Composition in the Interface between Clay Mineral Surfaces and

May 23, 2018 - These samples were prepared from very thin smears of centrifuged wet paste that were instantaneously chilled to liquid N2 temperature...
2 downloads 0 Views 954KB Size
Subscriber access provided by UNIVERSITY OF TOLEDO LIBRARIES

Interfaces: Adsorption, Reactions, Films, Forces, Measurement Techniques, Charge Transfer, Electrochemistry, Electrocatalysis, Energy Production and Storage

The composition in the interface between clay mineral surfaces and divalent cation electrolytes Stanislav Jelavi#, A. Rath Nielsen, S. L. S. Stipp, and N. Bovet Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.8b00370 • Publication Date (Web): 23 May 2018 Downloaded from http://pubs.acs.org on May 24, 2018

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

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

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

Langmuir

The composition in the interface between clay mineral surfaces and divalent cation electrolytes S. Jelavića*, A.R. Nielsena, S.L.S. Stippa, N. Boveta a

Nano-Science Center, Department of Chemistry, University of Copenhagen, Universitetsparken

5, Copenhagen 2100, Denmark *Corresponding author: [email protected]

KEYWORDS Illite, chlorite, pore mineral, oil, tertiary recovery, reservoir rock, sandstone, waterflooding.

ACS Paragon Plus Environment

1

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

Page 2 of 35

ABSTRACT

The interfacial free energy of a solid, which determines its adsorption properties, depends on interactions between the surface and the fluid. A change in surface composition can completely change the behaviour of the solid. Decades of work have explored adsorption and its effects at solid-fluid interfaces from the macroscopic perspective and using molecular modelling so the concept of the electric double layer (EDL) is well established in the community. However, direct, molecular level, experimental observations of the composition within the interface region, and its change with time and conditions, is not abundant. We used cryogenic X-ray photoelectron spectroscopy (cryoXPS) to observe the composition in the clay mineral-solution interface region as a function of bulk solution composition, on illite and chlorite in MgCl2 and CaCl2 electrolytes, over a range of concentrations (1-125 mM), in situ, on vitrified samples. These samples were prepared from very thin smears of centrifuged wet paste that were instantaneously chilled to liquid N2 temperature. They preserved the adsorbed solution in its amorphous state, maintaining the location of the ions and water with respect to the solid, without the disruption that occurs during drying, or the rearrangement that results as water crystallises during freezing. With decreasing ionic strength, we could directly monitor the loss of negative charge in the interface region, producing an anion deficiency, as predicted by theory. The Cl-/Me2+ ratio dropped below 1 for chlorite at 12-25 mM MeCl2 and for illite at 75-100 mM. In addition to better understanding of clay mineral behaviour in solution, this work demonstrates that only those clay minerals where surface charge density is the same or lower than for chlorite contribute to a low salinity enhanced oil recovery response (LS EOR). This explains many of the contradictory results from studies about the role of clay minerals in LS EOR.

ACS Paragon Plus Environment

2

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

Langmuir

INTRODUCTION Contamination that has resulted from industrialisation, as well as increased demands for clean water by a growing population, has created the need for reliable and inexpensive methods for remediating contaminated soil and groundwater. Deeper understanding of the controls on pore surface composition and behaviour, which determine interfacial free energy, would contribute to these efforts. A growing demand for energy and the desire to avoid the need for installing new oil production facilities in environmentally sensitive areas (such as the Arctic and very deep water) motivate the search for new ways to increase productivity from existing fields, where conventional recovery processes are reaching their limits. Low salinity enhanced oil recovery, LS EOR, is a method where additional oil is sometimes released by flooding the reservoir with water where salinity is reduced to < 5 000 ppm total dissolved solids (TDS).1-3 In some cases additional oil is not produced4 and in some reservoirs, flooding with very low salinity fluids results in substantial reduction or complete loss of rock permeability, because some clay minerals swell, a phenomenon described as formation damage.5,6 To minimise formation damage and maximise the LS response, it is important to define the conditions for which LS response does happen and ultimately, to understand the underlying mechanisms. Understanding the processes responsible for retention or release of organic molecules in general will provide clues for more effective groundwater remediation strategies as well and would find applications in industrial processes. Recent review by Jackson et al.7 reported that several conditions are needed for a LS response: i) the rock must contain phyllosilicates, ii) the pore fluid must contain divalent cations and iii) the oil must contain polar components. These conditions do not guarantee a positive LS response but they have been shown to be good indicators of it. Expansion of the electric double layer

ACS Paragon Plus Environment

3

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

Page 4 of 35

(EDL) coupled with ion exchange has been recognised as a likely mechanism for LS EOR.8-10 Introducing LS water into the reservoir results in change of surface species and a change in the electric potential of the interface, which causes the diffuse layer of the EDL to increase in thickness (EDL expansion). This effect would be enhanced by the exchange of cations that bridge between the oil and the pore surface, with other cations from the flooding fluid.10 This mechanism can also explain why the LS response is sometimes not observed, even though all conditions for it are met: the EDL behaves differently on different minerals because of their inherent composition and the EDL is influenced by the presence of different adsorbed organic compounds and different species in the fluid. Our goal was to test, at the molecular level, how EDL expansion affects attachment of organic material on mineral surfaces in reservoir rocks. We took into account the three requirements for observing LS response and we simplified the system using the following criteria: i) We simplified a reservoir rock to a representative mineral surface that is most likely to be in contact with the oil as a pore lining mineral, namely, clay minerals.11 This conveniently coincides with the observation that the clay minerals have to be present for a LS response. Kaolinite has been recognized as a clay mineral necessary for LS response2,3,12,13 and it has been demonstrated how EDL expansion and change in surface composition affect kaolinite surfaces.10 We explored the behaviour of two other common reservoir clay minerals, chlorite and illite. Both differ from kaolinite by having permanently charged basal surfaces (layers) that are made of tetrahedral silicon sheets. A negative layer charge is compensated by a positive interlayer charge: in illite this is K+ and in chlorites, it is (Mg,Fe)3(OH)6. As for kaolinite, illite and chlorite do not swell so they are less likely to cause formation damage during the LS flooding, in contrast to smectites.

ACS Paragon Plus Environment

4

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

Langmuir

ii) We simplified the aqueous solution (brine, seawater, LS flood) to its monocationic analogues, i.e. we tested with solutions of CaCl2 and then MgCl2 at a range of concentrations. This allowed us to examine the influence of only divalent cations on the EDL composition on the clay surfaces. This is important because the presence of divalent cations in the low salinity water is a condition for LS EOR. iii) We simplified the polar part of organic compounds to a single negative charge, Cl-. This simplification is convenient for our system because the components of oil are a complex mixture of mostly nonpolar hydrocarbons and in this case, the main part of interest is the charged fraction, which is found at the mineral-water-oil interface.14 In addition to the electrostatic interactions, other forces can influence the interaction between the clay minerals and the organic compounds, such as van der Waals forces or steric and hydrophobic forces that are particularly important for high molecular weight hydrocarbons. However, in this work, only electrostatic interactions are probed because of the importance of the interaction between the divalent cations and the polar organic compounds for LS EOR7 and for the interfacial processes in general. To investigate how the concentrations of positive and negative charges in EDL change with a change in bulk ionic strength, we analysed the samples with cryogenic X-ray photoelectron spectroscopy (cryoXPS). XPS is a spectroscopic technique that collects information from only a few nanometers into the surface of solids.15,16 With cryoXPS, where the adsorbed fluid is vitrified before analysis, one can analyse wet samples.17 This method has been significantly improved over the past decade18 and has been demonstrated to work on synthetic19-22 and natural samples,10,23,24 as well as biological materials.25 Under cryogenic conditions, XPS measures the composition of the material’s surface, which in this case, includes the solid surface and its thin

ACS Paragon Plus Environment

5

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

Page 6 of 35

layer of vitrified fluid, i.e. the solid-liquid interface. This allows us to monitor changes in the EDL as we change the composition of the aqueous solution. CryoXPS is able to simultaneously determine, at nearly in situ conditions, the concentration of aqueous cations and anions in the interface region. This is an advantage over traditional bulk adsorption studies that rely on measurements of the loss of species from solution. In this approach, the adsorbent has to be separated from the supernatant and the solutes in the supernatant must be analysed separately, using such methods as ion chromatography, atomic absorption or emission spectroscopy, or mass spectroscopy. Reconstructing the composition in the interface from such data is difficult from at least two perspectives: i) the composition in the interface region is calculated as a difference between the composition of the solution with the standardised concentration and the composition of the supernatant. This approach introduces large uncertainties in the concentrations of weakly adsorbing anions and ii) the analytical concentration sensitivity varies significantly for the various techniques, making the comparison of separately determined activities difficult and complicating interpretations of small compositional variations in the interface as a function of the bulk solution concentration. Other spectroscopic techniques for in situ, solid-liquid interface composition analysis usually suffer from nontrivial quantification difficulties and cannot provide the elemental information as it is the case for infrared and Raman spectroscopy or require flat, single crystal surfaces such as sum frequency generation spectroscopy; X-ray absorption and emission spectroscopies are not surface sensitive, ambient pressure XPS requires synchrotron radiation and liquid microjet XPS is only effective with particles that are several tens of nanometres in diameter. For these reasons, we chose cryoXPS for probing the composition in the interface region between the clay minerals and the aqueous solutions.

ACS Paragon Plus Environment

6

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

Langmuir

Clay minerals are among the strongest adsorbers among natural materials. Their ability to adsorb anything from simple cations to complex organic molecules is well known. In this work we provide direct evidence for how changes in bulk solution concentration affect the behaviour of the EDL and how this results in rearrangement (adsorption and/or desorption) of cations and anions close to the mineral surface. Different rearrangement, resulting from different composition and structure on the various clay mineral surfaces, could give a purely mineralogical explanation to questions about why oil sometimes is and sometimes is not recovered during the LS EOR. A mineral indicator for the LS response would greatly simplify core plug assessment for predicting LS effectiveness. It would also provide important molecular level understanding about fluid-solid interaction in general, for optimising remediation of contaminated soil and groundwater.

EXPERIMENTAL SECTION Characterisation of clay minerals. Illite (IMt-1) and chlorite (CCa-2, clinochlore) were obtained from the Source Clays Repository. The chlorite sample was gently hand crushed in an agate mortar. We worked with the fraction that passed a 50 µm sieve. Illite was gently crushed and the 18.2 MΩcm). We prepared suspensions of chlorite and illite in a range of CaCl2 and MgCl2 solutions (1, 6, 12, 25, 50, 75, 100 and 125 mM) keeping the specific surface area (N2BET SSA) to liquid ratio constant. This was possible because neither chlorite nor illite swell in contact with water so the N2-BET determined SSA is a fair representation of their surface area. The 15 ml PET centrifuge tubes containing the clay suspension were left to equilibrate for at least 16 hours on a rotary shaker and centrifuged at 5000 rpm for 10 minutes. Immediately prior to the analysis, the supernatant was poured out of the tubes and kept for pH measurements and the wet mineral paste was placed directly in the XPS entry chamber on the sample holder, which had been cooled with liquid nitrogen to approximately -170 °C. The wet paste freezes immediately, quenching the water film surrounding the particles, so the ions in the solution keep their position and the amount of water sublimated during analysis is minimised. The entry chamber was pumped to high vacuum, ~10-7 Torr. The sample was then moved to the analysis chamber, where the sample stage was also at liquid nitrogen temperature, and kept at ultrahigh vacuum (UHV) conditions during the analysis, ~10-9 Torr. Each specimen was analysed twice: once in the vitrified state (VS) and then later at room temperature (RT). For the RT measurement, the liquid N2 was turned off and the sample was allowed to warm to room temperature overnight in the analysis chamber, allowing the water to sublime from the paste. In this way, we could confirm that a water film had been present during analysis under cryogenic conditions and that the EDL had been preserved in the state it had been while in the solution. The presence of water in the vitrified state was verified for every

ACS Paragon Plus Environment

9

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

Page 10 of 35

measurement by comparing the binding energies for the O 1s and Si 2p peaks in the VS and RT samples and ensuring there was a water peak in the O 1s region (Figure S4, Supporting Information). This procedure also ensured that the solutes remained in the ionic state and had not precipitated during vitrification. This was tested by comparing the position of the Cl 2p peak in the sample with the Cl 2p peak position from solid CaCl2 and MgCl2 and from their solutions. The position of the Cl 2p peak from the liquid, where chloride was in its (ionic) state, was determined by pipetting a droplet of 1 M CaCl2 or MgCl2 solution directly onto a clean, empty copper sample holder that had been cooled to liquid N2 temperature in the XPS entry chamber. We have estimated uncertainty of our quantifications to be the same as the random uncertainty generally associated with XPS peak intensity, namely ± 12% for well defined, symmetrical peaks.28 Any systematic errors, e.g. in effective attenuation length, would not change the trends in the data because they are not expected to change from sample to sample. pH measurements. The pH of the supernatant was measured using a Metrohm 827 pH lab meter that was calibrated prior to every series of measurements using standardised Metrohm buffers. Temperature fluctuations in the laboratory were ≤ ±2 ºC. The pH of suspensions was never adjusted because this would change the ratio of anions and cations in the solution and effectively, their equilibrium surface concentrations. Between batches, the pH varied by ±0.3 units except for illite in CaCl2 solutions, where the variation was ±0.8 units (Fig. S5). However, these differences were completely uncorrelated with the trends we observed in our measurements.

ACS Paragon Plus Environment

10

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

Langmuir

RESULTS AND DISCUSSION Surface composition of clay mineral standards and photoelectron peak assignment. Typical XPS surface composition for powdered illite and chlorite is presented in Table 1 and the typical survey spectra, in Figure S6 (Supplementary Information). The surface composition of both samples is within natural variations for the bulk composition of both minerals.29,30 Table 2 presents the binding energy (BE) and full width at half maximum (FWHM) for the main elements for illite and chlorite determined from the high resolution spectra. Chlorite contains 0.1 at. % of Cl (Table 1). This Cl is not adsorbed contamination but it is a part of the clay mineral structure. Cl substitutes for hydroxyl in the metal octahedral sheets. Its position in the mineral structure is reflected in the lower BE of this Cl compared with the Cl from dry salts or vitrified pastes (Table 2). Table 2 also shows BE and FWHM for the Cl 2p3/2 measured for the dry MgCl2 and CaCl2, for their vitrified 1M solutions and for the adsorbed Cl- at illite and chlorite samples acquired under cryogenic conditions. BE and FWHM for the peaks that represent the adsorbed Mg2+ and Ca2+ 2p peaks could not be defined unambiguously because of the strong contribution of structural Mg in both illite and chlorite so we used only the Cl 2p region to demonstrate that solutes were in their ionic state during the analysis (Table 2, Figure 1). Cl 2p binding energies from clay mineral wet pastes were similar to the Cl 2p from the vitrified MgCl2 or CaCl2 solution and their binding energies are consistently lower (for 0.5 ± 0.1 eV) than peaks from the solid MgCl2 or CaCl2. The BE shifts for the Cl 2p3/2 peak in the wet pastes of chlorite (198.7 ± 0.1 eV) and illite (198.8 ± 0.1 eV) and the Cl 2p3/2 peak in the pure aqueous solution of CaCl2 (198.8 eV) and MgCl2 (198.7 eV) are negligible indicating that Cl in vitrified wet pastes and pure solution are present in their hydrated Cl- ionic state and not in the solid form as in the dry

ACS Paragon Plus Environment

11

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

Page 12 of 35

salt. These results provide a baseline for a further analysis of changes in the EDL as a result of change in composition of the solution. Table 1. XPS Elemental composition (in atom %) for the