SPE-184574-MS How Do Nanoparticles Stabilize Shale

SPE-184574-MS How Do Nanoparticles Stabilize Shale? Besmir Buranaj Hoxha, Eric van Oort, and Hugh Daigle, The University of Texas Copyright 2017, SPE International Conference on Oilfield Chemistry This paper was prepared for presentation at the SPE International Conference on Oilfield Chemistry held in Montgomery, Texas, USA, 3–5 April 2017. This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract The operational use of nanoparticles (NPs) in drilling and completion fluids is still limited at the present time, in part due to lack of consistent evidence for - and clarification of - NP interactions with rock formations, formation fluid, and other fluid additives. For instance, previous fluids research has emphasized that NPs bring about "pore plugging" that reduces pressure transmission, and in turn fluid inflow, into the shale pore matrix which ultimately helps stabilize the borehole. However, it is difficult to understand how pore plugging might be accomplished in the absence of any considerable filtration in shales considering the very low permeability of shales does not allow for any appreciable Darcy flow. This paper addresses the crucial question: "how, when, why do nanoparticles plug up shale pore throats?" Zeta Potential (ZP) measurements were carried out on the aqueous dispersions (NPs) and on intact shale thin sections exposed to the nanofluid in order to determine the degree of interaction behavior between NPs and shales. The experimental data was then used to calculate DLVO curves (describes the force between charged surfaces interacting through a liquid medium) in order to determine if the total potential energy was sufficient for NP's to diffuse through the repulsive barrier and attract (or overcome repulsion) to the shale surface. Estimated DLVO curves are used to demonstrate the NP's ability to contribute to borehole stability but are not directly correlated, and therefore, NP effects on shale stability were studied in detail using pore pressure transmission tests (PTT), which measure fluid pressure penetration in shales, and modified Thick Wall Collapse (TWC) tests, which explore the influence of NPs on the collapse pressure of shale samples. Our investigation shows that NPs can reduce fluid pressure penetration and delay borehole collapse in shales, but only under certain conditions. Electrostatic and electrodynamic interaction between NP's and shale surfaces, governed by DLVO forces, is the main mechanism that will lead to pore throat plugging, reducing pressure transmission, which in turn benefits borehole stability by slowing down near-wellbore pore-pressure elevation and effective stress reduction. For Mancos shale, it was shown that 20 nm nanosilica (anionic) are effective in partially plugging the pore throat system, depending on the pH of the nanofluid, which affects the surface potential and ZP of both NPs and shale. Furthermore, the positively charged nanosilica (cationic) showed better results for pore-plugging capabilities than the anionic nanosilica. The findings lead to some interesting challenges for the practical field application of NP-based drilling fluids for borehole stability, given that efficacy will depend on the specific type of shale, the specific type, size and concentration of NP, the interaction between NP-shale, and external factors such as pH,

2

SPE-184574-MS

salinity, temperature etc. NP use for practical shale stabilization therefore requires a dedicated, thoroughly engineered solution for each particular field application, and is unlikely to be "one size fits all".

Introduction The past decade has seen a great deal of interest in potential application of NPs in the oil and gas industry. Most use of NPs has focused on enhanced oil recovery (Rodriquez et al., 2009; Ahmadi et al., 2011; Yu et al., 2015), but increased focus on NP applications in drilling and completion fluids include:

• • • • •

Applications such as foaming / defoaming agents, gelation / thickening materials, and being incorporated in non-aqueous fluids to stabilize water-in-oil emulsions (Binks, 2002; Dai et al., 2015; Xue et al., 2015; Borisov et al., 2015). Use as fluid loss control and filter cake additives (Hussein, 2012; Amanullah et al., 2011; Hoelscher et al., 2013; Riley et al., 2012; Zakaria et al., 2012; Contreras et al., 2014A). Possible use in artificial wellbore strengthening (Contreras et al., 2014B&C; Nwaoji et al., 2013). Additives to promote wellbore stability in shales (Sensoy et al., 2009; Friedheim et al., 2011; Cai et al., 2012; Ji et al., 2012; Sharma et al., 2012; Ming Jung et al., 2013; Gao et al., 2016). Lubricity: Taraghikhah et al., 2015 and Riley et al., 2012.

The work presented herein concentrates on improved wellbore stability through the use of NP as agents in drilling fluid formulations. There are several concerns with the historical work published on this topic:

•

•

•

In the work by Sensoy et al. (2009), non-trivial reductions in pressure transmission rates were observed with NP laden fluids. However, sizeable reductions in pressure transmission rate compared to brine, sometimes by more than an order of magnitude, were also already seen with nothing but the simple water-based base muds tested prior to the NP laden fluids. This would imply that even without NP addition such simple muds would already have extra-ordinary shalestabilizing qualities all by themselves, a conclusion not supported by evidence in the field and the ongoing quest to find effective high-performance water-based muds. In fact, van Oort, showed that simple base muds should not show any reduction in pressure transmission rate. Therefore, the results by Sensoy et al. (2009), may have been compromised by the test artifacts (e.g. leaking fluid outside of the shale test plug, which were not actively confined during testing), which raises concerns about the validity of the NP results. High concentration of NPs, generally higher than 5% by weight, were used to get significant beneficial effects (Sensoy et al., 2009; Gao et al., 2016; Cai et al., 2012). Such high concentrations significantly raise the fluid cost, making them economically less viable. Moreover, handling and mixing such large amounts of NP would also create logistical difficulties and raise issues on operational feasibility. Furthermore, higher concentration of NP's in solution will reduce inter particle spacing and increase aggregation kinetics (i.e., increase rate of aggregation/agglomeration) and reduce colloidal stability (Metin et al., 2014). Additionally, since the particle size is below 100 nm, the particles begin to demonstrate a considerable amount of attractive force towards each other, especially more so at higher concentrations which results in unnecessary increase to the nanofluid viscosity (Metin et al., 2013). The mechanism by which NPs reduce pressure transmission to benefit shale stability has been described as "pore plugging", based primarily on an assessment of the average NP size compared to shale pore throats. However, many factors contribute to the NPs ability to attract and adhere to shale pore throats, as will be explained carefully in this publication. Furthermore, no actual mechanism nor explanation for such "pore plugging" has been proposed. Filtration of NPs across shale surfaces

SPE-184574-MS

3

is unlikely to be a significant mechanism, given the very low permeability of most shales (as low as 0.1 nD = 10−10 D = 10−22 m2 in the experiments by e.g. Sensoy et al., 2009, Ji et al., 2012). In this paper, it is shown that the adsorption kinetics and principles of NPs interacting with shale surfaces have not yet been properly explained in order to understand the effect of NPs on shale stability. In particular, the force interaction between charged rock surfaces and NPs, needs to be addressed.

Theory DLVO Theory - Particles of the Same Material NP adsorption onto surfaces has been studied widely for many years in other scientific fields, such as the biotechnology and biomedical industries. NPs exhibit phenomenal surface-active properties with their high surface-area-to-volume ratio, allowing for their customized attachment to compounds which can be utilized for "fit-for-purpose" applications. In our case, we are interested in the electrostatic and electrokinetic interactions between NPs and shale surfaces at the fluid interface. Fig. 1 gives a schematic overview of the factors that need to be considered: (1) surface charge density and surface potential of both the shale and the NPs, resulting in attractive and/or repulsive forces which will determine whether NPs can approach and adhere to the shale surface (2) repulsive and attractive forces between NPs themselves, which determine whether individual NPs will remain dispersed and/or suspended or as aggregated/agglomerated clusters floating/ settling (Israelachvili,1992; Elimelech et al., 1995).

Figure 1—Diagram of electrostatic interactions between charged NPs and shale. Adherence to the shale surface will result from many factors, one of which is the nanoparticle surface charge orientation and magnitude, causing interactions to oppositely charged (or depending on the differential charge) surface of the shale (influenced by the minerals/ clays contributing to the overall shale surface charge distribution). The overall surface charge distribution of the shale is dependent not only on the rock/mineral matrix, but also on the pore matrix that contributes to the heterogeneous surface charge distribution (the surface has many hydrophobic spaces characterized by no charge and/or hydrophilic spaces characterized by charge). The adsorption kinetics is dependent on this factor aforementioned as well as on the particle concentration, distance of particles away from each other and the shale surface and shale properties, such as but not limited to, CEC, TOC, and catalytic activity of the clay surface. The reader can refer to Celik et al., 2004 for more information on electrokinetic behavior of clay surface and shales.

Colloids or NPs in dispersion should experience a variety of attractive and repulsive forces due to the nature of the particles themselves as well as the properties of the nanofluid. In most literature, interparticle interactions are typically expressed in terms of potential Φ, which is the reversible work required

4

SPE-184574-MS

to bring two particles together to an arbitrary distance (Berg, 2010). In the classic DLVO theory, developed by Derjaguin and Landau (1941) and Verwey and Overbeek (1948), Φ is represented as the sum of two components, van der Waals interactions (ΦVDW) and electrostatic interactions (ΦE): (1) For spherical particles of radius a at a minimum distance S0, the van der Waals interaction potential is given by: (2) where A is the Hamaker constant. For a spherical particle and a flat surface, the van der Waals interaction potential is (3) In the case of particles with the same surface charges, or a particle and a flat surface with the same surface charge, the electrostatic interaction potential is given by: (4) for spherical particles of radius a, and, (5) for a spherical particle of radius a and a flat surface. Here, n∞ is the ion concentration in solution far away from the solid surfaces, z is the charge number of the electrolyte, e is the elementary charge, ψδ is the electrostatic potential on the solid surfaces, ε is the relative dielectric permittivity of the dispersion fluid, ε0 is the vacuum dielectric permittivity, k is Boltzmann's constant, and T is the absolute temperature. The ion concentration n∞ is related to the electrolyte concentration C in mol/L by n∞ = 1000NavC were Nav is Avogadro's number. The interested reader is referred to Berg (2010) for a more thorough development and discussion of Eqs. 2-5. Eqs. 2 through 5 can be combined to show how the interaction potential varies as a function of particle distance and dispersion salinity. Let us consider the situation of pure silica NPs (a = 5 nm, ψδ = −20 mV) in an aqueous NaCl solution (z = 1) at T = 20°C = 293 K. For the other parameters we will assume k = 1.381 × 10−23 J/K, ε0 = 8.854 × 10−12 F/m, e = 1.602 × 10−19 C, NAv = 6.022 × 1023 1/mol, and A = 8.33 × 1021 J (Hough and White, 1980). The relative dielectric permittivity of water was adjusted for salinity using the formulation of Gavish and Promislow (2016). Here, we do not account for variations in the Hamaker constant with salinity, though increasing electrolyte concentration can reduce the Hamaker constant somewhat (Berg, 2010). The results are shown in Figure 2 for the corresponding, calculated, DLVO forces vs S0 (separation distance). Positive values of Φ correspond to repulsion while negative values correspond to attraction. The potentials are generally repulsive at low salinity and move to more attractive at higher salinities. Repulsion is more pronounced between spheres and planes than between sphere-sphere, as should be expected by analogy with point charge sources versus planar charge sources (Morse and Feshbach, 1953). Note that

SPE-184574-MS

5

attractive interactions generally increase in importance as salinity increases. This effect can result in particle aggregation with elevated salinity.

Figure 2—(a) Sphere-sphere interaction potentials for salinity of 10 mM. (b) Sphere-sphere interaction potential for different salinities. (c) Sphere-plane interaction potential for different salinities. (d) Comparison of sphere-sphere and sphere-plane potentials at salinity of 10 mM. (e) Same as (d), but at salinity of 1000 mM. Note, at high amounts of salinity, the potentials are relatively similar.

6

SPE-184574-MS

DLVO Theory - Particles of Different Materials While the version of DLVO theory discussed above can be used to describe stability of nanoparticle dispersions and their attraction/ repulsion from a surface of the same composition, adsorption on shale surfaces requires a description of the potentials existing between NPs and substrates composed of different materials - specifically due to shales heterogenous composition. The van der Waals interactions are still described by Eq. 3, although a different Hamaker constant must be used to describe the interaction between the nanoparticle material and the shale substrate material. Shales tend to be predominantly composed of silicate, aluminosilicate, and carbonate minerals, and the variation in Hamaker constant for interactions among these materials through water is roughly a factor of 2 or 3, with the attraction between silica NPs and a silica substrate being the weakest interaction while interactions with calcite are the strongest (e.g., Hough and White, 1980; Bergström 1997). However, electrostatic interactions are more complicated. The following derivation is due to Hogg et al., 1966. Consider two flat plates separated by a distance D (Fig. 3). The electrical potential ψ between the plates must obey the Debye-Hückel equation: (6)

Figure 3—Electrical potential between two plates with different surface potentials.

where x is the orthogonal coordinate pointing from the surface of one plate into the space between the plates. Eq. 6 may be solved using the boundary conditions that ψ(0) = ψ1 and ψ(D) = ψ2. This yields:

(7)

The interaction potential between the plates due to the presence of the charged surfaces and associated electrical double layer is obtained by considering the difference in free energy required to form the partially overlapping electrical double layers between the plates (ΔGD) and the free energy required to form the individual double layers on the plates in isolation (ΔG∞1 and ΔG∞2): (8) For low surface potential (tens of mV), the free energy required to form an electrical double layer with potential ψ is −σψ/2, where σ is the surface charge density given by:

SPE-184574-MS

7

(9) The subscript "surface" refers to evaluating the derivative at the solid surface being considered. The free energies ΔG∞1 and ΔG∞2 are thus given as: (10) ΔGD is (11) where σ1 and σ2 are given by

(12)

(13)

Putting Eqs. 10-13 together with Eq. 8 yields the expression for interaction potential between two plates:

(14)

Because Eq. 14 expresses the interaction potential between two flat plates, it is of limited utility for considering nanoparticle interactions with shale surfaces. However, Derjaguin (1934) showed that the interaction potential ΦE, spheres between two spheres with different radii a1 and a2 may be determined from the interaction potential between two plates ΦE, plates with the same surface properties as the spheres: (15) where S0 is the distance of closest approach between the spheres and ξ is a variable of integration. Substituting ξ for D in Eq. 14 and carrying out the integration yields

8

SPE-184574-MS

(16)

To modify Eq. 16 for the case of a sphere interacting with a plate, we set a1 = a and a2 = ∞, which yields

(17)

This is the electrostatic interaction potential between a sphere with radius a and surface electrical potential ψ1 and a plate with surface electrical potential ψ2. The total potential (ΦVDW + ΦE) is plotted in Figure 4 for a nanoparticle interacting with a planar surface in 0.1 μM and 550 mM NaCl brine. These calculations use the same parameters as Figure 2 but different electrical potential on the planar surface. In the case of fresh water (Figure 4a), if the nanoparticle has a fixed charge of −20 mV, as the charge difference between the plane and nanoparticle increases, the total potential becomes less negative, indicating a reduction in attraction (or increase in repulsion). Additionally, this trend stalls or slightly reverses when the charge difference gets very large. However, the behavior, with any particular combination of charges, depends on what happens when the other terms in Eq.17 come into effect - this example is simplified to a case where the only variable is the charge of the planar surface. For example, the phenomenon, among other actors, is highly dependent on salinity, nonetheless. In Figure 4b, near seawater salinity, there is little difference in total potential as the charge difference increases.

Figure 4—DLVO potential between a silica nanoparticle and planar surface with different values of plane surface charge. (a) Aqueous dispersion salinity of 0.1 μM (0.0001mM). (b) Aqueous dispersion salinity of 550 mM. Note, the nanoparticle has a fixed charge of −20 mV, so as the charge on the plane becomes less negative the charge difference increases, indicating a reduction in attraction

SPE-184574-MS

9

Extended DLVO Theory The discussion up to this point has focused only on van der Waals and electrostatic interactions as described in classical DLVO theory. However, many studies have shown that other interactions need to be considered to fully understand interparticle interactions. These interactions include effects due to charge distribution, polarizability, ion adsorption, and pH (Yotsumoto and Yoon, 1993; Hoek and Agarwal, 2006; Dahirel and Jardat, 2010). While salinity and pH are probably the most important effects, a fundamental understanding of nanoparticle-shale interactions requires characterization of all these effects. In terms of the downhole environment, the properties that can be controlled are salinity and pH of drilling fluid, and nanoparticle surface chemistry through appropriate functionalization. These effects can best be understood through altering the zeta potential of the nanoparticle. Most NPs will acquire an electrical surface charge in aqueous solution. The surface chemistry of a nanoparticle is regulated by the surface charge density and the zeta potential of the colloid. It is possible to change the surface chemistry by altering the pH of the nanoparticle dispersion. If alkali (OH-) is added to a negatively charged NP dispersion, the NPs will acquire a more negative charge. In contrast, if acid (H+) is added, the nanoparticle ZP will near zero charge and possibly neutralize the surface charge (relatively non-ionic). If further acid is introduced, a buildup of positive cations (H+) will tend to accumulate around the surface of the NP and possibly make the NP positively charged. The behavior of NPs in drilling fluid must therefore be tuned in accordance with the characteristics of the shale formation being drilled through (mineralogy, native fluid composition, temperature etc.).

Experimental Sample Preparation & Shale Preservation Shale preservation and proper sample preparation is of essential importance when performing shale-fluid interaction testing. Schmitt et al. (1994) indicate that dehydration and air entrapment are the main concerns in sample preservation. van Oort et al. (2016) further clarified these concerns, describing that the native pore fluid of a poorly preserved shale evaporates from the pore space, which then fills with air. Subsequent air entrapment during shale-fluid compatibility testing can cause swelling test artifacts that should be avoided at all costs. Additionally, shale dehydration can cause cracking and reduce the stiffness/strength of the sample. Therefore, fluid-shale interaction studies require that rigorous shale preservation and sample preparations conditions need to be implemented in order to achieve realistic results that are representative of field conditions. Mancos shale was selected for testing, due to the fact that its composition does not vary as much as others shales (thus sample plugs are more replicable) with a low reactivity to water (CEC value of 6 meq/1bbg), allowing relative ease of performing borehole stability experimentation. Figure 5 shows the outcome of a series of shale characterization tests on shale mineralogy, native water activity, fluid content, pore size distribution, and physical properties for the Mancos shale tested in this study.

10

SPE-184574-MS

Figure 5—Mineralogy (via X-Ray Diffraction) and properties of outcrop Mancos shale sample.

The Mancos shale samples were stored in a controlled moisture environment (i.e. a desiccator). This technique ensures that the shale samples are reconstituted and hydrated to their equivalent water activity (Aw) conditions and preserved to their native in-situ water content. The Mancos shale samples were hydrated for 2 weeks prior to testing in a desiccator with 3.5%wt artificial seawater (Aw = 0.93) to achieve their inherent Aw of 0.87 – 0.93, which was found using a hygrometer and the adsorption isotherm technique. The permeability of Mancos has previously been reported up to 10nD (Sarker et al., 2010) and as low as 5nD (van Oort, 1994). TOC (total organic carbon) measurement was performed by LECO tests and displayed low values (1.3-1.5 wt%), and thus this factor does not directly impact the nanoparticle ZP value (it has been reported that nanoparticle adhesion might be adversely affected by the presence of organic matter since it presumably has neutral ZP, while most natural occurring minerals are typically negatively charged). NP Fluids Proper selection and testing of fluid additives leads to compatible synergies and ultimately enhance fluid properties and mud performance, which in turn will reduce overall drilling fluids cost. The experimental selections proposed in this paper are pragmatic and emphasize the use of fluid formulations that are more feasible for practical applications. For this reason, a 5% wt product concentration was selected as it compares to similar product concentration used for most wellbore stability additives (such as polyols, silicates, amines etc.). Additionally, high nanoparticle concentration is not desirable as it is inversely proportional to dispersion stability, by reducing inter-particle repulsion forces and accelerating aggregation kinetics. On the other hand, if the nanofluid is too dilute, it will have "spotty" NP concentration (NP colloids suspended at different sections of the nanofluid i.e., top vs. bottom) and provide poor nanofluid performance with inadequate wellbore stability. The nanosilica tested (see Table 1) were custom made by 3M specifically for this project and were intended for a "fit for purpose" design. The particle size was designed and selected to be 20 nm based on pore

SPE-184574-MS

11

throat characterization of Mancos shale via Mercury Injection Capillary Pressure (MICP) experimentation, which indicated that the pore size distribution of the Mancos shale samples were predominantly in the range of 5-50 nm. Nanosilica was chosen as the base core due to the wide range of use of silica, it is readily available, economically favorable, and ability be dispersed in polar solvents (water, ethanol, and other polyols). Additionally, our research has shown that nanosilica can be charged (from naturally anionic charge to a cationic charge) via various techniques undertaken by the manufacturer. Thus, allowing the capability to test the same base core at different charge orientation. Table 1—Nanoparticle design and properties. A Malvern Zetasizer dynamic light scattering (DLS) instrument was used to measure the particle size and the zeta potential of the NPs. For the particle size, the Z-average (d, nm) value was reported based on the ISO standard guideline for measuring particle size with a DLS machine for intensity-based measurements. The NP characterization measurements were additionally supported by using an Anton Paar Litesizer 500 DLS machine. Note, the modified nanosilica at pH 8.7 is the original nanofluid - one sample was kept at the original pH value while the other two samples (pH 11.9 and pH 2.5) were subjected to pH alteration via 1N NaOH and 0.2N H2SO4 and accordingly the ZP is altered. Note the custom-made nanosilica are amorphous (non-crystalline), non-porous and all made from the same batch of silica. Furthermore, the custom-made surface modified nanosilica has an acid compound pointing away from the nanoparticle surface (i.e. into the suspension) and not attached to the silica nanoparticle - this would be the stabilizer. The cationic nanosilica has no surface modifiers and the stabilizing ion has been exchanged, i.e stabilization has been achieved by electrostatics, not steric stabilization. Nanofluid (5 %wt)

Stabilizer

Surface Charge

Surface Modified?

Particle Size (nm)

Zeta Potential (mV)

pH

Bare Nanosilica (-)

Base/(Na+)

(-)

No

16.5

−34.1

9.7

Modified Nanosilica (pH = 8.7) - original

Acid

(-)

Yes

19.0

−31.0

8.7

Modified Nanosilica (pH = 11.9)

Ac id

(-)

Yes

23.0

−44.3

11.9

Modified Nanosilica (pH = 2.5)

Ac id

(-) approaching isoelectric point

Yes

35.3

−26.7

2.5

Bare Nanosilica (+)

Acid

(+)

NO

26.2

Cationic

8.0

NPs have the ability to form suspensions because the interaction of the particle's surface with the solvent is strong enough to overcome density (i.e. settling) effects. This same ability contributes to one of their major beneficial characteristics- large fraction of surface area per unit volume that will determine the interfacial surface area and specific surface area, all of which will influence their ability to interact with other solid surfaces. Calculations were performed to determine the surface properties and amount of NPs in the aqueous dispersion. Since the NPs investigated all have approximately the same density and particle diameter, their surface area was calculated to be 1256 nm2, specific surface area 125 m2/g, and number of NPs per 1 mL in 5%wt dispersion to be 5.1 × 1015. Zeta Potential Measurements of Shale Surface The zeta potential is an interfacial property (that takes into account electro-kinetic charge density) and is of great importance for understanding the behavior of solid materials, which will give insight into features for charge and adsorption characteristics of solid surfaces. While ZP is used to define the surface charge of a nanoparticle, it is an indirect method, but the only method to measure a nanoparticle's surface charge density. Surface charge is the difference in electrical potential between the inner and outer surface of the dispersed nanoparticle colloid. Zeta potential measurements of intact shale thin sections were performed by using a specially designed Malvern surface zeta potential cell that is capable of measuring the ZP of solid surfaces in aqueous solution via electrophoresis. Accordingly, it was also employed for the purpose of investigating the ability of NPshale charges to interact with one another. The measurement procedure is outlined in Fig. 6 below and described in more detail in Hoxha et al. (2016).

12

SPE-184574-MS

Figure 6—{bottom left} Shale thin section used for testing (dimensions: 4mm × 7mm × 1.5mm). {right} Surface zeta potential cell (with shale thin section sample attaced to it) wich is eventually immersed in a cuvette contanining the nanoparticle solution to be tested via the Malvern Zetasizer Instrument. {middle} Schematic of nanoparticle movement via electrokinetics (borrowed from Malvern Inc). For more inforation on surface zeta potential measurments, refer to Malvern Intruments publicaton - Corbett et al.,2011.

Pressure Transmission Testing The pressure transmission test (PTT) measures the tendency of a fluid filtrate, applied at overbalance pressure, to invade the shale matrix and elevate the near wellbore pore pressure (van Oort, 1994; van Oort et al., 1996). Fig. 7 shows a typical PTT set-up and all the major components. For more details on test protocol and data analysis, see van Oort et al. (2016). Simply explained, the downstream pressure buildup behavior due to pressure transmission through the shale sample is similar to the charging of a capacitor in a RC circuit. The pressure transmission is described as follows: (18) where, Po = initial pore pressure (Pa), Pm = upstream fluid pressure (Pa), P(l,t) = downstream pressure at sample end (Pa)  l = sample length (m) A = sample cross-sectional area (m2) V = volume of downstream reservoir (m3) β = fluid compressibility (Pa−1) μ = fluid viscosity (Pa.s)  k = relative shale permeability (m2)

SPE-184574-MS

13

Figure 7—Core holder, pore fluid lines, upstream and downstream assembly, pressure transducer and fluid accumulator. PTT equipment in a controlled temperature environment (35°C). The pore pressure transmission data is processed in accordance and fitted with a least- squares linear fit. From the slope of the fitted lines obtained for the pore fluid (first) cycle and the nanofluid (second) cycle, a delay factor can be characterized according to Eq. 19. The PTT simply compares the pressure transmission delays (pore fluid vs. test fluid), thus characterizing reduction in pressure transmission capability. The differential pressure is 250 psi and the confining pressure is at 1000 psi. The laminations on the shale samples are parallel to bedding - thus, fluid flows parallel to bedding in the test.

Tests are typically performed with two distinct cycles: a first cycle using pore fluid, to characterize rock permeability, and a second cycle (after re-equilibrating the rock sample to initial conditions) with NP test fluid. Since the viscosity μ and compressibility β of the filtrate of the test fluid/mud are generally unknown (or they can be uncertain calculations), a hydraulic conductivity k/μβ (m2/s) is characterized for each pore fluid cycle and subsequent nanofluid cycle - thus strictly measuring fluid pressure transmission. The fluids hydraulic conductivity measurements are than compared to yield a "delay factor" given by: (19) Also note that hydraulic diffusivity is inversely proportional to the shale's water activity - the higher the in situ pore fluid's water content, the lower the ability for hydraulic diffusivity across the shale.

Thick Walled Cylinder (TWC) Collapse Testing The downhole simulation cell (DSC) test (Simpson et al., 1989; Salisbury et al., 1991) is rightly regarded as a "gold standard" for shale borehole stability testing. In this paper, we are using a less sophisticated version of the DSC test based on a modification of the thick walled cylinder (TWC) test. A photograph and schematic of the test set-up is given in Figure 8. The details of this test are given in van Oort et al. (2016). Basically, a cylindrical shale sample is externally confined and exposed in its inner borehole to test fluid at overbalance for a period of a specific period of time (typically 24 hours). During this time, pore pressure elevation and effective stress reduction, as well as other destabilizing effects arising from e. g. chemical incompatibilities, may take place. At the end of the test, the confining stress is ramped until borehole failure is observed, as indicated by a sudden increase in volumetric strain (see Fig.8). The confining stress at failure gives an indication of the residual strength of the shale after fluid exposure. Note that native Mancos shale had a collapse strength of ~ 9,300 psi.

14

SPE-184574-MS

Figure 8—(top left) Schematic of TWC setup; (bottom left) Photograph of TWC equipment, based on a modification of a triaxial load frame; (top right) Photographs of cylindrical Mancos shale samples (1″ diameter, 2″ length) before coring the 0.3″ borehole, before TWC testing, and after TWC testing; (bottom right) Behavior of volumetric strain during a TWC test on Mancos shale as a function of ramping up the confining pressure; a sudden increase in volumetric strain signals borehole collapse (occurring at a maximum confining pressure of 8,843 psi for this example). Note, the laminations on the shale samples are perpendicular to the bedding.

Results & Discussion Fig.9 shows the ZP behavior of both the NPs themselves and the Mancos shale (immersed in NP fluid) as a function of pH. Clearly, the pH significantly influences the NP-ZP and shale-ZP values, especially at the extreme end of the pH spectrum. An increase in pH value (i.e. nanofluid becomes more alkaline) will generally increase the ZP of the negatively charged NP. By contrast, lowering the pH value i.e. (nanofluid becoming more acidic) the ZP of the NP will decrease and possibly neutralize the charge, reaching the isoelectric point, which is the point of zero charge (PZC). Approximately, at PZC, the ZP is exceedingly low and the NP dispersion is least stable: nanoparticles will quickly begin to attract, pack, and aggregate due to cohesive forces. Low colloidal stability is typically observed for ZP values less than 30 mV.

SPE-184574-MS

15

Figure 9—Plot of nanoparticle zeta potential and zeta potential of Mancos shale submersed in respective nanoparticle aqueous dispersions. Aggregation as well as other fluid interaction behavior (e.g., particle dissolution - although typically minimal but increasingly greater at higher and lower pH's) of the NP may alter the physiochemical properties, reactivity, and transport of the nanoparticles. Note, bare nanosilica (cationic) has the lowest differential charge whereas modified nanosilica (anionic) at pH 11.9 has the highest differential charge between NP and shale. The reader can refer to McDonald et al., 2016 to read further on how pH impacts silica dissolution at various pH range,

The values from Fig.9 and measurements of ionic strength of the NP aqueous dispersion were used to calculate the DLVO curves for ZP-shale interactions associated for each nanofluid. Fig. 10 shows a typical example of a PTT result, showing the first pressure build-up curve with native pore fluid / brine and the second pressure build-up curve (after sample re-equilibration) with a typical nanofluid. The delay factor (Eq.19) is derived by comparing the hydraulic conductivity observed for the pore fluid with that observed for the nano-fluid, which in this case for PTT measurement on bare nanosilica (anionic) at pH 9.7 amounts to 2.5.

Figure 10—Typical PTT pressure buildup curves for pore fluid (PF) and nanofluid (5% wt bare nanosilica, pH 9.7); (left) raw data; (right) processed results with linear fits from which a delay factor of 2.5 was derived.

16

SPE-184574-MS

Table 2 and Fig.11 shows the measured PTT delay factors and TWC testing, obtained for all NPs tested at a concentration of 5% wt. The following observations were made:

•

•

•

•

It was observed that the reduction in pressure transmission rate, as indicated by the magnitude of the delay factors, is not as large in value as reported in some literature (Sensoy et al., 2009; Cai et al., 2012, Ji et al., 2012, Jung et al., 2013). This is simply a consequence of testing NPs at lower concentrations. The delay factors, however, can still be meaningful, as they directly reflect a possible lengthening of trouble-free open-hole time in the field (see e.g. van Oort et al., 2017, where a delay factor of ~2 associated with a drilling fluid change-out resulted in significantly improved field performance). The bare (cationic) nanosilica demonstrated the highest apparent pore plugging and shale stabilizing capability according to the PTT & TWC results, as shown in Table 2 and Fig. 11. This is readily explained by the fact that this material displayed the greatest attraction to the shale surface (see Fig. 12), with a DLVO potential that is purely attractive. Apparently, for this type of NP, the surface charge and inter-molecular potential, particle size, morphology, salinity and temperature are all appropriate parameters to initiate effective pore-plugging. As expected, pH was found to influence borehole stability and signifies the importance of proper fluid development that needs to be compatible with field conditions. As shown in Table 2 and Fig. 11, the fluids with modified nanosilica at pH values of 8.7 and 2.5 showed second- and third-best overall performance in PTT and TWC tests. This result can again be explained by considering the DLVO curves for these fluids, as shown in Fig. 13. The DLVO curve for the pH = 2.5 nano-fluid is purely attractive at all distances (up to 10 nm in this calculation) away from the shale surface, and the curve for the pH = 8.7 fluid has a low repulsive region with the peak remaining below total potential energy value of 1 kT. In both cases, nano-particles in these fluids will experience limited repulsive obstruction to reaching the shale surface, leading to effective plugging. By comparison, the modified nanosilica fluid with pH = 11.9 shows notably worse PTT and TWC results than the pH = 2.5 and pH= 8.7 nano-fluids. This is also readily explained by considering the DLVO curve for this fluid: as shown in Fig. 13, for pH= 11.9 there is an extended repulsive region with total potential energy > 1 kT. Thus, clearly interfering with the ability of NPs at this alkaline pH level to reach the shale surface and cause shale plugging. This is consistent with literature information where Valdya and Fogler (1992) mention that total energy of interaction becomes repulsive at higher pH values due to the change of ZP on the surface of the shale. If pH is higher in the nanofluid medium, hydroxyl (OH–) anions assert a more negative charge to the shale surface, which may cause repulsion to like charges. All though this result is repulsive, it should be noted that small amounts of certain repulsion can still be overcome, however, particularly at elevated temperatures. Fig. 13 shows DLVO curves for modified nanosilica (pH 11.9) at elevated temperatures (up to 1250C). With an increase in temperature, the intermolecular potential apparently becomes less repulsive. This illustrates that NP plugging of shale surfaces may be more effective at downhole temperatures if the intermolecular repulsion reduces and there is an increase in nano-particle diffusion at elevated temperatures, assisting in overcoming the repulsion barrier.

SPE-184574-MS

Figure 11—Borehole stability capability curve, plot of TWC collapse pressure values vs. PTT delay factor values with a linear fit to data curve. The results show excellent correlation between higher PTT delay factors (meaning effective reduction of pressure transmission) and higher collapse factors observed in TWC tests. Multiple duplicate measurements were run on most shale sample to validate borehole stability data and lower the margin of experimental error. Mancos shale strength UCS (perpendicular to laminations) is ~9,300 psi (Fjaer et al., 2013, Torsaeter et al., 2012).

Figure 12—DLVO Curves. {left} DLVO curves for bare nanosilica (cationic) depicting van der Waals and electrostatic forces between NPs and shale surfaces - attraction and repulsion forces vs S0 (NP distance from the shale planar surface) exhibits the sum of total potential energy. {right} DLVO curves showing total potential energy for all 3 types of nanosilica (bare anionic, bare cationic, and modified anionic).

17

18

SPE-184574-MS

Figure 13—DLVO curves for {left) modified nanosilica (pH −11.9, 8.7, 2.5) and {right} modified nanosilica at pH 11.9 at different temperatures. It is evident that at in-situ temperature conditions, above 125 °C, will influence the nanofluid to attract to the shale surface. The reader is referred to theories such as "Brownian Motion" to further understand the influence of temperature on NPs diffusion behavior. Note, the pH is accounted for in 2 ways when calculating DLVO curves: pH adjustment changes NP dispersion ZP which is later used in DLVO calculations, and, pH adjustment changes ion concentration of NP dispersion. Table 2—Borehole stability testing of nanosilica, bare and modified, with different surface charges and at different pH values. Fluid Type

PTT Delay Factor

TWC Collapse Pressure (psi)

Bare Nanosilica (Cationic)

5.6

8,830

Modified Nanosilica (pH = 8.7)

2.9

8,634

Modified Nanosilica (pH = 2.5)

1.8

8,533

Bare Nanosilica (Anionic)

2.5

8,483

Modified Nanosilica (pH = 11.9)

1.6

8,323

Bare Nanoalumina (Non-spherical, cationic)

0.2

8,306

Although not fully discussed in the paper, testing was also performed on highly positive charged nanoalumina (Al2O3: +36mV in 5%wt aqueous dispersion). As can be seen in Table 2 and Fig. 11, the borehole stability values are significantly lower in comparison to the nanosilica (anionic and cationic). This in part is due to the morphology (non-spherical, rod-like, 20-200 nm) of the nano-alumina, which in a 5%wt aqueous dispersion will exhibit lower specific surface area and lower number of NPs per unit volume of nanofluid than the nanosilica dispersion. In addition, the rod-like structure of the particle makes it difficult for the nano-alumina particle to adhere into the shale pore throats. IT is evident that these results show that NPs can be effective in stabilizing shales, but should not be applied indiscriminately. Clearly, more work is needed to delineate the exact design criteria for optimizing the shale-stabilizing performance of NPs in actual field scenarios.

Summary & Conclusions The potential use of NPs in drilling and completion fluids is a fairly modern development that is both intriguing and puzzling at the same time. The resistance to more wide-scale adoption of nanoparticle fluids is due in part to a lack of reliable explanations of the behavior of NPs interactions with rock formations. Previous work has emphasized the potential of NPs to benefit borehole stability in shale, however, few studies have addressed the challenge of explaining the physical mechanism by which NPs and shales interact, and how this interaction might actually benefit shale stability. The preliminary work presented in this paper provides a more in-depth analysis and explanation of how NPs may enhance wellbore stability, by addressing the mechanisms that "drive" NPs to become attracted to - and block - shale pore throats by reducing fluid pressure transmission. The following key conclusions were reached:

SPE-184574-MS

19

1. Beneficial shale plugging by NPs is unlikely to be caused by filtration, given the very low matrix permeability of most shales. DLVO forces, i.e. van Waals attraction and electrostatic repulsion, need to be taken explicitly into account when explaining the interactions between NPs and charged shale/ clay surfaces. Important factors are (1) the surface charge of the shale (dependent on the surface charge distribution in the shale generated by its various mineralogical components); (2) surface charge density of the NPs; (3) external factors such as pH and ionic strength of the aqueous medium to which the shale and NPs are exposed. 2. To understand the shale-stabilizing effects of NPs, their inter-molecular (or inter-particle) interactions between each other and interactions with charged shale surfaces have to be considered. This is the main mechanism that will lead to pore throat plugging reducing pressure transmission, which in turn benefits borehole stability by slowing down near-wellbore pore-pressure elevation and effective stress reduction. Guiding principles include: a. The inter-molecular shale-NP interaction should allow for NPs to approach shale surfaces and be able to adhere to them. This means that electrostatic repulsion needs to be sufficiently low, such that NPs can "diffuse" through any repulsive barrier, if present, and reach the shale surface. b. The inter-particle repulsion between NP's themselves should be sufficiently high to prevent excessive NP aggregation. If NPs attract and aggregate, they will form larger particles that lose the ability to effectively plug pore throats (one of main reasons why micron-sized fluid loss additives and lost circulation materials fail to provide shale-stability, unless used to bridge micro-fractures). c. The NP size and concentration needs to be sufficient to get effective shale coverage and thereby significant shale stabilization. For Mancos shale, it was shown that 20 nm NPs are effective in partially plugging the pore throat system. However, 5 nm particles were found to be much less effective. It should be noted here that excess concentrations of NPs make application prohibitively costly and creates logistical difficulties given the large amount of NPs required. Instead, we advocate the use of NPs at optimized concentrations equal to - or less than – 5% by weight. 3. The above conclusions create some interesting challenges for the practical field application of NP fluids for borehole stability, given that efficacy will depend on: a. b. c. d.

The specific type of shale, particularly its surface charge in its specific aqueous environment; The specific type of nanoparticle and its surface charge in its specific aqueous environment; The specific interaction between the shale and the NPs, as governed by their DLVO forces; The average size, shape, and morphology of the NPs in relation to the average pore throat size of the shale; e. The concentration of the NPs in order to get sufficient shale coverage; f. External factors such as fluid pH / alkalinity, salt type / concentration / ionic strength, temperature etc. in the downhole environment where the NP fluids are applied.

Acknowledgements

First and foremost, we would like to thank every single participant in the Nanoparticles for Subsurface Engineering (NSE) Consortium for funding the "Nanoparticles in Drilling & Completion fluids Initiative". We would like to acknowledge Jimmie Baran and Wendy Thompson from 3M for their invaluable contribution in providing the custom-made nanoparticles, as well as providing technical knowledge to help define the scope of this project. We would like to thank John Southwell at Nissan Chemicals for his technical skills and assistance. Special thanks also goes to Carrie Schindler from Malvern Instruments for providing equipment as well as scientific expertise in electrochemistry. Personal thanks to Arthur Hale for

20

SPE-184574-MS

being a mentor to the lead author and providing vital chemistry and experimentation instructions. For their contributions in petrophysics knowledge and experimentation, a heartfelt thanks to Metarock Laboratories. Special thanks goes to Bence Toth, our visiting researcher, for his valued assistance with DLVO theory and calculations. Last but not least, a particular thank you to the exceptional scientists at the Texas Material Institute, Dr. Shouliang Zhang for his contribution with instrumentation assistance in nanoparticle-shale characterizations and Dr. Richard Piner for his contribution to particle physics.

Nomenclature BHS DLVO HC HP-WBM MICP NP(DF) OBM PTT PZC TWC ZP

References

= Borehole Stability = Derjaguin Landau Verwey Overbeek = Hydraulic Conductivity = High-Performance Water-Based Mud = Mercury Injection Capillary Pressure = Nano-Particle (Drilling Fluid) = Oil Based Mud = Pressure Transmission Test = Point Zero Charge = Thick Walled Cylinder Collapse Test = Zeta Potential References

Abbott, S. Nanocoating: Principles and Practices, from Research to Production TCNF Limited, UK. 2013. Ahmadi, M., Habibi, A., Pourafshary, P., & Ayatollahi, S. (2013, February 21). Zeta-Potential Investigation and Experimental Study of Nanoparticles Deposited on Rock Surface to Reduce Fines Migration. Society of Petroleum Engineers. doi:10.2118/142633-PA Amanullah, M., AlArfaj, M. K., & Al-abdullatif Ziad Abdullrahman. (2011, January 1). Preliminary Test Results of Nanobased Drilling Fluids for Oil and Gas Field Application. Society of Petroleum Engineers. doi: 10.2118/139534-MS Berg, J.C. 2010. An Introduction to Interfaces & Colloids: The Bridge to Nanoscience. Singapore: World Scientific. Bergstrom, L. 1997. Hamaker constants of inorganic materials. Adv. Colloid Interface Sci. 70: 125–169. Binks, P.B. Particles as surfactants - similarities and differences. Colloidal & Interface Science Journal. Elsevier. 2002. Borisov, A. A field application of nanoparticle-based invert emulsion drilling fluids. Journal of Nanoparticle Research. August 2015. DOI 10.1007/sl1051-015-3143-x Burkhardt, C. Micro and Nanopatterning of surface employing self-assembly nanoparticles and its application in biotechnology and biomedical engineering. Intech Open, pg 656. 2010. Cai, J., Chenevert, M. E., Sharma, M. M., & Friedheim, J. E. (2012, March 1). Decreasing Water Invasion into Atoka Shale Using Nonmodified Silica Nanoparticles. Society of Petroleum Engineers. doi:10.2118/146979-PA Celik, M. Electrokinetic behavior of clay surface. Journal of Interface Science and Technology. Amsterdam: Elsevier Academic Press. 2004. Chang, Raymond. Physical Chemistry for the Biosciences. Sausalito, CA. University Science Books, 2005. (498–500) Chenevert, M. E., & Amanullah, M. (2001, September 1). Shale Preservation and Testing Techniques for BoreholeStability Studies. Society of Petroleum Engineers. doi:10.2118/73191-PA Contreras, O., Hareland, G., Husein, M., Nygaard, R., & Al-Saba, M. (2014A, February 26). Application of In-House Prepared Nanoparticles as Filtration Control Additive to Reduce Formation Damage. Society of Petroleum Engineers. doi: 10.2118/168116-MS Contreras, O., Hareland, G., Husein, M., Nygaard, R., & Alsaba, M. (2014B, September 10). Wellbore Strengthening In Sandstones by Means of Nanoparticle-Based Drilling Fluids. Society of Petroleum Engineers. doi:10.2118/170263MS Contreras, O., Hareland, G., Husein, M., Nygaard, R., & Al-saba Mortadha T. (2014C, October 27). Experimental Investigation on Wellbore Strengthening In Shales by Means of Nanoparticle-Based Drilling Fluids. Society of Petroleum Engineers. doi:10.2118/170589-MS Corbett, J. Measuring surface zeta potential using phase analysis light scattering in a simple dip cell arrangement. Colloids and Surfaces A: Physicochem. Eng. Aspects 396 (2012) 169–176. Elsevier Journal. December 2011.

SPE-184574-MS

21

Dahirel, V. 2010. Effective interactions between charged nanoparticles in water: What is left from the DLVO theory? Curr. Opin. Colloid In. 15: 2–7. Dai, Caili. The first study of surface modified nanoparticles in pressure-decreasing application. Royal Society of Chemistry. 2015. Derjaguin, B.V. 1934. Friction and adhesion. IV. The theory of adhesion of small particles. Kolloid Z. 69: 155–164. Derjaguin, B.V., and Landau, L. 1941. Theory of the stability of strongly charged lyophobic sols and of the adhesion of strongly chargedparticles in solutions of electrolytes. Acta Physico-chim. USSR 14: 633–662. Dewhurst, D.N. Mechanics, physics, chemistry and shale to rock properties. ARMA. 47th US Rock Mechanics / Geomechanics Symposium. San Francisco, CA. June 2013. Elimelech, M. Particle deposition and aggregation. Butterworth Heinemann. Oxford UK. p 23–32. 1995. Fjaer, E. Strength anisotropy of Mancos shale. US Rock Mechanics Geomechanics Symposium. San Francisco, CA. June 2013. Friedheim, J. Testing and evaluation techniques for drilling fluids-shale interaction and shale stability. US Rock Mechanics Geomechanics Symposium. San Francisco, CA. June 2011. Gao, C., Miska, S. Z., Yu, M., Ozbayoglu, E. M., & Takach, N. E. (2016, September 14). Effective Enhancement of Wellbore Stability in Shales with New Families of Nanoparticles. Society of Petroleum Engineers. doi: 10.2118/180330-MS Gavish, N. 2016. Dependence of the dielectric constant of electrolyte solutions on ionic concentration: A microfield approach. Phys. Rev. E 94: 012611. Hoek, E.M.V., and Agarwal, G.K. 2006. Extended DLVO interactions between spherical particles and rough surfaces. J. Colloid Interface Sci. 298: 50–58. Hoelscher, K. P., De Stefano, G., Riley, M., & Young, S. (2012, January 1). Application of Nanotechnology in Drilling Fluids. Society of Petroleum Engineers. doi: 10.2118/157031-MS Hogg, R. Mutual coagulation of colloidal dispersions. Trans. Faraday Soc. 62: 1638–1651. Hough, D.B., and White, L.R. 1980. The calculation of Hamaker constants from Liftshitz theory with applications to wetting phenomena. Adv. Colloid Interface Sci. 14 (1): 3–41. Hoxha, B. B., Sullivan, G., van Oort, E., Daigle, H., & Schindler, C. (2016, May 30). Determining the Zeta Potential of Intact Shales via Electrophoresis. Society of Petroleum Engineers. doi:10.2118/180097-MS Hoxha, B.B. et al, Automated Particle Size Analysis using Advanced Analyzers. AADE-16-FTCE-78, Houston Texas, 2016. Hussein, M. Enhancing Drilling Fluid Performance using nanoparticles. Presentation given at the ACAMP Conventional Energy Seminar. Calgary, Canada. September 2012. Iler R.K. : "The chemistry of silica: solubility, polymerization colloid and surface properties, and biochemistry", WileyInterscience, New York, 1979. Israelachvili, J. N. Intermolecular and surface forces. London: Academic Press. (1992). Ji, L., Guo, Q., Friedheim, J. E., Zhang, R., Chenevert, M. E., & Sharma, M. M. (2012, January 1). Laboratory Evaluation and Analysis of Physical Shale Inhibition of an Innovative Water-Based Drilling Fluid with Nanoparticles for Drilling Unconventional Shales. Society of Petroleum Engineers. doi:10.2118/158895-MS Jung, C. M., Zhang, R., Chenevert, M., & Sharma, M. (2013, August 12). High-Performance Water-Based Mud Using Nanoparticles for Shale Reservoirs. Unconventional Resources Technology Conference. Lazaro, A. The influence of process condition and Ostwald ripening on the specific surface area of olvine nanosilica. Journal Paper. July 2013. McDonald, M. Silica gel as an economical and innovative alternative additive to bio-based polymers in aqueous hydraulic fracture fluids. AADE. Houston Texas. April 2016. Metin, C. Aggregation Kinetics and Shear Rheology of Aqueous Silica Suspension. 2014. Morse, P.M., and Feshbach, H. 1953. Methods of Theoretical Physics. New York: McGraw-Hill. Nwaoji, C. O., Hareland, G., Husein, M., Nygaard, R., & Zakaria, M. F. (2013, March 5). Wellbore Strengthening- NanoParticle Drilling Fluid Experimental Design Using Hydraulic Fracture Apparatus. doi: 10.2118/163434-MS Riley, M., Young, S., Stamatakis, E., Guo, Q., Ji, L., De Stefano, G., Friedheim, J. (2012, January 1). Wellbore Stability in Unconventional Shales - The Design of a Nano-Particle Fluid. Society of Petroleum Engineers. doi:10.2118/153729MS. Rodriquez, E. "Enhanced Migration of Surface-Treated nanoparticles in sedimentary rocks". ATCE, New Orleans, Louisiana, October 2009. Schlemmer, R. Chemical Osmosis, Shale, and Drilling Fluids. SPE Drilling and Completions Journal. 2003. Royer, Donald J. Bonding Theory. New York, NY.: McGraw-Hill, Inc., 1968. Salisbury, D. P., Ramos, G. G., & Wilton, B. S. (1991, January 1). Wellbore Instability of Shales Using a Downhole Simulation Test Cell. American Rock Mechanics Association.

22

SPE-184574-MS

Sarker, Rituparna, and Mike Batzle. "Anisotropic elastic moduli of the Mancos B shale-An experimental study." 2010 SEG Annual Meeting. Society of Exploration Geophysicists, 2010. Sensoy, T., Chenevert, M.E. and Sharma, M. 2009. "Minimizing Water Invasion in Shales Using Nanoparticles." SPE 124429, SPE Annual Technical Conference, New Orleans, 4-7 October. Sharma, M. A new family of nanoparticle based drilling fluids. ATCE. San Antonio, Texas. October 2012. Simpson, J. P. (1989, March 1). Downhole Simulation Cell Shows Unexpected Effects of Shale Hydration on Borehole Wall (includes associated papers 19519 and 19885). Society of Petroleum Engineers. doi:10.2118/17202-PA Taraghikhah, S. Multifunctional Nonadditive in Water Based Drilling Fluid for Improving Shale Stability. IPTC. Doha Qatar. December 2015. Torsaeter, M. Nanostructure vs. Macroscopic Properties of Mancos Shale. (2012). Canadian Unconventional resources Conference. November 2012. van Oort, E., A novel technique for the investigation of drilling fluid induced borehole instability in shales, SPE/ISRM 28064 presented at the 1994 SPE/ISRM Conference on Rock Mechanics in Petroleum Engineering, Delft, Aug. 29-31 van Oort, E., Hale, A. H., Mody, F. K., & Roy, S. (1996, September 1). Transport in Shales and the Design of Improved Water-Based Shale Drilling Fluids. Society of Petroleum Engineers. doi:10.2118/28309-PA van Oort, E.: On the Physical and Chemical Stability of Shales, J. Pet. Sci. & Engr., 1051 (2003), 1–23 van Oort, E., Hoxha, B., Hale, A.H., Aldin, M, and Patterson, R., How to test fluids for shale compatibility? AADE-16FTCE-77, Houston Texas, 2016 van Oort, E. Improved Wellbore Stability on Tor/Ekofisk Wells through Shale-Fluid Compatibility Optimization, paper SPE 184661, SPE IADC Drilling Conference 14-16 March 2017 in The Hague, The Netherlands Verwey, E.J.W., and Overbeek, J.T.G. 1948. Theory of the Stability of Lyophobic Colloids. Amsterdam: Elsevier. Xue, Z. "Viscosity and stability of ultra-high internal phase CO2 in water foams stabilized with surfactants and nanoparticles with or without polyelectrolyets". Journal of colloid and inrterface science. June 2015. http:// dx.doi.org/10.1016/jjcis.2015.08.031 Yotsumoto, H., and Yoon, R. 1993. Application of extended DLVO theory I. Stability of rutile suspensions. J. Colloid Interface Sci. 157 (2): 426–433. Yu, J., An, C., Mo, D., Liu, N., & Lee, R. L. (2012, January 1). Study of Adsorption and Transportation Behavior of Nanoparticles in Three Different Porous Media. Society of Petroleum Engineers. doi:10.2118/153337-MS Zakaria, M., Husein, M. M., & Harland, G. (2012, January 1). Novel Nanoparticle-Based Drilling Fluid with Improved Characteristics. Society of Petroleum Engineers. doi:10.2118/156992-MS