follows: Section 2 presents the

Z S

contact-angle-dependent models for capillary pressure and two-

w2 sds

phase relative permeabilities, which are then applied in Section 3 h 0

krw s Ss Z (4)

for derivation of analytical model for steady-state two-phase 1

ows in rocks with piecewise-constant contact angle. Laboratory w2 sds

0

data of two coreood tests obtained from literature are analysed for

the model validation in Section 4. Analysis of the results from

Z1

developed analytical model is then conducted in both linear and

radial ows (Section 5). Section 6 presents application of nano- w2 sds

technology to rock surface treatment for enhanced dewatering and krg s 1 Ssh ZS (5)

gas production from unconventional reservoirs. Discussions of the 1

w2 sds

model validity and conclusions on optimal contact angle nalise 0

the paper (Sections 7 and 8).

where h is the tortuosity index and w(S) pis the cleat height or

capillary radius, expressed as wS k=f=CS 1=a according to

2. Contact-angle-dependent model for capillary pressure and YoungeLaplace model (Barenblatt et al., 1990; Bedrikovetsky, 1993).

relative permeability In this model, the rock is assumed to be water-wet and water

occupies smaller cleats or pores. As water saturation increases, the

This section presents the models for water-saturation- and cleats or pores are lled by water in order of increase of their

contact-angle-dependent capillary pressure and relative perme- aperture (radius). Conversely, if the rock is hydrophobic, the water

ability. To introduce the capillary pressure model, let us dene the lling sequence occurs from the largest to the smallest cleats and

normalised water saturation S as: pores. While this model is applicable to uniformly-wet media, it

S. Naik et al. / Journal of Natural Gas Science and Engineering 26 (2015) 1573e1584 1575

cannot explain the behaviour of intermediately-wet natural rocks 3. Analytical model for steady-state two-phase ow

where the fractional or mixed wettability is more likely due to

surface roughness, mineral heterogeneity and different adsorbed In this section, analytical solution to the system of immiscible

compounds (Bradford et al., 1997). steady-state two-phase ow is derived, accounting for the effects of

In order to describe the relation between relative permeability contact angle on capillary pressure and relative permeability. The

and contact angle, we apply the model that accounts for wettability analytical model will be applied to the treatment of laboratory

heterogeneity of natural cores at the pore scale, which results in a coreood data (Section 4) and sensitivity study on gas-water pro-

difference in pore lling sequence for intermediate wettability ductivity (Section 5).

(McCaffery and Bennion (1974); see also Bradford et al. (1997) for

experimental validation of the model). An empirical weighting

function g(q) represents the portion of pores exhibiting hydro- 3.1. Analytical model

phobic surfaces:

Darcy's law applies to the ow of each phase (Barenblatt et al.,

gq 1 cosq=2 (6) 1990; Cinar and Riaz, 2014):

As it follows from Eqs (4) and (5), the contact-angle-dependent

kkrw s; q kkrg s; q

relative permeabilities for water and gas become: uw VPw ; ug VPg ; U uw ug

mw mg

Z S Z 1 (11)

1 gq w2 sds gq w2 sds

krw s; q Ssh 0

Z 1

1S

(7) where uw and ug stand for ow velocities of water and gas,

w2 sds respectively; U is the total velocity; krw and krg are relative per-

0 meabilities of water and gas, respectively; mw and mg are viscosities

of water and gas, respectively; Pw and Pg are phase pressures.

Z 1S Z1 Capillary pressure Pc is dened as the difference between the

gq w2 sds 1 gq w2 sds gas and water pressures:

0

krg s; q 1 Ssh Z 1

S

Pc Pg Pw (12)

2

w sds

0 Substituting Darcy equation for water in (11) and capillary

(8) pressure equation (12) into Darcy equation for gas in (11), we

obtain:

Particularly, for gas ow in tight formation, especially in nano-

!

porous media, molecular scale interactions can have signicant krg s; q krg s; qmw

impact on gas ow (Shapiro and Wesselingh, 2008). Gas velocity is k VPc uw 1 U (13)

mg krw s; qmg

non-zero at the distance from the pore wall less than the mean free

path of gas (Wu and Pruess, 1998). At this scale, the collision fre- Fractional ow function f is calculated as the ratio of water

quency between gas molecules and the pore walls increases if mobility krw/mw to the total mobility l:

compared with gas molecularemolecular collisions, yielding

higher gas ux than that calculated from no-slip condition at the krw s; q=mw krw s; q krg s; q

pore wall. This effect is referred to as the Klinkenberg (gas slip) f s; q ; ls; q (14)

ls; q mw mg

effect. It is evaluated by multiplying the Klinkenberg term

1 b= Pg to the gas permeability (8) as: Substituting Eqs. (14) and (3) into (13) results in:

Z 1S Z1

0 1 gq w2 sds 1 gq w2 sds

0

b A

krg s; q @1 1 Ssh Z 1

S

(9)

Pg 2

w sds

0

where Pg is the average gas pressure along the distance, and b is the

Zs Zy

Klinkenberg factor. For tight gas reservoirs with permeability below f s; qkrg s; qJ 0 s; q

c ds xa dx (15)

1 md, formula for Klinkenberg factor is given by Jones and Owen f s; q f 0

(1980) as: s0 y0

b 0:86k0:33 (10) p

where c is the capillary-viscous

p ratio, dened as c s kf=mg jUjL

in linear ow and c 2ps kf=mg jqj in radial ow. Water ux ratio

where k is the gas permeability at innite pressure.

f0 uw/U. For linear ow, distance y X, y0 0, a 0, and s0 is

Contact-angle-dependent models for capillary pressure and

water saturation at the core outlet or fracture face. For radial ow,

relative permeability presented in this section will be applied to the

y R, y0 Rw, a 1, and s0 is water saturation at wellbore. Eq. (15)

governing system of two-phase ow in the next section.

1576 S. Naik et al. / Journal of Natural Gas Science and Engineering 26 (2015) 1573e1584

and radial ows.

The ow direction is inverse to increasing distance y, from y 1

to y0. At the ow exit point y y0, capillary pressure is Pc 0. As

capillary pressure is a monotonic function of saturation, the exit

point saturation s0 can be determined from Pc (s0, q) 0 for various

contact angles. The saturation also tends to a certain value s0 at the

distance far away from the exit, i.e., as y / , s/s0. Inlet point

saturation s0 can be determined from f 0 f (s0, q) for various

contact angles.

The productivity index (PI) for each phase is calculated as the

ow rate divided by the phase pressure drawdown:

Fig. 2. Matching of the measured saturation proles by the analytical model.

in which i stands for either w (for water) or g (for gas). DP is the

pressure drop calculated from Darcy equation (11) by integration

over the distance from inlet to outlet: The region where y b is considered the unaltered zone and the

contact angle is set as the initial contact angle; q q0 (Fig. 1). The

Z1 solution procedure for piecewise-constant wettability case is

ui m i xa

DPi dx (17) similar to that for uniform wettability. Water saturation at ow

k kri sx; q

y0 outlet s0 is obtained from Pc(s0) 0. Then the saturation prole is

calculated from y0 up to b for given c and f0. Capillary pressure and

3.2. Solution procedure pressures for both phases are continuous across the boundary y b.

The saturation on the outer side of the boundary s, is calculated

For uniform wettability with known contact angle, the solution from the value on the inner side of the boundary s as: Pc(s,

procedure is as follows. First, water saturation at ow outlet s0 q0) Pc(s, qa). Afterwards, the saturation prole is calculated in the

is determined from capillary pressure function Pc (s0, q) 0. unaltered zone from y b to the inlet y 1. Pressure drops of water

Second, saturation prole is calculated from (15) for given c and and gas are calculated from (17). Finally, productivity indexes for

f 0. Third, pressure drops of water and gas are calculated from both phases are obtained from (16).

(17). Finally, productivity indexes for both phases are obtained

from (16). 4. Treatment of laboratory coreood data

Let us consider piecewise-constant wettability in which the

contact angle alteration is applied up to a fraction of the full length. In this section, the laboratory coreooding carried out by Bazin

The region y0 y < b is considered the altered zone, where b is the et al. (2010) under two ow rates are analysed in order to validate

altered zone length. The contact angle in this zone is q qa (Fig. 1). the model proposed in Sections 2 and 3.

Fig. 1. Schematic of piecewise-constant wettability inow cases: (a) linear ow in coreooding; (b) linear ow towards fracture face; (c) radial ow towards wellbore.

S. Naik et al. / Journal of Natural Gas Science and Engineering 26 (2015) 1573e1584 1577

Cylindrical Mo^liere sandstone core from tight gas reservoir increase in fractional ow function with contact angle (viscous

with diameter of 3.8 cm and length of 7 cm is used. The initial core effect), which increases water production and decreases gas

permeability is extremely low (0.01 md) and is typical for un- production.

conventional reservoirs; the porosity is 0.144. The tortuosity for The dimensionless total mobility is dened as the product of

tight sandstone is taken as h 2.0 (Barenblatt et al., 1990). The total mobility and water viscosity: L lmw. Fig. 3c presents

laboratory procedure is as follows: rst, the core was fully satu- dimensionless total mobility versus water saturation s at different

rated with water under vacuum. Second, nitrogen was injected contact angles. At small contact angle q, increase of saturation leads

into the core until irreducible water saturation was reached. Third, to increase of relative permeability for water krw and decrease of

water was injected into the core with constant pressure until relative permeability for gas krg. The dominance of gas permeability

breakthrough. Fourth, nitrogen backow was performed with due to large wateregas viscous ratio causes reduction of dimen-

constant pressure drawdown 10 bars in Test 1 and 15 bars in Test sionless total mobility L with saturation.

2, respectively. The measured saturation proles are shown in At large contact angle q, krw is high and krg is low at large water

Fig. 2. Measured irreducible saturation for water was swc 0.22 saturation. The effect of krw increase dominates over that of krg

and for gas was sgr 0.2. decrease, which yields increase of L at large water saturation.

The mathematical model presented in Sections 2 and 3 is Fig. 4 presents the saturation proles s(X) for different contact

applied to the calculation of saturation prole under the laboratory angles with small (Fig. 4a) and large (Fig. 4b) capillary-viscous ra-

conditions. s0 values are approximated from the saturation proles tios in linear ow. The curves start at X 0 from saturation value S0.

of each test; 0.320 and 0.295 for test 1 and test 2, respectively. The For X / , the proles tend to s0. The curves are monotonic. The

values of tuned parameters as obtained by least-square method conditions at X 0 and X / , as determined by the rock wetta-

(Chalk et al., 2012; You et al., 2013) are: capillary pressure model bility and by the rates of both phases, are independent. At a certain

parameters C 104, a 0.4 and contact angle q 57.3 . The ob- contact angle, s0 > s0 yields decreasing saturation proles (blue,

tained saturation proles for the two tests (blue and red curves (in green and red curves) and s0 < s0 yields increasing saturation

the web version) in Fig. 2) are in good agreement with measured proles (grey and cyan curves). The slope of s(X) increases with q,

proles (green and cyan points in Fig. 2). from low negative value at q 0 (blue curves) to high positive value

at q p (cyan curves). The capillary-viscous ratio c determines the

5. Analysis of the results signicance of capillary end effect. The magnitude of slope of s(X)

curve decreases with c (Fig. 4). The same behaviour occurs in radial

In this section, the analytical model developed in Section 3 is ow.

applied to investigating the effect of contact angle on the produc- The productivity index (Eq. (16)) can be used to evaluate the

tivity behaviour of both phases in linear and radial ows. The linear total effect of contact angle on productivity for each phase. The full

case applies to ow towards a fracture or coreood in the labora- range of contact angle from 0 to p is examined (Fig. 5). PI values for

tory; while the radial case applies to ow towards wellbore in the gas and water at contact angle of p/2 are used as scales to obtain

eld. Sensitivities of productivity for gas and water to the capillary- dimensionless PIs. The saturation prole is constant for all c values

viscous ratio and altered zone length are analysed. at this contact angle, which corresponds to the condition s0 s0

The values of rock and uid properties used in this study to (intersection between blue and green curves in Fig. 3b). Therefore,

portray unconventional reservoirs with low permeability are listed relative permeabilities (7, 8) are independent of c. PIs for both gas

in Table 1. The well radius is: rw 0.15 m. and water at q p/2 are independent of c (Fig. 5).

The contact angle affects boundary saturation and relative For large values of c, the effect of saturation decrease with

permeability curves. Sensitivity of the contact angle to fractional contact angle (capillary effect) dominates over the effect of frac-

ow function and boundary saturations s0 and s0 is analysed. tional ow increase (viscous effect) on PI. Increase of contact angle

Fractional ow function f(s) calculated from (14) is presented in leads to the increase of PI for gas (blue and green (in the web

Fig. 3a. The fractional ow decreases with contact angle for any version) curves in Fig. 5a) and decrease of PI for water (blue and

given saturation in the range swc < s < 1 sgr. green curves in Fig. 5b). As c decreases, the capillary and viscous

Fig. 3b shows the contact angle dependences of s0 calculated effects become comparable. The two competitive effects lead to an

from Pc(s0,q) 0 and s0 calculated from f(s0,q) f0. Both saturations optimal PI at an intermediate contact angle (red and cyan curves in

s0 and s0 decrease with contact angle; s0 and s0 are equal at an Fig. 5a and b). The optimal contact angle increases with c for gas

intermediate contact angle (Fig. 3b), which results in a constant and decreases with c for water. Larger value of c yields higher

saturation prole at this contact angle. increase in PI for both gas and water (Fig. 5).

Alteration of contact angle introduces two competitive effects Let us investigate the effect of altering the contact angle of a

on the productivity for water and gas. One effect is the reduction of fraction of the production zone on the PI-contact angle prole. We

saturation prole with contact angle (capillary effect), which de- consider an example where the rock surface is partially treated by a

creases water production and increases gas production; the other is wettability alteration agent from the ow exit y0 up to a certain

distance y1. The zone from y1 to the inlet y2 remains unaltered.

Without losing generality, the rock surface is assumed water wet

before treatment and the initial contact angle is zero. After the

Table 1

Values of rock and uid properties. treatment, the contact angle of the altered zone qa is shifted to a

chosen value ranging from 0 to p.

Parameter Value

Fig. 6 shows the saturation prole at different altered contact

k (md) 1 angles qa for low (Fig. 6a) and high (Fig. 6b) capillary-viscous ratios.

f 0.05

Saturation decreases with contact angle in the wettability alter-

h 0.75

s (N/m) 0.08 ation zone (0 < X < b). Capillary pressure continuity is satised

C 0.5 across the boundary between wettability altered and unaltered

a 0.9 zones (X b), which yields discontinuous saturation at this

mg(Pa s) 1.5 105 boundary (Fig. 6). Increase of contact angle in the zone near to core

mw (Pa s) 1 103

outlet or fracture face (0 < X < b) of water-wet rocks results in

1578 S. Naik et al. / Journal of Natural Gas Science and Engineering 26 (2015) 1573e1584

Fig. 3. Effects of contact angle on steady-state two-phase ow: (a) fractional ow curve for water at different contact angles; (b) boundary saturations versus contact angle at

different fractional ow for water f0; (c) dimensionless total mobility versus saturation at different contact angles.

decrease of saturation near the exit, but the capillary end effect still optimal contact angle for the maximum gas productivity (Fig. 7a).

occurs in the unaltered zone (X > b). The optimal contact angle is around p/2.

Restriction of gas ow due to capillary end effect occurs at the However, the water ow is only restricted at high contact an-

exit (X 0) for low qa and at the boundary between wettability gles. As capillary-viscous ratio increases, the end effect increases. It

altered and unaltered zones (X b) for high qa (Fig. 6). Despite the reduces the water productivity at high contact angles (Fig. 7b).

intention of reducing the end effect by wettability alteration in the Water production from water-wet reservoirs can be maximised by

zone near to wellbore or fracture face, end effect at the altered zone altering the contact angle in wettability alteration zone (0 < X < b)

boundary can still occur. Higher capillary-viscous ratio results in towards the optimal value qa < p/2, and can be minimised by

larger end effect. It reduces the productivity for gas at both high and altering the contact angle towards qa p (Fig. 7b).

low contact angles (Fig. 7a). Contact angle alteration affects satu- The sensitivity of productivity index to the altered zone length b

ration in the altered zone signicantly (Fig. 6). In this zone the is investigated. PI for gas versus contact angle in the case of

competing capillary and viscous effects yield an intermediate piecewise-constant contact angle in linear ow is shown in Fig. 8a.

Fig. 4. Saturation proles for different contact angles in linear ow (f0 0.01): (a) c 0.1; (b) c 10.

S. Naik et al. / Journal of Natural Gas Science and Engineering 26 (2015) 1573e1584 1579

Fig. 5. Productivity index versus contact angle at different capillary-viscous ratios c in linear ow (f0 0.01): (a) gas; (b) water.

Fig. 6. Saturation prole at different altered contact angle qa in linear ow with piecewise alteration of contact angle (f0 0.01, b 0.5): (a) c 0.1; (b) c 10.

Fig. 7. Productivity index versus piecewise alteration of contact angle in water-wet rock at different c in linear ow (f0 0.01, b 0.5): (a) gas; (b) water.

The maximum value of PI for gas increases with b, since the vari- Increase of contact angle in the zone near to wellbore (0 < R < b) of

ation of saturation and relative permeability in altered zone occurs water-wet rocks results in decrease of saturation near the exit, but

over a larger distance with b. The increase of b from 0.005 to 0.5 the capillary end effect still occurs in the unaltered zone (R > b)

results in the increase of productivity index for gas from 0.1% to 7% (Fig. 9). The pressure drop across the altered zone near the wellbore

at qa p/2 (green curve in Fig. 8b). is much higher compared to that in linear case, causing the altered

Saturation proles in radial ow towards well at different qa for zone properties to be more inuential over the total ow in radial

low (Fig. 9a) and high (Fig. 9b) c are presented in the case of than in linear case. Therefore, PI prole is more sensitive to b in

piecewise alteration of contact angle. Saturation decreases with radial case than that in linear case (Fig. 10a). Higher b leads to

contact angle qa in the wettability alteration zone (0 < R < b). higher productivity index; the increase of b from 5 104 to 0.05

1580 S. Naik et al. / Journal of Natural Gas Science and Engineering 26 (2015) 1573e1584

Fig. 8. Productivity index for gas in the case of piecewise-contact angle in linear ow (f0 0.01, c 0.1): (a) PIg versus qa; (b) PIg versus b.

Fig. 9. Saturation proles at different altered contact angle qa in radial ow with piecewise alteration of contact angle (f0 0.01, b 5.0 104): (a) c 0.01; (b) c 0.5.

Fig. 10. Productivity index for gas in the case of piecewise contact angle in radial ow (f0 0.01, c 0.01): (a) PIg versus qa; (b) PIg versus b.

results in the increase of productivity index for gas from 1% to 9% at and Torsaeter, 2014). The adsorption of nanoparticles onto rock

qa p/2 (green curve in Fig. 10b). surface is dependent on DLVO forces exerting on NPs (Li and

Torsaeter, 2015), which indicates that different types of NPs and

6. Nanoparticle injection for wettability alteration in surface materials exhibit different interactions. The wettability

unconventional rocks alteration of rock surface is affected by the types of NPs and the

underlying mineralogy, leading to different resultant contact angles

Presently, nanoparticles (NPs) are extensively applied to based on their combinations.

wettability alteration of reservoir rocks for waterooding (Maghzi The main minerals present in unconventional coal seam and

et al., 2012; Giraldo et al., 2013; Ju and Fan, 2013; Hendraningrat shale reservoirs are sand, clays and organic matter. Karimi et al.

S. Naik et al. / Journal of Natural Gas Science and Engineering 26 (2015) 1573e1584 1581

contact angle from 90 to 0 . Li and Torsaeter (2015) used

hydrophilic silica NPs to decrease contact angle of sandstone rock The analytical model for steady-state ow is based on equality of

surface from 54 to 22 . Hendraningrat and Torsaeter (2014) the pressure and capillary-pressure gradients. The exact solution

investigated three metal oxides NPs: aluminium (Al2O3), titanium for steady-state ow with piecewise-constant contact angle is

(TiO2) and silicon (SiO2). Nano uids used in the experiment expressed by implicit analytical formulae, allowing for explicit ex-

with the concentration 0.05 wt% result in the contact angles pressions for the well productivity index.

measured in brine (NaCl 3 wt%) lowered from 53.68 to 28.6 , The saturation prole for steady-state ow in the rock with

21.64 and 38.82 for Al2O3, TiO2 and SiO2, respectively. To in- constant contact angle starts at the boundary value s0, which cor-

crease the contact angle, lipophilic polysilicon NPs with surface responds to zero capillary pressure, and tends to constant s0 as X

modication can be used (Ju and Fan, 2009). Hydrophobic and tends to innity, where the constant saturation is determined by

neutral wet polysilicon nano uids with concentration 4 g/L suc- fractional ow. Size of the end effect zone, where the signicant

cessfully altered sandstone contact angle from 50 to 84.56 and fraction of variation s0 s0 occurs, is equal to cL. Therefore, at low

98.12 , respectively (Roustaei et al., 2012). Cupric oxide (CuO) NPs capillary-viscous ratio, the saturation prole is almost uniform,

increased water contact angles from less than 5 up to 130 and while the saturation varies along the distance at high capillary-

150 with amine- and thiol-modied surfaces, respectively (Basu viscous ratio. In the rocks with piecewise-constant contact angle,

et al., 2011). the above proles are valid in constant-contact-angle intervals; the

Lebedeva and Fogden (2011) investigated the wettability alter- proles in different intervals are matched by the continuity con-

ation of kaolinite clay surface, due to the deposition of asphaltene dition for capillary pressure yielding the saturation jumps.

nanoparticles. The clay surface was altered from hydrophilic to A plane-parallel-ow analytical model is applied to the treat-

hydrophobic with contact angle 140 . ment of coreood data, and also to ow and saturation prediction

To the best of our knowledge, application of NPs to the wetta- near to the fractured well. The axi-symmetric model is applied for

bility alteration of organic matter has not been reported in the vertical and horizontal wells.

contemporary literature. Let us discuss the limitations of the analytical model (15). It

Let us estimate the effects of contact angle alteration by assumes gas is incompressible. Since the gas density is a function of

applying different NPs using the analytical model (Section 3) for pressure, the incompressibility assumption is valid if the pressure

the above-mentioned cases. Productivity index values versus drawdown is much less than the reservoir pressure:

contact angle are shown in Figs. 5 and 7 for f0 0.01. For low

water fractional ow f0 0.01 and the contact angle altered from DP

1 (18)

50 to 90 (Roustaei et al., 2012), c 5 and 0.05 result in PIg P

increase by 2.0 and 1.08 times, respectively. For high water frac-

For linear ow during coreooding, substitution of Darcy's law

tional ow f0 0.95 and the contact angle altered from 0 to 130

U k=mDP=L into the condition (18) yields:

(Basu et al., 2011), gas productivity index PIg can be increased by

4.7 and 1.25 times for high (c 5) and low (c 0.05) capillary- DP UmL

viscous ratios, respectively. Given f0 0.01 with contact angle 1 (19)

P kP

alteration from 0 to 50 (Vafaei et al., 2006), water productivity

index PIw is increased by 2.1 and 1.96 times for c 5 and 0.05, or

respectively.

Different types of NPs affect wettability of shale- or coal- kp

U (20)

minerals differently; hence, a mixture (cocktail) of NPs must be mL

developed for multi-mineral rocks. The corresponding technol-

For core length L 0.1 m, permeability k varying from 0.01 to 10

ogy includes adding nanoparticles that increase the resulting

md, gas viscosity m varying from 105 to 4 105 Pa s, and pressure

contact angle, into a drilling or fracturing uid. The contact angle

P varying from 5 105 to 107 Pa, the maximum velocity where the

increase yields the pore-scale deviation of water into larger

incompressibility assumption is valid varies from 1.2 105 to 1 m/

pores, increasing water rate during dewatering, and decreasing

s. The typical ow velocities in cores have the order of magnitude

water saturation to facilitate the gas ow in the future. Injection

106e105 m/s (Barenblatt et al., 1990; Bazin et al., 2010). There-

of nanoparticles, making the rock more water-wet, after the

fore, gas compressibility is negligible under the conditions of these

dewatering, leads to water displacement into smaller pores with

laboratory studies by Bazin et al. (2010), so the incompressibility

the consequent gas rate increase. The fast pore pressure decrease

assumption is valid.

in both scenarios results in more intensive gas desorption. To

For fractured wells, velocity at the fracture face is calculated using

summarise, the well treatment described above includes adding

Eq. (20). For permeability variation k 0.01e10 md, m 105e4

NPs into drilling or fracturing uids with the consequent in-

105 Pa s, reservoir pressure P 5 106e108 Pa, and length of linear

crease of the contact angle (rst stage); at the second stage,

ow region towards the fracture L 10e250 m, the maximum ve-

hydrophilic NPs are injected into the reservoir in order to

locity where the incompressibility assumption is valid varies from

decrease the contact angle after dewatering during gas produc-

5 108 to 0.1 m/s. Typical velocities at fracture faces in gas reser-

tion. Therefore, two different cocktails must be developed for

voirs have the order of 105e104 m/s (see Shaoul et al., 2012).

two stages. The delivery of nanoparticles is performed by the

Therefore, there are some unconventional eld cases where the

injection together with fast deteriorating gas-based foam

incompressibility assumption is valid, and where it is not. Inequality

(Farajzadeh et al., 2011).

(20) denes the conditions where the analytical model (15) is valid.

The analytical model for steady-state gas-water ow in rocks

For radial ow towards well, the rate of incompressible uid per

with piecewise-constant contact angle can be used for design of

unit height of reservoir (Q) is calculated as:

the treatment technology, including dening volumes of injected

uids and their compositions. The analytical model can also be

2pkDP

applied to the upscaling of laboratory data for well index Q (21)

prediction. mlnre =rw

1582 S. Naik et al. / Journal of Natural Gas Science and Engineering 26 (2015) 1573e1584

Substitution of rate expression in vertical well into the condition For rocks with permeability 1 md (corresponding dp 0.1 mm),

(18) yields: the calculated Kn 0.1 for gas pressure P 6.8 105 Pa. For rocks

with low permeability 0.01 md (dp 0.01 mm), the calculated Kn 0.1

2pkp for P 6.8 106 Pa. From this we conclude that the model neglecting

Q (22)

mlnre =rw Knudsen diffusion is applicable only for reservoirs at high reservoir

pressures (P > 6.8 105 Pa for 1 md, P > 6.8 106 Pa for 0.01 md).

For k 0.01e10 md, m 105e4 105 Pa s, P 107e108 Pa, The YoungeLaplace model for capillary pressure is valid if the

ratio between drainage and well radii re/rw 103e104, the disjoining pressure is negligible and surface tension is constant. It is

maximum rate per unit height, where the incompressibility inappropriate for pore/channel sizes in the range of nanometres. In

assumption is valid varies from 1.7 105 to 0.9 m2/s. Some typical this scale, applying the augmented YoungeLaplace equation, the

rates for tight gas reservoirs vary in the range 103e0.1 m2/s (see capillary pressure can be calculated as the sum of the traditional

McGowen et al., 2007). Thus, the gas compressibility can be ignored YoungeLaplace term and the disjoining pressure in the thin water

in some cases, and cannot be ignored in other cases. lm which can wet the surface. The disjoining pressure effects have

For the linear ow case in this paper, k 1 md, m 2 105 Pa s been used to describe the wetting lm behaviour of a meniscus in

and length of linear ow region is 103 m. For pressure varying from regions around 0.1 mm (Kelly, 2013). If pore sizes are above this size,

5 106 to 108 Pa, the maximum velocity where the incompressibility the YoungeLaplace equation can be used to calculate capillary

assumption is valid varies from 2.5 106 to 5 105 m/s. The ve- pressure. For the synthetic case study in Section 4, the average pore

locities in the synthetic case (Section 5) are in the order of 107 m/s, radius estimated from KozenyeCarman model is 0.3 mm. Therefore,

therefore the incompressibility assumption is valid. For the radial the capillary pressure calculated from YoungeLaplace model is valid.

ow regime, we use the same permeability, viscosity and pressure Good agreement between two saturation proles as obtained

range. For re/rw 6.67 103, the maximum rate per unit height from coreoods with different rates, and the prediction by the

where the incompressibility assumption is valid varies from 2 103 analytical model (15) with three tuning parameters C, a and q has

to 3.5 102 m2/s. The rates in the synthetic case have the order of been observed (Fig. 2). The two measured saturation proles can be

104 m2/s, therefore the incompressibility assumption is valid. approximated by exponential functions with three degrees of

In the case that the gas compressibility cannot be neglected, freedom each: s(X) A BeCX, giving the two curves six degrees of

such as for high rates or high pressure drawdowns, equation of freedom totally. Good match by tuning only three parameters val-

state for gas rg(Pg) is included in gas rate equation (10). The nal idates the analytical model.

expression for steady-state solution is a system of two ordinary A simplied model for relative permeability and capillary

differential equations with respect to pressure and saturation. pressure (Section 2) has been applied in this paper. The model

The gas slip effect can be signicant in porous media with low corresponds to pore-space geometry as a bundle of parallel capil-

pressure or low permeability (Wu and Pruess, 1998; Civan et al., lary with different wettability. In reality, natural rocks consist of

2011). According to gas permeability Equation (9), the slip effect numerous minerals with different wettability and grid-like pore-

is negligible if gas pressure is much larger than the Klinkenberg space geometry. One way around is the percolation-effective media

factor b. For typical permeability k 0.001 and 1 md, the value of model for the rocks with spotted micro heterogeneity in both

b is 3.3 105 and 2.8 105 Pa, respectively. A typical reservoir capillary pressure and relative permeability (Barenblatt et al., 1990;

pressures in coal seams are 106e107 Pa (Gaurav et al., 2012). The Bedrikovetsky, 1993).

corresponding error in calculated relative permeability for gas The obtained steady-state solutions provide the long-term as-

using (9) is 2e6%, if gas slip effect is ignored. In shallow gas res- ymptotics for non-steady state problems of immiscible Newtonian

ervoirs, pressure can be even lower, which leads to higher effect of and Now-Newtonian displacements, which is important informa-

gas slip. For shallow reservoirs or reservoirs with ultra-low tion for the transient analysis (Bedrikovetsky, 1993; Morrow and

permeability, the gas slip effect can cause rate miscalculation Buckley, 2011; Fogden et al., 2011). Since the long-term asymp-

around 10% if it is ignored. It can be especially prevalent in core- totic states are in thermodymanic equilibrium, the solutions are

ooding tests, where low pressures are commonly used. applicable for partly-miscible ows also (Farajzadeh et al., 2011;

Knudsen diffusion occurs where the gas with low enough Cinar and Riaz, 2014).

density ows in small enough capillaries, such that the gas particles Two scenarios are investigated for wateregas ow with well

collide more frequently with the solid walls than with other par- treatment by NPs. One scenario involves increasing contact angle

ticles (Bhatia et al., 2011). Knudsen diffusion can be evaluated by under high water saturation which corresponds to the processes of

the Knudsen number Kn, dened as the ratio between the mean dewatering and the fracturing-uid ow-back. The other scenario

free path of gas molecule and the pore size: involves decreasing contact angle under low water saturation,

which occurs during gas production.

l k T

Kn ; l p B (23) Based on the analytical model developed, the strategy of gas

dp 2Ppd2g productivity enhancement is as follows: For rocks with uniform

wettability, alter contact angle of rock surface towards p/2 for low

where l is mean free path of gas, dp is pore diameter, kB is Boltz- capillary-viscous ratio and p for high capillary-viscous ratio. For

mann constant (1.38 1023 J/K), T is gas temperature, dg is piecewise wettability alteration of water-wet rocks, optimal contact

diameter of gas molecule. Knudsen diffusion is negligible if Kn 0.1 angle is p/2 for both low and high capillary-viscous ratios. The

(He et al., 2014). We consider methane molecule diameter strategy targeting maximum water productivity is contact angle

4.14 1010 m and typical reservoir temperature 373 K. Pore sizes alteration towards p/2 for low capillary-viscous ratio and zero for

in coal seam reservoirs can range from a few angstroms to over a high capillary-viscous ratio.

micron. For pressure varying from 5 106 to 108 Pa, Kn in coal

seams can vary from 7 105 to 3. Pore sizes in shales range from 8. Conclusions

0.005 to 0.1 mm (Nelson, 2009). For pressure varying from 5 106 to

108 Pa, Kn in shale reservoirs can vary from 7 104 to 0.3. Analytical modelling and its application to enhancing the well

Therefore, Knudsen diffusion effect can be signicant in tight for- productivity for gas and water by wettability alteration allow

mation; however, it is not always the case. drawing the following conclusions:

S. Naik et al. / Journal of Natural Gas Science and Engineering 26 (2015) 1573e1584 1583

Competition between the viscous and capillary effects on gas Greek letters

productivity results in non-monotonic contact-angle-de- a capillary model parameter

pendency of gas productivity index, suggesting the existence of b Altered zone length

optimal contact angle for enhanced gas production. c capillary-viscous ratio

For dewatering and fracturing uid ow-back, competition be- h tortuosity index

tween the viscous and capillary effects on water productivity q surface contact angle

yields the optimal contact angle which maximises the produc- L dimensionless total mobility

tivity index for water. l mean free path of gas, m; total mobility, 1/Pa s

In the case of wettability alteration over the full length, the m viscosity, Pa s

higher the capillary-viscous ratio c, the higher the capillary end r density, kg/m3

effect, the larger the end effect zone and the higher the s interfacial tension, N/m

maximum productivity indexes for gas and water. f porosity

In the case of wettability alteration on a fraction of water-wet

rock, increase of the capillary-viscous ratio c leads to the in- Abbreviations

crease of the maximum productivity index for water and DLVO DerjaguineLandaueVerweyeOverbeek

decrease for gas. This is because the capillary end effect at the NP nanoparticle

boundary between wettability altered and unaltered zones re- PI productivity index

duces gas ow more severely with higher c.

The larger the altered zone length b, the higher the contact angle Subscripts

effect on gas and water productivity index. This is due to alter- a altered

ation of saturation prole and relative permeability over a larger e drainage radius

distance. g gas

Productivity index is sensitive to small value of b in radial ow w water (for productivity index and relative permeability),

towards well, indicating that wettability alteration in small wetting (for saturation), well (radius)

neighbourhood of wellbore can effectively improve productivity 0 exit

indexes for gas and water.

Wettability alteration in the range achieved by nanoparticles,

Superscripts

leads to 2.0e4.7 times increase of well productivity for gas and

0 inlet

twice increase for water.

The maximum productivity index for water should be applied

References

during dewatering and slick water removal. The maximum pro-

ductivity index for gas must be used during gas production stage. Adibhatla, B., Mohanty, K., Berger, P., Lee, C., 2006. Effect of surfactants on wetta-

bility of near-wellbore regions of gas reservoirs. J. Pet. Sci. Eng. 52 (1), 227e236.

Acknowledgements Anderson, W.G., 1987. Wettability literature survey part 5: the effects of wettability

on relative permeability. J. Pet. Technol. 39 (11), 1453e1468.

Arab, D., Pourafshary, P., 2013. Nanoparticles-assisted surface charge modication of

Financial support from the Australian Research Council (ARC) the porous medium to treat colloidal particles migration induced by low

Discovery and Linkage Projects is gratefully acknowledged. Authors salinity water ooding. Colloids Surf. A Physicochem. Eng. Asp. 436, 803e814.

Assef, Y., Arab, D., Pourafshary, P., 2014. Application of nanouid to control nes

thank the Editor and three anonymous reviewers for their com- migration to improve the performance of low salinity water ooding and

ments that greatly improved the manuscript. alkaline ooding. J. Pet. Sci. Eng. 124, 331e340.

Bahrami, H., et al., 2011. Effect of water blocking damage on ow efciency and

productivity in tight gas reservoirs. In: SPE-142283. SPE Production and Oper-

Nomenclature

ations Symposium. Oklahoma, USA. 27e29 March.

Barenblatt, G.I., Entov, V.M., Ryzhik, V.M., 1990. Theory of Fluid Flows through

b Klinkenberg factor, Pa Natural Rocks. Kluwer Academic Publishers, Dordrecht.

Basu, M., Sinha, A.K., Pradhan, M., Sarkar, S., Negishi, Y., Pal, T., 2011. Fabrication and

C capillary model parameter

functionalization of CuO for tuning superhydrophobic thin lm and cotton

dg diameter of gas molecule, m wool. J. Phys. Chem. C 115, 20953e20963.

dp pore diameter, m Bazin, B., et al., 2010. In-situ water-blocking measurements and interpretation

f fractional ow function related to fracturing operations in tight gas reservoirs. SPE Prod. Oper. 25 (4),

431e437.

k permeability, m2 Bedrikovetsky, P.G., 1993. Mathematical Theory of Oil & Gas Recovery (With Ap-

kB Boltzmann constant, J/K plications to Ex-USSR Oil & Gas Condensate Fields). Kluwer Academic Pub-

kr relative permeability lishers, London-Boston-Dordrecht.

Bhatia, S.K., Bonilla, M.R., Nicholson, D., 2011. Molecular transport in nanopores: a

P pressure, Pa theoretical perspective. Phys. Chem. Chem. Phys. 13 (34), 15350e15383.

Pc capillary pressure, Pa Bradford, S.A., Abriola, L.M., Leij, F.J., 1997. Wettability effects on two- and three-

R dimensionless radius uid relative permeabilities. J. Contam. Hydrol. 28 (1e2), 171e191.

Bruining, J., Van Duijn, C.J., 2006. Traveling waves in a nite condensation rate

r radius, m model for steam injection. Comput. Geosci. 10 (4), 373e387.

S normalised water saturation Bryant, E., Bowman, R., Buckley, J., 2006. Wetting alteration of mica surfaces with

s water saturation polyethoxylated amine surfactants. J. Pet. Sci. Eng. 52 (1), 244e252.

Chalk, P., Gooding, N., Hutten, S., You, Z., Bedrikovetsky, P., 2012. Pore size distri-

sgr irreducible gas saturation bution from challenge coreood testing by colloidal ow. Chem. Eng. Res. Des.

swc connate water saturation 90, 63e77.

U total velocity, m/s Chaouche, M., Rakotomalala, N., Salin, D., Yortsos, Y., 1993. Capillary effects in

immiscible ows in heterogeneous porous media. EPL Europhys. Lett. 21 (1), 19.

u ow velocity of water or gas, m/s

Chen, D., Pan, Z., Liu, J., Connell, L.D., 2013. An improved relative permeability model

x distance, m for coal reservoirs. Int. J. Coal Geol. 109, 45e57.

T gas temperature, K Chen, P., Mohanty, K.K., 2014. Wettability alteration in high temperature carbonate

y0 ow exit distance, m reservoirs. In: SPE-169125. SPE Improved Oil Recovery Symposium, Tulsa,

Oklahoma, USA, 12e16 April.

y1 treated distance, m Cinar, Y., Riaz, A., 2014. Carbon dioxide sequestration in saline formations: part 2-

y2 ow entrance distance, m review of multiphase ow modeling. J. Pet. Sci. Eng. 124, 381e398.

1584 S. Naik et al. / Journal of Natural Gas Science and Engineering 26 (2015) 1573e1584

Civan, F., Rai, C.S., Sondergeld, C.H., 2011. Shale-gas permeability and diffusivity Mahadevan, J., Le, D., Hoang, H., 2009. Impact of capillary suction on fracture face

inferred by improved formulation of relevant retention and transport mecha- skin evolution in waterblocked wells. In: SPE-119585. SPE Hydraulic Fracturing

nisms. Transp. Porous Media 86 (3), 925e944. Technology Conference. The Woodlands, Texas, USA, 19e21 January.

Economides, M.J., Wood, D.A., 2009. The state of natural gas. J. Nat. Gas Sci. Eng. 1 Mahadevan, J., Sharma, M.M., Yortsos, Y.C., 2007. Capillary wicking in gas wells. SPE

(1), 1e13. J. 12 (4), 429e437.

Farajzadeh, R., Muruganathan, R.M., Rossen, W.R., Krastev, R., 2011. Effect of gas type McCaffery, F., Bennion, D., 1974. The effect of wettability on two-phase relative

on foam lm permeability and its implications for foam ow in porous media. permeabilities. J. Can. Pet. Technol. 13 (4), 42e53.

Adv. Colloid Interface Sci. 168 (1e2), 71e78. McGowen, J.M., Gilbert, J.V., Samari, E., 2007. Hydraulic fracturing down under. In:

Fogden, A., Kumar, M., Morrow, N.R., Buckley, J.S., 2011. Mobilization of ne particles SPE-106051. SPE Hydraulic Fracturing Technology Conference. College Station,

during ooding of sandstones and possible relations to enhanced oil recovery. Texas, USA, 29e31 January.

Energy Fuels 25 (4), 1605e1616. Morrow, N., Buckley, J., 2011. Improved oil recovery by low-salinity waterooding.

Gaurav, K., Akbar, A., Saada, T.H., Kumar, S., 2012. Performance analysis in coal seam J. Pet. Technol. 63 (5), 106e113.

gas. In: SPE-157696. SPETT 2012 Energy Conference and Exhibition, Port-of- Morsy, S., Gomaa, A., Sheng, J., 2014. Improvement of Mancos shale oil recovery by

Spain, Trinidad. 11e13 June. wettability alteration and mineral dissolution. In: SPE-169033. SPE Improved

Giraldo, J., Benjumea, P., Lopera, S., Cortes, F.B., Ruiz, M.A., 2013. Wettability alter- Oil Recovery Symposium, Society of Petroleum Engineers, Tulsa, Oklahoma,

ation of sandstone cores by alumina-based nanouids. Energy Fuels 27 (7), USA, 12e16 April.

3659e3665. Nasralla, R.A., Bataweel, M.A., Nasr-El-Din, H.A., 2013. Investigation of wettability

He, W., Lv, W., Dickerson, J.H., 2014. Gas diffusion mechanisms and models. In: Gas alteration and oil-recovery improvement by low-salinity water in sandstone

Transport in Solid Oxide Fuel Cells. Springer, pp. 9e17. rock. J. Can. Pet. Technol. 52 (2), 144e154.

Hendraningrat, L., Li, S., Torster, O., 2013. A coreood investigation of nanouid Nelson, P.H., 2009. Pore-throat sizes in sandstones, tight sandstones, and shales.

enhanced oil recovery. J. Pet. Sci. Eng. 111, 128e138. AAPG Bull. 93 (3), 329e340.

Hendraningrat, L., Torsaeter, O., 2014. Unlocking the potential of metal oxides O'Carroll, D.M., Abriola, L.M., Polityka, C.A., Bradford, S.A., Demond, A.H., 2005.

nanoparticles to enhance the oil recovery. In: OTC-24696, Offshore Technology Prediction of two-phase capillary pressureesaturation relationships in frac-

Conference Asia, Kuala Lumpur, Malaysia, 25e28 March. tional wettability systems. J. Contam. Hydrol. 77 (4), 247e270.

Huang, D.D., Honarpour, M.M., 1998. Capillary end effects in coreood calculations. Odumabo, S.M., Karpyn, Z., Ayala, H.,L., 2014. Investigation of gas ow hindrance

J. Pet. Sci. Eng. 19 (1), 103e117. due to fracturing uid leakoff in low permeability sandstones. J. Nat. Gas Sci.

Jones, F.O., Owens, W.W., 1980. A laboratory study of low-permeability gas sands. Eng. 17, 1e12.

J. Pet. Technol. 32 (9), 1631e1640. Roustaei, A., Moghadasi, J., Bagherzadeh, H., Shahrabadi, A., 2012. An experimental

Ju, B., Fan, T., 2009. Experimental study and mathematical model of nanoparticle investigation of polysilicon nanoparticles' recovery efciencies through changes

transport in porous media. Powder Technol. 192, 195e202. in interfacial tension and wettability alteration. In: SPE-156976. SPE Interna-

Ju, B., Fan, T., 2013. Experimental study on nanoparticles transport and its effects on tional Oileld Nanotechnology Conference, Noordwijk, The Netherlands, 12e14

two-phase ow behavior in porous networks. Part. Sci. Technol. 31 (2), 114e118. June 2012.

Karimi, A., Fakhroueian, Z., Bahramian, A., Pour Khiabani, N., Darabad, J.B., Azin, R., Shaoul, J.R., Spitzer, W., Dahan, M., 2012. Case study of unconventional gas-well

Arya, S., 2012. Wettability alteration in carbonates using zirconium oxide fracturing in Hungary. SPE Prod. Oper. 27 (2), 145e157.

nanouids: EOR implications. Energy Fuels 26 (2), 1028e1036. Shapiro, A.A., Wesselingh, J.A., 2008. Gas transport in tight porous media: gas ki-

Kathel, P., Mohanty, K.K., 2013. Wettability alteration in a tight oil reservoir. Energy netic approach. Chem. Eng. J. 142 (1), 14e22.

Fuels 27 (11), 6460e6468. Vafaei, S., Borca-Tasciuc, T., Podowski, M.Z., Purkayastha, A., Ramanath, G.,

Kelly, S., 2013. Experimental investigation of the inuence of molecular surface Ajayan, P.M., 2006. Effect of nanoparticles on sessile droplet contact angle.

interactions on imbibition in shale nano-pore proxies. In: SPE-167635. SPE Nanotechnology 17 (10), 2523e2527.

Annual Technical Conference and Exhibition, New Orleans, Louisiana, USA, 30 Van Duijn, C.J., Molenaar, J., De Neef, M.J., 1995. The effect of capillary forces on

September-2 October. immiscible two-phase ow in heterogeneous porous media. Transp. Porous

Kumar, K., Dao, E.K., Mohanty, K.K., 2008. Atomic force microscopy study of Media 21 (1), 71e93.

wettability alteration by surfactants. SPE J. 13 (2), 137e145. Van Lingen, P.P., Bruining, J., Van Kruijsdijk, C.P.J.W., 1996. Capillary entrapment

Lebedeva, E.V., Fogden, A., 2011. Wettability alteration of kaolinite exposed to crude caused by small-scale wettability heterogeneities. SPE Reserv. Eng. 11 (2),

oil in salt solutions. Colloids Surf. A Physicochem. Eng. Asp. 377 (1), 115e122. 93e99.

Li, G., Ren, W., Meng, Y., Wang, C., Wei, N., 2014. Micro-ow kinetics research on Wang, J.Y., Holditch, S.A., McVay, D.A., 2012. Effect of gel damage on fracture uid

water invasion in tight sandstone reservoirs. J. Nat. Gas Sci. Eng. 20, 184e191. cleanup and long-term recovery in tight gas reservoirs. J. Nat. Gas Sci. Eng. 9,

Li, S., Torsaeter, O., 2015. The impact of nanoparticles adsorption and transport on 108e118.

wettability alteration of intermediate wet berea sandstone. In: SPE-172943, SPE Wu, Y.-S., Pruess, K., 1998. Gas ow in porous media with Klinkenberg effects.

Middle East Unconventional Resources Conference and Exhibition, Muscat, Transp. Porous Media 32 (1), 117e137.

Oman, 26e28 January. Xie, X., Liu, Y., Sharma, M., Weiss, W.W., 2009. Wettability alteration to increase

Maghzi, A., Mohammadi, S., Ghazanfari, M.H., Kharrat, R., Masihi, M., 2012. Moni- deliverability of gas production wells. J. Nat. Gas Sci. Eng. 1 (1), 39e45.

toring wettability alteration by silica nanoparticles during water ooding to You, Z., Badalyan, A., Bedrikovetsky, P., 2013. Size-exclusion colloidal transport in

heavy oils in ve-spot systems: a pore-level investigation. Exp. Therm. Fluid Sci. porous media e stochastic modeling and experimental study. SPE J. 18,

40, 168e176. 620e633.