Comparative Study on the Static Adsorption Behavior of Zwitterionic

Energy Fuels , Article ASAP. DOI: 10.1021/acs.energyfuels.8b04013. Publication Date (Web): January 11, 2019. Copyright © 2019 American Chemical Socie...
0 downloads 0 Views 2MB Size
Subscriber access provided by Iowa State University | Library

Fossil Fuels

Comparative Study on the Static Adsorption Behavior of Zwitterionic Surfactants on Minerals in Middle Bakken Formation Xun Zhong, Hui Pu, Yanxia Zhou, and Julia Xiaojun Zhao Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.8b04013 • Publication Date (Web): 11 Jan 2019 Downloaded from http://pubs.acs.org on January 15, 2019

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 31 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

Energy & Fuels

Comparative Study on the Static Adsorption Behavior of Zwitterionic Surfactants on Minerals in Middle Bakken Formation Xun Zhong a, Hui Pu a *, Yanxia Zhou b, c, Julia Xiaojun Zhao b* a Department of Petroleum Engineering, University of North Dakota, Grand Forks, North Dakota 58202, United States b Department of Chemistry, University of North Dakota, Grand Forks, North Dakota 58202, United States c College of Petroleum Engineering, Northeast Petroleum University, Daqing, Heilongjiang 161318, PR China

ABSTRACT: Zwitterionic surfactants are promising additives especially for harsh reservoir conditions because of their high stability and good compatibility, as well as amazing interfacial activity, however, surfactant adsorption is always of great concern. In this paper, spectrophotometric method was applied to study the adsorption behavior of zwitterionic surfactants on complex Middle Bakken minerals at high temperature (105 ℃) and high salinity (TDS=289,820 mg/L) conditions, and the impacts of concentration, salinity, temperature, mineral types and surfactant types were investigated. Experimental results show that the adsorption isotherms of zwitterionic surfactant fit well with Langmuir adsorption model, with adsorption density increasing fast at lower concentrations and generally reaching the equilibrium. Salinity has varying influences on the adsorption of zwitterionic surfactants with different acidic and/or basic groups. Surfactants BW and CA, where -COO- functional groups have the potential to gain protons, showed an adsorption decrease of 2.06 ± 0.02 mg/g when Bakken formation brine was applied instead of deionized water, while surfactant CS50, which can neither be protonated nor deprotonated, shows a small increase of 0.35 mg/g due to the adsorption energy difference of different functional groups. Higher temperature causes desorption of zwitterionic surfactants, but chemical degradation or solubility difference may compensate for this gap at elevated temperature range of 80 ~ 105 ℃. Further data analysis indicated that, concentration, mineral types and the interaction effects of concentration and mineral types are the three dominant influential factors that affect zwitterionic surfactants adsorption. The driving forces for adsorption vary for different surfactants, and small changes in certain factors can lead to significant differences. Zwitterionic surfactant was found to have higher adsorption on Bakken minerals than nonionic surfactant HCS and anionic surfactant 964 regardless of the

ACS Paragon Plus Environment

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

salinity. 1. INTRODUCTION Bakken formation is a huge unconventional reserve spreading over Montana, North Dakota and Saskatchewan. It is mainly composed of Lower and Upper shale, and Middle Bakken member. Middle Bakken member, favorably positioned with respect to source and seal units, has been a research focus since its early exploration. However, its mineralogy is highly variable, ranging from a silty to very fine-grained dolomitic sandstone to a silty dolostone. Middle Bakken formation is characterized as a typical unconventional liquid reservoir (ULR) with narrow pores, low porosity (< 8.0 %), low permeability (< 0.1×10-3 μm2), high salinity (150,000 ~ 300,000 mg/L) and high temperature (80 ~ 120 ℃). Most importantly, its fairly low primary oil recovery (around 10.0 % OOIP) offers great opportunities for enhanced oil recovery (EOR) technologies.1-4 The purpose of EOR is to modify the physical and chemical properties of reservoirs (including minerals and fluids), thereby to increase oil recovery factor.5 Surfactants with excellent capability to lower the oil/water interfacial tension and/or alter rock wettability are now pretty popular in EOR area. Commonly used surfactants include cationic, anionic, zwitterionic and nonionic types, and most of them are stable and applicable at general conditions. However, when it comes to high temperature and high salinity conditions like Middle Bakken formation, cationic surfactants may possibly suffer from Hoffman elimination process, nonionic surfactants would precipitate due to weak hydrogen bonding with water molecules, and those frequently used anionic surfactants such as petroleum sulfonate, alpha olefin sulfonate and sodium dodecyl benzene sulfonate can hardly perform effectively and efficiently at temperature over 90 ℃ and salinity higher than 100,000 mg/L.6 Zwitterionic surfactants, containing both positive and negative charges in its hydrophilic part, recently have become a hot topic in chemical EOR especially for harsh reservoir conditions because of their excellent water solubility, insensitiveness towards salt and temperature, good biodegradation, low toxicity and remarkable interfacial activity.7 Wang et al.8 found that zwitterionic surfactant alone was able to generate ultralow IFTs, with a concentration of 50 ~ 3000 mg/L in the presence of over 200,000 mg/L salts.

ACS Paragon Plus Environment

Page 2 of 31

Page 3 of 31 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

Energy & Fuels

Meanwhile, zwitterionic surfactants have good compatibility with various ionic and nonionic surfactants, contributing to improved salt tolerance and strengthened interfacial properties. Li et al.9 and Zhao et al.10 showed that with proper addition of betaine, nonionic surfactant Triton X-200, anionic surfactants AOS16-18 and cationic surfactant quaternary ammonium salt all could produce desirable foams with higher stability. Moreover, adding betaine into anionic surfactant solution could lower the critical micelle concentration and the Krafft temperature.11 Most importantly, there are many indications showing their great potential in EOR applications, for example, by injecting 0.5 PV 500 mg/L zwitterionic surfactant solution, Kamal et al.12 were able to obtain an 8.0 % increase in oil recovery. Also, through the mechanisms of IFT reduction and wettability alteration, a lab synthesized zwitterionic surfactant was able to recover an additional 27.03 % oil.13 When used with polymer, alkyl-hydroxyl-sulfobetaine zwitterionic surfactant can extract extra 18.6 % more oil on the basis of water flooding.14 Thermal stability at given conditions, compatibility with reservoir fluids, low adsorption, IFT reduction capability, and low cost are some basic requirements of EOR surfactants. Plenty of researches have been conducted on the adsorption of traditional anionic, cationic and nonionic surfactants onto formation rocks under the effects of salinity, salt ions and temperature in the past several decades with either positively or negatively charged rocks and ionic or nonionic surfactants,15-19 and there is scarce information about the adsorption isotherms and kinetics of zwitterionic surfactants, except Li et al.20 studied the adsorption behavior of betainetype surfactant on quartz sand in 2011. Alvarez et al.5,

21

and Wang et al.7 figured out the adsorption

characteristic of sulfobetaine on limestone. Jian et al.22 researched the adsorption of betaine type surfactants on carbonates surfaces. However, most experiments were done at relatively ideal condition, with either simple mineral composition, low temperature or low salinity brine. Total adsorption is the cumulative effects of electrostatic interactions, chemical interactions (covalent bonding), hydrogen bonding, lateral associative interactions, solvation and desolvation. The driving forces can be easily influenced by physicochemical properties of solutions, surfactants and adsorbents.23-28 Small changes in certain factors may lead to significant

ACS Paragon Plus Environment

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

differences in adsorption density and isotherms. Proper surfactant adsorption on mineral surfaces can modify important interface properties thus to enhance oil recovery while excessive adsorption may result in high cost and limited effectiveness. Generally speaking, the requirement of surfactant adsorption loss in chemical combination flooding should be lower than 1.0 mg/g,29 and the adsorption threshold for carbonate reservoirs at high temperatures was around 1.0 mg/m2.6 Having a good knowledge of the adsorption behavior of zwitterionic surfactants can be beneficial to offer cost-effective choices for future development of surfactant formula. So far, to the best of our knowledge seldom has anyone done any researches on the adsorption behavior of zwitterionic surfactants on adsorbents with complex minerals at salinity higher than 250,000 mg/L (with divalent ions) and temperature higher than 100 ℃. In this study, spectrophotometric iodine method using KI-I2 solution as the color developing agent30 was applied to study and compare the adsorption behaviors of different zwitterionic surfactants on Middle Bakken minerals under Bakken condition. This method owes the advantages of high accuracy, easy operation, wide linear detection range with large regression coefficient, high stability in acid condition and low cost. 2. EXPERIMENTAL SECTION 2.1. Materials. Five surfactants including three zwitterionic surfactants (BW, CA, CS-50), one nonionic surfactant (HCS, cloud point=71 ℃) and one anionic surfactant (964, neutralization required) were used without further purification, details were tabulated in Table 1with some details remain undisclosed owing to confidential issues. Sodium chloride (NaCl), potassium chloride (KCl), magnesium chloride (MgCl2) and calcium chloride (CaCl2) were used to prepare Bakken formation brine31 (see Supporting Information Table S1, density= 1.17 g/cm3, pH=6.2±0.3). The four metal chlorides together with the iodine (I2) and potassium iodide (KI) purchased from VWR International were all of ACS grade. The main adsorbents are rock powders prepared by crushing Bakken rocks (Middle Bakken Formation, Mountrail County, ROSS Field) and Berea sandstone (Kocurek Industries, Inc., USA) with a ceramic mortar and pestle. Then, the resulting particles were sieved through 120 mesh (≤125 μm) steel wire screens. Other

ACS Paragon Plus Environment

Page 4 of 31

Page 5 of 31 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

Energy & Fuels

minerals include calcite (≤20 μm, Alfa Aesar) and clay (≤20 μm, Sigma-Aldrich). Adsorbents were not cleaned in order to simulate the original states and conditions. Table 1 Details of surfactants Surfactants

Type

Active, %

Manufacture

Molecular formula

BW

Zwitterionic

35.9

Lubrizol

C12H25N+(CH3)2CH2COO-

CA

Zwitterionic

29.4

Stepan

C11H23CO-NH-(CH2)3N+(CH3)2CH2COO-

CS-50

Zwitterionic

43.5

Stepan

C11H23CO-NH-(CH2)3N+(CH3)2CH2CH(OH)CH2SO3-

HCS

Nonionic

60.0

Stepan

CH3(CH2)m-O-(CH2CH2O)nH

964

Anionic

86.8

Sasol

CH3(CH2)m-O-(CH2CH(CH3)O)-(CH2CH2O)nCH2COOH

2.2. Characterization of Adsorbents. Scanning electron microscopy (HITACHI SU 8010) with minerals mounted on carbon tape was applied to study the surface morphology of adsorbents. X-ray diffraction (Rigaku Smartlab) based on observing the scattered intensity of an X-ray beam hitting a sample as a function of incident and scattered angle, polarization, and wavelength or energy was used to analyze mineral compositions. A nitrogen adsorption-based surface analyzer (Quantachrome Instruments Autosorb) was applied to measure the surface area of absorbents. Zeta potentials were obtained through a light scattering Zetasizer (Malvern). 2.3. Spectrophotometric Iodine Method. 2.3.1. Preparation and Characterization of KI-I2 Solution. In this research, 0.2 mM KI-0.1 mM I2 solution was adopted as the color developing agent. The absorbance spectra at different salinities in the wavelength range of 300 ~ 500 nm were recorded via a temperature compensated UV/vis/NIR spectrophotometer (Lambda 1050, PerkinElmer, USA) with a pathlength of 1 cm at the temperature of 298±0.2 K, ambient pressure. Pure saline or deionized water with the absence of KI-I2 were used as the blank controls. 2.3.2. Preparation of Calibration Curves. To plot the calibration curves of different surfactants, 4 ~ 6 surfactants standards of known concentrations were first prepared by serial dilution of stock solutions, then KI-

ACS Paragon Plus Environment

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

Page 6 of 31

I2 solution (with final concentrations of 0.2 mM and 0.1 mM, respectively) was added to start the reaction. The absorbance spectra were measured within 2 mins and the exact absorbance at specific characteristic peaks were recorded. 2.4. Static Adsorption. Static adsorption of surfactant on rocks is determined by batch equilibrium tests on crushed rock powders. Thirty milliliter surfactant solutions prepared by diluting the bulk solutions with Bakken formation brine (initial surfactant concentration = 100 ~ 1000 mg/L) is mixed with 2.0 g rock materials in a 40 mL glass vial. After 1 hr stirring at room temperature, samples were transferred to the oven with temperature setting at 20, 80 or 105 ℃, separately. Subsequently, after 24 hrs treatment, the solution was centrifuged for 20 min at 2500 rpm. Each sample was triplicated analyzed and the average value was used to calculate the equilibrium residue surfactant concentration.31 The exact adsorption amount (𝛤, mg/g) was calculated by the Equation 1. 𝛤=

(𝑐𝑖 ― 𝑐𝑒𝑞)𝑉𝑠𝑢𝑟𝑓 𝑚𝑎𝑑𝑠

∗ 10 ―3

(1)

Where 𝑐𝑖 is the initial surfactant concentration (mg/L), 𝑐𝑒𝑞 is the equilibrium surfactant concentration (mg/L), 𝑉𝑠𝑢𝑟𝑓 is the volume of the surfactant solution (mL), and 𝑚𝑎𝑑𝑠 is the total mass of the adsorbent (g). The adsorption density was also converted into mg/m2 based on the specific surface area of absorbents. 2.4.1. Stability of Surfactants. The compatibility of surfactants with Bakken formation brine was assessed at the concentration of 1000 mg/L, ambient pressure and various temperatures (20, 80 and 105 ℃). Those surfactants whose concentrations did not undergo sharp decrease within one week were regarded as stable. 2.4.2. Fitting of Adsorption Models. Adsorption models are necessary to predict the loading on the adsorption matrix at a certain concentration of the component. Langmuir isotherm and Freundlich isotherm are two commonly used equilibrium adsorption models to correlate the equilibrium adsorption density (𝛤𝑒, mg/g) and concentration (𝑐𝑒, mg/L).32 Langmuir isotherm (Equation 2), describes the adsorption behavior of adsorbates on an ideal homogeneous

ACS Paragon Plus Environment

Page 7 of 31 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

Energy & Fuels

adsorbent surface.33 Usually a linear relationship between 1/𝛤𝑒 and 1/𝑐𝑒 can be observed if Langmuir isotherm is applicable to depict the adsorption equilibrium. Values of 𝛤𝑚𝑎𝑥 (maximum adsorption, mg/g) and 𝐾𝐿 (equilibrium constant, L/mg) can be obtained through curve fitting from the slope and the intercept, separately. 𝛤𝑒 =

𝛤𝑚𝑎𝑥𝐾𝐿𝑐𝑒 1 + 𝐾𝐿𝑐𝑒

(2)

Freundlich assumed that the adsorbent has a heterogeneous surface composed of different classes of adsorption sites 34 and proposed Freundlich isotherm. Generally, it was applied to predict reversible adsorption and was not limited to monolayer adsorption, as expressed by Equation 3, where 𝐾𝐹 and 𝑛 are Freundlich constants. 𝛤𝑒 = 𝐾𝐹𝑐1/𝑛 𝑒

(3)

3. RESULTS & DISCUSSIONS 3.1. Physico-Chemical Properties of Adsorbents. SEM analysis (Figure 1) and XRD measurement (Figure 2) were performed to characterize the basic physico-chemical properties of various adsorbents. Berea rock is a typical sandstone where quartz is the abundant mineral (79.4 wt%) while the composition of Bakken rock is more complicated, containing 41.5 wt% quartz, 16.9 wt% clay (illite and kaolinite), 14.1 wt% feldspar (albite and sanidine), 12.2 wt% calcite, 11.5 wt% dolomite and a few pyrite. Moreover, small chips are common on larger particle surfaces, which provide more adsorption sites for chemicals. Laminated clay particles have the largest BET specific surface area (18.63 m2/g), followed by Bakken minerals (7.54 m2/g), Berea sandstone (2.85 m2/g) and calcite (2.12 m2/g). Usually, quartz and clay are negatively charged while carbonate minerals (calcite and dolomite) are positively charged.

ACS Paragon Plus Environment

Energy & Fuels

(a)

(b)

(c)

(d)

Figure 1. SEM images of various adsorbents, Scale bar=10.0 μm. (a) Bakken minerals. (b) Berea sandstone. (c) Calcite. (d) Clay. 90

45 41.5

Bakken minerals

40

80

35

70

30

60

Content, wt.%

Content, wt.%

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

Page 8 of 31

25 20 15.8 13.4

15 10

11.9 7.9

Berea sandstone

79.6

50 40 30 20

6.2 5

2.1

10.2

10

1.1 0

1.1

2.9

3.4

Dolomite

Albite

Muscovite

2.8

0

Quartz

Calcite Dolomite Albite Sanidine Pyrite

Illite

Kaolinite

Quartz

(a)

(b)

ACS Paragon Plus Environment

Sanidine

Kaolinite

Page 9 of 31

Figure 2. Compositions of Bakken minerals and Berea sandstone. (a) Bakken minerals. (b) Berea sandstone.

3.2. Characterization of KI-I2 Solution. The color developing agent is composed of 0.2 mM KI and 0.1 mM I2. The absorption spectra of the KI-I2 mixture in the wavelength range of 300 ~ 500 nm were shown in Figure 3a, where the absorption peaks at wavelength of 351 nm and 460 nm can be attributed to the characteristic absorption of I3- and I2, respectively.35 However, when the high salinity Bakken formation brine was used, the typical I3- peak at 351 nm was would be replaced by a new peak at 433 nm, which could be explained by the equilibrium shift of the reversible reaction between KI and I2 (KI+I2

⇌KI3), resulting in

increasing free iodine and decreasing triiodide ions. 0.7

0.36 0.33

Deionized water Bakken Brine

-

I3 , λ=352 nm

0.30

2 mg/L 656 30 mg/L BW 100 mg/L CA 100 mg/L CS-50 6 mg/L HCS 100 mg/L 964

0.6

0.27

0.5

Absorbance, a.u.

0.24

Absorbance, a.u.

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

Energy & Fuels

0.21 0.18

I2, λ=460 nm

0.15

λ=433 nm

0.12 0.09 0.06

428 nm

0.4

365 nm 427 nm

0.3

426 nm

0.2

0.1

362 nm

0.03 0.00 300

320

340

360

380

400

420

440

460

480

500

0.0 300

320

340

360

380

400

420

440

460

480

500

Wavelength, nm

Wavelength, nm

(a)

(b)

Figure 3. Absorption spectra of KI-I2 mixtures in the presence Bakken formation brine and/or surfactants. (a) In the absence of Bakken formation brine alone. (b) In the presence of Bakken formation brine and different surfactants.

3.3. Static Adsorption. 3.3.1. Preparation of Calibration Curves. Calibration curves are necessary to quantify surfactant concentration in certain solutions. Therefore, the absorbance at wavelength of 365 nm for surfactant BW, 362 nm for surfactants CA and CS-50, 427 nm for surfactant HCS and 426 nm for surfactant 964 were recorded to prepare the calibration curves at Bakken salinity (see Supporting Information Figure S1). The absorbance at characteristic wavelengths for different surfactants all showed good linear relationships towards surfactant

ACS Paragon Plus Environment

Energy & Fuels

concentration, with regression coefficients higher than 0.99. 3.3.2. Stability of Surfactants. The concentration change of active components in different surfactant samples versus time were measured according to the pre-plot calibration curves, as given in Figure 4. The stability of nonionic surfactant and anionic surfactant were only tested at room temperature because in most cases, the effects of oxygen on the stability of ethoxylated surfactants (such as surfactants HCS and 964) can never be neglected at elevated temperatures. Surfactants BW and CA showed higher stability than surfactant CS-50, and no precipitation or phase separation was observed within 60 days, satisfying the long-term stability requirement for field application. 1400 1300 1200 1100

Original 20 ℃ -7d 80 ℃ -7d 105 ℃ -7d

1000

Concentration, mg/L

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

Page 10 of 31

900 800 700 600 500 400 300 200 100 0 BW

CA

CS-50

HCS

964

Figure 4. Effective surfactants concentrations after aging in given conditions for 7 days.

3.3.3. Optimization of Adsorption Parameters. Suitable liquid/solid ratio and adsorption cycle are important to study the equilibrium adsorption density. Thus, the adsorption of surfactant BW (1000 mg/L) was first measured at different liquid/solid (Berea sandstone) ratios ranging from 4:1 to 20:1 after 24 hrs at 20 ℃, Bakken formation brine, as illustrated in Figure 5a. When the liquid/solid ratio is lower than 8:1, a sharp increase in adsorption density was observed, then the increasing rate decreased with a further increase in this ratio. Once the liquid/solid ratio was higher than 15:1, the adsorption density remained nearly unchanged, indicating an equilibrium adsorption state. Thus, liquid/solid ratio 15:1 was used in the following experiments. The adsorption duration also has an impact. In this part, BW concentration and liquid/solid ratio were 1000

ACS Paragon Plus Environment

Page 11 of 31

mg/L and 15:1, respectively, and the aging time varied from 1 hr to 50 hrs, as shown in Figure 5b. The adsorption of BW on Berea sandstone in the first 5 hrs showed a linear increase, then reached a plateau at around 24 hrs, therefore, 24 hrs was selected as an adsorption cycle. 12

12 11

11

10 9

Adsorption density, mg/g

10

Adsorption density, mg/g

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

Energy & Fuels

9 8 7 6

8 7 6 5 4 3 2

5

1 0

4 4℃ 1

6℃ 1

8℃ 1

10:1

12:1

15:1

20:1

0

5

10

15

20

25

30

35

40

45

50

55

Experimental period, h

Liquid/solid ratio

(a)

(b)

Figure 5. Optimization of static adsorption parameters. (a) Liquid/solid ratio. (b) Experimental period.

3.3.4. Effects of Salinity. In this part, the influence of salinity on adsorption density of zwitterionic surfactants on Bakken minerals were studied at 20 ℃, ambient pressure, as shown in Figure 6. All the three surfactants experienced a sharp increase when initial surfactant concentration is relatively low, followed by a reduced slope when surfactant concentration increases further and then gradually reached the adsorption equilibrium. To further study the adsorption equilibrium of surfactants BW and CA in the presence of deionized water and surfactant CS-50 in the presence of both deionized water and Bakken formation brine, the initial surfactant concentration was increased up to 2000 mg/L considering the cost. The difference between the adsorption density of 1000 mg/L surfactant and equilibrium adsorption density was measure to be less than 1.0 mg/g.

ACS Paragon Plus Environment

Energy & Fuels

14

14

BW CA CS-50

12

BW CA CS-50

12

10

Adsorption density, mg/g

Adsorption density, mg/g

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

Page 12 of 31

8

6

4

10

8

6

4

2

2

0

0 0

25

50

75

100

125

150

175

200

225

250

275

300

0

50

100

Equilibrium concentration, mg/L

(a)

150

200

250

300

350

400

450

Equilibrium concentration, mg/L

(b)

Figure 6. Effects of salinity on adsorption density. (a) Deionized water (TDS ≈ 500 mg/L). (b) Bakken formation brine (TDS=289,820 mg/L).

The adsorption of CS-50 on Bakken minerals in deionized water and Bakken formation brine were similar, both around 12.5 mg/g (1.66 mg/m2) for a 1000 mg/L surfactant solution. Because of the differences in steric factors of different functional groups, the adsorption was slightly higher when the interactions take place by the sulfonic group. While the adsorption density of surfactants BW and CA in the presence of Bakken formation brine were 8.75 mg/g (1.16 mg/m2) and 9.33 mg/g (1.24 mg/m2), respectively, both were around 2.06 ± 0.02 mg/g lower than those of 10.83 mg/g (1.44 mg/m2) and 11.37 mg/g (1.51 mg/m2) in deionized water. It is worth mentioning that the zeta potential of Bakken minerals reversed to 1.6 ± 0.5 mV from -21.2 ± 0.5 mV when Bakken formation brine was applied due to the presence of large Ca2+ and Mg2+ cations. Different from surfactant CS-50, the acidic parts of surfactants BW and CA are carboxylate group, which has higher tendency to gain protons in acidic solutions and generally turn the original zwitterionic into a fully protonated form, leaving the cationic part plays the dominate role. Herein, both deionized water and Bakken formation brine are slightly acidic, as a consequence, part of carboxylate groups would be deprotonated, implying basic groups (cationic parts) partial overweigh the acidic groups (anionic parts). Thus, comparing with positively charged adsorbents, BW and CA have higher adsorption on negative charged minerals. Though

ACS Paragon Plus Environment

Page 13 of 31 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

Energy & Fuels

a decrease in adsorption density was observed in high salinity conditions, the equilibrium adsorption densities of surfactants BW and CA were still higher than the threshold adsorption density proposed at high temperatures (1.0 mg/m2). Possible explanations for this higher adsorption are zwitterionic property of surfactants and large clay and quartz contents in adsorbents, and details were illustrated in the subsequent paragraphs. Feasible strategies to reduce the surfactant adsorption are to add sacrificial agents like polymers6 or nanoparticles36. Furthermore, the experimental data at 100 ~ 1000 mg/L initial surfactant concentration and Bakken salinity were matched by Langmuir and Freundlich adsorption model, separately, and relevant parameters are listed in Table 2 and Table 3. Considering the values of regression coefficient (R2) and mean square error, Langmuir model is slightly better fitted and can be more accurate to describe the adsorption isotherm of zwitterionic surfactants on the Bakken minerals. Although high salinity is reported to be favorable for vesicle formation, the relatively low initial surfactant concentration used gave no similar vesicle induced variation trend5 in our study. Table 2. Parameters of Langmuir adsorption model Surfactant BW CA CS-50

Correlation

R2

KL, *102 L/mg

𝛤max

25.495 + 0.0555 𝑐𝑒 16.806 1/𝛤𝑒 = + 0.0527 𝑐𝑒 3.4583 1/𝛤𝑒 = + 0.0516 𝑐𝑒

0.9818

0.22

18.02

0.9888

0.31

18.98

0.985

1.49

19.38

1/𝛤𝑒 =

Table 3. Parameters of Freundlich adsorption model Surfactant

Correlation

R2

KF

1/n

BW

𝛤𝑒 = 0.062𝑐0.8248 𝑒

0.9658

0.062

0.8248

CA

𝛤𝑒 = 0.1345𝑐0.7444 𝑒

0.9684

0.1345

0.7444

CS-50

𝛤𝑒 = 0.5618𝑐0.6522 𝑒

0.9633

0.5618

0.6522

3.3.5. Effects of Temperature. The adsorption at higher temperatures (80 ℃ and 105 ℃) were also studied

ACS Paragon Plus Environment

Energy & Fuels

(Figure 7). The equilibrium adsorption density at 80 ℃ is slightly smaller than that of 20 ℃ when the original surfactant concentration is low (100 mg/L), and the difference is more prominent at higher concentration (1000 mg/L). Adsorption is an exothermic process, so in ideal situations when active surfactant components remain constant, the adsorption density of zwitterionic surfactants at higher temperatures should be lower. However, for our case, surfactant degradation and/or solubility change at temperature as high as 105 ℃ and salinity around 290,000 mg/L are inevitable and noticeable especially for surfactant CS-50, so the variation trend in the temperature range of 80 ~ 105 ℃ was more complicated. 1.6 1.4

14

20℃ 80℃ 105℃

12

Adsorption density, mg/g

1.2

Adsorption density, mg/g

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

Page 14 of 31

1.0 0.8 0.6 0.4

20℃ 80℃ 105℃

10

8

6

4

2

0.2 0.0

0 BW

CA

CS-50

BW

(a)

CA

CS-50

(b)

Figure 7. Effects of temperature on adsorption density. (a) Original surfactant concentration=100 mg/L. (b) Original surfactant concentration=1000 mg/L.

3.3.6. Effects of Mineral Types. The composition of Bakken minerals is complicated, which can be divided into three types, quartz, carbonate and clay. Herein, Berea sandstone, calcite and clay were used to represent different mineral groups, and the adsorption of 1000 mg/L surfactant solutions on individual mineral group was studied separately (Figure 8). The zeta potential of different minerals at different salinities are presented in Table S2. Experimental results showed that the surfactant adsorption on clay minerals in this case is far away from the equilibrium state.

ACS Paragon Plus Environment

Page 15 of 31

20 18 16

Adsorption density, mg/g

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

Energy & Fuels

14

Bakken-deionized water Bakken-formation brine Berea-deionized water Berea-formation brine Calcite-deionized water Calcite-formation brine Clay-deionized water Clay-formation brine

12 10 8 6 4 2 0 BW

CA

CS

Figure 8. Effects of mineral types on adsorption density.

Zwitterionic surfactants have higher loss on clay and quartz than on calcite surfaces no matter of the salinity, which was just in accordance with the trend of specific surface area, clay has the largest surface area of 18.63 m2/g, followed by Berea sandstone of 2.85 m2/g and calcite of 2.12 m2/g. Results demonstrate that the adsorption of zwitterionic surfactants on Bakken minerals are strongly influenced by clay and quartz ratios. Surfactant loss in Bakken formation due to adsorption might increase with increasing clay and quartz contents. Apart from large clay and quartz contents, the configuration of surfactant molecules also matters. Normally, zwitterionic molecules have three possible configurations, 1) when mineral is strongly negatively charged, the cationic quaternary nitrogen group would orient to the surfaces with anionic part moves away (Mode 1). 2) when surface negative charge is reduced by the adsorption of cations, oblique configuration appeared with sulfonate or carboxylate group getting closer to the mineral surfaces (either because of reduced repulsion interaction or increased attraction interactions with absorbed Ca2+ and Mg2+) and cationic parts remain attracted by the net negative surface charge (Mode 2), and 3) vertical configuration when minerals become absolute positively charged. Cationic parts are repulsed and anionic parts are attracted (Mode 3). Thereinto, oblique configuration contributes to the minimum adsorption because of the largest contact area of a single molecule19, , as illustrated in Figure 9. In addition, sulfobetaine with sulfonate group was observed to have higher

21

adsorption than that of betaine with carboxylate group, which was contrary to the observations of Li20. The

ACS Paragon Plus Environment

Energy & Fuels

main reason lies in the surfactant structures. Different structures have different interaction energies and higher negative interaction energies result in higher adsorption.

(a)

(b)

(c)

Figure 9. Schematic models of suggested adsorption mechanism of zwitterionic surfactants at different salinities. (a) Deionized water. (b) Low salinity with few divalent cations. (c) High salinity with large divalent cations.

Obviously, the salinity of Bakken formation brine is very high, and the adsorption configuration shift to mode 3 from mode 1 directly, which offers a good opportunity for later research on finding the optimal salinity so as to minimize the adsorption of zwitterionic surfactants. 3.3.7. Effects of Surfactant Types. Besides zwitterionic surfactants, nonionic surfactants and anionic surfactants are also popular in unconventional reservoirs. In this section, the nonionic surfactant HCS and the anionic surfactant 964 were applied for comparison (Figure 10). Relevant experiments were implemented at room temperature, ambient pressure with initial surfactant concentration of 1000 mg/L. 14

12

Adsorption density, mg/g

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

Page 16 of 31

Deionized water Formation brine

10

8

6

4

2

0 BW

CA

CS-50

HCS

ACS Paragon Plus Environment

964

Page 17 of 31 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

Energy & Fuels

Figure 10. Effects of surfactant types on adsorption density.

Zwitterionic surfactants have higher adsorption than either nonionic or anionic surfactant regardless of the salinity on Bakken minerals with large clay and quartz contents. Middle Bakken mineralogy is known to vary from dolomitic sandstone to silty dolostone, so according to our results, zwitterionic surfactants may be more suitable to be used as the main component in a surfactant flooding formula for dolostone formations while used as an additive in sandstone formations to help increase the salt tolerance, thermal tolerance and efficiency of other surfactants. The main driving force varies with surfactants structures. For ionic surfactants (Figure 11), electrostatic interactions play the leading role, therefore, the adsorption of ionic surfactant might be easily affected by environmental salinity and other factors that would change the surface potential of adsorbents. While for nonionic surfactants (Figure 12), hydrogen bonding and hydrophobic interactions are believed to be the main adsorption mechanisms. Under such circumstance, salinity will not significantly affect the distribution of surface hydroxy but may contribute to a more compact adsorption pattern. Surfactant adsorption is closed related to chemical structures, environmental salinity and temperature, adsorbents, etc. Small changes can result in huge differences.

(a)

(b)

Figure 11. Schematic models of suggested adsorption mechanism of ionic surfactant at different salinities. (a) Deionized water. (b) High salinity with large divalent cations.

ACS Paragon Plus Environment

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

(a)

Page 18 of 31

(b)

Figure 12. Schematic models of suggested adsorption mechanism of nonionic surfactant at different salinities. (a) Deionized water. (b) High salinity with large divalent cations.

Surfactant adsorption can be affected by initial surfactant concentration, salinity, temperature, chemical structures, as well as mineral types. But which factor is more influential and what are the interaction effects between various factors are still obscure. Sometimes, studying the effects of a single factor while fixing other parameters is not sufficient to draw professional conclusions. So, in this paper, a thoroughly statistical analysis was done through Minitab 2016 (a data processing software). To be more specifically, the 2k full factorial design was applied. The salinity was set to be Bakken salinity and other details were shown in Supporting Information Table S3. In order to make accurate predictions, the initial surfactant concentration range was divided into two parts, 100 ~ 600 mg/L and 600 ~ 1000 mg/L. Herein, surfactants BW and CA were selected because of higher stability. After primary screening, only one-way interaction and two-way interactions (interactions between every two factors) were studied. Their specific effects on equilibrium adsorption density were calculated directly, and their ranking was clarified, as presented in Table 4. Either in Set 1 or Set 2, the most significant factors remain to be initial surfactant concentration, mineral (adsorbents) and the interaction effects of concentration and mineral. In addition, other interaction effects such as surfactant and mineral, temperature and mineral also matter. For Set 2, when initial surfactant concentration is relatively high, the impacts of temperature, interaction effects of surfactant and concentration, surfactant and temperature were enlarged while the effects of temperature-mineral and temperature-concentration interactions were weakened. In this paper, the influence of surfactant structure is not so obvious for surfactants BW and

ACS Paragon Plus Environment

Page 19 of 31 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

Energy & Fuels

CA. To predict the approximate adsorption at some other conditions, the contour plots of adsorption versus mineral, concentration, and temperature for different simulation sets were plotted, as presented in Figures 13 and 14. The x and y axials of the red dots shown on the plot represent the experimental conditions and the values are experimental data (real values), which are consistent with the predicted ranges. Table 4. Effects and ranking of different influencing factors Factor

Set 1

Set 2

Effect

Ranking

Effect

Ranking

Surfactant

0.2022

7

0.303

7

Concentration

3.7647

1

1.732

2

Temperature

-0.2200

6

-0.773

4

Mineral

-1.6492

2

-4.023

1

Surfactant * Concentration

0.1515

9

-0.301

8

Surfactant * Temperature

0.1235

10

0.304

6

Surfactant * Mineral

-0.2806

5

-0.315

5

Concentration * Temperature

-0.1847

8

0.048

10

Concentration * Mineral

-1.3734

3

-1.195

3

Temperature * Mineral

-0.2979

4

0.050

9

ACS Paragon Plus Environment

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

(a)

Page 20 of 31

(b)

Figure 13. Representative contour plots for simulation Set 1. (a) Adsorption vs. Mineral and Concentration. (b) Adsorption vs. Temperature and Concentration.

(a)

(b)

Figure 14. Representative contour plots for simulation Set 2. (a) Adsorption vs. Mineral and Concentration. (b) Adsorption vs. Temperature and Concentration.

4. CONCLUSIONS (1) When Bakken formation brine was used, the adsorption of BW and CA showed a 2.06 ± 0.02 mg/g decrease while CS-50 increased a little bit compared with those in deionized water. (2) The adsorption of zwitterionic surfactants increases with increasing surfactant concentration at a restrained range. All the three researched zwitterionic surfactants showed higher increasing rate in the range of 100 ~ 600 mg/L than that of 800 ~ 1000 mg/L at Bakken conditions. (3) Higher temperature leads to decreased adsorption of zwitterionic surfactants, but considering the degradation phenomenon and/or solubility difference at elevated temperatures, the trend is not obvious for the temperature range of 80 ~ 105 ℃. (4) Comparing with nonionic surfactant HCS and anionic surfactant 964, zwitterionic surfactant has higher adsorption on Bakken minerals with large clay and quartz contents.

ACS Paragon Plus Environment

Page 21 of 31 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

Energy & Fuels

(5) Zwitterionic surfactants have higher loss on clay and quartz than calcite. The high adsorption of researched zwitterionic surfactants on Bakken minerals regardless of salinity may be ascribed to the fairly large clay and quartz contents. (6) The main driving force for zwitterionic surfactants adsorption is electrostatic interaction. CORRESPONDING AUTHORS * Hui Pu Tel: (+1) 701-777-6861 E-mail: [email protected]. * Julia Xiaojun Zhao Tel: (+1) 701-777-3610 E-mail: [email protected]. ACKNOWLEDGEMENTS The authors would like to thank the North Dakota Industrial Commission Oil and Gas Research Program (Contract No. G-041-081) for the financial support. Acknowledgement is extended to the surfactant provider. Also thank North Dakota Geological Survey Wilson M. Laird Core and Sample Library for providing Bakken rock samples. REFERENCES (1) Janet, K.P.; Leigh, C.P. and Julie, A.L. U.S. Geological Survey Professional Paper 2001, 1653, 1-24. (2) Sonnenberg, S. A.; Vickery, J.; Theloy, C. and Sarg, J.F. Proceedings of the 2011 American Association of Petroleum Geologists (AAPG) Annual Convention and Exhibition; Houston, Texas, Apr. 10-13, 2011; SPE Paper 50449. (3) Rui, Z.; Cui, K.; Wang, X.; et al. J. Pet. Sci. Eng. 2018, 166, 900-905. (4) Wang, X.X.; Hou, J.G.; Song, S.H.; et al. J. Pet. Sci. Eng. 2018, 171, 353-361. (5) Nieto-Alvarez, D.A.; Zamudio-Rivera, L.S.; Luna-Rojero, E.E.; et al. Langmuir 2014, 30 (41), 12243-

ACS Paragon Plus Environment

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

12249. (6) Da, C.; Alzobaidi, S.; Jian, G. Q.; et al. J. Pet. Sci. Eng. 2018, 16, 880-890. (7) Wang, Y.; Ge, J.J.; Zhang, G.C.; et al. RSC Adv. 2015, 5, 59738-59744. (8) Wang, D.; Liu, C.; Wu, W.; Wang, G. Proceedings of the 2010 Society of Petroleum Engineers (SPE) Enhanced Oil Recovery (EOR) Conference at Oil and Gas West Asia; Muscat, Oman, April 11-13, 2010; SPE Paper 127452. (9) Li, R. F.; Hirasaki, G.; Miller, C. A. and Masalmeh, S. K. SPE J. 2012, 17(4), 1207-1220. (10) Zhao, L.; Li, A. F.; Chen, K.; et al. J. Pet. Sci. Eng. 2012, 81, 18-23. (11) Prajapati, R. R.; Bhagwat, S. S. J. Chem. Eng. Data 2012, 57 (3), 869-874. (12) Kamal, M. S.; Hussain, S. M. S. and Fogang, L. T. J. Surfact. Deterg. 2018, 21, 165-174. (13) Kumar, A. and Mandal, A. J. Mol. Liq. 2017, 243, 61-71. (14) Guo, S. F.; Wang, H. Y.; Shi, J.; et al. J. Petrol Explor Prod Technol. 2015, 5, 321-326 (15) Koopal, L.K.; Lee, E.M.; Bohmer, M. R. J. Colloid Interface Sci. 1995, 170, 85-97. (16) Xu, Z.; Yang, X. and Yang, Z. J. Phys. Chem. B 2008, 112, 13802-13811. (17) Ahmadi, M.A.; Shadizadeh, S. R. Fuel 2013, 104, 462-467. (18) Amirianshoja, T.; Junin, R.; Idris, A. K. and Rahmani, O. A. J. Pet. Sci. Eng. 2013, 101, 21-27. (19) Duran-Alvarez, A.; Moldonado-Dominguez, M; Gonzalez-Antonio, O.; et al. Langmuir 2016, 32(11), 2608-2616. (20) Li, N.; Zhang, G.; Ge, J.; et al. Energy Fuels 2011, 25, 4430-4437. (21) Nieto-Alvarez, D. A.; Marn-Leon, A.; Luna-Rojero, E. E.; et al. Ind. Eng. Chem. Res. 2018, 57(6), 20752082. (22) Jian, G. Q.; Puerto, M.; Wehowsky, A.; et al. J. Colloid Interface Sci. 2018, 513(1), 684-692. (23) Ball, B.; Fuerstenau, D. W. Discuss. Faraday Soc. 1971, 52, 361-371. (24) Paria, S.; Kartic, C. K. Adv. Colloid Interface Sci. 2004, 110, 75-95.

ACS Paragon Plus Environment

Page 22 of 31

Page 23 of 31 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

Energy & Fuels

(25) Cui, G.; Wang, Y.; Chen, B.; et al. Energy 2018, 155, 281-296. (26) Cui, G.; Ren, S.; Ezekiel, J.; et al. Appl. Energy 2018, 227. 49-63. (27) Wasan, D. T.; Ginn, M.; Shah, D. O. New York: Marcel Dekker, 1988, pp. 52-86. (28) Wei, B.; Li, Q.; Jin, F.; et al. Energy Fuels 2016, 30, 2882-2891. (29) Jian, G. Q.; Puerto, M. C.; Wehosky, A.; et al. Langmuir 2016, 32(40), 10244-10252. (30) Halt, S. K.; Moulik, S. P. J. Surf. Deterg. 2001, 4, 303. (31) Wang, D. M.; Butler, R.; Liu, H. and Ahmed, S. Proceedings of the 2011 Society of Petroleum Engineers (SPE) Annual Technical Conference and Exhibition; Denver, Colorado, Oct. 30-Nov. 2, 2011; SPE Paper 145510. (32) Ahmadi, M. A. and Shadizadeh, S. R. Energy Fuels 2012, 26, 4655-4663. (33) Langmuir, I. J. Am. Chem. Soc. 1916, 38, 2221-2295. (34) Rosen, M. J. John Wiley & Sons, Inc.: New York, 1989, pp. 33-63. (35) Kireev, S. V. and Shnyrev, S. L. Laser Phys. 2015, 25, 1-11. (36) Wu, Y. N.; Chen, W. X.; Dai, C. L.; et al. J. Pet. Sci. Eng. 2017, 153, 283-287.

ACS Paragon Plus Environment

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

Page 24 of 31

Comparative Study on the Static Adsorption Behavior of Zwitterionic Surfactants on Minerals in Middle Bakken Formation Xun Zhong a, Hui Pu a *, Yanxia Zhou b, c, Julia Xiaojun Zhao b* a Department of Petroleum Engineering, University of North Dakota, Grand Forks, North Dakota 58202, United States b Department of Chemistry, University of North Dakota, Grand Forks, North Dakota 58202, United States c College of Petroleum Engineering, Northeast Petroleum University, Daqing, Heilongjiang 161318, PR China

(a)

(b)

(c)

(d)

Figure 1. SEM images of various adsorbents, Scale bar=10.0 μm. (a) Bakken minerals. (b) Berea sandstone. (c) Calcite. (d) Clay.

ACS Paragon Plus Environment

Page 25 of 31

90

45

Bakken minerals

40

80

35

70

30

60

Content, wt.%

Content, wt.%

41.5

25 20 15.8 13.4

15

11.9

10

Berea sandstone

79.6

50 40 30 20

7.9 6.2

5

10.2

10

2.1 1.1

0

1.1

2.9

3.4

Dolomite

Albite

Muscovite

2.8

0 Quartz

Calcite Dolomite Albite Sanidine Pyrite

Illite

Kaolinite

Quartz

(a)

Sanidine

Kaolinite

(b)

Figure 2. Compositions of Bakken minerals and Berea sandstone. (a) Bakken minerals. (b) Berea sandstone. 0.7

0.36 0.33

Deionized water Bakken Brine

-

I3 , λ=352 nm

0.30

2 mg/L 656 30 mg/L BW 100 mg/L CA 100 mg/L CS-50 6 mg/L HCS 100 mg/L 964

0.6

0.27

0.5

Absorbance, a.u.

0.24

Absorbance, a.u.

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

Energy & Fuels

0.21 0.18

I2, λ=460 nm

0.15

λ=433 nm

0.12

428 nm

0.4

365 nm 427 nm

0.3

426 nm

0.2 0.09 0.06

0.1

362 nm

0.03 0.00 300

320

340

360

380

400

420

440

460

480

500

0.0 300

320

340

360

380

400

420

440

460

480

500

Wavelength, nm

Wavelength, nm

(a)

(b)

Figure 3. Absorption spectra of KI-I2 mixtures in the presence Bakken formation brine and/or surfactants. (a) In the absence of Bakken formation brine alone. (b) In the presence of Bakken formation brine and different surfactants.

ACS Paragon Plus Environment

Energy & Fuels

1400

Original 20 ℃-7d 80 ℃-7d 105 ℃-7d

1300 1200 1100

Concentration, mg/L

1000 900 800 700 600 500 400 300 200 100 0 BW

CA

CS-50

HCS

964

Figure 4. Effective surfactants concentrations after aging in given conditions for 7 days. 12

12 11

11 10 9

Adsorption density, mg/g

10

Adsorption density, mg/g

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

Page 26 of 31

9 8 7 6

8 7 6 5 4 3 2

5 1 0

4 4:1

6:1

8:1

10:1

12:1

15:1

20:1

0

5

10

15

20

25

30

35

40

45

Experimental period, h

Liquid/solid ratio

(a)

(b)

Figure 5. Optimization of static adsorption parameters. (a) Liquid/solid ratio. (b) Experimental period.

ACS Paragon Plus Environment

50

55

Page 27 of 31

14

14

BW CA CS-50

BW CA CS-50

12

10

Adsorption density, mg/g

Adsorption density, mg/g

12

8

6

4

10

8

6

4

2

2

0

0 0

25

50

75

100

125

150

175

200

225

250

275

0

300

50

100

150

200

250

300

350

400

450

Equilibrium concentration, mg/L

Equilibrium concentration, mg/L

(a)

(b)

Figure 6. Effects of salinity on adsorption density. (a) Deionized water (TDS≈500 mg/L). (b) Bakken formation brine (TDS=289,820 mg/L). 1.6 1.4

14

20℃ 80℃ 105℃

12

20℃ 80℃ 105℃

Adsorption density, mg/g

1.2

Adsorption density, mg/g

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

Energy & Fuels

1.0 0.8 0.6 0.4

10

8

6

4

2

0.2 0.0

0 BW

CA

CS-50

BW

(a)

CA

CS-50

(b)

Figure 7. Effects of temperature on adsorption density. (a) Original surfactant concentration=100 mg/L. (b) Original surfactant concentration=1000 mg/L.

ACS Paragon Plus Environment

Energy & Fuels

20 18

Adsorption density, mg/g

16 14

Bakken-deionized water Bakken-formation brine Berea-deionized water Berea-formation brine Calcite-deionized water Calcite-formation brine Clay-deionized water Clay-formation brine

12 10 8 6 4 2 0 BW

CA

CS

Figure 8. Effects of mineral types on adsorption density.

(a)

(b)

(c)

Figure 9. Schematic models of suggested adsorption mechanism of zwitterionic surfactants at different salinities. (a) Deionized water. (b) Low salinity with few divalent cations. (c) High salinity with large divalent cations. 14

12

Adsorption density, mg/g

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

Page 28 of 31

Deionized water Formation brine

10

8

6

4

2

0 BW

CA

CS-50

HCS

964

Figure 10. Effects of surfactant types on adsorption density.

ACS Paragon Plus Environment

Page 29 of 31 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

Energy & Fuels

(a)

(b)

Figure 11. Schematic models of suggested adsorption mechanism of ionic surfactant at different salinities. (a) Deionized water. (b) High salinity with large divalent cations.

(a)

(b)

Figure 12. Schematic models of suggested adsorption mechanism of nonionic surfactant at different salinities. (a) Deionized water. (b) High salinity with large divalent cations.

(a)

(b)

Figure 13. Representative contour plots for simulation Set 1. (a) Adsorption vs. Mineral and Concentration. (b) Adsorption vs. Temperature and Concentration.

ACS Paragon Plus Environment

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

Page 30 of 31

(a)

(b)

Figure 14. Representative contour plots for simulation Set 2. (a) Adsorption vs. Mineral and Concentration. (b) Adsorption vs. Temperature and Concentration.

Table 1 Details of surfactants Surfactants

Type

Active, %

Manufacture

Molecular formula

BW

Zwitterionic

35.9

Lubrizol

C12H25N+(CH3)2CH2COO-

CA

Zwitterionic

29.4

Stepan

C11H23CO-NH-(CH2)3N+(CH3)2CH2COO-

CS-50

Zwitterionic

43.5

Stepan

C11H23CO-NH-(CH2)3N+(CH3)2CH2CH(OH)CH2SO3-

HCS

Nonionic

60.0

Stepan

CH3(CH2)m-O-(CH2CH2O)nH

964

Anionic

86.8

Sasol

CH3(CH2)m-O-(CH2CH(CH3)O)-(CH2CH2O)nCH2COOH

Table 2. Parameters of Langmuir adsorption model Surfactant

Correlation

R2

KL, *102 L/mg

𝛤 max

BW

1/𝛤𝑒 =

25.495 + 0.0555 𝑐𝑒

0.9818

0.22

18.02

CA

1/𝛤𝑒 =

16.806 + 0.0527 𝑐𝑒

0.9888

0.31

18.98

ACS Paragon Plus Environment

Page 31 of 31 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

Energy & Fuels

CS-50

1/𝛤𝑒 =

3.4583 + 0.0516 𝑐𝑒

0.985

1.49

19.38

Table 3. Parameters of Freundlich adsorption model Surfactant

Correlation

R2

KF

1/n

BW

𝛤𝑒 = 0.062𝑐𝑒0.8248

0.9658

0.062

0.8248

CA

𝛤𝑒 = 0.1345𝑐𝑒0.7444

0.9684

0.1345

0.7444

CS-50

𝛤𝑒 = 0.5618𝑐𝑒0.6522

0.9633

0.5618

0.6522

Table 4. Effects and ranking of different influencing factors Factor

Set 1

Set 2

Effect

Ranking

Effect

Ranking

Surfactant

0.2022

7

0.303

7

Concentration

3.7647

1

1.732

2

Temperature

-0.2200

6

-0.773

4

Mineral

-1.6492

2

-4.023

1

Surfactant * Concentration

0.1515

9

-0.301

8

Surfactant * Temperature

0.1235

10

0.304

6

Surfactant * Mineral

-0.2806

5

-0.315

5

Concentration * Temperature

-0.1847

8

0.048

10

Concentration * Mineral

-1.3734

3

-1.195

3

Temperature * Mineral

-0.2979

4

0.050

9

ACS Paragon Plus Environment