- Email: [email protected]
Synthesis of carboxymethylcelluloses with different degrees of substitution and their performance as renewable stabilizing agents for aqueous ceramic suspensions
Industrial Crops & Products 107 (2017) 54–62
Contents lists available at ScienceDirect
Industrial Crops & Products journal homepage: www.elsevier.com/locate/indcrop
Synthesis of carboxymethylcelluloses with different degrees of substitution and their performance as renewable stabilizing agents for aqueous ceramic suspensions B.M. Cerruttia, M. Zambona, J.D. Megiatto Jr.b, E. Frollinia,
MARK
⁎
a Macromolecular Materials and Lignocellulosic Fibers Group, Center for Science and Technology of BioResources, Institute of Chemistry of São Carlos, University of São Paulo, PO Box 780, 13560-970, São Carlos, SP, Brazil b Institute of Chemistry, University of Campinas (UNICAMP), PO Box 6154, 13083-970, Campinas, SP, Brazil
A R T I C L E I N F O
A B S T R A C T
Keywords: Carboxymethylcellulose Bio-based polyelectrolytes Ceramic suspensions
Low average molar mass cellulose has been submitted to carboxymethylation reactions to yield carboxymethylcellulose (CMC) materials with different degrees of substitution (DS) that have been investigated as alternative renewable stabilizing agents for aqueous alumina suspensions to usual synthetic polyelectrolyte additives. The CMC materials were characterized by infrared, X-ray diffraction and 1H nuclear magnetic resonance (NMR) spectroscopies, as well as by size exclusion chromatography (SEC), thermogravimetry (TG) and differential scanning calorimetry (DSC). All CMC materials reported (DS of 0.7, 1.3 and 1.8 as estimated by 1H NMR) proved to be good additives to stabilize aqueous alumina suspensions with high solid concentrations (60%, w/w). Addition of low amounts of CMC (from 0.10% to 0.20%, w/w) produced suspensions with small and uniformly distributed particle sizes, thereby yielding colloids with lower viscosity, negative zeta potential values and longer sedimentation times. The present work demonstrates the viability of substituting synthetic fossil-based polyelectrolytes in traditional industrial activities with renewable cellulosic biomass-based ones.
1. Introduction The rich and complex macromolecular structure of cellulose, hemicelluloses and lignin allows a large number of chemical modifications to be carried out in order to change their properties. However, it is imperative that chemical manipulations do not significantly alter the molecular structure of the biomacromolecules so that microorganisms can still recognize and degrade the chemically modified materials once they are discharged in the environment (Fischer and Mackwitz, 2016). In this context, an alkali-O2 oxidized lignin derivative has been investigated as a renewable cement dispersant (Kalliola et al., 2015), whereas lignopolyurethanes, which have been prepared from oxypropylated lignosulfonates and diphenylmethane, have been studied as renewable polymeric matrices in sisal reinforced composites (Oliveira et al., 2012).Bendahou and co-workers reported hybrid aerogels prepared from zeolites and cellulose nanofibers that presented superior insulating properties (Bendahou et al., 2015). The microcrystalline cellulose investigated in the present study has been modified via imidazole-catalyzed acylation reactions with carboxylic acid anhydrides (Pires et al., 2015) to yield bio-based ionic liquids as well as to prepare nanocrystalline cellulose via ultrasonic-assisted enzymatic ⁎
Corresponding author. E-mail address: [email protected] (E. Frollini).
http://dx.doi.org/10.1016/j.indcrop.2017.05.029 Received 21 October 2016; Received in revised form 29 March 2017; Accepted 17 May 2017 0926-6690/ © 2017 Elsevier B.V. All rights reserved.
hydrolysis (Cui et al., 2016). Our group has been involved in a continuous effort aiming at replacing non-renewable fossil-based derivatives in material design and processing with renewable biomacromolecules. For example, we have shown that lignins can partially replace phenol in the preparation of phenol-formaldehyde thermoset composites reinforced with several lignocellulosic fibers (Botaro et al., 2010; Megiatto et al., 2007a,b, 2008, 2009, 2010; Ramires et al., 2010a,b,c; Trindade et al., 2005). Another industrial activity that might benefit from the use of biomacromolecules as renewable alternatives to petrochemicals is the colloidal ceramic processing industry. In the colloidal ceramic processing, suspensions with high solid concentration and with uniform particle size distribution are necessary to yield ceramic objects with higher density, lower defects and excellent mechanical and thermal properties. However, thermal collisions among the fine particles in suspensions cause formation of larger particle aggregates with random size distributions that ultimately lead to ceramic objects with poor mechanical and thermal performances. To avoid such problems, synthetic polyelectrolytes (charged macromolecules) able to interact with the surface of the particles are added to the suspensions to create electrostatic and/or steric barriers at the solid/liquid interface that
Industrial Crops & Products 107 (2017) 54–62
B.M. Cerrutti et al.
2. Experimental
prevent particle aggregation upon thermal collision (Ohtsuka et al., 2011; Wisniewska et al., 2014). Aqueous alumina suspensions are extensively used in ceramic processing. When dispersed in water, alumina particles acquire surface charges that are pH dependent. In acidic medium, the alumina surface hydroxyl groups are protonated and the overall particle surface charge is positive. On the other hand, the alumina surface hydroxyl groups are deprotonated in alkaline pH regime, thereby the net surface charge of the particles in suspension is negative. At those two extreme pH regimes, alumina suspensions are stabilized by the electrostatic mechanism as the mutual electrostatic repulsion among the surface likecharged particles minimize formation of aggregates. However, in intermediate pH regimes, some surface hydroxyl groups are protonated, while others are deprotonated. This uneven surface charge distribution creates attractive electrostatic forces among the particles in suspension that drive aggregation. There is a specific pH value in which the number of protonated and deprotonated surface hydroxyl groups is roughly the same. At this pH value, namely isoelectric point (IEP), particle aggregation is most favored due to the mutual attractive electrostatic interactions among the particles in suspension (Cerrutti et al., 2010, 2012, 2013; Megiatto et al., 2016; Ohtsuka et al., 2011; Wisniewska et al., 2014) To circumvent aggregation in intermediate pH regimes, synthetic polyelectrolytes such as sodium polyacrylates are used as stabilizing agents for aqueous alumina suspensions in ceramic processing. Polyelectrolytes interact with the alumina particle surface by van-derWaals forces, formation of hydrogen bonds and electrostatic interactions. Lignosulfonates and chemically modified lignins have proved to be excellent polyelectrolytes to stabilize aqueous alumina suspensions in a wide pH range (Megiatto et al., 2016). Continuing our efforts to expand the scope of biomacromolecules in material processing, herein is described the investigation of carboxymethylcelluloses (CMC) (Fig. 1) with three distinct degrees of substitution (DS) as polyelectrolytes for the stabilization of aqueous alumina suspensions. CMCs are versatile platforms that have been used as the carbon precursor to prepare carbon microspheres with ordered, mesoporous structures (Wu et al., 2015), or blended with carboxymethyl-carrageenan to be used as host to prepare green polymer electrolytes (Rudhziah et al., 2015). In the present work, microcrystalline cellulose (MC) was submitted to carboxymethylation reactions in heterogeneous medium to afford CMCs with relatively high DS values. The use of MC as starting material is based on its low average molecular mass compared to other celluloses. The low average molecular mass enables easier access of reactants to the hydroxyl functionalities on the MC chemical structure, thereby facilitating the heterogeneous carboxymethylation reactions to yield materials with high DS. Furthermore, the low average molecular mass of the resulting CMCs might provide a benefit in the stabilization process of aqueous alumina colloids. CMCs with high average molecular masses are composed of long cellulose chains that might interact with several alumina particles in suspension at the same time (“bridging effects”) to promote aggregation instead of dispersion. Congruently, CMCs with shorter cellulose chains should cover single alumina particles in suspension to create an electrosteric barrier that might prevent aggregation and collapse of the colloid (Cerrutti et al., 2012; Cerrutti et al., 2013; Wisniewska et al., 2014).
2.1. Materials Alumina was kindly provided by Treibacher Schleifmittel Brasil Ltda, São Paulo, Brazil. This material is an α-alumina, mesh325 (particle size < 440 μm), with superficial area of 2.4 m2/g (BET). Microcrystalline cellulose Avicel(MC) was kindly provided by Valdequímica – Comércio e Importação, São Paulo, Brazil. MC had average molecular mass of 20,500 g mol−1 as measured by the Ubbelohde capillary (Φ = 0.53 mm) AVS-350 viscometer from SchottGeräte, which was equipped with Visco Doser AVS 20 Piston Burette also from Schott-Geräte. Isopropanol, ethanol, sodium hydroxide, sodium chloride, acetic acid and hydrochloric acid were purchased from Synth, whereas sodium monochloroacetate was purchased from Acros Organics and deuterated sulfuric acid and deuterated water were purchased from Merck. All purchased chemicals were analytical grade and used as received. Unless otherwise noted, deionized water was used throughout the investigation.
2.2. Synthesis (Barba et al., 2002; Heinze et al., 1999; Heinze and Koschella, 2005; Kutsenko et al., 2002; Ramos et al., 2005; Vashney et al., 2006) 2.2.1. Procedure for the synthesis of carboxymethylcellulose with 0.7 degree of substitution (CMC07) Five grams of MC were dispersed in 130 mL of isopropanol under mechanical stirring for 20 min at room temperature. To the resulting suspension, it was added NaOH (34.00 g, 0.85 mol) and sodium monochloroacetate (12.00 g, 0.1 mol) and the reaction mixture was heated at 55 °C for 3.5 h. After cooling, the crude product was neutralized with acetic acid and precipitated with 500 mL of methanol. The afforded solid was copiously washed with an ethanol/water mixture (9:1, v/v) and re-dissolved in 1000 mL of water. The solution was mechanically stirred for 12 h at room temperature. Addition of 1000 mL of aqueous NaCl 0.2 mol/L promoted precipitation of impurities, which were separated by filtration and discarded. The product in solution was precipitated with ethanol and filtered off. The solid was washed with several portions of ethanol to yield the final product CMC07, which was dried at room temperature until constant mass.
2.2.2. Procedure for the synthesis of carboxymethylcellulose with 1.3 degree of substitution (CMC13) Five grams of CMC07 were submitted to recarboxylation under the same reaction conditions and purification processes as described in 2.2.1 to yield CMC13.This recarboxylation procedure was necessary to yield materials with degree of substitution higher than 0.7.
2.2.3. Procedure for the synthesis of carboxymethylcellulose with 1.8 degree of substitution (CMC18) Five grams of CMC13 were submitted to recarboxylation under the same reaction conditions and purification processes as described in 2.2.1 to yield CMC18. This re-recarboxylation procedure was necessary to yield materials with degree of substitution higher than 1.3.
Fig. 1. Cellulose and carboxymethylcellulose (CMC) chemical structures.
55
Industrial Crops & Products 107 (2017) 54–62
B.M. Cerrutti et al.
2.4.5. Thermal analyses Differential scanning calorimetry (DSC) experiments were carried out using a Shimadzu DSC model 50 piece of equipment. About 6.0 mg of each material was placed in an appropriate platinum sealed container and heated from 20 to 500 °C, at 20 °C min−1, under N2atmosphere (20 mL min−1). Thermogravimetric (TG) studies were carried out using a Shimadzu model TGA-50TA piece of equipment. About 6.0 mg of each material was placed in an appropriate platinum container and heated from 25to 700 °C at 20 °C min−1, under N2atmosphere (20 mL min−1).
2.3. Preparation of aqueous alumina suspensions For viscometer investigations, the aqueous alumina suspension was prepared by adding 60 g of alumina into 40 mL of water to yield a suspension with 60% (w/w) concentration. For zeta potential and particle size investigations, 2.0 g of alumina were added to 98.0 mL of water to yield a suspension with 2% (w/w) concentration, which was the maximum concentration allowed by both pieces of equipment used in the present investigation (Brookhaven, 1999). The pH of the alumina suspensions was adjusted by adding the appropriate volumes of 0.1 molL−1HCl or NaOH aqueous solutions. The pH values were measured using a Micronal digital pHmeter. The CMC materials were added as solids to the alumina suspensions in order to investigate the CMC properties as dispersing agents. After addition of the suitable volumes of acid or base solutions and the desired mass of CMC material, the resulting suspension was kept under mechanical stirring for 1 h, followed by viscometer, zeta potential and particle size measurements.
2.5. Aqueous alumina suspension investigation 2.5.1. Viscometry Viscometer measurements were carried out at room temperature using a Brookfield Viscometer DV II model equipped with a spindle number 3 operating at 3 rpm. The freshly prepared aqueous alumina suspensions were placed into appropriate glass vessels and the viscosity value was read in centipoise (cps).
2.4. CMC – characterization
2.5.2. Zeta potential measurements Zeta potential values were afforded using a ZETAPALS equipment from Brookhaven Instruments Corporation. Aliquots of about 1.5 mL of the suitable suspension was placed in an appropriate cuvette and the zeta potential was read in millivolts (mV).
2.4.1. 1H Nuclear Magnetic Resonance Spectroscopy (1H NMR) 1 H NMR analyses were performed as described by Gronski and Hellmann (Gronski and Hellmann, 1987). Accordingly, CMC samples (50 mg) were dissolved in 1 mL of a deuterated solvent mixture composed of D2SO4/D2O (1:4, v/v). The solution was heated at 90 °C overnight to promote the CMC acid depolymerization reaction. After cooling to room temperature, the hydrolyzed product was transferred to a regular NMR tube without any further purification. The 1H NMR spectra were acquired on a Bruker AMX 250 spectrometer using the deuterated solvents as lock. The spectra were collected at 25 °C and chemical shifts (δ, ppm) were referenced to the residual solvent peak. A total of 16 accumulations were acquired for each sample.
2.5.3. Particle size measurements The average particle diameter was determined by Photon Correlation Spectroscopy using a Fiber Optical Quasi-Elastic Light Scattering “FOQELS” piece of equipment from Brookhaven Instruments Corporation operating with aneodymium laser (50 mW, λ = 532 nm, θ = 155°) and resolution between 2 and 4000 nm. Aliquots of about 3 mL of the suitable suspension were placed in an appropriate vessel, which was introduced into the FOQELS equipment and the particle diameter was read in nanometers (nm). For the experiments of average particle size as function of time, 5 min elapsed between one measurement and the next.
2.4.2. Fourier Transform Infrared Spectroscopy (FTIR) FTIR spectra were collected using a BomemMB-102 spectrometer. The investigated materials were mixed as solids with KBr (ratio 1:100 mg) using a mortar and pestle. The resulting solid mixture was compressed under vacuum to form pellets, which were immediately analyzed.
2.5.4. Scanning electron microscopy – SEM Sedimentation of pristine and CMC treated suspensions yielded wet solids after filtration, which were dried under vacuum at room temperature until constant weight. The resulting dried solids were placed in appropriate containers and coated with a thin gold layer in a sputter-coat system prior to the SEM analyses. The piece of equipment used in the present investigation was a Zeiss-Leica electron microscopy, model 440, operating at 20 kV electron acceleration.
2.4.3. Size-Exclusion Chromatography (SEC) The average molecular mass (Mw) of the CMC samples were investigated using a Shimadzu SCL-10A system equipped with a refractive-index detector, RID-6A. A PLgel guard column and a PLgel Mixed column (10 mm) (300 × 8 mm) were used as stationary phases. Aqueous NaNO3 (0.1 mol L−1) was the mobile phase, flowing at 1 mL/ min, under 27 kgf/cm2pressure and at 35 °C. The CMC materials were dissolved in water at reflux under magnetic stirring for 1.5 h prior to the analyses to yield solutions with 4 mg mL−1 concentration. After cooling, the resulting solution was filtered using Glassfaser Mikrofilter GMF-3 membranes with 47 mm diameter and 1.2 μm pore size. Pullulan polysaccharides were used as standards for calibration (Mw = 1,600,000; 380,000; 212,000; 100,000; 48,000; 23,700; 12,200; 5800; 738 and 180 g mol−1).
3. Results and discussion 3.1. Synthesis of the CMC materials The degree of substitution (DS) is an indirect measurement of the efficiency of the cellulose carboxymethylation reactions. DS can be estimated by 1H NMR spectroscopy of the depolymerized CMC material (Gronski and Hellmann, 1987; see Supporting Information, S.I). The 1H NMR analyses (spectra not shown) of CMC07, CMC13 and CMC18 yielded DS = 0.7, 1.3 and 1.8, respectively (Table 1), indicating partial carboxymethylation of the cellulose chains. Infrared spectroscopy confirmed the partial introduction of carboxymethyl groups onto the cellulose chains. The infrared spectrum of CMC18 (Fig. S1 in Supporting Information, S.I.) showed two carboxymethyl diagnosis bands at 1520 and at 950 cm−1, which were assigned to the symmetrical axial COOe bond stretching and the CHeOeCH bond deformation, respectively (Biswal, 2004; Britto and Assis, 2007; Heinze et al., 1999). Similar infrared data were afforded for CMC07 and CMC13. The average molar mass (Mw) of the CMC materials was investigated by SEC (Fig. S2 in S.I.). The average molar mass of the MC
2.4.4. Crystalline index (Ic) The crystalline index (Ic) of the cellulose and CMC samples were determined by X-ray diffraction using a RIGAKU Rotaflex model RU–200 B diffractometer. The diffractometer was operated at 40 kV, 20 mA and λ (Cu Kα) = 1.5418 Å. Crystalline index of all samples were calculated using the Buschle-Diller and Zeronian equation Ic = (1–I1)/ I2, where I1 is the minimum value for intensity between 2θ = 18° and 19° and I2 is the maximum value for intensity of the crystalline peak between 2θ = 22° and 23° (Buschle-Diller and Zeronian, 1992). 56
Industrial Crops & Products 107 (2017) 54–62
B.M. Cerrutti et al.
DSC analyses (Fig. S5 in S.I.) agreed with the dTG data. A steep exothermic peak (maximum at 390 °C) due to cellulose depolymerization was observed for MC (Fig. S5, top-left, in S.I.), whereas the chemically modified materials yielded lofty exothermic peaks at lower temperatures for the same process. Interestingly, thermal depolymerization events on the modified materials yielded exothermic peaks with two maximums at 292 °C and 302 °C for CMC07 (Fig. S5, top-right, in S.I.), 300 °C and 313 °C for CMC13 (Fig. S5, bottom-left, in S.I.) and 296 °C and 314 °C for CMC18 (Fig. S5, bottom-right, in S.I.). Those two maximums on the DSC traces corroborated the idea that the CMC materials were composed of cellulose chains containing pristine and carboxymethyl glucose subunits, in which the latter decomposed at lower temperatures than the former. The endothermic peaks observed below 100 °C on the DSC profiles of all materials were due to evaporation of residual solvent molecules, most likely ethanol and/or water molecules, whereas the exothermic process at 340 °C observed only on the MC DSC trace probably corresponded to evaporation of water molecules strongly interacting with cellulose chains, also called “structural water”. The present CMC materials were intended to be used as stabilizing agents for aqueous alumina colloids, and their thermal degradation at lower temperature is an asset for the ceramic colloidal processing as the organic additive has to be removed from the final product. Usually, this is done by heating the ceramic object before sintering. Accordingly, CMC thermal decomposition at lower temperatures means that less energy has to be applied to burn off the CMC additives from the ceramic item.
Table 1 Physical and chemical properties of the investigated materials. Material
DSa
Ic (%)b
Average Molar Mass (g mol−1)
Mw/Mn
MC CMC07 CMC13 CMC18
0.0 0.7 1.3 1.8
80 60 40 43
20,500c 127,880d 177,000d 254,500d
– 2.5 3.7 2.1
a
Degree of substitution estimated from NMR data. Crystalline index calculated from the Buschle-Diller and Zeronian equation Ic = (1–I1)/I2, where I1 is the minimum value for intensity between 2θ = 18° and 19° and I2 is the maximum value for intensity of the crystalline peak between 2θ = 22° and 23° (Buschle-Diller and Zeronian, 1992) (Fig. S3 in S.I). c Average molar mass (Mw) afforded from viscometer measurements, see section 2.1. d Average molar mass (Mw) afforded from SEC analyses (Fig. S2 in S.I). b
starting material was 20,500 g mol−1 as measured by viscometry. However, this average molar mass afforded via viscometry should not be directly compared with those afforded via SEC for the CMC ones (Table 1). Apart from the differences between the respective chemical structures, which yield different hydrodynamic radii for the materials in solution, the solvents used in the viscometry and SEC investigations had to be different in our study due to solubility issues (see Experimental 2.4.3). Different solvents might yield distinct hydrodynamic radii even for materials with virtually identical macromolecular structures. Furthermore, the CMC average molar masses obtained via SEC are values relative to Pullulan standards used to calibrate the SEC chromatograms. Congruently, only the values of average molar masses afforded for the CMC materials via SEC can be compared with each other and are discussed as following. SEC investigation of the CMC materials (Table 1 and Fig. S2 in S.I.) indicated that the larger the number of carboxymethyl groups on the cellulose chains the larger is the Mw of the CMC material. This is expected since substitution of the hydroxyl hydrogen nuclei with large carboxymethyl groups must increase cellulose Mw. Most importantly, those findings suggested that depolymerization of the cellulose chains was negligible under the conditions used for the carboxymethylation reactions. Therefore, the CMC materials could work as polyelectrolytes upon dissolution. X-ray analysis of the CMC materials revealed the expected reduction of cellulose crystallinity after the carboxymethylation reaction when compared to the MC starting material (Table 1 and Fig. S3 in S.I.). Random distribution of carboxymethyl side-groups along the cellulose chains reduces the efficiency of the macromolecular packing, encouraging crystallite disorganization. Those structural changes upon carboxymethylation significantly improved dissolution of the CMC materials in water. This high solubility is fundamental to our further investigation about using those polyelectrolytes as stabilizing agents for aqueous alumina suspensions (vide infra). All materials revealed usual cellulose thermal decomposition processes (Fig. S4 in S.I). However, those common thermal events took place at lower temperature ranges for the modified materials. The thermal data clearly showed that the higher the DS the lower is the temperature range in which cellulose backbone thermal decomposition occur. For example, thermolysis of the cellulose glycoside bonds to produce levoglucosan (Britto and Assis, 2007; Ciacco et al., 2010; Ramos et al., 2005; Scheirs et al., 2001), which yielded the largest mass losses on the dTG traces (Fig. S4 in S.I.), occurred between 325 and 400 °C (maximum at 374 °C) for MC, whereas that same thermal event happened between 270 and 350 °C for the CMC materials, with maximum at 322 °C, 308 °C and 305 °C for CMC07, CMC13 and CMC18, respectively. Carbonization of CMC07and CMC13 yielded very broad mass losses at T > 400 °C, whereas carbonization of CMC18 yielded a distinct band centered at 535 °C. CMC18 had the highest DS, therefore carbonization through decarbonylation and decarboxylation of the carboxymethyl groups were more evident for that CMC material when compared to the others.
3.2. CMC materials as stabilizing agents for aqueous alumina suspensions When dispersed in aqueous medium, alumina particles acquire surface charges due to acid/base interactions with water molecules. Accordingly, the alumina net surface charge is pH dependent. In the pH region of the IEP, mutual attractive electrostatic interactions among the fine dispersed alumina particles in conjunction with Brownian motion induce particle aggregation. Formation of large alumina aggregates with random size distribution dramatically increases the suspension viscosity, and ultimately leads to colloid collapse to yield sedimentation. Congruently, the pH value at which maximum viscosity is observed is an estimate of the IEP of the suspension. The IEP is different for each suspension; thereby IEP determination of suspensions is a critical factor in ceramic colloidal processing. Therefore, we first established the IEP for our pristine alumina suspension by viscometry (Fig. 2). The data unambiguously revealed a maximum viscosity value about pH 7.0, which was established as the IEP for the present alumina suspension. Zeta potential and average particle size distribution studies confirmed the IEP value as 7.0 for the alumina suspension investigated
Fig. 2. Pristine aqueous alumina suspension viscosity as function of pH. Suspension with 60% (w/w) of alumina concentration, spindle number 3, rotation speed 3 rpm, room temperature. cps = centipoise.
57
Industrial Crops & Products 107 (2017) 54–62
B.M. Cerrutti et al.
(vide infra). We next investigated the stabilizing effects of CMC addition on the suspension viscosity at critical pH 7.0 (Fig. 3). An ideal stabilizing agent should prevent particle aggregation in a broad pH region, mainly at the IEP, to afford suspension with low viscosity. As the data unfolded, addition of any of the three reported CMC materials steadily decreased the suspension viscosity until reaching a minimum value, which did not change upon further CMC addition. Those were clear evidences for suspension stabilization promoted by the CMC materials. For CMC07 and CMC13, the minimum viscosity values were reached after addition of 0.10% (w/w), whereas CMC18 required 0.20% (w/w) to do the same. It has been reported (Rödel et al., 2012) that addition of 0.10–0.20% (w/w) of sodium polyacrylates reduce the viscosity and produces high negative surface potentials in alumina suspensions at 60% concentration similar to those investigated in the present work. Therefore, the amounts of the CMC materials reported herein are in the range of the that of the benchmark and were established as optimum CMC dosages to achieve suspension stabilization under the conditions investigated. Those optimum amounts were used throughout the present investigation. Studies of viscosity as function of pH were carried out using the optimum amount established for each CMC material to verify their additive properties in acidic and alkaline pH regimes (Fig. 4). All three CMC polyelectrolytes were able to reduce suspension viscosity throughout the pH range investigated. CMC07 and CMC13 performed well and similarly in all pH regimes, whereas CMC18was a good additive at
acidic pH, but not as good as in the IEP and alkaline regions. A possible explanation for CMC18 poorer performance in pH > 5 is its higher DS when compared to the others. The pKa of the carboxymethyl groups is about 4.0; therefore, the appended carboxymethyl groups are deprotonated in the IEP and alkaline regions. The negatively charged carboxymethyl groups on the cellulose backbone bring about repulsive electrostatic forces within a cellulose chain, which prompt the macromolecule to adopt extended conformations in solution rather than the usual statistical random-coil ones. Extended conformations expose a larger number of COOe appended groups to the aqueous medium, allowing the same cellulose chain to interact with several alumina particles in suspension. Hence, bridging them and promoting aggregation. Furthermore, cellulose stretched chains adsorbed onto a single alumina particle should yield a thin electrosteric barrier, which might not be robust enough to preclude the particles to come closer together upon collisions. This effect is more dramatic in high concentration suspensions like those of the present work. If that is true for CMC18, depletion effects became important and desorption of the CMC macromolecules ensued, resulting in spontaneous formation of large alumina aggregates, mainly at the IEP region. Large aggregates in suspension mean higher viscosity. Zeta potential measurements as function of pH were carried out to investigate interactions between alumina particles and CMC macromolecules (Fig. 5). Pristine alumina suspension showed the usual zeta potential behavior, with high positive and negative values in acidic (pH < 6.0) and alkaline medium (pH ≥ 9.0), respectively, indicating the particles had significant net surface charges at those pH regions. Between pH 6.0–8.0, potentials were negligible for the pristine suspension, suggesting equilibrium of surface charges. Zeta potential value was equal to zero at pH 7.10, confirming the IEP for the pristine alumina suspension. Addition of the established amount of any of the present CMC materials led to significant reduction of the zeta potential between 3.0 < pH < 9.0, except for CMC13. In this pH range, the CMC materials bear non-negligible negative charges as the pKa of the carboxymethyl groups is about 4.0. Since the zeta technique only measures potentials at interfaces, the reduction of zeta potential values indicated adsorption of CMC macromolecules onto the surface of the alumina particles in the pH interval investigated. The CMC adsorption partially canceled out the positive surface charges on the pristine alumina particles, thereby reducing the zeta values in pH ≤ 8.0. Nonetheless, suspensions treated with CMC13 showed similar positive zeta potential values in the 3.0–5.0 pH range. Among the CMC materials investigated, CMC13 had the highest polydispersity (Table 1). This heterogeneous molecular size distribution should yield appended carboxymethyl groups with nonuniform pKa values. In proteins, the close proximity of buried carboxylate groups is known to dramatically increase their pKa values due to dipole-dipole and
Fig. 4. Viscosity of pristine and CMC treated alumina suspensions as function of pH. Total suspension mass = 100 g. Alumina mass = 60 g (concentration = 60%, w/w). Conditions: spindle 3, rotation speed 3 rpm, room temperature. cps = centipoise.
Fig. 5. Zeta potential as function of pH for the pristine suspension as well as suspensions treated with CMCs. Conditions: alumina concentration = 2% (w/w), room temperature. mV = millivolts.
Fig. 3. Aqueous alumina suspension viscosity as function of CMC added masses (%, w/w). Total suspension mass = 100 g. Alumina mass = 60 g (concentration = 60%, w/w). Conditions: spindle 3, rotation speed 3 rpm, room temperature. cps = centipoise.
58
Industrial Crops & Products 107 (2017) 54–62
B.M. Cerrutti et al.
Fig. 6. Average particle size distribution of pristine and CMC treated alumina suspensions at pH 10. Conditions: alumina concentration = 2% (w/w), room temperature. a.u. = arbitrary unit.
Fig. 7. Average particle size as function of time for pristine and CMC treated suspensions at pH 7.0. Conditions: alumina concentration = 2% (w/w), room temperature.
the stabilizing effects promoted by addition of CMC to the colloids (Figs. 8–10 ). In the three pH values investigated, the pristine suspension yielded particles with irregular size and shape, which induced poor packing of particles upon sedimentation. Indeed, a large number of voids could be seen on the images afforded from the pristine suspension. On the other hand, CMC treated colloids amended the heterogeneous size and shape distribution of the alumina particles with variable degree of success, ultimately yielding solids with a somewhat better packing. By SEM analyses (Figs. 8–10), CMC13 was the best additive when compared to the other two CMC materials. CMC13 images displayed particles with better size and shape distribution, which allow them to pack up smoothly to yield solids with evident lower porosity. It should be noted however that the SEM images were afforded from dry materials, i.e. after sedimentation of the suspension, followed by filtration and vacuum drying of the resulting materials. Therefore, the morphological aspects revealed by SEM should be taken as indirect evidences of the real state of the suspensions.
hydrogen bonding interactions (Lau et al., 2010). In the present case, the cellulose chains with larger molecular sizes in CMC13 should adopt folded conformations with some carboxymethyl groups buried in their interior. The pKa values of the buried groups in CMC13 should be larger than 4 and are therefore in the neutral form at the 3.0-5.0 pH range. Congruently, suspensions treated with CMC13 yielded positive zeta-potentials at those pH regimes. For pH > 8.0, the zeta potentials for pristine and stabilized suspensions were virtually the same. This finding would suggest no adsorption of CMC onto alumina particles at pH > 8.0. However, viscosity measurements as function of pH (Fig. 4) for the CMC stabilized suspensions were lower compared to that of the pristine colloid at pH > 8.0, informing that adsorption had occurred. To settle the dispute between those two opposing hypothesis, average particle size distribution using the FOQELS technique were carried out. Fig. 6 shows the average particle size distribution afforded for the pristine and CMC stabilized suspensions at pH 10. The data clearly revealed that the CMC treated colloids have smaller particles in suspension than the pristine one, suggesting colloid stabilization that could only occur in alkaline medium through CMC adsorption onto the particles. Although attractive electrostatic interactions between the biopolyelectrolyte and alumina particles are unlike at pH 10.0, CMC carboxylate groups might engage in Lewis acid/base interactions with aluminum atoms on the particle surface to produce thermodynamically stable adducts that drive adsorption in alkaline medium. Furthermore, deprotonated hydroxyl groups on the particle surface can easily form hydrogen bonds with the hydroxyl groups present on the CMC cellulose chains to provide extra driving force for the adsorption process. Those interactions might promote CMC adsorption onto alumina particles in suspension in the alkaline regime. CMC18 had the poorer performance among the CMC additives, yielding colloids with relatively larger particles in suspension if compared to the other two CMCs. This large average particle size distribution observed for CMC18 nicely corroborated our hypothesis that CMC materials with higher DS promote particle aggregation in alkaline medium most likely through “bridging” and/or “depletion effects” (vide supra). Experiments of average particle size as function of time were carried out at the critical IEP region (pH 7.0) to gather information about aggregation and sedimentation kinetics (Fig. 7). Pristine alumina particles in suspension promptly formed large aggregates upon Brownian collision, which reached the maximum particle size detection limit of our equipment (4000 nm) in about 20 min. Conversely, suspension containing particles decorated with CMC macromolecules had smaller and near constant size (400–700 nm) in the time range investigated (80 min), suggesting the CMC polyelectrolytes indeed challenged particle aggregation in the IEP region. Scanning electron microscopy of the alumina suspension after sedimentation followed by filtration and vacuum drying highlighted
4. Conclusions Introduction of carboxymethyl groups onto cellulosic chains decreases crystallinity and thermal stability when compared to pristine cellulose. Reduction of crystallinity improves solubility, which favors CMC materials to act as bio-based polyelectrolytes in ceramic colloidal processing. Thermolysis at lower temperatures is also advantageous as thermal decomposition of the CMC additives before sintering of the ceramic object will require less energy to take place. CMC materials are able to stabilize aqueous alumina colloids with high solid content in a wide pH range. The carboxymethyl groups appended to the cellulose chains can engage in hydrogen bonding formation, electrostatic and Lewis acid/base interactions with the hydroxyl groups and aluminum atoms on the surface of alumina particles in suspension. Those interactions promote strong adsorption of polyelectrolyte species onto the particles that create electrosteric barriers at the solid/liquid interface that prevent aggregation. As a result, suspension with small particle size and low viscosity are afforded upon addition of 0.1–0.2% (w/w) of CMC. However, the degree of substitution (DS) is a crucial factor in the preparation of CMC polyelectrolytes. Introduction of too many carboxymethyl groups onto the cellulosic backbone leads to materials with poorer performance as stabilizing agents. Our investigation reveals that DS values as high as 1.8 induce some particle aggregation if compared to materials with lower ones. On the other hand, CMC materials with DS of about 1.3 exhibit excellent properties as additives, opening up the possibility of using those materials as biorenewable stabilizing agents in alumina colloidal processing. Thus, careful control of the carboxymethylation reaction conditions is fundamental to afford CMC additives with the right DS and consistent dispersing properties. 59
Industrial Crops & Products 107 (2017) 54–62
B.M. Cerrutti et al.
Fig. 8. SEM images of the pristine and CMC treated alumina suspensions at pH 3.0 after sedimentation, filtration and drying.
Fig. 9. SEM images of the pristine and CMC treated alumina suspensions at pH 7.0 after sedimentation, filtration and drying.
suspensions treated with CMCh and CML under similar conditions. The CMC better performance is probably due to its lower average molecular mass, which favors its adsorption onto the surface of single alumina particles in suspension, which in turn leads to a better stabilization of the colloids.
In our previous works, we investigated carboxymethylchitosan (CMCh) (Cerrutti et al., 2013), and carboxymethyllignins (CML) (Cerrutti et al., 2012) as stabilizing agents for aqueous alumina suspensions. Comparing those materials with the present ones, it is possible to conclude that the CMCs reported in the present investigation perform the best, especially CMC13. Suspensions treated with CMCs show lower viscosity values, higher negative zeta potentials and small particle sizes in a broad pH range when compared to the alumina 60
Industrial Crops & Products 107 (2017) 54–62
B.M. Cerrutti et al.
Fig. 10. SEM images of the pristine and CMC treated alumina suspensions at pH 10.0 after sedimentation, filtration and drying. Crops. Prod. 83, 346–352. Fischer, H., Mackwitz, H., 2016. Renewable raw material for chemical technical products, conversion of biomass in the non-energy fields. http://www.sinanco.com/biomassarticle1.html. (Accessed 2 June 2016). Gronski, W., Hellmann, G., 1987. NMR-spektroskopische charakterisierung von carboxymethylcellulose. Papier (Darmstadt) 41, 668–672. Heinze, T., Koschella, A., 2005. Carboxymethyl ethers of cellulose and starch. A review. Macromol. Symp. 223, 13–39. Heinze, T., Liebert, T., Klufers, P., Meister, F., 1999. Carboxymethylation of cellulose in unconvencional média. Cellulose 6, 153–165. Kalliola, A., Vehmas, T., Liitia, T., Tamminen, T., 2015. Alkali-O2 oxidized lignin: a biobased concrete plasticizer. Ind. Crops Prod. 74, 150–157. Kutsenko, L.I., Ivanova, N.P., Karetnikova, E.B., Bobasheva, A.S., Bochek, A.M., Panarin, E.F., 2002. Characteristics of aqueous solutions methyl-cellulose and polymethacrylamidoglusose mixtures. Russ. J. Appl. Chem. 75, 314–318. Lau, W.L., DeGrado, W.F., Roder, H., 2010. The effects of pKa tuning on the thermodynamics and kinetics of folding: design of a solvent-shielded carboxylate pair at the a-position of a coiled-coil. Biophys. J. 99, 2299–2308. Megiatto Jr., J.D., Oliveira, F.B., Rosa, D.S., Castellan, A., Frollini, E., 2007a. Renewable resources as reinforcement of polymeric matrices: composites based on phenolic thermosets and chemically modified sisal fibers. Macromol. Biosci. 7, 1121–1131. Megiatto Jr., J.D., Hoareau, W., Gardrat, C., Frollini, E., Castellan, A., 2007b. Sisal fibers: surface chemical modification using reagent obtained from a renewable source; characterization of hemicellulose and lignin as model study. J. Agric. Food Chem. 55, 8576–8584. Megiatto Jr., J.D., Silva, C.G., Rosa, D.S., Frollini, E., 2008. Sisal chemically modified with lignins: correlation between fibers and phenolic composites properties. Polym. Degrad. Stab. 93, 1109–1121. Megiatto Jr., J.D., Silva, C.G., Ramires, E.C., Frollini, E., 2009. Thermoset matrix reinforced with sisal fibers: effect of the cure cycle on the properties of the biobased composite. Polym. Test. 28, 793–800. Megiatto Jr., J.D., Ramires, E.C., Frollini, E., 2010. Phenolic matrices and sisal fibers modified with hydroxy terminated polybutadiene rubber: impact strength, water absorption, and morphological aspects of thermosets and composites. Ind. Crop. Prod. 31, 178–184. Megiatto Jr., J.D., Cerrutti, B.M., Frollini, E., 2016. Sodium lignosulfonate as a renewable stabilizing agent for aqueous alumina suspensions. Int. J. Biol. Macromol. 82, 927–932. Ohtsuka, H., Mizutani, H., Iio, S., Asai, K., Kiguchi, T., Satone, H., Mori, T., Tsubaki, J., 2011. Effects of sintering additives on dispersion properties of Al2O3 slurry containing polyacrylic acid dispersant. J. Eur. Ceram. Soc. 31, 517–522. Oliveira, F., Ramires, E.C.A., Frollini, E., Belgacem, M.N., 2012. Lignopolyurethanic materials based on oxypropylated sodium lignosulfonate and castor oil blends. Ind. Crop. Prod. 72, 77–86. Pires, P.A.R., Teixeira, T.C., Bioni, T.A., Nawaz, H., El Seoud, O.A., 2015. Imidazolecatalyzed esterification of cellulose in ionic liquid/molecular solvents: a multitechnique approach to probe effects of the medium. Ind. Crops. Prod. 77, 180–189. Rödel, C., Müller, M., Glorius, M., Potthoff, A., Michaelis, A., 2012. Effect of varied powder processing routes on the stabilizing performance and coordination type of
Acknowledgements The authors are grateful to FAPESP (The State of São Paulo Research Foundation, Brazil) and to CNPq (National Council for Scientific and Technological Development, Brazil) for financial support. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.indcrop.2017.05.029. References Barba, C., Montané, D., Rinaudo, M., Farriol, X., 2002. Synthesis and characterizations of carboxymethylcellulose (CMC) from nom-wood fibers I: Accessibility of cellulose fibers and CMC synthesis. Cellulose 9, 319–326. Bendahou, D., Bendahou, A., Seantier, B., Grohens, Y., Kaddami, H., 2015. Nanofibrillatedcellulose-zeolites based new hybrid composites aerogels with super thermal insulating properties. Ind. Crop. Prod. 65, 374–382. Biswal, D.R., 2004. Characterization of carboxymethyl cellulose and polyacrylamide graft copolymer. Carbohyd. Polym. 57, 379–387. Botaro, V.R., Siqueira, G., Megiatto Jr., J.D., Frollini, E., 2010. Sisal fibers treated with NaOH and benzophenonetetracarboxylicdianhydride as reinforcement of phenolic matrix. J. Appl. Polym. Sci. 115, 269–276. Britto, D., Assis, O.B.G., 2007. Synthesis and mechanical properties of quaternary salts of chitosan-based films for food applications. Int. J. Biol. Macromol. 41, 198–203. Brookhaven Instruments Corporation, 1999. Instruction Manual for ZetaPALS, Zeta Potential Analyzer. Brookhaven Instruments Corporation, New York pp. 23. Buschle-Diller, G., Zeronian, S.H., 1992. Enhancing the reactivity and strength of cotton fibers. J. Appl. Polym. Sci. 45, 967–979. Cerrutti, B.M., Lamas, J.C., Britto, D., Campana-Filho, S.P., Frollini, E., 2010. Derivatives of biomacromolecules as stabilizers of aqueous alumina suspensions. J. Appl. Polym. Sci. 117, 58–66. Cerrutti, B.M., Souza, C.S., Castellan, A., Ruggiero, R., Frollini, E., 2012. Carboxymethyl lignin as stabilizing agent in aqueous ceramic suspensions. Ind. Crop. Prod. 36, 108–115. Cerrutti, B.M., Lamas, J.C., Campana-Filho, S.P., Frollini, E., 2013. Carboxymethyl chitosan: preparation and use in colloidal ceramic processing. J. Polym. Environ. 21, 816–825. Ciacco, G.T., Morgado, D.L., Frollini, E., Possidonio, S., Seoud El, O.A., 2010. Some aspects of acetylation of untreated and mercerized sisal cellulose. J. Braz. Chem. Soc. 21, 71–77. Cui, S., Zhang, S., Xiong, L., Sun, Q., 2016. Green preparation and characterization of size controlled nanocrystalline cellulose via ultrasonic-assisted enzymatic hydrolysis. Ind.
61
Industrial Crops & Products 107 (2017) 54–62
B.M. Cerrutti et al.
application. Ind. Crops. Prod. 72, 133–141. Scheirs, J., Camino, G., Tumiati, W., 2001. Overview of water evolution during the thermal degradation of cellulose. Eur. Polym. J. 37, 933–942. Trindade, W.G., Hoareau, W., Megiatto Jr., J.D., Razera, I.A.T., Castellan, A., Frollini, E., 2005. Thermoset phenolic matrices reinforced with unmodified and surface-grafted furfuryl alcohol sugar cane bagasse and curaua fibers: properties of fibers and composites. Biomacromolecules 6, 2485–2496. Vashney, V.K., Gupta, P.K., Naithani, S., Khullar, R., Bhatt, A., Soni, P.L., 2006. Carboxymethylationof α-cellulose isolated from Lantana camara with respect to degree of substitution and rheological behavior. Carbohyd. Polym. 63, 40–45. Wisniewska, M., Urban, T., Grza̧dka, E., Zarko, V.I., Gunko, V.M., 2014. Comparison of adsorption affinity of polyacrylic acid for surfaces of mixed silica-alumina. Colloid. Polym. Sci. 292, 699–705. Wu, Q., Li, W., Wu, Y., Zong, G., Liu, S., 2015. Effect of reaction time on structure of ordered mesoporous carbon microspheres prepared from carboxymethyl cellulose by soft-template method. Ind. Crops. Prod. 76, 866–872.
polyacrylate in alumina suspensions. J. Eur. Ceram. Soc. 32, 363–370. Ramires, E.C., Megiatto Jr., J.D., Gardrat, C., Castellan, A., Frollini, E., 2010a. Valorization of an industrial organosolv-sugarcane bagasse lignin: characterization and use as a matrix in biobased composites reinforced with sisal fibers. Biotechnol. Bioeng. 107, 612–621. Ramires, E.C., Megiatto Jr., J.D., Gardrat, C., Castellan, A., Frollini, E., 2010b. Biobased composites from glyoxal-phenol matrices reinforced with microcrystalline cellulose. Polimeros 20, 126–133. Ramires, E.C., Megiatto Jr., J.D., Gardrat, C., Castellan, A., Frollini, E., 2010c. Biobased composites from glyoxal-phenolic resins and sisal fibers. Bioresour. Technol. 101, 1998–2006. Ramos, L.A., Frollini, E., Heinze, T.H., 2005. Carboxymethylation of cellulose in the new solvent dimethyl sulfoxide/tetrabutylammonium fluoride. Carbohyd. Polym. 60, 259–267. Rudhziah, S., Rani, M.S.A., Ahmad, A., Mohamed, N.S., Kaddami, H., 2015. Potential of blend of kappacarrageenan and cellulose derivatives for green polymer electrolyte
62
Contents lists available at ScienceDirect
Industrial Crops & Products journal homepage: www.elsevier.com/locate/indcrop
Synthesis of carboxymethylcelluloses with different degrees of substitution and their performance as renewable stabilizing agents for aqueous ceramic suspensions B.M. Cerruttia, M. Zambona, J.D. Megiatto Jr.b, E. Frollinia,
MARK
⁎
a Macromolecular Materials and Lignocellulosic Fibers Group, Center for Science and Technology of BioResources, Institute of Chemistry of São Carlos, University of São Paulo, PO Box 780, 13560-970, São Carlos, SP, Brazil b Institute of Chemistry, University of Campinas (UNICAMP), PO Box 6154, 13083-970, Campinas, SP, Brazil
A R T I C L E I N F O
A B S T R A C T
Keywords: Carboxymethylcellulose Bio-based polyelectrolytes Ceramic suspensions
Low average molar mass cellulose has been submitted to carboxymethylation reactions to yield carboxymethylcellulose (CMC) materials with different degrees of substitution (DS) that have been investigated as alternative renewable stabilizing agents for aqueous alumina suspensions to usual synthetic polyelectrolyte additives. The CMC materials were characterized by infrared, X-ray diffraction and 1H nuclear magnetic resonance (NMR) spectroscopies, as well as by size exclusion chromatography (SEC), thermogravimetry (TG) and differential scanning calorimetry (DSC). All CMC materials reported (DS of 0.7, 1.3 and 1.8 as estimated by 1H NMR) proved to be good additives to stabilize aqueous alumina suspensions with high solid concentrations (60%, w/w). Addition of low amounts of CMC (from 0.10% to 0.20%, w/w) produced suspensions with small and uniformly distributed particle sizes, thereby yielding colloids with lower viscosity, negative zeta potential values and longer sedimentation times. The present work demonstrates the viability of substituting synthetic fossil-based polyelectrolytes in traditional industrial activities with renewable cellulosic biomass-based ones.
1. Introduction The rich and complex macromolecular structure of cellulose, hemicelluloses and lignin allows a large number of chemical modifications to be carried out in order to change their properties. However, it is imperative that chemical manipulations do not significantly alter the molecular structure of the biomacromolecules so that microorganisms can still recognize and degrade the chemically modified materials once they are discharged in the environment (Fischer and Mackwitz, 2016). In this context, an alkali-O2 oxidized lignin derivative has been investigated as a renewable cement dispersant (Kalliola et al., 2015), whereas lignopolyurethanes, which have been prepared from oxypropylated lignosulfonates and diphenylmethane, have been studied as renewable polymeric matrices in sisal reinforced composites (Oliveira et al., 2012).Bendahou and co-workers reported hybrid aerogels prepared from zeolites and cellulose nanofibers that presented superior insulating properties (Bendahou et al., 2015). The microcrystalline cellulose investigated in the present study has been modified via imidazole-catalyzed acylation reactions with carboxylic acid anhydrides (Pires et al., 2015) to yield bio-based ionic liquids as well as to prepare nanocrystalline cellulose via ultrasonic-assisted enzymatic ⁎
Corresponding author. E-mail address: [email protected] (E. Frollini).
http://dx.doi.org/10.1016/j.indcrop.2017.05.029 Received 21 October 2016; Received in revised form 29 March 2017; Accepted 17 May 2017 0926-6690/ © 2017 Elsevier B.V. All rights reserved.
hydrolysis (Cui et al., 2016). Our group has been involved in a continuous effort aiming at replacing non-renewable fossil-based derivatives in material design and processing with renewable biomacromolecules. For example, we have shown that lignins can partially replace phenol in the preparation of phenol-formaldehyde thermoset composites reinforced with several lignocellulosic fibers (Botaro et al., 2010; Megiatto et al., 2007a,b, 2008, 2009, 2010; Ramires et al., 2010a,b,c; Trindade et al., 2005). Another industrial activity that might benefit from the use of biomacromolecules as renewable alternatives to petrochemicals is the colloidal ceramic processing industry. In the colloidal ceramic processing, suspensions with high solid concentration and with uniform particle size distribution are necessary to yield ceramic objects with higher density, lower defects and excellent mechanical and thermal properties. However, thermal collisions among the fine particles in suspensions cause formation of larger particle aggregates with random size distributions that ultimately lead to ceramic objects with poor mechanical and thermal performances. To avoid such problems, synthetic polyelectrolytes (charged macromolecules) able to interact with the surface of the particles are added to the suspensions to create electrostatic and/or steric barriers at the solid/liquid interface that
Industrial Crops & Products 107 (2017) 54–62
B.M. Cerrutti et al.
2. Experimental
prevent particle aggregation upon thermal collision (Ohtsuka et al., 2011; Wisniewska et al., 2014). Aqueous alumina suspensions are extensively used in ceramic processing. When dispersed in water, alumina particles acquire surface charges that are pH dependent. In acidic medium, the alumina surface hydroxyl groups are protonated and the overall particle surface charge is positive. On the other hand, the alumina surface hydroxyl groups are deprotonated in alkaline pH regime, thereby the net surface charge of the particles in suspension is negative. At those two extreme pH regimes, alumina suspensions are stabilized by the electrostatic mechanism as the mutual electrostatic repulsion among the surface likecharged particles minimize formation of aggregates. However, in intermediate pH regimes, some surface hydroxyl groups are protonated, while others are deprotonated. This uneven surface charge distribution creates attractive electrostatic forces among the particles in suspension that drive aggregation. There is a specific pH value in which the number of protonated and deprotonated surface hydroxyl groups is roughly the same. At this pH value, namely isoelectric point (IEP), particle aggregation is most favored due to the mutual attractive electrostatic interactions among the particles in suspension (Cerrutti et al., 2010, 2012, 2013; Megiatto et al., 2016; Ohtsuka et al., 2011; Wisniewska et al., 2014) To circumvent aggregation in intermediate pH regimes, synthetic polyelectrolytes such as sodium polyacrylates are used as stabilizing agents for aqueous alumina suspensions in ceramic processing. Polyelectrolytes interact with the alumina particle surface by van-derWaals forces, formation of hydrogen bonds and electrostatic interactions. Lignosulfonates and chemically modified lignins have proved to be excellent polyelectrolytes to stabilize aqueous alumina suspensions in a wide pH range (Megiatto et al., 2016). Continuing our efforts to expand the scope of biomacromolecules in material processing, herein is described the investigation of carboxymethylcelluloses (CMC) (Fig. 1) with three distinct degrees of substitution (DS) as polyelectrolytes for the stabilization of aqueous alumina suspensions. CMCs are versatile platforms that have been used as the carbon precursor to prepare carbon microspheres with ordered, mesoporous structures (Wu et al., 2015), or blended with carboxymethyl-carrageenan to be used as host to prepare green polymer electrolytes (Rudhziah et al., 2015). In the present work, microcrystalline cellulose (MC) was submitted to carboxymethylation reactions in heterogeneous medium to afford CMCs with relatively high DS values. The use of MC as starting material is based on its low average molecular mass compared to other celluloses. The low average molecular mass enables easier access of reactants to the hydroxyl functionalities on the MC chemical structure, thereby facilitating the heterogeneous carboxymethylation reactions to yield materials with high DS. Furthermore, the low average molecular mass of the resulting CMCs might provide a benefit in the stabilization process of aqueous alumina colloids. CMCs with high average molecular masses are composed of long cellulose chains that might interact with several alumina particles in suspension at the same time (“bridging effects”) to promote aggregation instead of dispersion. Congruently, CMCs with shorter cellulose chains should cover single alumina particles in suspension to create an electrosteric barrier that might prevent aggregation and collapse of the colloid (Cerrutti et al., 2012; Cerrutti et al., 2013; Wisniewska et al., 2014).
2.1. Materials Alumina was kindly provided by Treibacher Schleifmittel Brasil Ltda, São Paulo, Brazil. This material is an α-alumina, mesh325 (particle size < 440 μm), with superficial area of 2.4 m2/g (BET). Microcrystalline cellulose Avicel(MC) was kindly provided by Valdequímica – Comércio e Importação, São Paulo, Brazil. MC had average molecular mass of 20,500 g mol−1 as measured by the Ubbelohde capillary (Φ = 0.53 mm) AVS-350 viscometer from SchottGeräte, which was equipped with Visco Doser AVS 20 Piston Burette also from Schott-Geräte. Isopropanol, ethanol, sodium hydroxide, sodium chloride, acetic acid and hydrochloric acid were purchased from Synth, whereas sodium monochloroacetate was purchased from Acros Organics and deuterated sulfuric acid and deuterated water were purchased from Merck. All purchased chemicals were analytical grade and used as received. Unless otherwise noted, deionized water was used throughout the investigation.
2.2. Synthesis (Barba et al., 2002; Heinze et al., 1999; Heinze and Koschella, 2005; Kutsenko et al., 2002; Ramos et al., 2005; Vashney et al., 2006) 2.2.1. Procedure for the synthesis of carboxymethylcellulose with 0.7 degree of substitution (CMC07) Five grams of MC were dispersed in 130 mL of isopropanol under mechanical stirring for 20 min at room temperature. To the resulting suspension, it was added NaOH (34.00 g, 0.85 mol) and sodium monochloroacetate (12.00 g, 0.1 mol) and the reaction mixture was heated at 55 °C for 3.5 h. After cooling, the crude product was neutralized with acetic acid and precipitated with 500 mL of methanol. The afforded solid was copiously washed with an ethanol/water mixture (9:1, v/v) and re-dissolved in 1000 mL of water. The solution was mechanically stirred for 12 h at room temperature. Addition of 1000 mL of aqueous NaCl 0.2 mol/L promoted precipitation of impurities, which were separated by filtration and discarded. The product in solution was precipitated with ethanol and filtered off. The solid was washed with several portions of ethanol to yield the final product CMC07, which was dried at room temperature until constant mass.
2.2.2. Procedure for the synthesis of carboxymethylcellulose with 1.3 degree of substitution (CMC13) Five grams of CMC07 were submitted to recarboxylation under the same reaction conditions and purification processes as described in 2.2.1 to yield CMC13.This recarboxylation procedure was necessary to yield materials with degree of substitution higher than 0.7.
2.2.3. Procedure for the synthesis of carboxymethylcellulose with 1.8 degree of substitution (CMC18) Five grams of CMC13 were submitted to recarboxylation under the same reaction conditions and purification processes as described in 2.2.1 to yield CMC18. This re-recarboxylation procedure was necessary to yield materials with degree of substitution higher than 1.3.
Fig. 1. Cellulose and carboxymethylcellulose (CMC) chemical structures.
55
Industrial Crops & Products 107 (2017) 54–62
B.M. Cerrutti et al.
2.4.5. Thermal analyses Differential scanning calorimetry (DSC) experiments were carried out using a Shimadzu DSC model 50 piece of equipment. About 6.0 mg of each material was placed in an appropriate platinum sealed container and heated from 20 to 500 °C, at 20 °C min−1, under N2atmosphere (20 mL min−1). Thermogravimetric (TG) studies were carried out using a Shimadzu model TGA-50TA piece of equipment. About 6.0 mg of each material was placed in an appropriate platinum container and heated from 25to 700 °C at 20 °C min−1, under N2atmosphere (20 mL min−1).
2.3. Preparation of aqueous alumina suspensions For viscometer investigations, the aqueous alumina suspension was prepared by adding 60 g of alumina into 40 mL of water to yield a suspension with 60% (w/w) concentration. For zeta potential and particle size investigations, 2.0 g of alumina were added to 98.0 mL of water to yield a suspension with 2% (w/w) concentration, which was the maximum concentration allowed by both pieces of equipment used in the present investigation (Brookhaven, 1999). The pH of the alumina suspensions was adjusted by adding the appropriate volumes of 0.1 molL−1HCl or NaOH aqueous solutions. The pH values were measured using a Micronal digital pHmeter. The CMC materials were added as solids to the alumina suspensions in order to investigate the CMC properties as dispersing agents. After addition of the suitable volumes of acid or base solutions and the desired mass of CMC material, the resulting suspension was kept under mechanical stirring for 1 h, followed by viscometer, zeta potential and particle size measurements.
2.5. Aqueous alumina suspension investigation 2.5.1. Viscometry Viscometer measurements were carried out at room temperature using a Brookfield Viscometer DV II model equipped with a spindle number 3 operating at 3 rpm. The freshly prepared aqueous alumina suspensions were placed into appropriate glass vessels and the viscosity value was read in centipoise (cps).
2.4. CMC – characterization
2.5.2. Zeta potential measurements Zeta potential values were afforded using a ZETAPALS equipment from Brookhaven Instruments Corporation. Aliquots of about 1.5 mL of the suitable suspension was placed in an appropriate cuvette and the zeta potential was read in millivolts (mV).
2.4.1. 1H Nuclear Magnetic Resonance Spectroscopy (1H NMR) 1 H NMR analyses were performed as described by Gronski and Hellmann (Gronski and Hellmann, 1987). Accordingly, CMC samples (50 mg) were dissolved in 1 mL of a deuterated solvent mixture composed of D2SO4/D2O (1:4, v/v). The solution was heated at 90 °C overnight to promote the CMC acid depolymerization reaction. After cooling to room temperature, the hydrolyzed product was transferred to a regular NMR tube without any further purification. The 1H NMR spectra were acquired on a Bruker AMX 250 spectrometer using the deuterated solvents as lock. The spectra were collected at 25 °C and chemical shifts (δ, ppm) were referenced to the residual solvent peak. A total of 16 accumulations were acquired for each sample.
2.5.3. Particle size measurements The average particle diameter was determined by Photon Correlation Spectroscopy using a Fiber Optical Quasi-Elastic Light Scattering “FOQELS” piece of equipment from Brookhaven Instruments Corporation operating with aneodymium laser (50 mW, λ = 532 nm, θ = 155°) and resolution between 2 and 4000 nm. Aliquots of about 3 mL of the suitable suspension were placed in an appropriate vessel, which was introduced into the FOQELS equipment and the particle diameter was read in nanometers (nm). For the experiments of average particle size as function of time, 5 min elapsed between one measurement and the next.
2.4.2. Fourier Transform Infrared Spectroscopy (FTIR) FTIR spectra were collected using a BomemMB-102 spectrometer. The investigated materials were mixed as solids with KBr (ratio 1:100 mg) using a mortar and pestle. The resulting solid mixture was compressed under vacuum to form pellets, which were immediately analyzed.
2.5.4. Scanning electron microscopy – SEM Sedimentation of pristine and CMC treated suspensions yielded wet solids after filtration, which were dried under vacuum at room temperature until constant weight. The resulting dried solids were placed in appropriate containers and coated with a thin gold layer in a sputter-coat system prior to the SEM analyses. The piece of equipment used in the present investigation was a Zeiss-Leica electron microscopy, model 440, operating at 20 kV electron acceleration.
2.4.3. Size-Exclusion Chromatography (SEC) The average molecular mass (Mw) of the CMC samples were investigated using a Shimadzu SCL-10A system equipped with a refractive-index detector, RID-6A. A PLgel guard column and a PLgel Mixed column (10 mm) (300 × 8 mm) were used as stationary phases. Aqueous NaNO3 (0.1 mol L−1) was the mobile phase, flowing at 1 mL/ min, under 27 kgf/cm2pressure and at 35 °C. The CMC materials were dissolved in water at reflux under magnetic stirring for 1.5 h prior to the analyses to yield solutions with 4 mg mL−1 concentration. After cooling, the resulting solution was filtered using Glassfaser Mikrofilter GMF-3 membranes with 47 mm diameter and 1.2 μm pore size. Pullulan polysaccharides were used as standards for calibration (Mw = 1,600,000; 380,000; 212,000; 100,000; 48,000; 23,700; 12,200; 5800; 738 and 180 g mol−1).
3. Results and discussion 3.1. Synthesis of the CMC materials The degree of substitution (DS) is an indirect measurement of the efficiency of the cellulose carboxymethylation reactions. DS can be estimated by 1H NMR spectroscopy of the depolymerized CMC material (Gronski and Hellmann, 1987; see Supporting Information, S.I). The 1H NMR analyses (spectra not shown) of CMC07, CMC13 and CMC18 yielded DS = 0.7, 1.3 and 1.8, respectively (Table 1), indicating partial carboxymethylation of the cellulose chains. Infrared spectroscopy confirmed the partial introduction of carboxymethyl groups onto the cellulose chains. The infrared spectrum of CMC18 (Fig. S1 in Supporting Information, S.I.) showed two carboxymethyl diagnosis bands at 1520 and at 950 cm−1, which were assigned to the symmetrical axial COOe bond stretching and the CHeOeCH bond deformation, respectively (Biswal, 2004; Britto and Assis, 2007; Heinze et al., 1999). Similar infrared data were afforded for CMC07 and CMC13. The average molar mass (Mw) of the CMC materials was investigated by SEC (Fig. S2 in S.I.). The average molar mass of the MC
2.4.4. Crystalline index (Ic) The crystalline index (Ic) of the cellulose and CMC samples were determined by X-ray diffraction using a RIGAKU Rotaflex model RU–200 B diffractometer. The diffractometer was operated at 40 kV, 20 mA and λ (Cu Kα) = 1.5418 Å. Crystalline index of all samples were calculated using the Buschle-Diller and Zeronian equation Ic = (1–I1)/ I2, where I1 is the minimum value for intensity between 2θ = 18° and 19° and I2 is the maximum value for intensity of the crystalline peak between 2θ = 22° and 23° (Buschle-Diller and Zeronian, 1992). 56
Industrial Crops & Products 107 (2017) 54–62
B.M. Cerrutti et al.
DSC analyses (Fig. S5 in S.I.) agreed with the dTG data. A steep exothermic peak (maximum at 390 °C) due to cellulose depolymerization was observed for MC (Fig. S5, top-left, in S.I.), whereas the chemically modified materials yielded lofty exothermic peaks at lower temperatures for the same process. Interestingly, thermal depolymerization events on the modified materials yielded exothermic peaks with two maximums at 292 °C and 302 °C for CMC07 (Fig. S5, top-right, in S.I.), 300 °C and 313 °C for CMC13 (Fig. S5, bottom-left, in S.I.) and 296 °C and 314 °C for CMC18 (Fig. S5, bottom-right, in S.I.). Those two maximums on the DSC traces corroborated the idea that the CMC materials were composed of cellulose chains containing pristine and carboxymethyl glucose subunits, in which the latter decomposed at lower temperatures than the former. The endothermic peaks observed below 100 °C on the DSC profiles of all materials were due to evaporation of residual solvent molecules, most likely ethanol and/or water molecules, whereas the exothermic process at 340 °C observed only on the MC DSC trace probably corresponded to evaporation of water molecules strongly interacting with cellulose chains, also called “structural water”. The present CMC materials were intended to be used as stabilizing agents for aqueous alumina colloids, and their thermal degradation at lower temperature is an asset for the ceramic colloidal processing as the organic additive has to be removed from the final product. Usually, this is done by heating the ceramic object before sintering. Accordingly, CMC thermal decomposition at lower temperatures means that less energy has to be applied to burn off the CMC additives from the ceramic item.
Table 1 Physical and chemical properties of the investigated materials. Material
DSa
Ic (%)b
Average Molar Mass (g mol−1)
Mw/Mn
MC CMC07 CMC13 CMC18
0.0 0.7 1.3 1.8
80 60 40 43
20,500c 127,880d 177,000d 254,500d
– 2.5 3.7 2.1
a
Degree of substitution estimated from NMR data. Crystalline index calculated from the Buschle-Diller and Zeronian equation Ic = (1–I1)/I2, where I1 is the minimum value for intensity between 2θ = 18° and 19° and I2 is the maximum value for intensity of the crystalline peak between 2θ = 22° and 23° (Buschle-Diller and Zeronian, 1992) (Fig. S3 in S.I). c Average molar mass (Mw) afforded from viscometer measurements, see section 2.1. d Average molar mass (Mw) afforded from SEC analyses (Fig. S2 in S.I). b
starting material was 20,500 g mol−1 as measured by viscometry. However, this average molar mass afforded via viscometry should not be directly compared with those afforded via SEC for the CMC ones (Table 1). Apart from the differences between the respective chemical structures, which yield different hydrodynamic radii for the materials in solution, the solvents used in the viscometry and SEC investigations had to be different in our study due to solubility issues (see Experimental 2.4.3). Different solvents might yield distinct hydrodynamic radii even for materials with virtually identical macromolecular structures. Furthermore, the CMC average molar masses obtained via SEC are values relative to Pullulan standards used to calibrate the SEC chromatograms. Congruently, only the values of average molar masses afforded for the CMC materials via SEC can be compared with each other and are discussed as following. SEC investigation of the CMC materials (Table 1 and Fig. S2 in S.I.) indicated that the larger the number of carboxymethyl groups on the cellulose chains the larger is the Mw of the CMC material. This is expected since substitution of the hydroxyl hydrogen nuclei with large carboxymethyl groups must increase cellulose Mw. Most importantly, those findings suggested that depolymerization of the cellulose chains was negligible under the conditions used for the carboxymethylation reactions. Therefore, the CMC materials could work as polyelectrolytes upon dissolution. X-ray analysis of the CMC materials revealed the expected reduction of cellulose crystallinity after the carboxymethylation reaction when compared to the MC starting material (Table 1 and Fig. S3 in S.I.). Random distribution of carboxymethyl side-groups along the cellulose chains reduces the efficiency of the macromolecular packing, encouraging crystallite disorganization. Those structural changes upon carboxymethylation significantly improved dissolution of the CMC materials in water. This high solubility is fundamental to our further investigation about using those polyelectrolytes as stabilizing agents for aqueous alumina suspensions (vide infra). All materials revealed usual cellulose thermal decomposition processes (Fig. S4 in S.I). However, those common thermal events took place at lower temperature ranges for the modified materials. The thermal data clearly showed that the higher the DS the lower is the temperature range in which cellulose backbone thermal decomposition occur. For example, thermolysis of the cellulose glycoside bonds to produce levoglucosan (Britto and Assis, 2007; Ciacco et al., 2010; Ramos et al., 2005; Scheirs et al., 2001), which yielded the largest mass losses on the dTG traces (Fig. S4 in S.I.), occurred between 325 and 400 °C (maximum at 374 °C) for MC, whereas that same thermal event happened between 270 and 350 °C for the CMC materials, with maximum at 322 °C, 308 °C and 305 °C for CMC07, CMC13 and CMC18, respectively. Carbonization of CMC07and CMC13 yielded very broad mass losses at T > 400 °C, whereas carbonization of CMC18 yielded a distinct band centered at 535 °C. CMC18 had the highest DS, therefore carbonization through decarbonylation and decarboxylation of the carboxymethyl groups were more evident for that CMC material when compared to the others.
3.2. CMC materials as stabilizing agents for aqueous alumina suspensions When dispersed in aqueous medium, alumina particles acquire surface charges due to acid/base interactions with water molecules. Accordingly, the alumina net surface charge is pH dependent. In the pH region of the IEP, mutual attractive electrostatic interactions among the fine dispersed alumina particles in conjunction with Brownian motion induce particle aggregation. Formation of large alumina aggregates with random size distribution dramatically increases the suspension viscosity, and ultimately leads to colloid collapse to yield sedimentation. Congruently, the pH value at which maximum viscosity is observed is an estimate of the IEP of the suspension. The IEP is different for each suspension; thereby IEP determination of suspensions is a critical factor in ceramic colloidal processing. Therefore, we first established the IEP for our pristine alumina suspension by viscometry (Fig. 2). The data unambiguously revealed a maximum viscosity value about pH 7.0, which was established as the IEP for the present alumina suspension. Zeta potential and average particle size distribution studies confirmed the IEP value as 7.0 for the alumina suspension investigated
Fig. 2. Pristine aqueous alumina suspension viscosity as function of pH. Suspension with 60% (w/w) of alumina concentration, spindle number 3, rotation speed 3 rpm, room temperature. cps = centipoise.
57
Industrial Crops & Products 107 (2017) 54–62
B.M. Cerrutti et al.
(vide infra). We next investigated the stabilizing effects of CMC addition on the suspension viscosity at critical pH 7.0 (Fig. 3). An ideal stabilizing agent should prevent particle aggregation in a broad pH region, mainly at the IEP, to afford suspension with low viscosity. As the data unfolded, addition of any of the three reported CMC materials steadily decreased the suspension viscosity until reaching a minimum value, which did not change upon further CMC addition. Those were clear evidences for suspension stabilization promoted by the CMC materials. For CMC07 and CMC13, the minimum viscosity values were reached after addition of 0.10% (w/w), whereas CMC18 required 0.20% (w/w) to do the same. It has been reported (Rödel et al., 2012) that addition of 0.10–0.20% (w/w) of sodium polyacrylates reduce the viscosity and produces high negative surface potentials in alumina suspensions at 60% concentration similar to those investigated in the present work. Therefore, the amounts of the CMC materials reported herein are in the range of the that of the benchmark and were established as optimum CMC dosages to achieve suspension stabilization under the conditions investigated. Those optimum amounts were used throughout the present investigation. Studies of viscosity as function of pH were carried out using the optimum amount established for each CMC material to verify their additive properties in acidic and alkaline pH regimes (Fig. 4). All three CMC polyelectrolytes were able to reduce suspension viscosity throughout the pH range investigated. CMC07 and CMC13 performed well and similarly in all pH regimes, whereas CMC18was a good additive at
acidic pH, but not as good as in the IEP and alkaline regions. A possible explanation for CMC18 poorer performance in pH > 5 is its higher DS when compared to the others. The pKa of the carboxymethyl groups is about 4.0; therefore, the appended carboxymethyl groups are deprotonated in the IEP and alkaline regions. The negatively charged carboxymethyl groups on the cellulose backbone bring about repulsive electrostatic forces within a cellulose chain, which prompt the macromolecule to adopt extended conformations in solution rather than the usual statistical random-coil ones. Extended conformations expose a larger number of COOe appended groups to the aqueous medium, allowing the same cellulose chain to interact with several alumina particles in suspension. Hence, bridging them and promoting aggregation. Furthermore, cellulose stretched chains adsorbed onto a single alumina particle should yield a thin electrosteric barrier, which might not be robust enough to preclude the particles to come closer together upon collisions. This effect is more dramatic in high concentration suspensions like those of the present work. If that is true for CMC18, depletion effects became important and desorption of the CMC macromolecules ensued, resulting in spontaneous formation of large alumina aggregates, mainly at the IEP region. Large aggregates in suspension mean higher viscosity. Zeta potential measurements as function of pH were carried out to investigate interactions between alumina particles and CMC macromolecules (Fig. 5). Pristine alumina suspension showed the usual zeta potential behavior, with high positive and negative values in acidic (pH < 6.0) and alkaline medium (pH ≥ 9.0), respectively, indicating the particles had significant net surface charges at those pH regions. Between pH 6.0–8.0, potentials were negligible for the pristine suspension, suggesting equilibrium of surface charges. Zeta potential value was equal to zero at pH 7.10, confirming the IEP for the pristine alumina suspension. Addition of the established amount of any of the present CMC materials led to significant reduction of the zeta potential between 3.0 < pH < 9.0, except for CMC13. In this pH range, the CMC materials bear non-negligible negative charges as the pKa of the carboxymethyl groups is about 4.0. Since the zeta technique only measures potentials at interfaces, the reduction of zeta potential values indicated adsorption of CMC macromolecules onto the surface of the alumina particles in the pH interval investigated. The CMC adsorption partially canceled out the positive surface charges on the pristine alumina particles, thereby reducing the zeta values in pH ≤ 8.0. Nonetheless, suspensions treated with CMC13 showed similar positive zeta potential values in the 3.0–5.0 pH range. Among the CMC materials investigated, CMC13 had the highest polydispersity (Table 1). This heterogeneous molecular size distribution should yield appended carboxymethyl groups with nonuniform pKa values. In proteins, the close proximity of buried carboxylate groups is known to dramatically increase their pKa values due to dipole-dipole and
Fig. 4. Viscosity of pristine and CMC treated alumina suspensions as function of pH. Total suspension mass = 100 g. Alumina mass = 60 g (concentration = 60%, w/w). Conditions: spindle 3, rotation speed 3 rpm, room temperature. cps = centipoise.
Fig. 5. Zeta potential as function of pH for the pristine suspension as well as suspensions treated with CMCs. Conditions: alumina concentration = 2% (w/w), room temperature. mV = millivolts.
Fig. 3. Aqueous alumina suspension viscosity as function of CMC added masses (%, w/w). Total suspension mass = 100 g. Alumina mass = 60 g (concentration = 60%, w/w). Conditions: spindle 3, rotation speed 3 rpm, room temperature. cps = centipoise.
58
Industrial Crops & Products 107 (2017) 54–62
B.M. Cerrutti et al.
Fig. 6. Average particle size distribution of pristine and CMC treated alumina suspensions at pH 10. Conditions: alumina concentration = 2% (w/w), room temperature. a.u. = arbitrary unit.
Fig. 7. Average particle size as function of time for pristine and CMC treated suspensions at pH 7.0. Conditions: alumina concentration = 2% (w/w), room temperature.
the stabilizing effects promoted by addition of CMC to the colloids (Figs. 8–10 ). In the three pH values investigated, the pristine suspension yielded particles with irregular size and shape, which induced poor packing of particles upon sedimentation. Indeed, a large number of voids could be seen on the images afforded from the pristine suspension. On the other hand, CMC treated colloids amended the heterogeneous size and shape distribution of the alumina particles with variable degree of success, ultimately yielding solids with a somewhat better packing. By SEM analyses (Figs. 8–10), CMC13 was the best additive when compared to the other two CMC materials. CMC13 images displayed particles with better size and shape distribution, which allow them to pack up smoothly to yield solids with evident lower porosity. It should be noted however that the SEM images were afforded from dry materials, i.e. after sedimentation of the suspension, followed by filtration and vacuum drying of the resulting materials. Therefore, the morphological aspects revealed by SEM should be taken as indirect evidences of the real state of the suspensions.
hydrogen bonding interactions (Lau et al., 2010). In the present case, the cellulose chains with larger molecular sizes in CMC13 should adopt folded conformations with some carboxymethyl groups buried in their interior. The pKa values of the buried groups in CMC13 should be larger than 4 and are therefore in the neutral form at the 3.0-5.0 pH range. Congruently, suspensions treated with CMC13 yielded positive zeta-potentials at those pH regimes. For pH > 8.0, the zeta potentials for pristine and stabilized suspensions were virtually the same. This finding would suggest no adsorption of CMC onto alumina particles at pH > 8.0. However, viscosity measurements as function of pH (Fig. 4) for the CMC stabilized suspensions were lower compared to that of the pristine colloid at pH > 8.0, informing that adsorption had occurred. To settle the dispute between those two opposing hypothesis, average particle size distribution using the FOQELS technique were carried out. Fig. 6 shows the average particle size distribution afforded for the pristine and CMC stabilized suspensions at pH 10. The data clearly revealed that the CMC treated colloids have smaller particles in suspension than the pristine one, suggesting colloid stabilization that could only occur in alkaline medium through CMC adsorption onto the particles. Although attractive electrostatic interactions between the biopolyelectrolyte and alumina particles are unlike at pH 10.0, CMC carboxylate groups might engage in Lewis acid/base interactions with aluminum atoms on the particle surface to produce thermodynamically stable adducts that drive adsorption in alkaline medium. Furthermore, deprotonated hydroxyl groups on the particle surface can easily form hydrogen bonds with the hydroxyl groups present on the CMC cellulose chains to provide extra driving force for the adsorption process. Those interactions might promote CMC adsorption onto alumina particles in suspension in the alkaline regime. CMC18 had the poorer performance among the CMC additives, yielding colloids with relatively larger particles in suspension if compared to the other two CMCs. This large average particle size distribution observed for CMC18 nicely corroborated our hypothesis that CMC materials with higher DS promote particle aggregation in alkaline medium most likely through “bridging” and/or “depletion effects” (vide supra). Experiments of average particle size as function of time were carried out at the critical IEP region (pH 7.0) to gather information about aggregation and sedimentation kinetics (Fig. 7). Pristine alumina particles in suspension promptly formed large aggregates upon Brownian collision, which reached the maximum particle size detection limit of our equipment (4000 nm) in about 20 min. Conversely, suspension containing particles decorated with CMC macromolecules had smaller and near constant size (400–700 nm) in the time range investigated (80 min), suggesting the CMC polyelectrolytes indeed challenged particle aggregation in the IEP region. Scanning electron microscopy of the alumina suspension after sedimentation followed by filtration and vacuum drying highlighted
4. Conclusions Introduction of carboxymethyl groups onto cellulosic chains decreases crystallinity and thermal stability when compared to pristine cellulose. Reduction of crystallinity improves solubility, which favors CMC materials to act as bio-based polyelectrolytes in ceramic colloidal processing. Thermolysis at lower temperatures is also advantageous as thermal decomposition of the CMC additives before sintering of the ceramic object will require less energy to take place. CMC materials are able to stabilize aqueous alumina colloids with high solid content in a wide pH range. The carboxymethyl groups appended to the cellulose chains can engage in hydrogen bonding formation, electrostatic and Lewis acid/base interactions with the hydroxyl groups and aluminum atoms on the surface of alumina particles in suspension. Those interactions promote strong adsorption of polyelectrolyte species onto the particles that create electrosteric barriers at the solid/liquid interface that prevent aggregation. As a result, suspension with small particle size and low viscosity are afforded upon addition of 0.1–0.2% (w/w) of CMC. However, the degree of substitution (DS) is a crucial factor in the preparation of CMC polyelectrolytes. Introduction of too many carboxymethyl groups onto the cellulosic backbone leads to materials with poorer performance as stabilizing agents. Our investigation reveals that DS values as high as 1.8 induce some particle aggregation if compared to materials with lower ones. On the other hand, CMC materials with DS of about 1.3 exhibit excellent properties as additives, opening up the possibility of using those materials as biorenewable stabilizing agents in alumina colloidal processing. Thus, careful control of the carboxymethylation reaction conditions is fundamental to afford CMC additives with the right DS and consistent dispersing properties. 59
Industrial Crops & Products 107 (2017) 54–62
B.M. Cerrutti et al.
Fig. 8. SEM images of the pristine and CMC treated alumina suspensions at pH 3.0 after sedimentation, filtration and drying.
Fig. 9. SEM images of the pristine and CMC treated alumina suspensions at pH 7.0 after sedimentation, filtration and drying.
suspensions treated with CMCh and CML under similar conditions. The CMC better performance is probably due to its lower average molecular mass, which favors its adsorption onto the surface of single alumina particles in suspension, which in turn leads to a better stabilization of the colloids.
In our previous works, we investigated carboxymethylchitosan (CMCh) (Cerrutti et al., 2013), and carboxymethyllignins (CML) (Cerrutti et al., 2012) as stabilizing agents for aqueous alumina suspensions. Comparing those materials with the present ones, it is possible to conclude that the CMCs reported in the present investigation perform the best, especially CMC13. Suspensions treated with CMCs show lower viscosity values, higher negative zeta potentials and small particle sizes in a broad pH range when compared to the alumina 60
Industrial Crops & Products 107 (2017) 54–62
B.M. Cerrutti et al.
Fig. 10. SEM images of the pristine and CMC treated alumina suspensions at pH 10.0 after sedimentation, filtration and drying. Crops. Prod. 83, 346–352. Fischer, H., Mackwitz, H., 2016. Renewable raw material for chemical technical products, conversion of biomass in the non-energy fields. http://www.sinanco.com/biomassarticle1.html. (Accessed 2 June 2016). Gronski, W., Hellmann, G., 1987. NMR-spektroskopische charakterisierung von carboxymethylcellulose. Papier (Darmstadt) 41, 668–672. Heinze, T., Koschella, A., 2005. Carboxymethyl ethers of cellulose and starch. A review. Macromol. Symp. 223, 13–39. Heinze, T., Liebert, T., Klufers, P., Meister, F., 1999. Carboxymethylation of cellulose in unconvencional média. Cellulose 6, 153–165. Kalliola, A., Vehmas, T., Liitia, T., Tamminen, T., 2015. Alkali-O2 oxidized lignin: a biobased concrete plasticizer. Ind. Crops Prod. 74, 150–157. Kutsenko, L.I., Ivanova, N.P., Karetnikova, E.B., Bobasheva, A.S., Bochek, A.M., Panarin, E.F., 2002. Characteristics of aqueous solutions methyl-cellulose and polymethacrylamidoglusose mixtures. Russ. J. Appl. Chem. 75, 314–318. Lau, W.L., DeGrado, W.F., Roder, H., 2010. The effects of pKa tuning on the thermodynamics and kinetics of folding: design of a solvent-shielded carboxylate pair at the a-position of a coiled-coil. Biophys. J. 99, 2299–2308. Megiatto Jr., J.D., Oliveira, F.B., Rosa, D.S., Castellan, A., Frollini, E., 2007a. Renewable resources as reinforcement of polymeric matrices: composites based on phenolic thermosets and chemically modified sisal fibers. Macromol. Biosci. 7, 1121–1131. Megiatto Jr., J.D., Hoareau, W., Gardrat, C., Frollini, E., Castellan, A., 2007b. Sisal fibers: surface chemical modification using reagent obtained from a renewable source; characterization of hemicellulose and lignin as model study. J. Agric. Food Chem. 55, 8576–8584. Megiatto Jr., J.D., Silva, C.G., Rosa, D.S., Frollini, E., 2008. Sisal chemically modified with lignins: correlation between fibers and phenolic composites properties. Polym. Degrad. Stab. 93, 1109–1121. Megiatto Jr., J.D., Silva, C.G., Ramires, E.C., Frollini, E., 2009. Thermoset matrix reinforced with sisal fibers: effect of the cure cycle on the properties of the biobased composite. Polym. Test. 28, 793–800. Megiatto Jr., J.D., Ramires, E.C., Frollini, E., 2010. Phenolic matrices and sisal fibers modified with hydroxy terminated polybutadiene rubber: impact strength, water absorption, and morphological aspects of thermosets and composites. Ind. Crop. Prod. 31, 178–184. Megiatto Jr., J.D., Cerrutti, B.M., Frollini, E., 2016. Sodium lignosulfonate as a renewable stabilizing agent for aqueous alumina suspensions. Int. J. Biol. Macromol. 82, 927–932. Ohtsuka, H., Mizutani, H., Iio, S., Asai, K., Kiguchi, T., Satone, H., Mori, T., Tsubaki, J., 2011. Effects of sintering additives on dispersion properties of Al2O3 slurry containing polyacrylic acid dispersant. J. Eur. Ceram. Soc. 31, 517–522. Oliveira, F., Ramires, E.C.A., Frollini, E., Belgacem, M.N., 2012. Lignopolyurethanic materials based on oxypropylated sodium lignosulfonate and castor oil blends. Ind. Crop. Prod. 72, 77–86. Pires, P.A.R., Teixeira, T.C., Bioni, T.A., Nawaz, H., El Seoud, O.A., 2015. Imidazolecatalyzed esterification of cellulose in ionic liquid/molecular solvents: a multitechnique approach to probe effects of the medium. Ind. Crops. Prod. 77, 180–189. Rödel, C., Müller, M., Glorius, M., Potthoff, A., Michaelis, A., 2012. Effect of varied powder processing routes on the stabilizing performance and coordination type of
Acknowledgements The authors are grateful to FAPESP (The State of São Paulo Research Foundation, Brazil) and to CNPq (National Council for Scientific and Technological Development, Brazil) for financial support. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.indcrop.2017.05.029. References Barba, C., Montané, D., Rinaudo, M., Farriol, X., 2002. Synthesis and characterizations of carboxymethylcellulose (CMC) from nom-wood fibers I: Accessibility of cellulose fibers and CMC synthesis. Cellulose 9, 319–326. Bendahou, D., Bendahou, A., Seantier, B., Grohens, Y., Kaddami, H., 2015. Nanofibrillatedcellulose-zeolites based new hybrid composites aerogels with super thermal insulating properties. Ind. Crop. Prod. 65, 374–382. Biswal, D.R., 2004. Characterization of carboxymethyl cellulose and polyacrylamide graft copolymer. Carbohyd. Polym. 57, 379–387. Botaro, V.R., Siqueira, G., Megiatto Jr., J.D., Frollini, E., 2010. Sisal fibers treated with NaOH and benzophenonetetracarboxylicdianhydride as reinforcement of phenolic matrix. J. Appl. Polym. Sci. 115, 269–276. Britto, D., Assis, O.B.G., 2007. Synthesis and mechanical properties of quaternary salts of chitosan-based films for food applications. Int. J. Biol. Macromol. 41, 198–203. Brookhaven Instruments Corporation, 1999. Instruction Manual for ZetaPALS, Zeta Potential Analyzer. Brookhaven Instruments Corporation, New York pp. 23. Buschle-Diller, G., Zeronian, S.H., 1992. Enhancing the reactivity and strength of cotton fibers. J. Appl. Polym. Sci. 45, 967–979. Cerrutti, B.M., Lamas, J.C., Britto, D., Campana-Filho, S.P., Frollini, E., 2010. Derivatives of biomacromolecules as stabilizers of aqueous alumina suspensions. J. Appl. Polym. Sci. 117, 58–66. Cerrutti, B.M., Souza, C.S., Castellan, A., Ruggiero, R., Frollini, E., 2012. Carboxymethyl lignin as stabilizing agent in aqueous ceramic suspensions. Ind. Crop. Prod. 36, 108–115. Cerrutti, B.M., Lamas, J.C., Campana-Filho, S.P., Frollini, E., 2013. Carboxymethyl chitosan: preparation and use in colloidal ceramic processing. J. Polym. Environ. 21, 816–825. Ciacco, G.T., Morgado, D.L., Frollini, E., Possidonio, S., Seoud El, O.A., 2010. Some aspects of acetylation of untreated and mercerized sisal cellulose. J. Braz. Chem. Soc. 21, 71–77. Cui, S., Zhang, S., Xiong, L., Sun, Q., 2016. Green preparation and characterization of size controlled nanocrystalline cellulose via ultrasonic-assisted enzymatic hydrolysis. Ind.
61
Industrial Crops & Products 107 (2017) 54–62
B.M. Cerrutti et al.
application. Ind. Crops. Prod. 72, 133–141. Scheirs, J., Camino, G., Tumiati, W., 2001. Overview of water evolution during the thermal degradation of cellulose. Eur. Polym. J. 37, 933–942. Trindade, W.G., Hoareau, W., Megiatto Jr., J.D., Razera, I.A.T., Castellan, A., Frollini, E., 2005. Thermoset phenolic matrices reinforced with unmodified and surface-grafted furfuryl alcohol sugar cane bagasse and curaua fibers: properties of fibers and composites. Biomacromolecules 6, 2485–2496. Vashney, V.K., Gupta, P.K., Naithani, S., Khullar, R., Bhatt, A., Soni, P.L., 2006. Carboxymethylationof α-cellulose isolated from Lantana camara with respect to degree of substitution and rheological behavior. Carbohyd. Polym. 63, 40–45. Wisniewska, M., Urban, T., Grza̧dka, E., Zarko, V.I., Gunko, V.M., 2014. Comparison of adsorption affinity of polyacrylic acid for surfaces of mixed silica-alumina. Colloid. Polym. Sci. 292, 699–705. Wu, Q., Li, W., Wu, Y., Zong, G., Liu, S., 2015. Effect of reaction time on structure of ordered mesoporous carbon microspheres prepared from carboxymethyl cellulose by soft-template method. Ind. Crops. Prod. 76, 866–872.
polyacrylate in alumina suspensions. J. Eur. Ceram. Soc. 32, 363–370. Ramires, E.C., Megiatto Jr., J.D., Gardrat, C., Castellan, A., Frollini, E., 2010a. Valorization of an industrial organosolv-sugarcane bagasse lignin: characterization and use as a matrix in biobased composites reinforced with sisal fibers. Biotechnol. Bioeng. 107, 612–621. Ramires, E.C., Megiatto Jr., J.D., Gardrat, C., Castellan, A., Frollini, E., 2010b. Biobased composites from glyoxal-phenol matrices reinforced with microcrystalline cellulose. Polimeros 20, 126–133. Ramires, E.C., Megiatto Jr., J.D., Gardrat, C., Castellan, A., Frollini, E., 2010c. Biobased composites from glyoxal-phenolic resins and sisal fibers. Bioresour. Technol. 101, 1998–2006. Ramos, L.A., Frollini, E., Heinze, T.H., 2005. Carboxymethylation of cellulose in the new solvent dimethyl sulfoxide/tetrabutylammonium fluoride. Carbohyd. Polym. 60, 259–267. Rudhziah, S., Rani, M.S.A., Ahmad, A., Mohamed, N.S., Kaddami, H., 2015. Potential of blend of kappacarrageenan and cellulose derivatives for green polymer electrolyte
62