pH-Responsive Water-Based Drilling Fluids Containing Bentonite and

Feb 6, 2018 - Consequently, the MMT platelets would not be able to distribute in a layer-by-layer manner in the suspension, leading to the deteriorati...
0 downloads 5 Views 3MB Size
Research Article Cite This: ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

pubs.acs.org/journal/ascecg

pH-Responsive Water-Based Drilling Fluids Containing Bentonite and Chitin Nanocrystals Mei-Chun Li,† Qinglin Wu,*,† Kunlin Song,† Alfred D. French,‡ Changtong Mei,§ and Tingzhou Lei∥ †

School of Renewable Natural Resources, Louisiana State University AgCenter, Baton Rouge, Louisiana 70803, United States Southern Regional Research Center, United States Department of Agriculture, New Orleans, Louisiana 70124, United States § College of Materials Science and Engineering, Nanjing Forestry University, Nanjing, 210037, China ∥ Key Biomass Energy Laboratory of Henan Province, Zhengzhou, 450008, Henan, China ‡

S Supporting Information *

ABSTRACT: The forthcoming era of sustainable development demands the utilization of naturally abundant, sustainable, biodegradable, and environmentally friendly resources for a wide variety of applications. Herein, we report the isolation of chitin nanocrystals (ChNCs) from speckled swimming crab shell waste and their novel utilization as rheology modifiers in bentonitewater-based drilling fluids (BT-WDFs). The positively charged functional groups on the surface of ChNCs enable them to attach to the negatively charged face surfaces of BT by electrostatic attraction, leading to a notable improvement in the rheological property that is responsible for the ability of drilling fluids to carry away drill cuttings. Moreover, owing to the reversible protonation and deprotonation of ChNCs and BT platelets, ChNC/BT-WDFs can form distinctive dispersion states and nanostructured surface interactions under different pH conditions. As a result, ChNC/BT-WDFs exhibit interesting pHresponsive rheological and filtration performance, especially when the pH is changed from neutral to acidic. Finally, the combined use of ChNCs with a conventional filtration control agentpolyanionic cellulose(PAC) leads to remarkable improvement in both rheological and filtration performance of BT-WDFs. It is expected that these observations not only enrich the high-value utilization of crab shell waste but also advance the development of smart nanofluids with pH-responsiveness for various applications. KEYWORDS: Drilling fluids, Chitin nanocrystals, Rheology, Filtration, pH-responsive, Interface



Despite the oil-based and synthetic-based drilling fluids performing better in well excavation, especially for high pressure/high temperature wells, the bentonite-water-based drilling fluids (BT-WDFs) are more popular in drilling fields because of their low cost and low environmental impact.6 Approximately 80% of wells in the world were excavated using the BT-WDFs. In recent years, the low-solid, smart/nano BT-WDFs have attracted considerable attention from both academic researchers and drilling engineers due to several advantages, e.g., high

INTRODUCTION

The energy industry has been gaining benefits from the development of drilling technology. Drilling operations involving the pumping of drilling fluids into the wellbore and subsequently transporting the drill cuttings out of the wellbore are critical for successful oil and gas exploration and production. Therefore, drilling fluids are usually designed as shear-thinning non-Newtonian fluids, which have high viscosity at low shear rates to suspend/transport drill cuttings to the surface but low viscosity at high shear rates in order to be rapidly pumped into the wellbore. In addition to superior carrying capacity for drilling cuttings, drilling fluids provide other functions like lubricating and cooling the drilling bit, minimizing formation damage, and stabilizing the wellbore.1−5 © XXXX American Chemical Society

Received: November 9, 2017 Revised: January 24, 2018 Published: February 6, 2018 A

DOI: 10.1021/acssuschemeng.7b04156 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering drilling rate, low friction in all drilling equipment, thin filter cake, less pipe sticking, good shale stability, and tailor-made rheological and filtration properties.1 The major drawback for low-solid BT-WDFs has always been their limited carrying capacity for drill cuttings because of the low BT content. Much effort has been devoted to improving the carrying capacity for drill cuttings by adding rheology modifiers, e.g., inorganic nanoparticles (graphene oxide, carbon nanotube, SiO2, Fe3O4, CuO, and ZnO),3,7−11 synthetic polymers,12−14 and their hybrid nanocomposites.15−17 Despite these rheology modifiers allowing effective improvement in the rheological performance of BT-WDFs, their sustainability has been a big concern. This has driven us to keep exploring naturally abundant, inexpensive, sustainable, and environmentally friendly resources as rheology modifiers for the formulation of low-solid, smart/nano BTWDFs. In previous work, we demonstrated that cellulose nanocrystals (CNCs) acted as efficient rheology modifiers in BT-WDFs due to the attachment of CNCs on the edge surfaces of BT.18 Nowadays, about 6−8 million tons of lobster, shrimp, and crab shells are produced worldwide annually.19 In most countries, the shells are commonly discarded as industrial or food waste in landfills or at sea without effective utilization. The shells are rich in calcium carbonate, protein, and chitin, all of which are valuable chemicals, but the extraction of these chemicals from the shell waste on an industrial scale is being ignored. Chitin, β-(1−4)-poly-N-acetyl-D-glucosamine, is a semicrystalline polysaccharide present in the form of microfibrils in the shells of crustaceans.20−22 The weight of chitin in the shells of crustaceans ranges from 10% to 30% depending on the source. Through a series of treatments, i.e., deproteinization, demineralization, and depigmentation using acid and alkali, chitin can be isolated from the shells.23 Further acid hydrolysis of the isolated chitin cleaves the amorphous regions, producing highly crystalline and rod-like nanoparticles−chitin nanocrystals (ChNCs).24,25 ChNCs are similar to CNCs in shape and dimension, whereas they have different surface characteristics. ChNCs inherently possess a large number of amino and positively charged NH3+ groups on the backbone,26 whereas CNCs contain tiny negatively charged carboxylate or sulfate groups on the backbone that were introduced during the preparation process using TEMPO oxidation or sulfuric acid hydrolysis, respectively.27,28 Therefore, the natural presence of abundant amino and positively charged NH3+ groups on the surfaces of ChNCs allowed them to form strong interfacial interactions with other active materials (e.g., BT) through electrostatic attraction, which might be one important advantage of ChNCs compared with CNCs. Herein, we report the isolation of ChNCs from speckled swimming crab shells and their utilization as rheology modifiers for the formulation of low-solid, smart/nano BT-WDFs. The speckled swimming crab, Arenaeus cribrarius−a common species widely distributed all along Atlantic, Caribbean, and Gulf of Mexico shorelines, was used as starting materials. Through a multistep procedure, including grinding, deproteinization, demineralization, and depigmentation, chitin was successfully isolated from the speckled swimming crab shells. ChNCs were then prepared from the extracted chitin via the combination treatments of hydrochloric acid hydrolysis and high pressure homogenization. The pH-responsive behaviors of ChNCs were comprehensively investigated using several analytic techniques, i.e., transmission electron microscope (TEM), zeta potential, visual inspection, ultraviolet−visible

(UV−vis) spectroscopy, and rheometry. Afterward, the ChNCs obtained were used to formulate low-solid, smart/nano ChNC/ BT-WDFs, and their rheological and filtration performance were studied as a function of pH value. Distinctive phenomena were observed and interpreted in terms of dispersion state and surface interaction between ChNCs and BT platelets. Finally, a conventional filtration control agentpolyanionic cellulose (PAC)was further added into the formulated ChNC/BTWDFs. The synergic effect of PAC and ChNC in improving rheological and filtration performance of BT-WDFs was achieved.



EXPERIMENTAL SECTION

Materials. Fresh speckled swimming crabs were collected from Pensacola Beach, Florida, USA. The crab meat was removed carefully after steaming in a pot. The resultant shell wastes were washed with water and then dried in an oven. Wyoming sodium bentonite (AQUAGEL GOLD SEAL, dry-powdered, 200 Mesh) was provided from Baroid Industrial Drilling Products Inc., Houston, TX, USA. Polyanionic cellulose (PAC) was purchased from Halliburton Company, Houston, TX, USA. Hydrochloric acid (HCl, assay 36.5− 38%) was provided from VWR Company, West Chester, PA, USA. Sodium hydroxide (NaOH) pellets were obtained from Sigma-Aldrich St. Louis, MO, USA. Ethanol (assay 95%) was supplied from Carolina Biological Supply Company, Burlington, NC, USA. Preparation of Chitin and Chitin Nanocrystals (ChNCs) from the Speckled Swimming Crab Shells. The isolation of chitin from speckled swimming crab shells is schematically illustrated in Figure 1,

Figure 1. Schematic illustration of the extraction of ChNCs from speckled swimming crab shells. following the protocols reported previously.23,29 First, the crab shells were ground into fine powders using a laboratory high-speed rotor mill (Columbia International Technical Equipment & Supplies LLC, Irmo, SC, USA). Afterward, the fine powders were chemically treated in sodium hydroxide, hydrochloride acid, and ethanol solutions to eliminate proteins, calcium carbonate, and pigments, respectively. In detail, the powders were treated with 5 wt % NaOH solution at 65 °C for 6 h to remove the proteins. The suspension was then washed with excess deionized water to a neutral pH and filtered. Subsequently, demineralization was carried out in 7 wt % HCl solution at 25 °C for 3 days. After the washing and filtering with excess deionized water, the B

DOI: 10.1021/acssuschemeng.7b04156 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

Figure 2. Chemical structure of crab shell powders (CSP), chitin, and ChNCs: (a) FTIR spectra and (b) XRD patterns. samples were further treated with 5 wt % NaOH solution at 65 °C for another 2 days in order to completely remove the residual proteins. The resultant mixtures were further filtered and washed with deionized water for several times to a neutral pH. Finally, the pigment composition was removed using 95% ethanol for 6 h at 25 °C. ChNCs were extracted from the purified chitin using a 3 N HCl solution at 100 °C for 3 h in accordance with the literature with some modifications.30 To terminate the acid hydrolysis, excess deionized water was added into the suspension. The suspension was then washed by centrifugation, followed by dialysis against deionized water for several days until a neutral pH was achieved. In order to further decrease the particle size of ChNCs, high-pressure homogenization treatment (Microfluidizer M-110P, Microfluidics Corp., Newton, MA, USA) was carried out under an operating pressure of 138 MPa three times. Finally, the pH of the ChCN suspension was adjusted to 3.4, 7.0, and 9.0 using the diluted HCl or NaOH solution for further use. Characterization of Purified Chitin and ChNCs. The extracted chitin and ChNC suspensions were dried in an oven under vacuum for several days to obtain the solid films. The chemical and crystalline structure of chitin and ChNC films were studied using Fourier transform infrared spectroscopy (FTIR) and X-ray diffraction (XRD). FTIR spectra were recorded in the transmittance mode using a Bruker FTIR analyzer (Tensor-27, Bruker Optics Inc., Billerica, MA), equipped with a Zn/Se attenuated total reflectance crystal accessory. The wavenumber increased from 600 to 4000 cm−1 with a resolution of 4.0 cm−1. XRD patterns were collected in a Bruker Siemens D5000 X-ray diffractometer using the Cu−Kα radiation (λ = 0.154 nm). Measurements were performed at 40 kV and 30 mA in the range from 5° to 35°. Zeta potential values of 0.1 wt % ChNC suspensions were measured using a ZetaTrac analyzer (MicroTrac Inc., Largo, FL, USA). The morphology of ChNCs was examined using TEM (JEM 1400, JEOL) at an accelerating voltage of 120 kV. Prior to TEM observations, ChNC suspensions were diluted to 0.02 wt %, followed by magnetic stirring (Item CG-1994-V-50, Chemgalss) at s speed of 1000 rpm for 30 min. Glow discharge was applied to treat the copper grids (CF-400-CU, Electron Microscopy Sciences, Hatfield, PA, USA) before the dropping ChNC suspension onto the grids. In order to improve the contrast, the samples were stained using a 2 wt % uranyl acetate solution for 2 min. Visual appearance of 0.1 wt % ChNC suspensions in 20 mL vials were recorded by a digital camera. Prior to recording, ChNC suspensions were vigorously stirred using a magnetic bar at s speed of 1000 rpm for 5 min. After 30 s of stabilization, their digital photos were taken. UV−vis spectra of 0.1 wt % ChNC suspensions were collected using a UV−vis spectrophotometer (Evolution 600, Thermo Electron Corp., USA) with the wavelength ranging from 200 to 900 nm. Rheological properties of 1 wt % ChNC suspensions were studied using a stress-controlled rheometer (AR 2000, TA Instrument Inc., New Castle, DE, USA) with a cone-and-

plate geometry (cone angle 20; diameter 40 mm; truncation 56 μm) at 25 °C. The steady-state viscosity versus shear rate was conducted in the shear rate region from 1 to 100 s−1. Prior to dynamic frequency sweeps, dynamic strain sweeps were carried out at a fixed frequency of 1 Hz in the strain region from 0.1 to 100% in order to determine the linear viscoelastic region. Dynamic frequency sweeps were performed in the angular frequency region from 0.1 to 100 rad/s within the linear viscoelastic region. Formulation of BT, chitin/BT, ChNC/BT, PAC/BT, and PAC/ ChNC/BT-WDFs. BT-WDFs with 3 wt % BT (base on the weight of aqueous solution) were prepared by adding BT powders into water, followed by mechanically stirring at a rate of 10 000 rpm for 1 h. In case of chitin/BT and ChNC/BT-WDFs, chitin and ChNC suspensions were first diluted into 0.1 wt %. Then, 3 wt % BT powders (based on the weight of chitin or ChNC suspension) were slowly added into the resultant suspensions. The mixture was vigorously stirred at a rate of 10 000 rpm for 1 h, too. PAC/BT and PAC/ChNC/BT-WDFs were also formulated by compounding 0.2 and 0.1 wt % PAC into the resultant BT and ChNC/BT-WDFs under the same conditions, respectively. The pH values of WDFs were adjusted to 3.4, 7.0, and 9.0 using the diluted HCl or NaOH solution prior to measurements. Characterization of BT, chitin/BT, ChNC/BT, PAC/BT, and PAC/ChNC/BT-WDFs. Rheological properties of WDFs was examined using a stress-controlled rheometer (AR 2000, TA Instrument, New Castle, DE, USA) equipped with a DIN concentric cylinder geometry containing a stainless steel cup (30.38 mm in diameter) and a rotator (28.03 mm in diameter). Shear flow measurements were performed at 25 °C in the shear rate region from 0.1 to 1200 s−1. Dynamic time sweeps were carried out at a constant frequency of 1 Hz and strain of 10%. Before each test, the sample was presheared at a shear rate of 10 s−1 for 30 s and, then, stood for 2 min to obtain the same shear history. Filtration performance of WDFs were evaluated using the standard API filter press (Model No. 30201, Fann Instrument Co., Houston, TX) under 100 psi at room temperature following the API standard (API recommended Practice 13B-1, 2003).31 The volume of filtrate was recorded at 1.0, 7.5, 15, 20, 25, and 30 min. After the filtration measurements, the deposited filter cakes were carefully removed from the cell and their thickness was measured. The filter cakes were then naturally air-died at room temperature, and their fractured surface morphology was observed using FE-SEM (FEI QuantaTM 3D FEG dual beam SEM/FIB system, Hillsboro, OR). Prior to FE-SEM observation, the samples were coated with a thin layer of gold. The filtrate rate and permeability of fresh filter cakes was measured using the liquid flowing through already formed cake method, as reported previously.5,32 C

DOI: 10.1021/acssuschemeng.7b04156 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

Figure 3. Morphology of ChNCs at pH of (a) 9.0, (b) 7.0, and (c) 3.4; and the corresponding size distribution of ChNCs at a pH of 3.4: (d) width and (e) length. Scale bar: 1 μm.



RESULTS AND DISCUSSION The multistep procedure for isolation of chitin from the speckled swimming crab shells is schematically illustrated in Figure 1. It is worth noting that the grinding of shells into fine powder prior to chemical treatments facilitated the elimination of nonchitin components. Afterward, during the chemical treatments using sodium hydroxide, hydrochloric acid, and ethanol, the color of crab shell powder suspension gradually changed from yellow to tan to off-white to milk-white, indicating the gradual removal of protein, calcium carbonate, and pigments. The yield of chitin was about 12.0 wt % with respect to the original amount of crab shell powders used, close to the values reported for crab (Carcinus mediterraneus) shells (10.0 wt %)33 and red king crab (Paralithodes camtschaticus) shells (12.1 wt %).29 Successful extraction of chitin from speckled swimming crab shells was confirmed by FTIR and XRD analyses (Figure 2). For example, after the multistep extraction, the strong absorption peak at 1403 cm−1 corresponding to protein was absent.29 On the other hand, the amide I bands at 1661 and 1621 cm−1, amide II band at 1551 cm−1 and amide III band at 1310 cm−1, which are characteristic absorption peaks of chitin, were present (Figure 2a).29,34 From the FTIR spectrum, the allomorph of the chitin, e.g., the α- or β-form could be also determined. It was reported that the amide I band in the spectrum of α-chitin splits into two components due to the formation of two different intermolecular hydrogen bonds; whereas for β-chitin, its amide I band appears as a single peak.35,36 The present chitin displays two absorption peaks at 1661 and 1621 cm−1, indicating the α allomorphic crystalline form, i.e., their molecular chains are arranged in antiparallel

association with strong intra- and intermolecular hydrogen bonding among the carbonyl and amide groups. XRD analysis of crab shell powders (CSP) and extracted chitin further confirmed the purification and crystalline structure of α-chitin (Figure 2b).37 In particular, the intense diffraction peak at 29.6° originating from calcium carbonate in the crab shell pattern completely disappeared after the multistep treatments, suggesting the successful removal of minerals from crab shell powders. Furthermore, the purified chitin fiber has several diffraction peaks at 9.3° (020), 12.4° (021), 19.3° (110), 20.7° (120), 23.3° (130), and 26.3° (013) that are characteristic peaks of α-chitin.29 ChNCs were further produced through the cleavage of chitin microfibrils using hydrochloric acid hydrolysis and subsequent high pressure homogenization (Figure 1). The yield of ChNCs was about 56.7 wt %, which was calculated by the weight of ChNCs obtained divided by the initial weight of purified chitin. The ChNCs showed the analogous characteristic peaks as purified chitin in both FTIR spectra and XRD patterns (Figure 2), suggesting that the original functional groups and crystal structure of α-chitin were well maintained after the acid hydrolysis and high pressure homogenization. The differences in intensities of chitin and ChNC despite the retained peak positions are ascribed to differences in the degree and type of preferred orientation in the XRD samples. Rietveld analysis of both the chitin and ChNC indicated crystallite sizes ranging from 7 to 9 nm, depending on the simulated background and possible amorphous content. The ChNCs have rod-like morphology similar to CNCs, as shown in TEM micrographs (Figure 3). Interestingly, it was found that their dispersion state in aqueous suspensions greatly depended on the pH value. At pH values of 9.0 and 7.0, ChNCs D

DOI: 10.1021/acssuschemeng.7b04156 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

Figure 4. pH-Responsive behaviors of ChNC suspension: (a−c) visual appearance of 0.1 wt % ChNC suspension at pH of 9.0, 7.0 and 3.4, respectively; (d) UV−vis spectra of 0.1 wt % ChNC suspension; (e) steady-state viscosity versus shear rate of 1 wt % ChNC suspension; (f) G′ (solid symbols) and G′′ (open symbols) versus strain of 1 wt % ChNC suspension; and (g) G′ (solid symbols) and G′′ (open symbols) versus angular frequency of 1 wt % ChNC suspension.

were treated twice using 5 wt % NaOH solution for 2 days and 6 h (Figure 1). Such NaOH treatment might also decrease the degree of acetylation and thus exposure more amine groups, producing highly positive zeta potential value (+32.57 ± 1.09) even at neutral conditions. Because of the more dispersed state of ChNCs under acidic conditions, their dimensions, including width and length were measured based on 50 randomly chosen ChNCs from Figure 3c using ImageJ 1.47 software, and the corresponding size distribution plots are shown in Figure 3d and e. The width and length of ChNCs were in the range of 3− 25 nm and 100−700 nm, with an average value of 14.8 ± 3.4 and 378.9 ± 98.1 nm, respectively. The width of ChNCs observed from TEM observation did not match the crystallite sizes ranging from 7 to 9 nm as defined by XRD Rietveld analysis. This result suggests that the ChNC at acidic condition was still composed of 2−3 crystallites. The average aspect ratio (i.e., length/width) of ChNC particles was calculated as 25.6, much higher than that of the hydrochloric acid-hydrolyzed chitin nanocrystals from yellow squat lobster (∼5)42 and similar to that of the hydrochloric acid-hydrolyzed chitin nanocrystals from crab shell flakes (∼20)43 and shrimp shell powder (20.2).44 Contrastingly, it was lower than those of the TEMPOoxidized chitin nanocrystals from crab shell powders (∼42.5)45 and shrimp shell powders (∼42.2).44 Indeed, the origin of chitin and preparation methods have critical influence on the resulting dimensions of ChNCs. A comprehensive comparison of the dimensions of ChNCs prepared from different chitin resources using different methods is provided in Table S1.

were significantly aggregated into large crystalline clusters due to the high specific surface area as well as the strong hydrogen bonding created between neighboring ChNCs (Figure 3a and b). In contrast, at a pH of 3.4, ChNCs were more individualized on the nanoscale and homogeneously distributed in aqueous suspension (Figure 3c). Chitin has amino groups (NH2) on the backbone that are protonated to form NH3+ in acid media.38 Reported pK values of chitin range from 6.2 to 7.0, depending on the conditions of measurement and the degree of acetylation.39 Therefore, we speculated that positively charged NH3 + groups in ChNC suspensions at pH 3.4 repel neighboring ChNCs by electrostatic forces. This disrupts the clustering of crystallites, leading to a more uniform distribution of individual particles. Zeta potentials were measured to confirm this assumption. As expected, the ChNC suspension at pH of 9.0 and 7.0 had an average zeta potential value of −10.10 ± 0.37 and 32.57 ± 1.09 mV, respectively; which dramatically increased to 61.00 ± 1.86 mV when the pH was decreased to 3.4. The pKa value of ChNCs is greatly dependent on the degree of acetylation of chitin and ionic strength of suspension as well as the measurement methods. For example, Tzoumaki et al. reported that the ChNCs isolated from crude chitin from shrimp shells by acid hydrolysis (3 N HCl, 95 °C, 90 min) have pKa value of 7;40 whereas Pereira et al. found that the ChNCs isolated from commercial chitin from crab shells by acid hydrolysis (3 N HCl, 100 °C, 90 min) have pKa value of 7.6, which increased from 7.6 to 8.2 and 8.8 after deacetylation using NaOH treatment.41 It is worth noting that in the present work, in order to completely remove protein, the crab shells E

DOI: 10.1021/acssuschemeng.7b04156 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

Figure 5. Rheological properties of WDFs: (a) viscosity and (b) shear stress versus shear rate of BT and ChNC/BT-WDFs at different pH values and (c) viscosity and (d) shear stress versus shear rate of BT, chitin/BT, and ChNC/BT-WDFs at pH 9.0.

ChNC suspensions behave essentially as Newtionian fluids, i.e., the viscosity is independent of the applied shear rate; while the shear-thinning behavior was still observed from alkaline and neutral ChNC suspensions (Figure S1). Dynamic rheological analysis (Figure 4f and g) was also performed, providing supporting information to the steady-state viscosity investigations. Prior to frequency sweep, strain sweep (Figure 4f) was carried out to determine the linear viscoelastic region at a fixed frequency of 1 Hz. Under alkaline and neutral conditions, the ChNC suspension exhibited a nearly linear viscoelastic region from 0.1 to 20 s−1, beyond which both the storage modulus (G′) and loss modulus (G′′) decreased with increase in the strain due to the disruption of ChNC network. The ChNC suspension under acidic condition exhibited distinctive viscoelastic-strain curve with that under alkaline and neutral condition. Particularly, both G′ and G′′ became independent of strain. It is believed that due to the homogeneous dispersion state of ChNCs under acidic media, it is difficult to create an effective network, and thus no critical transition could be observed. Based on the strain sweep results, a fixed strain of 1% in the linear viscoelastic domain was employed to perform frequency sweep (Figure 4g). For the ChNC suspensions at pH of 9.0 and 7.0, the G′ was much higher than G′′, indicating the solid-like viscoelastic behavior. Interestingly, the viscoelastic behavior of ChNC suspension became viscous fluid-like as the pH value decreased to 3.4, i.e., the G′ and G′′ curves were almost overlapped within the investigated strain range. The observed rheological phenomena confirmed the pH-responsive behaviors of ChNC suspensions. The change in the dispersion state and the absence/presence of electrostatic repulsive forces, as a result of the deprotonation/protonation of amino groups

The pH-responsive behaviors of ChNC suspension were further investigated with several techniques, including visual observation, UV−vis spectroscopy and rheometry (Figure 4). Owing to the formation of large crystalline clusters, ChNC suspensions showed turbid appearance under alkaline and neutral conditions (Figure 4a and b). As the pH decreased to 3.4, the protonated ChNCs in the crystalline clusters repelled each other, yielding a more transparent and uniform ChNC suspension (Figure 4c). The UV−vis spectra gave a more quantitative analysis of the transmittance (Figure 4d). Apparently, a remarkable increase in the transmittance was achieved by acidifying the ChNC suspension. For example, at a wavelength of 700 nm, the transmittance values of ChNC suspension at pH of 9.0, 7.0, and 3.4 were 43.56, 47.23, and 63.73%, respectively. Rheological behaviors of ChNC suspension under alkaline, neutral, and acidic conditions are depicted in Figure 4e−g. The steady-state viscosity measurements (Figure 4e) indicate that the ChNC suspension have notable shear-thinning behavior under alkaline and neutral conditions, i.e., the viscosity decreases as the shear rate increases. It is believed that at increasing shear rates, the crystalline clusters in ChNC suspension progressively undergo aggregation, disintegration, and orientation, giving rise to the observed shear-thinning behavior. However, because of the uniform dispersion state of ChNC suspension under acidic condition, such progressive change in the dispersion of ChNCs would be largely restrained. As a result, acidic ChNC suspensions behave less significant shear-thinning behavior with much lower viscosity over the whole studied shear rate range (Figure 4e). Moreover, when the ChNC suspensions were diluted from 1 to 0.1 wt %, acidic F

DOI: 10.1021/acssuschemeng.7b04156 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

Figure 6. Proposed dispersion states and surface interactions in BT and ChNC/BT-WDFs under alkaline, neutral, and acidic conditions.

both BT and ChNC/BT-WDFs slightly increased. However, when the pH value was further decreased to 3.4, pronounced improvement was observed. For example, at a shear rate of 0.1 s−1, the viscosities of BT and ChNC/BT-WDFs at pH of 3.4 are 2.08 and 4.06 Pa·s, respectively. Similar phenomena were observed in the shear stress plots upon addition of ChNCs and change in drilling fluid pH values (Figure 5b). The Herschel−Bulkley model, normally used to simulate the relationship between shear stress (τ) and shear rate (γ̇) for drilling fluids, has the following form:46

on the surface of ChNCs, were responsible for the observed pH-responsive behaviors. Drilling fluids are normally designed as shear-thinning fluids that have high viscosity at low shear rates in order to suspend and transport drill cuttings from the wellbores, but low viscosity at high shear rates to ensure good pumpability and low energy consumption. Carrying capacity is considered as one of the most essential functions for drilling fluids, a function that is directly related to their rheological properties. The observed shear-thinning of ChCN suspension (Figure 4e, S1) has inspired us the utilization of ChCNs as rheology modifiers for the formulation of low-solid, smart/nano BT-WDFs. To exemplify the effectiveness of ChNCs in improving the rheological performance of BT-WDFs, only 0.1 wt % ChNCs was mixed with 3 wt % BT in an aqueous solution. It is also anticipated that the observed pH-responsive behaviors of ChNCs can help produce drilling fluids with smart functionality, i.e., tailor-made rheological and filtration performance through pH adjustment. This would allow drilling engineers to precisely control the rheology and filtration of drilling fluids by tuning pH value in order to meet the requirements of different well conditions, resulting in efficient and safe drilling operations. The influence of ChNCs on the steady-state viscosity of BTWDFs under alkaline, neutral, and acidic conditions is shown in Figure 5a. As expected, because of the low solids content, neat BT-WDFs under alkaline conditions exhibited a typical Newtonian rheological behavior with very low viscosities at any investigated shear rate, predicting extremely poor carrying capacity for drill cuttings. With the presence of 0.1 wt % ChNCs, the rheological behavior of BT-WDFs was changed from Newtonian to shear-thinning, which is advantageous for drilling fluids. Furthermore, remarkable improvement in the viscosity of BT-WDFs in the low shear rate region was observed by adding 0.1 wt % ChNCs. For example, at a low shear rate of 0.1 s−1, BT and ChNC/BT-WDFs at pH of 9.0 had viscosity values of 3.18 × 10−3 and 1.04 × 10−1 Pa·s, respectively. The incorporation of only 0.1 wt % ChNCs resulted in a roughly 33 times improvement in the viscosity value. Such a dramatic increase indicated a considerable development in the carrying capacity for drill cuttings, i.e., the superior wellbore cleaning efficiency of ChNC/BT-WDFs over BT-WDFs. As the pH value deceased to 7, the viscosity of

τ = τ0 + Kγ ṅ

(1)

where τ0 is the yield point, K is the flow consistency coefficient, and n is the flow behavior index. The derived τ0 values for BT and ChNC/BT-WDFs at pH 9.0 are 0.01 and 0.20 Pa, respectively; which were slightly shifted to 0.02 and 0.24 Pa at pH 7 and dramatically jumped to 0.27 and 0.61 Pa at pH 3.4 (Table S3). The yield point (τ0), the minimum stress required to move the WDFs, is also an indicator of their carrying capacity. The higher τ0 values of ChNC/BT-WDFs over BTWDFs at any pH values as well as the gradual increase in τ0 values with decrease in pH values further validated the effectiveness of ChNCs in enhancing carrying capacity of BTWDFs and their pH-responsiveness. Moreover, it should be pointed out that the particle size of chitin is vital to the improvement in the rheology of BT-WDFs. When the microsized chitin was used, no shear-thinning behavior could be observed at any loadings (Figure 5c and d), indicating that microsized chitin could not act as rheology modifier for BTWDFs. The nanoscale dimension as well as high surface area of ChNCs played a critical role in improving the rheology of BTWDFs. The effectiveness of ChNCs in improving rheology of BTWDFs is ascribed to the formation of nanostructured surface interactions between ChNC and BT. However, their dispersion state and surface interaction patterns are greatly dependent on the pH value arising from the reversible protonation and deprotonation of BT and ChNC, as schematically illustrated in Figure 6. It is well-known that BT particles consist of superimposed planar montmorillonite (MMT) platelets with two types of surface charges.47 A permanent negative charge is formed on the face of MMT platelets due to isomorphic substitution of Si4+ by Al3+ within the Si−O4 tetrahedral layer, G

DOI: 10.1021/acssuschemeng.7b04156 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

As the pH value further decreased from 7.0 to 3.4, on one hand, more amino groups on the surface of ChNCs were protonated to become positively charged NH3+ as indicated by increased zeta potential from 32.57 ± 1.09 to 61.00 ± 1.86 mV; on the other hand, the edge surfaces of MMT platelets were also protonated to be positively charged Al−OH2+. We believe that besides the formation of house of cards structures by MMT platelets, the deprotonated ChNCs would link MMT platelets as well, producing more rigid networks and larger flocculates (Figure 6f), yielding the highest viscosity and shear stress values (Figure 5a and b). The proposed surface interactions in ChNC/BT-WDFs under different pH conditions were further confirmed by dynamic rheological measurements, as shown in Figure 7. At a

whereas their edge charge is pH-dependent because of the reversible protonation and deprotonation of hydroxyl groups (−OH) at the edge surface.48 Therefore, three possible association modes, i.e., face-to-face, face-to-edge, and edge-toedge can be produced between adjacent MMT platelets via van der Waals and electrostatic repulsion/attraction forces, depending on the pH value, and ionic strength.49,50 Under alkaline conditions, the MMT platelets are negative on both the face and edge. The electrostatic repulsion forces occurred between face and face, edge and face, and edge and edge and, thus, predominated throughout the suspension, leading to a uniform dispersion of MMT platelets in aqueous solution (Figure 6a), which was responsible for the observed extremely low viscosity and shear stress (Figure 5a and b). Under neutral conditions, the MMT platelets are negative on the face and neutral at the edge. The negatively charged faces repelled each other significantly until achieving the minimum repulsion forces, i.e., when the MMT platelets were attached closely through edge-to-edge association model (Figure 6b). At this time, the van der Waals attraction forces became pronounced between edge and edge, since the van der Waals force is a distancedependent force.51 The formation of edge-to-edge attachment via van der Waals forces between MMT platelets resulted in slight increase in viscosity and shear stress (Figure 5a and b). Under acidic conditions, the amphoteric Al−OH groups at the edge surface were fully protonated to form positively charged Al−OH2+. The face-to-edge associations were created via electrostatic attraction between negatively charged face and positively charged edge, forming a “house of cards” structure (Figure 6c).50 The house of cards structure has strong resistance against flow, reflected in high viscosity and shear stress (Figure 5a and b). The ChNC suspension had average zeta potential value of −10.10 ± 0.37, 32.57 ± 1.09, and 61.00 ± 1.86 mV at pH of 9.0, 7.0, and 3.4, respectively. The incorporation of negatively charged ChNCs in BT suspension at pH of 9.0 might disturb the uniform dispersion of MMT through the occurrence of entropic depletion forces. The entropic depletion forces usually arise between large colloidal particles that are suspended in a dilute solution of small, nonadsorbing particles, normally known as depletants.52,53 The origin of depletion forces was first interpreted by Asakura and Oosawa,54 who suggested that the depletants are preferentially excluded from the vicinity of the large particles, leading to an increase in osmotic pressure of the surrounding solution on the large colloidal particles. Therefore, it is speculated that once the negatively charged ChNCs were added into the negatively charged BT suspension at pH of 9.0, ChNCs acted depletants, which pushed the MMT platelets to link together via the entropic depletion forces (Figure 6d). This also led to improvement of the rheological properties of BT suspension (Figure 5a and b). Similar phenomena were observed from the CNC/bacteria system by Sun et al.,55 who reported that the presence of negatively charged CNCs in the negatively charged bacteria suspension caused flocculation of bacteria due to the entropic depletion effect. However, under neutral conditions, the positively charged ChNCs were attached to the negatively charged face of MMT platelets via electrostatic attraction (Figure 6e). Consequently, ChNCs acted in “bridge” roles, connecting two or more MMT platelets and producing rigid networks. The complexation of ChNCs with MMT platelets is thus responsible for the enhanced rheological properties of BTWDFs with the presence of ChCNs at pH of 7.0 (Figure 5a,b).

Figure 7. Dynamic time sweep curves for ChNC/BT-WDFs at different pH conditions. Note: In order to determine the influence of shearing on the microstructures of ChNC/BT-WDFs, vigorous shearing was applied a speed of 1000 s−1 ranging from 5 to 6 min for 1 min.

pH of 9.0, ChNC/BT-WDFs showed fluidlike viscous rheological behaviors, i.e., G′′ ≫ G′ and tan δ > 1.56,57 At pH of 7.0, ChNC/BT-WDFs still exhibited fluidlike viscoelastic behavior; whereas the difference between G′ and G′′ was negligible. On the contrary, the solidlike elastic rheological behaviors were observed at pH of 3.4, i.e., G′ ≫ G′′ and tan δ < 1. Meanwhile, it was observed that the initial G′ values gradually increased from 0.014 to 0.083 and 7.50 MPa as the pH decreased from 9.0 to 7.0 and 3.4, respectively. All these observations demonstrated that more rigid networks were created in ChNC/BT-WDFs with change in the suspension conditions from alkaline to acidic. During drilling operations, drilling fluids are pumped into a well at high shear speeds. It is of great practical significance to understand whether vigorous shearing disrupts the microstructures of the fluids, leading to the deterioration on their carrying capacity of drill cuttings. A well-designed fluids should be able to rebuild the microstructures disrupted by shearing. To demonstrate the process for ChNC/BT-WDFs, the fluids after 1 min of vigorous shearing at a speed of 1000 s−1 were progressively monitored as a function of time for 1 h (Figure 7). In the case of ChNC/BT-WDFs at pH of 9.0, it immediately returned to the initial G′ value, indicating that the shearing had no obvious influence on the BT networks. H

DOI: 10.1021/acssuschemeng.7b04156 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

Figure 8. Filtration performance of WDFs: (a) fluid loss volume versus time of BT and ChNC/BT-WDFs at different pH, (b) filtrate rate determination of the already formed filter cakes deposited from BT and ChNC/BT-WDFs at different pH, (c) fluid loss volume versus time of BT, chitin/BT, ChNC/BT, PAC/BT, and PAC/ChNC/BT-WDFs at pH of 9.0, and (d) filtrate rate determination of the already formed filter cakes deposited from BT, chitin/BT, ChNC/BT, PAC/BT and PAC/ChNC/BT-WDFs at pH of 9.0.

API fluid loss value of 28.7 and 30.5 mL, respectively; which were increased to 31.0 and 34.5 mL with the addition of 0.1 wt % ChNCs. When the pH was decreased to 3.4, the API fluid loss values of BT and ChNC/BT-WDFs were further increased to 32.9 and 40.0 mL, respectively, showing pH-dependent filtration performance. It is believed that upon the addition of ChNCs or the change in the drilling fluid pH from alkaline to neutral to acidic, the flocculation of MMT platelets gradually occurred as discussed previously (Figure 6). Consequently, the MMT platelets would not be able to distribute in a layer-bylayer manner in the suspension, leading to the deterioration on the barrier function of MMT platelets against water penetration. However, the influence of microsized chitin on the filtration performance of BT-WDFs was very limited (Figure 8c and d), further highlighting the vital role of nanostructured surface interactions in the flow and filtration performance of ChNC/BT-WDFs. During the drilling operations, the solids in drilling fluids spontaneously deposited, forming a layer of solids described as “filter cake”. The characteristics of deposited filter cake, i.e., permeability and microstructure, also had a critical influence on the fluid loss value. Generally, superior filtration performance comes along with the formation of compact filter cake with low permeability. To calculate the permeability (Kc) of filter cake, we employed Darcy’s Law, which has the following form:59

Indeed, under the alkaline condition, there were few networks created, since little surface interaction occurred between ChNCs and MMT platelets, and the primary attraction in the system was the entropic depletion force between MMT platelets, which is very weak attraction force. In case of ChNC/BT-WDFs at pH of 7.0 and 3.4, significant deviations were observed after the vigorous shearing at 1000 s−1 for 1 min with their initial G′ value decreased from 0.083/7.50 to 0.021/ 3.01 MPa, yielding 74.7% and 59.9% reduction, respectively. These phenomena suggested that the networks created between ChNCs and MMT platelets via electrostatic attraction at pH of 7.0 and 3.4 were broken by shearing. The destroyed networks were readily reformed with time. As shown in Figure 7, it took approximately 47 and 10 min for ChNC/BT-WDFs at pH of 7.0 and 3.4 to return their original states, respectively. In contrast to ChNC/BT-WDFs under neutral condition, ChNC/BT-WDFs under acidic condition showed more stable microstructures as well as faster relaxation speed, as judged from the lower reduction percentage as well as shorter relaxation time. This is probably attributed to the enhanced electrostatic interaction between the highly protonated ChNCs and the negatively charged face of MMT platelets, when the pH was decreased to 3.4 (Figure 6f).58 The formation of different dispersion states and nanostructured surface interactions arising from the reversible protonation and deprotonation of MMT platelets and ChNCs also led to pH-responsive filtration performance. As shown in Figure 8a, the BT-WDFs at pH of 9.0 and 7.0 had an

Kc = I

μt cdV ΔPA dt

=

μt cq ΔPA

(2) DOI: 10.1021/acssuschemeng.7b04156 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

completely consistent with the API fluid loss results, confirming that the filtration performance of drilling fluids was substantially governed by the permeability of the deposited filter cakes. It was proposed that the flocculation of MMT platelets in drilling fluids and the high permeability of filter cake accounted for the deterioration of the filtration capability of BT-WDFs with the addition of ChNCs and change in pH values. These hypotheses were confirmed by the morphological observations on the fracture surface of deposited filter cake using FE-SEM (Figure 9). For example, a well-aligned multilayered microstructure was observed on the fracture surface of filter cake deposited from BT-WDFs at neutral pH (Figure 9a and e). As discussed previously, the MMT platelets at pH 7 were mainly associated in an “edge-to-edge” mode due to van der Waals forces (Figure 6b). The edge-to-edge association mode allowed the MMT platelets to deposit layer-by-layer and stack during the filtration measurement, producing the observed wellaligned multilayered microstructure. It is believed that this compact, layer-by-layer stacking pattern mostly blocked the fluid penetration pathway, giving rise to relatively superior filtration performance. When the pH of BT-WDFs was lowered to 3.4, the well-aligned multilayered microstructure disappeared; instead, the characteristic house of cards structure was observed (Figure 9b and f). This observation validated our proposed dispersion state and surface interaction of MMT platelets in BT-WDFs at acidic pH, as schematically illustrated in Figure 6c. The change in the microstructure of filter cake from layer-by-layer stacking to house of cards opened more channels for fluid penetration, and therefore the BT-WDFs at acidic pH exhibited higher permeability and fluid loss compared to that under neutral condition. In case of filter cake deposited from neutral ChNC/BT-WDFs, a porous microstructure was observed on the fracture surface (Figure 9c and g). The attachment of positively charged ChNCs on the negatively charged face surface of MMT platelets resulted in the flocculation of BT (Figure 6e). During the filtration measurement, the BT flocculates are thought to randomly deposit and stack to form a porous microstructure. That porous microstructure had less capacity to block fluid leakage, resulting in higher permeability, thicker filter cake and higher fluid loss

where μ is the viscosity of filtrate at 25 °C (1 cP); tc is the thickness of filter cake (cm); ΔP is the pressure difference (6.8 atm); A is the cross-section area (45.8 cm2); and q = dV/dt is the filtrate rate (cm3/s). The filtrate rate (q) was determined using the liquid flowing through already formed cake, as reported previously.5,32 The corresponding experimental results are plotted in Figure 8b and d. It is observed that the filtration volume increases linearly with time. Through simple linear regression (filtration volume versus time), the slope value of each line was calculated, corresponding to the filtrate rate (q). The filtrate rate (q) and thickness (tc) of filter cake deposited from BT and ChNC/BT-WDFs under alkaline, neutral, and acidic conditions are summarized in Table 1. Under the same Table 1. Filtrate Rate (q), Thickness (tc), and Permeability (Kc) of Filter Cake from BT, chitin/BT, ChNC/BT, PAC/ BT, and PAC/ChNC/BT-WDFs samples BT3 pH 9.0 BT3 pH 7.0 BT3 pH 3.4 ChNC0.1/BT3 pH 9.0 ChNC0.1/BT3 pH 7.0 ChNC0.1/BT3 pH 3.4 Chitin0.1/BT3 pH 9.0 Chitin0.5/BT3 pH 9.0 PAC0.1/ChNC0.1/BT3 pH 9.0 PAC0.2/BT3 pH 9.0

q × 10−3 (cm3/s)

tc (cm)

Kc × 10−3 (mD)

7.89 8.29 9.82 7.93 8.72 10.42 7.87 7.89 5.82

0.11 0.12 0.15 0.13 0.18 0.26 0.11 0.12 0.10

2.79 3.19 4.73 3.31 5.04 8.70 2.78 3.04 1.87

5.21

0.07

1.17

conditions, the ChNC/BT-WDFs had higher q and tc values than BT-WDFs. Furthermore, the change in pH value from 9.0 to 3.4 also resulted in faster filtration rate and thicker filter cake. By substituting these values into eq 2, the Kc values were determined (Table 1). Kc increased in the following order: BTWDFs at pH 9.0 < BT-WDFs at pH 7.0 < ChNC0.1/BT3 at pH 9.0 < BT-WDFs at pH 3.4 < ChNC/BT-WDFs at pH 7.0 < ChNC/BT-WDFs at pH 3.4. These observations were

Figure 9. Fractured surface morphology of filter cakes deposited from different WDFs: (a, e) BT3 at pH 7.0, (b, f) BT3 at pH 3.4, (c, g) ChNC0.1/ BT3 at pH 7.0, (d, h) ChNC0.1/BT3 at pH 3.4. Scale bar: (a−d) 20 μm and (e−h) 5 μm. J

DOI: 10.1021/acssuschemeng.7b04156 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

were 0.20 and 0.78 Pa for ChNC/BT and PAC/ChNC/BTWDFs, respectively; showing notable improvement in the carrying capacity of drill cuttings (Table S2). The improved rheology of ChNC/BT-WDFs by PAC was also ascribed to the entropic depletion effect, as discussed previously. With the addition of PAC, MMT platelets could be further pushed to attach layer-by-layer, giving rise to notable increase in rheology. Moreover, it was observed that PAC0.1/ChNC0.1/BT3-WDFs exhibited superior rheological performance over PAC0.2/BT3WDFs at the same loading of additive. Overall, the combined use of ChNCs and PAC resulted in remarkable improvement of both rheological and filtration performance of BT-WDFs.

compared to BT-WDFs. When the pH of ChNC/BT-WDFs was reduced from 7.0 to 3.4, an even more porous microstructure was produced (Figure 9d and h). In that situation, the larger BT flocculates were more randomly stacked (Figure 6f), giving the worst filtration performance. Overall, it is demonstrated that ChNCs primarily acted as rheological modifiers in BT-WDFs rather than filtration reducer. The best rheological and filtration performance of ChNC/BT-WDFs were achieved under acidic and alkaline conditions, respectively. ChNC/BT-WDFs exhibited promising pH-dependent rheological and filtration performance, especially when the pH was changed from neutral to acidic. In other words, the rheological and filtration performance of ChNC/ BT-WDFs can be well tailored by controlling the pH value, demonstrating their small functionality. Some previous work has reported the formulation of smart/nano drilling fluids with tailor-made performance, i.e., their rheological and filtration performance could be well tuned by shear rate, nanoparticle loading or magnetic strength.1,60,61 In the present study, a novel smart/nano drilling fluid with pH-tunable rheological and filtration performance was formulated. It is anticipated that the unique capacity to modulate the rheology and filtration of ChNC/BT-WDFs through pH adjustment could enable drilling engineers to obtain the desirable performance under diverse operating environments, and thus facilitate the drilling operations. The leakage of water into rock formations and the deposition of thick, highly permeable filter cake from drilling fluids can cause serious problems, such as formation collapse, wellbore damage, and stuck drill pipe, leading to poor production efficiency and expensive stimulation treatments.1−5 For practical application of the ChNC/BT-WDFs, superior filtration performance, i.e., the penetration of low fluid and the deposition of thin, impermeable filter cake are strongly desirable. Polyanionic cellulose (PAC), a water-soluble anionic cellulose ether, is a commercial filtration control agent used in BT-WDFs. Therefore, we further investigated whether the combined use of PAC in ChNC/BT-WDFs could improve its filtration performance at alkaline condition. Satisfyingly, the presence of only 0.1 wt % PAC remarkably reduced the API fluid loss value and permeability of ChNC/BT-WDFs from 31.0 to 18.9 mL and from 3.31 × 10−3 to 1.87 × 10−3 mD, respectively (Figure 8c and Table 1). The developed filtration performance of ChNC/BT-WDFs by adding PAC was also inferred from FE-SEM observation on the fracture surfaces of deposited filter cake (Figure S2a). It is obvious that with the addition of PAC, the highly porous microstructure disappeared and a well-aligned layer-by-layer stacking microstructure was created. This compact and multilayered microstructure can resist the fluid penetration effectively, resulting in better filtration performance. The effectiveness of PAC in improving filtration performance of ChNC/BT-WDFs might be ascribed to the highly negatively charged surface characteristics of PAC, which could strongly repel the negatively charged face surface of BT platelets, leading to the layer-by-layer deposition of BT platelets (Figure S2). The presence of PAC further enhanced the rheology of ChNC/BT-WDFs at pH of 9.0, as shown in Figure S3. For example, at a low shear rate of 0.1 s−1, ChNC/BT and PAC/ ChNC/BT-WDFs had viscosity values of 0.1 and 2.3 Pa·s, respectively. The addition of only 0.1 wt % PAC further yielded a roughly 23 times improvement in the viscosity value. The derived yield stress values using the Herschel−Bulkley model



CONCLUSIONS We have extracted valuable ChNCs from speckled swimming crab shells and utilized them as rheological modifiers in lowsolid, smart/nano BT-WDFs. Similar to ChNCs from other sources produced by hydrolysis with hydrochloric acid, our ChNCs are 14.8 ± 3.4 nm in width and 378.9 ± 98.1 nm in length, and have an interesting pH-responsive behavior due to protonation/deprotonation of the amino group on the surface of ChNCs. These ChNCs are effective rheology modifiers for BT-WDFs even at an extremely low loading of 0.1 wt % due to the entropic depletion effect at alkaline condition and the complexation of positively charged ChNCs with negatively charged face surface of MMT platelets at neutral and acidic conditions. Both BT and ChNC/BT-WDFs displayed pHresponsive rheological and filtration behaviors because of the occurrence of different dispersion states and surface interactions arising from the reversible protonation and deprotonation of MMT platelets and ChNCs, especially when the pH was altered from neutral to acidic. Under acidic conditions, in addition to the formation of house of cards structure by MMT platelets, the highly positively charged ChNCs further linked MMT platelets, which resulted in more rigid networks and larger flocculates. However, the presence of ChNCs slightly deteriorated the filtration performance of BT-WDFs. The poor layer-by-layer stacking capacity and the high permeability of deposited filter cake were responsible for the deteriorated filtration performance. Fortunately, such poor filtration performance of ChNC/BT-WDFs can be overcome by adding a small amount of filtration control agentPAC. The highly negatively charged PAC repelled MMT platelets to attach layerby-layer, which not only led to thin, compact, and low permeable filter cakes but also further improved the rheological performance of ChNC/BT-WDFs. This research demonstrates the utilization of ChNCs isolated from crab shell wastes as effective rheological modifiers in BT-WDFs, providing us an alternative strategy to turn agriculture and food wastes into valuable industrial products. We also expect that the unique pH-responsive rheological and filtration behaviors driven by nanostructured surface interactions will give us a better understanding how the pH value as well as the interfacial interactions between additives and BT would affect the overall performance of their drilling fluids.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acssuschemeng.7b04156. K

DOI: 10.1021/acssuschemeng.7b04156 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering



(11) Vryzas, Z.; Kelessidis, V. C.; Bowman, M. B. J.; Nalbantian, L.; Zaspalis, V.; Mahmoud, O.; Nasr-El-Din, H. A. Smart Magnetic Drilling Fluid With In-Situ Rheological Controllability Using Fe3O4 Nanoparticles. SPE Middle East Oil & Gas Show and Conference, Kingdom of Bahrain, Manama, March 1−3, 2017; SPE Paper No. 183906. (12) Yan, L.; Wang, C.; Xu, B.; Sun, J.; Yue, W.; Yang, Z. Preparation of a novel amphiphilic comb-like terpolymer as viscosifying additive in low-solid drilling fluid. Mater. Lett. 2013, 105, 232−235. (13) Liu, F.; Jiang, G.; Peng, S.; He, Y.; Wang, J. Amphoteric Polymer as an Anti-calcium Contamination Fluid-Loss Additive in Water-Based Drilling Fluids. Energy Fuels 2016, 30, 7221−7228. (14) Xie, B.; Liu, X. Thermo-thickening behavior of LCST-based copolymer viscosifier for water-based drilling fluids. J. Pet. Sci. Eng. 2017, 154, 244−251. (15) Mao, H.; Qiu, Z.; Shen, Z.; Huang, W. Hydrophobic associated polymer based silica nanoparticles composite with core−shell structure as a filtrate reducer for drilling fluid at utra-high temperature. J. Pet. Sci. Eng. 2015, 129, 1−14. (16) Ponmani, S.; Nagarajan, R.; Sangwai, J. S. Effect of Nanofluids of CuO and ZnO in Polyethylene Glycol and Polyvinylpyrrolidone on the Thermal, Electrical, and Filtration-Loss Properties of Water-Based Drilling Fluids. SPE J. 2016, 21, 405−415. (17) Wu, Y.; Wang, Z.; Yan, Z.; Zhang, T.; Bai, Y.; Wang, P.; Luo, P.; Gou, S.; Guo, Q. Poly(2-acrylamide-2-methylpropanesulfonic acid)Modified SiO2 Nanoparticles for Water-Based Muds. Ind. Eng. Chem. Res. 2017, 56, 168−174. (18) Li, M.-C.; Wu, Q.; Song, K.; Qing, Y.; Wu, Y. Cellulose Nanoparticles as Modifiers for Rheology and Fluid Loss in Bentonite Water-based Fluids. ACS Appl. Mater. Interfaces 2015, 7, 5006−5016. (19) Yan, N.; Chen, X. Don’t waste seafood waste: Turning cast-off shells into nitrogen-rich chemicals would benefit economies and the environment. Nature 2015, 524, 155−158. (20) Ifuku, S.; Morooka, S.; Norio Nakagaito, A.; Morimoto, M.; Saimoto, H. Preparation and characterization of optically transparent chitin nanofiber/(meth)acrylic resin composites. Green Chem. 2011, 13, 1708−1711. (21) Gao, X.; Chen, X.; Zhang, J.; Guo, W.; Jin, F.; Yan, N. Transformation of Chitin and Waste Shrimp Shells into Acetic Acid and Pyrrole. ACS Sustainable Chem. Eng. 2016, 4, 3912−3920. (22) Liu, Y.; Liu, M.; Yang, S.; Luo, B.; Zhou, C. Liquid Crystalline Behaviors of Chitin Nanocrystals and their Reinforcing Effect on Natural Rubber. ACS Sustainable Chem. Eng. 2018, 6, 325. (23) No, H. K.; Meyers, S. P.; Lee, K. S. Isolation and characterization of chitin from crawfish shell waste. J. Agric. Food Chem. 1989, 37, 575−579. (24) Goodrich, J. D.; Winter, W. T. α-Chitin Nanocrystals Prepared from Shrimp Shells and Their Specific Surface Area Measurement. Biomacromolecules 2007, 8, 252−257. (25) Wang, Y.; Zhu, L.; You, J.; Chen, F.; Zong, L.; Yan, X.; Li, C. Catecholic Coating and Silver Hybridization of Chitin Nanocrystals for Ultrafiltration Membrane with Continuous Flow Catalysis and Gold Recovery. ACS Sustainable Chem. Eng. 2017, 5, 10673−10681. (26) Fan, Y.; Saito, T.; Isogai, A. Individual chitin nano-whiskers prepared from partially deacetylated α-chitin by fibril surface cationization. Carbohydr. Polym. 2010, 79, 1046−1051. (27) Montanari, S.; Roumani, M.; Heux, L.; Vignon, M. R. Topochemistry of Carboxylated Cellulose Nanocrystals Resulting from TEMPO-Mediated Oxidation. Macromolecules 2005, 38, 1665− 1671. (28) Dong, X. M.; Kimura, T.; Revol, J.-F.; Gray, D. G. Effects of Ionic Strength on the Isotropic−Chiral Nematic Phase Transition of Suspensions of Cellulose Crystallites. Langmuir 1996, 12, 2076−2082. (29) Ifuku, S.; Nogi, M.; Abe, K.; Yoshioka, M.; Morimoto, M.; Saimoto, H.; Yano, H. Preparation of Chitin Nanofibers with a Uniform Width as α-Chitin from Crab Shells. Biomacromolecules 2009, 10, 1584−1588.

Steady-state viscosity versus shear rate of 0.1 wt % ChNC suspension at pH of 9.0, 7.0, and 3.4 (Figure S1); Fractured surface morphology of filter cakes deposited from PAC0.1/ChNC0.1/BT3 and PAC0.2/BT3-WDFs (Figure S2); Viscosity and shear stress versus shear rate of ChNC0.1/BT3, PAC0.1/ChNC0.1/BT3 and PAC0.2/BT3-WDFs at pH of 9.0 (Figure S3); Comprehensive comparison of the dimensions of ChNCs from different resources using different extraction methods (Table S1); and Derived rheological parameters for BT, chitin/BT, ChNC/BT, PAC/BT, and PAC/ChNC/BT-WDFs using the Herschel-Bulkley model (Table S2) (PDF)

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (Q.W.). ORCID

Qinglin Wu: 0000-0001-5256-4199 Kunlin Song: 0000-0003-2141-1020 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This collaborative study was carried out with support from the US Endowment and USDA Forest Service [P3Nano E17-23] and the Louisiana Board of Regents [LEQSF (2016-2017)ENH-TR-01, LEQSF(2015-17)-RD-B-01].



REFERENCES

(1) Vryzas, Z.; Kelessidis, V. C. Nano-Based Drilling Fluids: A Review. Energies 2017, 10, 540. (2) Kelessidis, V. C.; Poulakakis, E.; Chatzistamou, V. Use of Carbopol 980 and carboxymethyl cellulose polymers as rheology modifiers of sodium-bentonite water dispersions. Appl. Clay Sci. 2011, 54, 63−69. (3) Kosynkin, D. V.; Ceriotti, G.; Wilson, K. C.; Lomeda, J. R.; Scorsone, J. T.; Patel, A. D.; Friedheim, J. E.; Tour, J. M. Graphene Oxide as a High-Performance Fluid-Loss-Control Additive in WaterBased Drilling Fluids. ACS Appl. Mater. Interfaces 2012, 4, 222−227. (4) Meng, X.; Zhang, Y.; Zhou, F.; Chu, P. K. Effects of carbon ash on rheological properties of water-based drilling fluids. J. Pet. Sci. Eng. 2012, 100, 1−8. (5) Li, M.-C.; Wu, Q.; Song, K.; Lee, S.; Jin, C.; Ren, S.; Lei, T. Soy Protein Isolate As Fluid Loss Additive in Bentonite−Water-Based Drilling Fluids. ACS Appl. Mater. Interfaces 2015, 7, 24799−24809. (6) Sönmez, A.; Verşan Kök, M.; Ö zel, R. Performance analysis of drilling fluid liquid lubricants. J. Pet. Sci. Eng. 2013, 108, 64−73. (7) Jung, Y.; Son, Y.-H.; Lee, J.-K.; Phuoc, T. X.; Soong, Y.; Chyu, M. K. Rheological Behavior of Clay−Nanoparticle Hybrid-Added Bentonite Suspensions: Specific Role of Hybrid Additives on the Gelation of Clay-Based Fluids. ACS Appl. Mater. Interfaces 2011, 3, 3515−3522. (8) William, J. K. M.; Ponmani, S.; Samuel, R.; Nagarajan, R.; Sangwai, J. S. Effect of CuO and ZnO nanofluids in xanthan gum on thermal, electrical and high pressure rheology of water-based drilling fluids. J. Pet. Sci. Eng. 2014, 117, 15−27. (9) Ismail, A. R.; Aftab, A.; Ibupoto, Z. H.; Zolkifile, N. The novel approach for the enhancement of rheological properties of water-based drilling fluids by using multi-walled carbon nanotube, nanosilica and glass beads. J. Pet. Sci. Eng. 2016, 139, 264−275. (10) Barry, M. M.; Jung, Y.; Lee, J.-K.; Phuoc, T. X.; Chyu, M. K. Fluid filtration and rheological properties of nanoparticle additive and intercalated clay hybrid bentonite drilling fluids. J. Pet. Sci. Eng. 2015, 127, 338−346. L

DOI: 10.1021/acssuschemeng.7b04156 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering (30) Gopalan Nair, K.; Dufresne, A. Crab Shell Chitin Whisker Reinforced Natural Rubber Nanocomposites. 1. Processing and Swelling Behavior. Biomacromolecules 2003, 4, 657−665. (31) American Petroleum Institute Recommended Practice for Field Testing of Water-Based Drilling Fluids, 3rd ed.; ANSI/API 13B-1; American Petroleum Institute: Washington, DC, 2003. (32) Li, W.; Kiser, C.; Richard, Q. Development of A Filter Cake Permeability Test Methodology; American Filtration & Separations Society Conferences, Ann Arbor, MI, September 2005. (33) Hajji, S.; Younes, I.; Ghorbel-Bellaaj, O.; Hajji, R.; Rinaudo, M.; Nasri, M.; Jellouli, K. Structural differences between chitin and chitosan extracted from three different marine sources. Int. J. Biol. Macromol. 2014, 65, 298−306. (34) Kasaai, M. R. A review of several reported procedures to determine the degree of N-acetylation for chitin and chitosan using infrared spectroscopy. Carbohydr. Polym. 2008, 71, 497−508. (35) Kumirska, J.; Czerwicka, M.; Kaczyński, Z.; Bychowska, A.; Brzozowski, K.; Thö ming, J.; Stepnowski, P. Application of Spectroscopic Methods for Structural Analysis of Chitin and Chitosan. Mar. Drugs 2010, 8, 1567−1636. (36) Rinaudo, M. Chitin and chitosan: Properties and applications. Prog. Polym. Sci. 2006, 31, 603−632. (37) Sikorski, P.; Hori, R.; Wada, M. Revisit of α-Chitin Crystal Structure Using High Resolution X-ray Diffraction Data. Biomacromolecules 2009, 10, 1100−1105. (38) Tzoumaki, M. V.; Karefyllakis, D.; Moschakis, T.; Biliaderis, C. G.; Scholten, E. Aqueous foams stabilized by chitin nanocrystals. Soft Matter 2015, 11, 6245−6253. (39) Guo, B.; Finne-Wistrand, A.; Albertsson, A.-C. Facile Synthesis of Degradable and Electrically Conductive Polysaccharide Hydrogels. Biomacromolecules 2011, 12, 2601−2609. (40) Tzoumaki, M. V.; Karefyllakis, D.; Moschakis, T.; Biliaderis, C. G.; Scholten, E. Aqueous foams stabilized by chitin nanocrystals. Soft Matter 2015, 11, 6245−6253. (41) Pereira, A. G. B.; Muniz, E. C.; Hsieh, Y.-L. Chitosan-sheath and chitin-core nanowhiskers. Carbohydr. Polym. 2014, 107, 158−166. (42) Salaberria, A. M.; Diaz, R. H.; Labidi, J.; Fernandes, S. C. M. Role of chitin nanocrystals and nanofibers on physical, mechanical and functional properties in thermoplastic starch films. Food Hydrocolloids 2015, 46, 93−102. (43) Naseri, N.; Algan, C.; Jacobs, V.; John, M.; Oksman, K.; Mathew, A. P. Electrospun chitosan-based nanocomposite mats reinforced with chitin nanocrystals for wound dressing. Carbohydr. Polym. 2014, 109, 7−15. (44) Butchosa, N.; Brown, C.; Larsson, P. T.; Berglund, L. A.; Bulone, V.; Zhou, Q. Nanocomposites of bacterial cellulose nanofibers and chitin nanocrystals: fabrication, characterization and bactericidal activity. Green Chem. 2013, 15, 3404−3413. (45) Fan, Y.; Saito, T.; Isogai, A. Chitin nanocrystals prepared by TEMPO-mediated oxidation of α-chitin. Biomacromolecules 2008, 9, 192−198. (46) Herschel, W. H.; Bulkley, R. Konsistenzmessungen von gummibenzollösungen. Colloid Polym. Sci. 1926, 39, 291−300. (47) Galindo-Rosales, F. J.; Rubio-Hernández, F. J. Structural breakdown and build-up in bentonite dispersions. Appl. Clay Sci. 2006, 33, 109−115. (48) Furukawa, Y.; Watkins, J. L.; Kim, J.; Curry, K. J.; Bennett, R. H. Aggregation of montmorillonite and organic matter in aqueous media containing artificial seawater. Geochem. Trans. 2009, 10, 2. (49) Luckham, P. F.; Rossi, S. The colloidal and rheological properties of bentonite suspensions. Adv. Colloid Interface Sci. 1999, 82, 43−92. (50) Benna, M.; Kbir-Ariguib, N.; Magnin, A.; Bergaya, F. Effect of pH on Rheological Properties of Purified Sodium Bentonite Suspensions. J. Colloid Interface Sci. 1999, 218, 442−455. (51) Jönsson, B.; Labbez, C.; Cabane, B. Interaction of Nanometric Clay Platelets. Langmuir 2008, 24, 11406−11413.

(52) Weroński, P.; Walz, J. Y.; Elimelech, M. Effect of depletion interactions on transport of colloidal particles in porous media. J. Colloid Interface Sci. 2003, 262, 372−383. (53) Mao, Y.; Cates, M. E.; Lekkerkerker, H. N. W. Depletion force in colloidal systems. Phys. A 1995, 222, 10−24. (54) Asakura, S.; Oosawa, F. On Interaction between Two Bodies Immersed in a Solution of Macromolecules. J. Chem. Phys. 1954, 22, 1255−1256. (55) Sun, X.; Danumah, C.; Liu, Y.; Boluk, Y. Flocculation of bacteria by depletion interactions due to rod-shaped cellulose nanocrystals. Chem. Eng. J. 2012, 198−199, 476−481. (56) Kavanagh, G. M.; Ross-Murphy, S. B. Rheological characterisation of polymer gels. Prog. Polym. Sci. 1998, 23, 533−562. (57) Li, M.-C.; Wu, Q.; Song, K.; Lee, S.; Qing, Y.; Wu, Y. Cellulose Nanoparticles: Structure−Morphology−Rheology Relationships. ACS Sustainable Chem. Eng. 2015, 3, 821−832. (58) Kim, Y.-R.; Cornillon, P.; Campanella, O. H.; Stroshine, R. L.; Lee, S.; Shim, J.-Y. Small and Large Deformation Rheology for Hard Wheat Flour Dough as Influenced by Mixing and Resting. J. Food Sci. 2008, 73, E1−E8. (59) Darcy, H. Les Fontaines Publiques De La Ville De Dijon; Dalmont: Paris, 1856. (60) Amanullah, M.; Al-Tahini, A. M. Nano-Technology−Its Significance in Smart Fluid Development for Oil and Gas Field Application. SPE Saudi Arabia Section Technical Symposium, Al-Khobar, Saudi Arabia, May 9−11, Society of Petroleum Engineers, 2009; SPE Paper No. 126102. (61) Vryzas, Z.; Kelessidis, V. C.; Bowman, M. B. J.; Nalbantian, L.; Zaspalis, V.; Mahmoud, O.; Nasr-El-Din, H. A. Smart Magnetic Drilling Fluid With In-Situ Rheological Controllability Using Fe3O4 Nanoparticles. SPE Middle East Oil & Gas Show and Conference, Manama, Kingdom of Bahrain, March 6−9, Society of Petroleum Engineers, 2017; SPE Paper No. 183906.

M

DOI: 10.1021/acssuschemeng.7b04156 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX