Development of self-assembled nanocrystalline cellulose as a promising practical adsorbent for methylene blue removal

Development of self-assembled nanocrystalline cellulose as a promising practical adsorbent for methylene blue removal

Carbohydrate Polymers 199 (2018) 92–101 Contents lists available at ScienceDirect Carbohydrate Polymers journal homepage:

1MB Sizes 0 Downloads 8 Views

Carbohydrate Polymers 199 (2018) 92–101

Contents lists available at ScienceDirect

Carbohydrate Polymers journal homepage:

Development of self-assembled nanocrystalline cellulose as a promising practical adsorbent for methylene blue removal


Kok Bing Tana,b, Alavy Kifait Rezaa,e, Ahmad Zuhairi Abdullahc, Bahman Amini Horria,d, ⁎ Babak Salamatiniaa, a

Chemical Engineering Discipline, School of Engineering, Monash University Malaysia, Jalan Lagoon Selatan, 47500, Bandar Sunway, Selangor, Malaysia Centre of Pre-U Studies, HELP College of Arts and Technology, Jalan Metro Pudu, Fraser Business Park, 55100, Kuala Lumpur, Malaysia c School of Chemical Engineering, Universiti Sains Malaysia, 14300, Nibong Tebal, Penang, Malaysia d Department of Chemical & Process Engineering, Faculty of Engineering and Physical Sciences, University of Surrey, Surrey, GU2 7XH, UK e Research and Development Cell, BangladeshKnitwear Manufacturers & Exporters Association (BKMEA), 13/A Sonargaon Road, Banglamotor, Dhaka, 1000, Bangladesh b



Keyword: Nano crystalline cellulose Self assembly Promising green adsorbent Separation Adsorption

This study is focused on nanocrystalline cellulose (NCC) flakes for methylene blue (MB) removal via adsorption. NCC flakes exhibit a high adsorption capacity (188.7 mg/g fixed at 0.7 g/L adsorbent dosage, 25 °C and pH 6) compared to other nanomaterials, such as carbon nanotube and other cellulosic materials, such as coffee husks. Unlike NCC powder, it was observed that NCC flakes can be easily separated from wastewater containing MB. Further adsorption studies were conducted on NCC flakes, and it was found that 0.7 g/L was the optimum adsorbent dosage, which fitted well with the Langmuir Isotherm. The mean free energy value from DubininRadushkevich isotherm was less than 8 kJ/mol. ΔGo values at different temperatures were within the -20 kJ/mol to 0 kJ/mol range. In conclusion, NCC flakes is a promising and practical ‘green’ nanomaterial that can be further developed for industrial applications.

1. Introduction The textile industry has been a major contributor to the world economy. However, pollution level from dyes is rising and sustainability concerns are being addressed at each point of the supply chain. Most dyeing factories have a high polluting footprint, and it is estimated that up to 200,000 tons of dyes escape in effluents that are released into water bodies due to inefficient wastewater treatment process (Chequer et al., 2013). Methylene blue (MB) is a common cationic dye that is widely used to dye textiles, such as cotton, cellulose, wood, and silk (Rafatullah, Sulaiman, Hashim, & Ahmad, 2010). Most dyes in the industry are potentially injurious to human health and MB was reported to cause chronic toxicity (Gillman, 2006), particularly to central nervous system (Vutskits et al., 2008). Hence, there is an urgent need to remove these pollutants for a healthier environment (Alswat, Ahmad, & Saleh, 2016). Adsorption is reported to be the process of choice for dye treatment as it is inexpensive, simple, and easy to be operated (Saleh, Sarı, & Tuzen, 2017; Sani et al., 2017; Vakili et al., 2017). Industrial processes commonly use activated carbon as an adsorbent due to its high porosity and surface area (Matsis & Grigoropoulou,

2008). Commercial activated carbon is relatively expensive due to its high production cost as regeneration requires high pressure steam (Aksu, Açıkel, Kabasakal, & Tezer, 2002). Recently, interest in the application of nanomaterials as adsorbents is growing (Alansi, Alkayali, Al-qunaibit, Qahtan, & Saleh, 2015). The future development and commercialization of nanomaterial-based adsorbents for dye removal are wrought with challenges, particularly on environmental concern as nanomaterials, such as carbon nanotube, titania nanotube, and nano zerovalent iron are toxic (Tan et al., 2015). Cellulose is the most abundant polysaccharide in the world, which has been proven to be non-toxic, low cost, sustainable, and renewable (Ali, Rachman, & Saleh, 2017). By exposing cellulose to acid, particularly sulfuric acid, its amorphous region would be selectively degraded, leaving the rod-shape crystalline region of the nano-size cellulose intact (Zhou & Wu, 2012). The nanocrystalline cellulose (NCC) is an anionic nano-size cellulose that is unlike regular cellulose. It is reported to be tougher than steel (Mitchell, 2004), and more effective towards removal of cationic dyes, such as MB dye. As an adsorbent, NCC is relatively cheaper compared to other nanomaterials, such as carbon nanotube (Huang & Rodrigue, 2015), and more environmentally friendly (Peng, Dhar, Liu, & Tam, 2011). Hence, environmentally friendly, and

Corresponding author. E-mail address: [email protected] (B. Salamatinia). Received 4 January 2018; Received in revised form 19 June 2018; Accepted 3 July 2018 Available online 05 July 2018 0144-8617/ © 2018 Elsevier Ltd. All rights reserved.

Carbohydrate Polymers 199 (2018) 92–101

K.B. Tan et al.

2.3. Dye uptake experiment

effective NCC-based adsorbents for dye removal could be developed as an alternative to activated carbon. Even with this advantage, all nanomaterials share a similar drawback. Nanomaterials will form either stable suspension or dispersion in treated wastewater, making separation a major challenge. NCC adsorbent could not be separated from treated water, despite being exposed to vigorous centrifugation. The use of calcium chloride (CaCl2) is reported in the separation of NCC adsorbent from MB water, which will cause aqua-toxicity in excessive amount (Mallick, Mohapatra, & Sarangi, 2014). This drawback proves the necessity of developing an alternative approach. Thermal treatment as a self-assembly method for NCC could overcome the separation problem by taking advantage of its anisotropic property (Peng et al., 2011). While self-assembly of NCC is widely used in optical application and protein immobilization (Peng et al., 2011), to the authors’ knowledge, there are no reported studies on thermal treated NCC for adsorption application. In addition, the effect of heat on self-assemble curing of NCC on dye uptake has not been reported. Efficient production of selfassembled NCC film can be potentially increase by reducing the curing time. There is a knowledge gap on the changes that occur under thermal conditions for intensified self-assembled NCC flakes. The properties of the flake could affect the performance of the material, in particular for water treatment applications. This highlights the necessity of characterizing the NCC films to assume unchanged properties. Therefore, in this work, NCC powder had self-assembled into NCC flakes via the evaporation method, whereby curing conditions, namely, curing temperature and curing time during the self-assembly process, were varied to study their effects on MB uptake. The NCC flakes were characterized using several techniques. These include Fourier Emission Scanning Electron Microscope (FE-SEM) for surface morphology, Fourier Transform Infrared Spectroscopy (FTIR) for chemical functional group, mercury porosimeter for surface area, and zeta potential for surface charge at different pHs. In addition, the adsorption isotherm and thermodynamics of the removal of MB could provide insights on the adsorption mechanism of the self-assembled NCC adsorbent. A performance comparison with other adsorbents was also conducted to determine any changes to the adsorption mechanism by the self-assembled NCC.

For the adsorption studies, 0.035 g of NCC flakes was added into 50 mL of 100 ppm (parts per million) MB solution. The MB solution was shaken using an orbital shaker at 300 rpm and 1 mL samples were drawn at 0.5 min intervals until no changes were observed in the removal percentage. The treated dye solution was then centrifuged at 6000 rpm using a micro-centrifuge (Vitaris ScansSeed Mini Personal Micro-centrifuge, Switzerland) for 1 min to separate the NCC flakes from the dye wastewater. The final absorbance number was measured at 663 nm using a UV–vis spectrophotometer (Thermo Genesys 10UV, USA), followed by obtaining the final dye concentration via a calibration curve prepared beforehand. Dye uptake was then calculated using the following equation:

Qe = [(Co − C f ) V]/ m


where, Co represents the initial dye concentration, Cf represents the final dye concentration, Qe represents dye uptake in mg/g, V represents the volume of the dye solution, and m represents the mass of NCC flakes. All data points were repeated 3 times to obtain average values. Similar procedure and protocol for measurements of dye concentration have also been reported by Dil et al. (2017). To compare the behavior of NCC powder in MB wastewater with the self-assembled NCC flakes, adsorption experiment was also conducted. However, instead of curing, 0.292 mL of the NCC dispersion was directly added into 50 mL of 100 ppm MB solution, as proposed by Batmaz et al. (2014) and was shaken at 300 rpm using an orbital shaker. Upon equilibrium, the solution was added into a centrifuge tube and 0.44 mL of CaCl2 (0.09 M) solution was added to separate NCC from the dye. The mixture was centrifuged using the Thermo-fisher Heraeus Multifuge X3R (USA) machine for 2 min at 6000 rpm. 2.3.1. Isotherm studies and thermodynamics Analysis of adsorption isotherm data for different materials is essential to predict the adsorption mechanism. To describe the adsorption isotherm, four widely used models were applied, namely, the linear forms of the Langmuir, Freundlich, Temkin, and Dubinin-Radushkevich (Chang, Lai, & Lee, 2016). Thermodynamics studies are essential to study the spontaneity of the process with temperature, by taking into consideration both energy and entropy. This can be performed using the Van Hoff’s equation (Tan, Abdullah, Horri, & Salamatinia, 2016).

2. Materials and methods 2.1. Materials

2.3.2. Effects of adsorbent dosage, initial dye concentration, and dye pH The adsorbent dosage was varied between 0.1 and 1.3 g/L, while the initial dye concentration was varied from 10 to 1000 ppm. The pH of the dye was adjusted using 0.01 M of nitric acid and sodium hydroxide, in the range of pH 2 to pH 10. Similar procedure and protocol were also reported by Asfaram et al. (2016).

NCC powder (sulfur content 0.89 wt%) was supplied by University of Maine, USA. MB powder (98% purity) was supplied by SigmaAldrich, while nitric acid and sodium hydroxide were supplied by Merck, Germany. Calcium chloride (CaCl2) was supplied by R&M Chemicals, Malaysia. All chemicals were of analytical grade.

2.4. Characterization 2.2. Preparation of adsorbent The morphology of the NCC flakes was observed using FE-SEM (Hitachi SU8000, Japan) at an accelerating voltage of 2–5.7 kV. The samples were coated with platinum using a coater (Quarum Q150R S) prior to conducting the analysis. Energy-dispersive X-ray (EDX) spectroscopy was also conducted for the elemental analysis of the NCC flakes. The sample was scanned for 3 times at the respective location, and the best results have been chosen for the discussion. The chemical functional groups of the samples were analyzed using an FTIR Spectroscope (Thermo Scientific Nicolet IS10, USA), at the range of 4000 cm−1 to 800 cm−1. (Recoreded at 0.4 cm-1 resolution and averaged by 64 scans) under transmittance mode using ATR technique. X-ray diffraction (XRD) analysis was done using XRD diffractometer (D8 Discover from Bruker, Germany) to understand the crystallinity NCC flakes. The analysis was run at 40 kV and 4 mA with a copper tube as the radiation source (λ = 1.54060 Å). Scan range of 5–50 with a step

Initially, 4.8 g of pure NCC powder was added into 40 mL of deionized water. The NCC mixture was sonicated using a sonicator (Qsonica, USA) for 5 min, with 3 s of pulse-on, and 1 s of pulse-off at 60% amplitude in an ice bath to avoid any rise of temperature by means of ultrasonic. The NCC dispersion was poured into 50 mL centrifuge tubes and was centrifuged at 6000 rpm (revolution per minute) using a centrifuge (Thermofisher Heraeus Multifuge X3R, USA) until all bubbles were removed. Then, 40 mL of NCC dispersion was poured into standard size glass petri dishes (9.5 cm in diameter, 3.8 cm in height). The samples were then cured at varying ranges of temperature (50–90 °C) and time (5–25 h). The resulting NCC films were pealed from the petri dishes and grounded using a blender (Panasonic MX GM-1011, Japan) for 10 min to form NCC flakes. The NCC flakes were stored in an incubator prior to adsorption studies. 93

Carbohydrate Polymers 199 (2018) 92–101

K.B. Tan et al.

Fig. 1. FE-SEM Images of NCC Flakes Cured at: (a) 50 °C; and (b) 90 °C, Fixed at Curing Time of 5 h and (c) EDX spectra of the Sample Cured at 50 °C.

size of 0.02 and rotation of 15 min−1. Surface charges or zeta potentials of the NCC flakes in a pH range of 2–10 were measured using a zetasizer (Malvern Zetasizer Series, UK). Mercury porosimeter (Thermofisher Pascal, USA) was used to investigate changes in the pore size with varied curing conditions. Tests using the Pascal Series Mercury Intrusion Porosimeters (Thermo Scientific-Pascal 140, USA) followed a two-step process; first, the samples were filled with mercury in, whereby pressure was increased from vacuum to 400 kPa at a ramping rate of 4 kPa per minute. Next, the samples were moved to Pascal 440, whereby pressure was increased from 400 kPa to 400 MPa, with a ramping rate of 4 MPa per minute.

assembly of NCC molecules. As a result, the NCC macrostructure could be separated from the solution much easier (Mancell-Egala et al., 2017). The resulting structure of the NCC flake is observed in Fig. 1(a). Fig. 1(b) shows FE-SEM image of NCC flakes cured at higher temperature of 90 °C for 5 h. Denser structures were observed compared to that was shown in Fig. 1(a). The large difference between bulk temperature and surface temperature had resulted in higher convection, thus causing larger pressure difference between the outside and the inside of the capillary bridge. Thus, stronger capillary force and stronger hydrodynamics force pushed the NCC molecules closer to each other. This promotes a sintering effect between adjacent NCC molecules, which explains the formation of denser-packed structures. Similarly, Pivetta, Pacchioni, Fernandes, and Brune (2015) showed that selfassembly between adjacent dicarbonitrile-pentaphenyl molecules became denser when the temperature was increased. In addition, (Landsmann, Luka, & Polarz, 2012) found that the assembly of polyoxometalate on membrane became compact when the assembling temperature was increased. Fig. 1(c) shows EDX analysis, for the sample treated at 50 °C. Carbon and oxygen are predominantly present in NCC, which is verified from the chemical structure of. Platinum peak is also observed due to the platinum coating during the preparation process for EDX and FE-SEM. In addition, sulfur peak noted is due to formation of sulfate group on NCC upon sulfuric acid hydrolysis.

3. Results and discussion 3.1. Surface morphology Fig. 1(a) shows the FE-SEM image of NCC flakes cured at 50 °C for 5 h. Unlike the distinct individual rod-shaped NCC powder, as presented by other researchers (Bondeson, Mathew, & Oksman, 2006; Pranger & Tannenbaum, 2008), rod-shaped NCC molecules that were stacked together are observed as a result of the self-assembly of NCC. During this process, evaporation of water from NCC dispersion occurred. Volume fraction of the particles increased, while liquid level decreased due to water evaporation caused by the differences in the bulk and surface temperatures (Fleck, McMeeking, & Kraus, 2015). Simultaneously, convective hydrodynamic flux occurred, which lead to hydrodynamic forces (Kralchevsky & Denkov, 2001). Meanwhile, a capillary bridge was formed between the two adjacent NCC molecules as a result of pressure differences between the inside and outside surface of the bridge (Thorkelsson, Bai, & Xu, 2015). Convection during evaporation gave rise to the larger pressure differences between the inside and the outside of the bridge (Tang et al., 2015). The electrostatic repulsion between the NCC molecules was overcome by the combined effect of hydrodynamic and capillary forces, leading to the self-

3.2. FTIR Fig. 2(a) shows the FTIR spectra of: (i) NCC Powder; (ii) NCC flakes cured at 50 °C; and (iii) NCC flakes cured at 90 °C, for 5 h. The peak at 1150 cm−1 represents the covalent band of sulfate (SO32-) group, which exists due to sulfuric acid hydrolysis reaction during the production of NCC forming sulfate (SO32-) group on some of the NCC monomers (Habibi, Lucia, & Rojas, 2010). The sulfur element from the sulfate group is also verified by EDX results. Fig. 2(ai) shows the spectrum of NCC powder, where −OH peak can be observed at 3325 cm−1. Self94

Carbohydrate Polymers 199 (2018) 92–101

K.B. Tan et al.

attributed to the lesser amount of sulfate groups compared to hydroxyl groups. Based on the data given by the supplier, only 0.89 wt% of the NCC polymer consists of sulfur. Therefore, the shift in peak would be insignificant. The self-assembly during a larger difference in bulk temperature and surface temperature had resulted in a higher convection, producing stronger capillary force that caused adjacent NCC molecules to be packed closer together. As they got closer, stronger interactions between adjacent NCC molecules created more hydrogen bonds between them. However, it could be seen that there was no significant changes within the functional groups present on the treated NCC, which confirms that the chemical properties of the NCC have remained unchanged under the thermal treating conditions. Fig. 2(b) shows the XRD peaks for (i) NCC flakes cured at 50 °C as well as (ii) the pure NCC powder. According to the results from the XRD peaks the initial NCC included two polymorphs of cellulose I and II (Peng et al., 2013). Looking into the general structure of both Fig. 2(bi) and (bii) it is observed that there are not significant changes in terms of the peak appearance. However there are some changes observed in the intensity of the peaks which correspond to the shift from cellulose Type I to Cellulose Type II via the thermal treatment. Two consecutive peaks can be seen at 2θ values at 14.6° and 16.6°, while another significant peak is observed in 22.4°. These peaks correspond to the −110, 110 and 200 crystallographic planes of monolithic cellulose type I, respectively (Gaspar et al., 2014). Those peaks at 2θ = 12.1°, 19.8°, and 22.0° for (−110), (110), and (200) planes are assigned to cellulose II crystal (Qi, Cai, Zhang, & Kuga, 2009). It is worth noting that crystalline planes of 200 (at 2θ = 22.6) and 004 (at 2θ = 34.5) were noticed in both types of celluloses at the same location of the XRD spectra. The diffraction signals of crystalline planes of from cellulose I (200 and 004) and cellulose II (020 and 004) in the NCC are observed because of the sulfuric acid treatment during the manufacturing processes which was responsible for the generation of cellulose II (Peng et al., 2013). The presence of Sulfur was also confirmed by the EDX results. The thermal treatment did not affect the crystallinity of the NCC and the change was found to be from 62.5% for NCC powder to 64.4% in NCC flakes. Hence, it can be concluded that there is no change in chemical compound as the NCC powder was self-assembled into NCC flakes. Fig. 2(c) shows the chemical interaction between NCC molecules upon self-assembly. There are positive and negative dipole moments on the hydrogen and oxygen molecules, respectively, attached to the hydroxyl group (Waigh, 2014). Hence, hydrogen in the hydroxyl group could form hydrogen bond with oxygen in the hydroxyl group of other NCC molecules, and vice-versa (Ek, Gellerstedt, & Gunnar, 2009). In addition, hydrogen attached to the hydroxyl group is also able to form hydrogen bond with double-bonded oxygen attached to the sulfate group (Daintith, 2004) of adjacent NCC molecules. In addition, hydrogen is also able to form ion-dipole interaction with oxygen ion attached to the sulfate group of other NCC molecules (Lemke, Williams, Roche, & Zito, 2013). However, since the amount of sulfate relative to the hydroxyl group is negligible, the chemical interaction in the selfassembly process is dominated by the hydrogen bonds between adjacent hydroxyl groups.

Fig. 2. (a) FTIR Spectra of: (i) NCC Powder; (ii) NCC Flakes treated at 50 °C for 5 h; (iii) NCC Flakes Treated at 90 °C for 5 h; and (b) XRD Pattern of (i) NCC Flakes Cured at 50 °C and (ii) NCC powder and (c) Chemical Interaction between Self-assembled NCC.

assembled NCC flakes cured at 50 °C showed−OH peak located at a higher wavenumber of 3357 cm−1 on its spectrum, as shown in Fig. 2(aii). This indicates that hydrogen bonding had occurred between the −OH functional groups of adjacent NCC molecules during self-assembly process (Suratago et al., 2015). Fig. 2 (aiii) shows −OH peak at 3375 cm−1 for NCC flakes cured at 90 °C. Meanwhile, the bands at 2900 cm−1 represent the C–H group stretching vibrations (Asfaram et al., 2016), while the band at 892 cm−1 is attributed to the CeOeC stretching of β-1-4 glycosidic linkages (Jabli, Tka, Ramzi, & Saleh, 2018). Sulfates could also participate in the self-assembly of NCC via hydrogen bond with adjacent hydroxyl group (Hay, Dixon, Lumetta, Vargas, & Garza, 2004). However, there is no shift in peaks attributed to the sulfate groups observed in all the FTIR spectra. This can be

3.3. Effect of curing temperature and time on dye uptake Fig. 3 shows the effect of curing temperature on dye uptake. It was found that at any fixed curing time, the dye uptake decreased with increasing temperature. This trend can be explained based on the surface morphology as presented by the FE-SEM images in Fig. 1(a) and (b), the structure of the self-assembled NCC became denser when the curing temperature was increased, which subsequently decreases the pore width of the resulting NCC flakes. This was experimentally verified via the mercury porosimeter test, whereby at a fixed curing time, the pore width of the self-assembled NCC had decreased from 115 to 97 Å, when the curing temperature was increased from 50 °C to 90 °C. Subsequently, the surface area was found to decrease from 78 to 32 m2/g. 95

Carbohydrate Polymers 199 (2018) 92–101

K.B. Tan et al.

curing temperatures could have been too low for sintering to occur. Hence, the self-assembly was purely based on hydrogen bonding between adjacent NCC molecules, as induced by convection. Meanwhile, self-assembly was also driven by inter- and intra-molecular forces that drove the molecules into a stable low-energy state (Boeckl & Graham, 2006). As a result, the self-assembled NCC appeared as packed as possible until it reached a stable low-energy state. Therefore, it is postulated that by 5 h, NCC flakes cured at 50 and 60 °C have achieved stable low-energy state due to the relatively low temperature. Furthermore, dry NCC film has been formed within 5 h. Thus, there was no external force to push the NCC molecules closer together as the evaporation rate became insignificant. Therefore, further increase in the curing time towards 10 h and longer did not cause the NCC molecules to be packed closer. The gap between NCC molecules would remain the same, so that the diffusion of incoming MB molecules and dye uptake would also remain constant. Table 1 lists the self-assembly mechanism of NCC at higher temperature at 70, 80, and 90 °C. By 5 h of curing time, dry NCC has been formed at fixed curing temperatures of 70 to 90 °C. Although there should not be any more external force to push the NCC molecules closer together when the evaporation rate became negligible, it is postulated that self-assembly continued to occur because a stable low-energy state has not been achieved at such high temperatures. It is proposed that the self-assembly is attributed to the partial sintering effect, which might have occurred simultaneously, even when the water was still undergoing evaporation from the NCC dispersion (Kang, 2010), as well as after the formation of NCC film. When partial sintering occurs, adjacent NCC molecules would undergo grain growth, which would eventually combine them into one molecule. This would effectively reduce the contact area between MB molecules and NCC flakes (Cheung & Darvell, 2002). This explains the formation of dense structures shown in Fig. 1(b) at curing conditions of 90 °C and 5 h. At longer than 5 h of curing time at fixed curing temperatures of 60 to 90 °C, partial sintering continued to occur because it was utilizing the excess energy received from the heat (German, 2010). Thus, the contact area between MB molecule and NCC flakes would become smaller, which explains the significant downtrend of dye uptake when the curing time was increased at these curing temperatures. However, it is postulated that at least by 10 h of curing time, low-stable energy would

Fig. 3. Effects of Curing Temperature (50–90 °C) and Curing Time (5–25 h) of Self-assembled NCC Flakes on Dye Uptake.

Similarly, Tang et al. (2015) reported that the pore width of self-assembled silver/silica nanostructures had decreased when the curing temperature was increased. Consequently, the diffusion of the incoming MB molecules and the contact areas on the surface of the NCC flakes was hindered at higher temperatures, resulting in lower dye uptakes (Sathishkumar et al., 2008). In addition, more hydroxyl groups on the NCC structure were involved in the hydrogen bonds at higher temperature due to higher capillary force that was pushing the NCC molecules closer to each other. Therefore, lesser adsorption sites from the hydroxyl groups on the NCC molecules were available to interact with MB molecules. Fig. 3 also shows the effect of curing time on MB uptake. It was found that curing time was almost independent of dye uptake at fixed curing temperatures of 50 °C and 60 °C. However, different results were observed at higher fixed temperatures of 70, 80, and 90 °C. This could be attributed to the partial sintering effect that occurred upon hydrogen bonding during the self-assembly process at higher temperatures. In order to explain this trend of results, analysis of the self-assembly mechanisms was proposed. Table 1 describes the self-assembly mechanism of NCC at 50 and 60 °C. It should be noted that each material has a minimum sintering temperature. Therefore, at 50 and 60 °C, the Table 1 Self-assembly Mechanism of NCC at Different Curing Temperatures. Curing Temperature

50 °C and 60 °C

70 °C, 80 °C and 90 °C

During Self-Assembly Process

After 5 h

After 10 h

10 h and longer


Carbohydrate Polymers 199 (2018) 92–101

K.B. Tan et al.

observed as it approached 100%. Thus, the optimum adsorbent dosage was determined to be 0.7 g/L for efficient dye removal and was used for subsequent experiments. By comparison, the optimum dosage for NCC flakes was lower than for NCC powder as adsorbent, which was reported to be 20 g/L. This was due to the different sources of NCC used in both works, whereby the sulfur content in the NCC might have been different. The sulfur content in the NCC used by Batmaz et al. (2014) was calculated by Beck, Bouchard, and Berry (2012) to be 0.736 wt%. Meanwhile, the sulfur content of the NCC used in this research was determined to be at 0.89 wt%. Higher sulfur content would indicate that more sulfate (SO32−) groups exists on the NCC structure, and therefore, offered more available adsorption sites for the incoming MB molecules. Fig. 4(b) shows the removal percentage of MB by NCC flakes at different initial concentrations, from 10 to 1000 ppm at the optimum dosage of 0.7 g/L, where adsorption equilibrium was achieved in less than 1 min. At low dye initial concentration of 10–50 ppm, as well as moderate initial concentration of 100 ppm, the adsorbent has sufficient amount of adsorption sites to accommodate the low amount of MB molecules, which could explain the high percentage removal of 88–96% (Roosta, Ghaedi, & Asfaram, 2015). However, due to the low dosage at 0.7 g/L, the amount of adsorption sites were limited and could not accommodate large amount of MB molecules (Weng & Pan, 2007) at 500 ppm and 1000 ppm, which explains the significantly low dye removal at 27% and 14%, respectively. This was found to be in stark contrast with the work of Batmaz et al. (2014), where the percentage of MB removal by NCC powder was found to be 75%, even at a high dye initial concentration of 1000 ppm. This could be due to the significantly higher adsorbent dosage of 10 g/L. Hence, the adsorbent would have more adsorption sites for incoming dye molecules. Despite that, similar trends were also found for the adsorption of MB by other adsorbents, such as MCC (Tan et al., 2016), activated clay (Weng & Pan, 2007), and coffee husk (Oliveira, Franca, Alves, & Rocha, 2008), whereby the maximum removal percentage is decreased when the initial dye concentration is increased.

Fig. 4. (a) Effect of Adsorbent Dosage on the Adsorption of MB fixed at 100 ppm, 25 °C, and pH 6; (b) Effect of initial MB concentration on its removal percentage, fixed at adsorbent dosage of 0.7 g/L, 25 °C, and pH 6.

have been achieved. As a result, no excess energy could be utilized for significant sintering to occur. Hence, the downtrend of dye uptake became less significant when the curing time was prolonged. Based on Fig. 3, the highest dye uptake was achieved with a curing temperature of 50 °C for 5 h, with a dye uptake of 123.3 mg/g. By comparison, MB uptake due to the adsorption of NCC powder was determined to be 142.0 mg/g. As the NCC flakes were self-assembled, some of the adsorption sites were involved in the hydrogen bonding between adjacent NCC molecules. This effectively reduces the adsorption sites for MB molecules. However, NCC powder could not be easily separated from MB wastewater without the assistance of chemical, while NCC flakes could easily be separated from MB wastewater. Hence, NCC flakes were found to be more practical to be packed into an adsorption column compared to NCC powder. Based on this reasoning, further adsorption experiments were conducted on the developed NCC flakes to determine its interaction and adsorption performance with MB.

3.5. Effect of pH Apart from the isotherm and thermodynamics studies, the effect of pH on dye removal could also offer an overview of the adsorption mechanism between NCC flakes and MB. Fig. 5(a) shows the effect of different pH ranging between 2 to 10 on the percentage removal of MB. However, Fig. 5(b) shows that despite the sharp increase in negative zeta potential from −15 to −55 mV when the pH value was increased from 2 to 6, the effect of pH was found to be almost independent from the percentage removal of MB. The percentage removal was only increased from 76 to 88% when the pH was increased from 2 to 6. This indicates that electrostatic attraction was not the main adsorption mechanism for the interaction between NCC flakes and MB molecules. It is possible that the dominant adsorption mechanism could be attributed to the hydrogen bonding between NCC flakes and MB molecules, which will be explained in details in the next section.

3.4. Effects of adsorbent dosage and initial concentration

3.6. Isotherm and thermodynamics studies

Fig. 4(a) shows the effect of different NCC adsorbent dosages on dye removal at equilibrium. This was done to determine the optimum adsorbent dosage to be used for isotherm and thermodynamics studies. For a fixed concentration of dye at 100 ppm at adsorbent dosage of 0.1 g/L, dye removal was found to be at 10%. This is because at low adsorbent dosage, the adsorption sites could have easily became saturated (Bagheri, Ghaedi, Asfaram, Jannesar, & Goudarzi, 2017). However, percentage of dye removal increased significantly from 10 to 88% when the adsorbent dosage was increased from 0.1 to 0.7 g/L. When the adsorbent dosage was increased, more adsorption sites were available for MB molecules, thus leading to a higher dye removal efficiency (Yagub, Sen, Afroze, & Ang, 2014). It was found that at adsorbent dosage of higher than 0.7 g/L, insignificant increase in dye removal was

Isotherm studies were conducted at different initial dye concentrations (10–1000 ppm) and at different temperatures that ranged between 25 to 45 °C. This was done to determine the maximum adsorption capacity and the adsorption mechanism of the system. These were fitted into Langmuir, Freundlich, Temkin, and Dubinin-Radushkevich isotherms, which are presented in the Supplementary Data (Figs. S1–S5). The data collected from the isotherm fittings are tabulated in Table 2. The equilibrium data were found to fit Langmuir isotherm very well at all temperatures, with the coefficient of correlation (R2) ranging from 0.996 to 0.999, compared to with Freundlich, Temkin, and DubininRadushkevich isotherms. Thus, the adsorption of MB using NCC flakes 97

Carbohydrate Polymers 199 (2018) 92–101

K.B. Tan et al.

temperature, which explains the decrease in the value of the Langmuir isotherm constant, kL. Based on Table 2, the value of adsorption heat, ΔHT, which was obtained from the Temkin isotherm, was determined to be -0.1 kJ/mol at all temperatures, which were larger than -20 kJ/mol. This suggests that physical adsorption is the dominant effect for the adsorption of MB on NCC flakes (Itodo & Itodo, 2010). The values of the mean free energy of adsorption (Ea) that were obtained from the Dubinin-Radushkevich isotherm at 25, 35, and 45 °C were found to be 2.9, 0.6, and 0.3 kJ/mol, respectively: all of which were less than 8 kJ/mol. These values also suggests that physical adsorption is the dominating effect (Ibrahim & Sani, 2014). It should be noted that adsorption would have occurred at an extremely rapid rate because by 0.5 min, equilibrium would have been achieved. It was very difficult to collect data of the percentage removal of MB at a narrower time interval. Since adsorption kinetics require more data points before the adsorption equilibrium to plot pseudo-first order and pseudo-second order models. Therefore, the kinetics study was not perform for the adsorption of MB using NCC flakes. Meanwhile, Table 2 displays the thermodynamic data based on the graph plotted in Supplementary Data. The values of enthalpy ΔH and entropy ΔS are obtained from the intercept and slope of the plot respectively. It was found that the Gibbs Free energy, ΔGo values at all temperatures were negative, confirming the spontaneous and feasible adsorption of MB dye on NCC flakes (El Boujaady, Mourabet, BennaniZiatni, & Taitai, 2014). Furthermore, the ΔGo values at different temperatures were within the -20 kJ/mol and 0 kJ/mol. This verified that physical adsorption was the dominant adsorption mechanism behind the adsorption of MB dye on NCC (Mahmoodi, Hayati, Arami, & Lan, 2011).

Fig. 5. (a) Effect of pH on the Adsorption of MB fixed at 100 ppm, 0.7 g/L of NCC flakes, and 25 °C; (b) Zeta Potential of NCC Flakes at a Wide Range of pH.

3.7. Adsorption mechanism

Table 2 Isotherm and Thermodynamics Data for the Adsorption of MB using NCC Flakes at 25 °C, 35 °C, and 45 °C. Parameters

The chemical structures of MB molecules and NCC flakes, which are shown in the Supplementary Data (Fig. S6), show that MB is a cationic compound, whereby the positive charge could exist in the sulfur element inside the aromatic ring (Machulek et al., 2013). Meanwhile, NCC has a similar structure to cellulose, which consists of many hydroxyl groups. Additional sulfate groups linked to the methyl groups on some of the NCC monomers were also detected (Araki, 2013). Table 3 shows the possible adsorption mechanisms that could occur during the adsorption of MB molecules onto NCC flakes. One of the possible adsorption mechanisms is the hydrogen bonding between NCC flakes and MB molecules. Since an MB molecule has lone pairs of electrons on the nitrogen element inside and on both tail ends of the aromatic rings, these electrons will attract the hydrogen part of the hydroxyl group in the NCC structure. As a result, hydrogen bonds would be formed, which would induce the adsorption of MB molecules onto NCC flakes (Douissa, Dridi-Dhaouadi, & Mhenni, 2016). In addition, a lone electron pair can be found at the double-bonded oxygen on the sulfate group. This pair is also able to interact with the hydrogen element to form hydrogen bond (Douissa et al., 2016). However, since there is no hydrogen attached to the MB molecule that would be able to form hydrogen bond, the sulfate group does not participate in the hydrogen bonding between NCC flakes and MB molecule. Another possible adsorption mechanism is the ion-dipole interaction. As shown in Table 3, there will be a negative dipole moment on the oxygen attached to the hydroxyl group of the NCC flakes. Thus, it is proposed that ion-dipole interaction could occur between the oxygen element on the hydroxyl group and the positively charged sulfur inside the aromatic ring in MB molecule. In addition, the oxygen that is attached to the hydroxyl group on the NCC flakes is able to form iondipole interaction with the S+ ion within the aromatic ring of MB molecule, as shown in Table 3. There is another possible ion-dipole interaction between the S]O bond on the sulfate group attached to the NCC, and the S+ ion within the aromatic ring of the MB. Hence, these interactions could have induced the adsorption of MB molecules onto

Temperature 25 °C

35 °C

45 °C

Langmuir Isotherm

qmax (mg/g) kL (L/mg) R2

188.7 0.09 0.9997

172.4 0.05 0.9955

156.3 0.02 0.9982

Freundlich Isotherm

kf (L1/n/g) 1/n R2

43.0 0.3 0.8491

19.9 0.4 0.7573

8.4 0.5 0.8634

Temkin Isotherm

BT (L/mg) kT ΔHT (kJ/mol) R2

20.2 26.1 −0.1 0.9823

23.7 1.9 −0.1 0.9311

25.5 0.5 −0.1 0.9364

Dubinin-Radushkevich Isotherm

qm (mg/g) B (mol2/kJ2) Ea (kJ/mol) R2

140.3 0.06 2.9 0.8924

120.9 1.5 0.6 0.8983

98.8 4.3 0.3 0.8122


ΔGo (kJ/mol) ΔH (kJ/mol) ΔS (J/mol)

−5.82 −4.54 −63.09 −191.51


was concluded to be a monolayer adsorption at the homogenous surface of NCC flakes, which is also similar to NCC powder (Batmaz et al., 2014). This also indicates that the adsorbed MB molecules had no interaction with each other at neighboring adsorption sites. It was found that when temperature was decreased, the value of maximum adsorption capacity, qmax, had also decreased. This observation can be attributed to the weakening hydrogen bonds between the MB molecules and the hydroxyl functional groups on the NCC flakes caused by the increase in molecular motion (Chatterjee, Chatterjee, Chatterjee, & Guha, 2007). Subsequently, the bonding energy between NCC flakes and MB molecules would decrease with increasing 98

Carbohydrate Polymers 199 (2018) 92–101

K.B. Tan et al.

Table 3 Possible Adsorption Mechanisms of MB molecule onto NCC Flakes. Bond

Adsorption Mechanism

Hydrogen Bond

Ion-Dipole Interaction

Electrostatic Interaction

the surface of NCC flakes. As shown in Table 3, electrostatic attraction could also occur between the S+ on the MB molecule and the O− attached to the sulfate group on the NCC. These three possible adsorption mechanisms could occur simultaneously or individually to drive the adsorption of MB molecules onto NCC flakes. However, it is unclear which adsorption mechanism is the dominant interaction between NCC flakes and MB molecules. Nevertheless, the effect of pH on dye removal, as shown in Fig. 5(a), could provide an insight into the dominant adsorption mechanism. At pH 2, the H + ion concentration in the wastewater would be high (Alipanahpour Dil et al., 2017). Thus, the H+ ion could interact with the oxygen attached to hydroxyl group and the double-bonded oxygen attached to the sulfate group via the ion-dipole interaction, as well as via electrostatic attraction to the O− attached to the sulfate group, which can be found in Supplementary Data (Fig. S7). Changes in pH might hinder the iondipole interaction and the electrostatic attraction between NCC flakes and MB molecules, and subsequently might hinder the hydrogen bonding. However, since NCC flakes that were cured at 50 °C for 5 h have a surface area of 78 m2/g, there could have been a stronger interaction between MB and NCC flakes. Therefore, stronger and more hydrogen bonds can be formed without being affected by the hindrance on ion-dipole interaction and electrostatic attraction, which explains the high percentage removal at 75% even at pH 2. When the pH was increased to 3, the concentration of H+ ion had decreased. As a result, more ion-dipole interactions and electrostatic attractions occurred between MB molecules and NCC flakes, leading to the formation of more hydrogen bonds. Therefore, the percentage removal was increased to 82%, which is only a 6% increase from pH 2. Hence, it is deduced that ion-dipole interaction and electrostatic attraction contributed lesser interaction between NCC flakes and MB molecule compared to hydrogen bonding. Henceforth, physical adsorption due to hydrogen bonding is considered as the dominant adsorption mechanism behind the interaction between NCC flakes as adsorbent and MB molecules as adsorbate. This observation was also verified from the results obtained

under isotherms and thermodynamics studies. 3.8. Comparison of performances Results of the maximum adsorption capacity of NCC flakes towards MB molecules obtained from the Langmuir isotherm model were compared with other reported work listed in Table 4. Self-assembled NCC flakes have the highest maximum adsorption capacity (188.87 mg/g) compared to other more toxic nanomaterials, such as carbon nanotube (59.7 mg/g) (Wang, Ng, Wang, Li, & Hao, 2012), and titania nanotube (133.3 mg/g) (Xiong et al., 2010), which were difficult to separate from dye wastewater. Meanwhile, the adsorption performance of NCC flakes was determined to be better than magnetic adsorbents that could easily be separated from dye wastewater using external magnetic force. Magnetic adsorbents may include ilmenite FeTiO3 (Chen, 2011) and nano-Fe3O4 (Iram, Guo, Guan, Ishfaq, & Liu, 2010), which have maximum adsorption capacities of 71.9 mg/g and 93.1 mg/g, respectively. Both adsorbents were made of toxic nanomaterials (Tan et al., 2015). The maximum adsorption capacity for this current work was also determined to be higher than other reported nanomaterial adsorbents, such as self-assembled carbon nanotube-graphene composite (maximum adsorption capacity of 43.1 mg/g), where the self-assembled method was utilized to solve separation problems (Vijwani, Nadagouda, Namboodiri, & Mukhopadhyay, 2015). The performance of the NCC flakes was also more efficient compared to activated carbon nanocomposites, such as manganese CuS/ZnS activated carbon-loaded nanocomposite (Asfaram, Ghaedi, Azqhandi, Goudarzi, & Hajati, 2017). In addition, the self-assembled method utilized in this work is greener because it only need water to self-assemble the NCC flakes compared to self-assembled carbon nanotubes, which require additional chemicals (Ai & Jiang, 2012). Overall, it was found that the adsorption performance of the selfassembled NCC flakes was more efficient compared to low cost materials and other cellulosic materials, such as coffee husks (90.1 mg/g) 99

Carbohydrate Polymers 199 (2018) 92–101

K.B. Tan et al.

Table 4 Comparisons between NCC flakes and other materials on the removal of MB dye. Materials

Maximum Adsorption Capacity (mg/g)


Self-Assembled NCC flakes NCC Powder Carbon Nanotube Self-assembled carbon nanotube-graphene-oxide composite Self-assembled carbon nanotube-reticulated carbon foam nano-composite Titania Nanotube Ilmenite FeTiO3 Nano-Fe3O4 Activated Clay Sludge Ash Activated carbon coated palygorskite Dimethyl diallyl ammonium chloride and diallylamin co-polymer modified bio-film derived from palm dates Manganese CuS/ZnS activated carbon-loaded nanocomposite Microcrystalline Cellulose Coffee Husks Orange Peel Posidonia oceanica (L.) fibers Clay Pyrolyzed petrified sediment

188.7 118.0 59.7 81.9 43.1 133.3 71.9 93.1 109.1 1.1 351.0 150.0

Current work Batmaz et al. (2014) Wang et al. (2012) Ai and Jiang (2012) Vijwani et al. (2015) Xiong et al. (2010) Chen (2011) Iram et al. (2010) Weng and Pan (2007) Weng and Pan (2006) Zhang et al. (2015) Jabli et al. (2017)

126.42 5.0 90.1 18.6 5.6 6.3 2.4

Asfaram et al. (2017)) Tan et al. (2016) Oliveira et al. (2008) Annadurai et al. (2002) Ncibi et al. (2009) Gürses et al. (2004) Aroguz, Gulen, and Evers, (2008)

Appendix A. Supplementary data

(Oliveira et al., 2008), and orange peel (18.6 mg/g) (Annadurai, Juang, & Lee, 2002). Self-assembled NCC flakes were also better than toxic nano-adsorbents, self-assembled adsorbents, and magnetic adsorbents, as previously explained. In addition, the NCC flakes’ adsorption performance was also more efficient than modified cellulosic materials, such as dimethyl diallyl ammonium chloride and diallylamin copolymer modified bio-film derived from palm dates (150.0 mg/mg) (Jabli, Saleh, Sebeia, Tka, & Khiari, 2017). Thus, these results suggest that self-assembled NCC flakes are a promising and practical green nanomaterial that could be further developed to be used as adsorbent in the industry.

Supplementary material related to this article can be found, in the online version, at doi: References Ai, L., & Jiang, J. (2012). Removal of methylene blue from aqueous solution with selfassembled cylindrical graphene–carbon nanotube hybrid. Chemical Engineering Journal, 192(0), 156–163. Aksu, Z., Açıkel, Ü., Kabasakal, E., & Tezer, S. (2002). Equilibrium modelling of individual and simultaneous biosorption of chromium(VI) and nickel(II) onto dried activated sludge. Water Research, 36(12), 3063–3073. Alansi, A. M., Alkayali, W. Z., Al-qunaibit, M. H., Qahtan, T. F., & Saleh, T. A. (2015). Synthesis of exfoliated polystyrene/anionic clay MgAl-layered double hydroxide: Structural and thermal properties. RSC Advances, 5(87), 71441–71448. Ali, S. A., Rachman, I. B., & Saleh, T. A. (2017). Simultaneous trapping of Cr(III) and organic dyes by a pH-responsive resin containing zwitterionic aminomethylphosphonate ligands and hydrophobic pendants. Chemical Engineering Journal, 330, 663–674. Alipanahpour Dil, E., Ghaedi, E., Asfaram, M., Zare, A., Mehrabi, F., & Sadeghfar, F. (2017). Comparison between dispersive solid-phase and dispersive liquid–liquid microextraction combined with spectrophotometric determination of malachite green in water samples based on ultrasound-assisted and preconcentration under multi-variable experimental design optimization. Ultrasonics Sonochemistry, 39, 374–383. Alswat, A. A., Ahmad, M. B., & Saleh, T. A. (2016). Zeolite modified with copper oxide and iron oxide for lead and arsenic adsorption from aqueous solutions. Journal of Water Supply: Research and Technology - Aqua, 65(6), 465–479. Annadurai, G., Juang, R.-S., & Lee, D.-J. (2002). Use of cellulose-based wastes for adsorption of dyes from aqueous solutions. Journal of Hazardous Materials, 92(3), 263–274. Araki, J. (2013). Electrostatic or steric?–preparations and characterizations of well-dispersed systems containing rod-like nanowhiskers of crystalline polysaccharides. Soft Matter, 9(16), 4125–4141. Aroguz, A. Z., Gulen, J., & Evers, R. H. (2008). Adsorption of methylene blue from aqueous solution on pyrolyzed petrified sediment. Bioresource Technology, 99(6), 1503–1508. Asfaram, A., Ghaedi, M., Azqhandi, M. H. A., Goudarzi, A., & Hajati, S. (2017). Ultrasound-assisted binary adsorption of dyes onto [email protected] CuS/ZnS-NC-AC as a novel adsorbent: Application of chemometrics for optimization and modeling. Journal of Industrial and Engineering Chemistry, 54, 377–388. Asfaram, A., Ghaedi, M., Ghezelbash, G. R., Dil, E. A., Tyagi, I., Agarwal, S., ... Gupta, V. K. (2016). Biosorption of malachite green by novel biosorbent Yarrowia lipolytica isf7: Application of response surface methodology. Journal of Molecular Liquids, 214, 249–258. Bagheri, A. R., Ghaedi, M., Asfaram, A., Jannesar, R., & Goudarzi, A. (2017). Design and construction of nanoscale material for ultrasonic assisted adsorption of dyes: Application of derivative spectrophotometry and experimental design methodology. Ultrasonics Sonochemistry, 35, 112–123. Batmaz, R., Mohammed, N., Zaman, M., Minhas, G., Berry, R., & Tam, K. (2014). Cellulose nanocrystals as promising adsorbents for the removal of cationic dyes. Cellulose, 21(3), 1655–1665. Beck, S., Bouchard, J., & Berry, R. (2012). Dispersibility in Water of dried nanocrystalline cellulose. Biomacromolecules, 13(5), 1486–1494. Boeckl, M., & Graham, D. (2006). Self-assembled monolayers: Advantages of pure alkanethiols. Materials Matters, 1(2), 3–7.

4. Conclusion This study has evaluated a more environmentally friendly and sustainable way of removing dyes from wastewater. Thermally-cured NCC that turn into self-assembled NCC flakes had been a successful solution for the separation problem of NCC adsorbent from dye solution. This study found that at curing conditions of 50 °C and 5 h, the adsorption of MB dye by NCC flakes demonstrated the highest uptake at 123.3 mg/g. The adsorption fitted well in Langmuir isotherm, with R2 exceeding 0.996. Results for the Dubinin-Radushkevich isotherm, thermodynamics and effect of pH suggest that physical adsorption due to hydrogen bonding was the dominant adsorption mechanism. Thermodynamics studies also showed that the adsorption on NCC flakes was exothermic in nature. These properties were found to be similar to the properties of NCC powder, suggesting that the self-assembled NCC flakes had not altered the adsorption properties and mechanism, but merely the physical structure to solve the separation problem. The maximum adsorption capacity was achieved by 188.7 mg/g at 0.7 g/L of NCC dosage. Thus, it was concluded that this work has solved the NCC separation problem, and offers a practical and promising NCCbased green nanomaterial adsorbent that is more efficient than other non-green nanomaterials on the removal of MB dye.

Acknowledgments The authors gratefully acknowledge the Ministry of Education Malaysia (MOE) for providing research funding under the FRGS scheme, grant number FRGS/2/2013 TK05/MUSM/03/1. The authors also highly appreciate the assistance of Dr. Oh Pei Ching from Universiti Teknologi Petronas (UTP) during the mercury porosity test. 100

Carbohydrate Polymers 199 (2018) 92–101

K.B. Tan et al.

limitations through correlation with floc size and shape. Water Research, 109, 54–68. Matsis, V. M., & Grigoropoulou, H. P. (2008). Kinetics and equilibrium of dissolved oxygen adsorption on activated carbon. Chemical Engineering Science, 63(3), 609–621. Mitchell, B. S. (2004). An introduction to materials engineering and science. The structure of materials. John Wiley & Sons, Inc1–135. Ncibi, M. C., Hamissa, A. M. B., Fathallah, A., Kortas, M. H., Baklouti, T., Mahjoub, B., ... Seffen, M. (2009). Biosorptive uptake of methylene blue using Mediterranean green alga Enteromorpha spp. Journal of Hazardous Materials, 170(2–3), 1050–1055. Oliveira, L. S., Franca, A. S., Alves, T. M., & Rocha, S. D. F. (2008). Evaluation of untreated coffee husks as potential biosorbents for treatment of dye contaminated waters. Journal of Hazardous Materials, 155(3), 507–512. Peng, B. L., Dhar, N., Liu, H. L., & Tam, K. C. (2011). Chemistry and applications of nanocrystalline cellulose and its derivatives: A nanotechnology perspective. The Canadian Journal of Chemical Engineering, 89(5), 1191–1206. Peng, Y., Gardner, D. J., Han, Y., Kiziltas, A., Cai, Z., & Tshabalala, M. A. (2013). Influence of drying method on the material properties of nanocellulose I: Thermostability and crystallinity. Cellulose, 20(5), 2379–2392. Pivetta, M., Pacchioni, G. E., Fernandes, E., & Brune, H. (2015). Temperature-dependent self-assembly of NC–Ph5–CN molecules on Cu(111). The Journal of Chemical Physics, 142(10), 101928. Pranger, L., & Tannenbaum, R. (2008). Biobased nanocomposites prepared by in situ polymerization of furfuryl alcohol with cellulose whiskers or montmorillonite clay. Macromolecules, 41(22), 8682–8687. Qi, H., Cai, J., Zhang, L., & Kuga, S. (2009). Properties of films composed of cellulose nanowhiskers and a cellulose matrix regenerated from alkali/urea solution. Biomacromolecules, 10(6), 1597–1602. Rafatullah, M., Sulaiman, O., Hashim, R., & Ahmad, A. (2010). Adsorption of methylene blue on low-cost adsorbents: A review. Journal of Hazardous Materials, 177(1), 70–80. Roosta, M., Ghaedi, M., & Asfaram, A. (2015). Simultaneous ultrasonic-assisted removal of malachite green and safranin O by copper nanowires loaded on activated carbon: Central composite design optimization. RSC Advances, 5(70), 57021–57029. Saleh, T. A., Sarı, A., & Tuzen, M. (2017). Optimization of parameters with experimental design for the adsorption of mercury using polyethylenimine modified-activated carbon. Journal of Environmental Chemical Engineering, 5(1), 1079–1088. Sani, H. A., Ahmad, M. B., Hussein, M. Z., Ibrahim, N. A., Musa, A., & Saleh, T. A. (2017). Nanocomposite of ZnO with montmorillonite for removal of lead and copper ions from aqueous solutions. Process Safety and Environmental Protection, 109, 97–105. Sathishkumar, M., Vijayaraghavan, K., Binupriya, A. R., Stephan, A. M., Choi, J. G., & Yun, S. E. (2008). Porogen effect on characteristics of banana pith carbon and the sorption of dichlorophenols. Journal of Colloid and Interface Science, 320(1), 22–29. Suratago, T., Taokaew, S., Kanjanamosit, N., Kanjanaprapakul, K., Burapatana, V., & Phisalaphong, M. (2015). Development of bacterial cellulose/alginate nanocomposite membrane for separation of ethanol–water mixtures. Journal of Industrial and Engineering Chemistry, 32, 305–312. Tan, K. B., Abdullah, A. Z., Horri, B. A., & Salamatinia, B. (2016). Adsorption mechanism of microcrystalline cellulose as green adsorbent for the removal of cationic methylene blue dye. Journal of The Chemical Society of Pakistan, 36(5). Tan, K. B., Vakili, M., Horri, B. A., Poh, P. E., Abdullah, A. Z., & Salamatinia, B. (2015). Adsorption of dyes by nanomaterials: Recent developments and adsorption mechanisms. Separation and Purification Technology, 150, 229–242. Tang, Y., Yang, M., Dong, W., Tan, L., Zhang, X., Zhao, P., ... Wang, G. (2015). Temperature difference effect induced self-assembly method for Ag/SBA-15 nanostructures and their catalytic properties for epoxidation of styrene. Microporous and Mesoporous Materials, 215, 199–205. Thorkelsson, K., Bai, P., & Xu, T. (2015). Self-assembly and applications of anisotropic nanomaterials: A review. Nano Today, 10(1), 48–66. Vakili, M., Rafatullah, M., Salamatinia, B., Ibrahim, M. H., Ismail, N., & Abdullah, A. Z. (2017). Adsorption studies of methyl tert-butyl ether from environment. Separation & Purification Reviews, 46(4), 273–290. Vijwani, H., Nadagouda, M. N., Namboodiri, V., & Mukhopadhyay, S. M. (2015). Hierarchical hybrid carbon nano-structures as robust and reusable adsorbents: Kinetic studies with model dye compound. Chemical Engineering Journal, 268(0), 197–207. Vutskits, L., Briner, A., Klauser, P., Gascon, E., Dayer, A. G., Kiss, J. Z., ... Morel, D. R. (2008). Adverse effects of methylene blue on the central nervous system. The Journal of the American Society of Anesthesiologists, 108(4), 684–692. Waigh, T. A. (2014). The physics of living processes: A mesoscopic approach. Manchester, UK: Wiley. Wang, S., Ng, C. W., Wang, W., Li, Q., & Hao, Z. (2012). Synergistic and competitive adsorption of organic dyes on multiwalled carbon nanotubes. Chemical Engineering Journal, 197, 34–40. Weng, C.-H., & Pan, Y.-F. (2006). Adsorption characteristics of methylene blue from aqueous solution by sludge ash. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 274(1–3), 154–162. Weng, C.-H., & Pan, Y.-F. (2007). Adsorption of a cationic dye (methylene blue) onto spent activated clay. Journal of Hazardous Materials, 144(1–2), 355–362. Xiong, L., Yang, Y., Mai, J., Sun, W., Zhang, C., Wei, D., ... Ni, J. (2010). Adsorption behavior of methylene blue onto titanate nanotubes. Chemical Engineering Journal, 156(2), 313–320. Yagub, M. T., Sen, T. K., Afroze, S., & Ang, H. M. (2014). Dye and its removal from aqueous solution by adsorption: A review. Advances in Colloid and Interface Science, 209, 172–184. Zhang, X., Cheng, L., Wu, X., Tang, Y., & Wu, Y. (2015). Activated carbon coated palygorskite as adsorbent by activation and its adsorption for methylene blue. Journal of Environmental Sciences, 33, 97–105. Zhou, C., & Wu, Q. (2012). Recent development in applications of cellulose nanocrystals for advanced polymer-based nanocomposites by novel fabrication strategies.

Bondeson, D., Mathew, A., & Oksman, K. (2006). Optimization of the isolation of nanocrystals from microcrystalline celluloseby acid hydrolysis. Cellulose, 13(2), 171–180. Chang, Y., Lai, J. Y., & Lee, D. J. (2016). Thermodynamic parameters for adsorption equilibrium of heavy metals and dyes from wastewaters: Research updated. Bioresource Technology, 222, 513–516. Chatterjee, S., Chatterjee, S., Chatterjee, B. P., & Guha, A. K. (2007). Adsorptive removal of congo red, a carcinogenic textile dye by chitosan hydrobeads: Binding mechanism, equilibrium and kinetics. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 299(1–3), 146–152. Chen, Y. H. (2011). Synthesis, characterization and dye adsorption of ilmenite nanoparticles. Journal of Non-Crystalline Solids, 357(1), 136–139. Chequer, F. M., Oliveira, G. A., Ferraz, E. A., Cardoso, J. C., Zanoni, M. V., & Olivera, D. P. (2013). Textile dyes: Dyeing process and environmental impact. Eco-friendly textile dyeing and finishing. London, United Kingdom: Intech151–176. Cheung, K. C., & Darvell, B. W. (2002). Sintering of dental porcelain: Effect of time and temperature on appearance and porosity. Dental Materials, 18(2), 163–173. Daintith, J. (2004). The facts on file dictionary of organic chemistry. New York: Market House Book. Dil, E. A., Ghaedi, M., Asfaram, A., Hajati, S., Mehrabi, F., & Goudarzi, A. (2017). Preparation of nanomaterials for the ultrasound-enhanced removal of Pb2+ ions and malachite green dye: Chemometric optimization and modeling. Ultrasonics Sonochemistry, 34, 677–691. Douissa, N. B., Dridi-Dhaouadi, S., & Mhenni, M. F. (2016). Spectrophotometric investigation of the interactions between cationic (CI basic Blue 9) and anionic (CI acid Blue 25) dyes in adsorption onto extracted cellulose from Posidonia oceanic in single and binary system. Water Science and Technology, 73(9), 2211–2221. Ek, M., Gellerstedt, G., & Gunnar, H. (2009). Paper products physics and technology. Berlin, Germany: De Gruyter. El Boujaady, H., Mourabet, M., Bennani-Ziatni, M., & Taitai, A. (2014). Adsorption/ desorption of Direct Yellow 28 on apatitic phosphate: Mechanism, kinetic and thermodynamic studies. Journal of the Association of Arab Universities for Basic and Applied Sciences, 16, 64–73. Fleck, N. A., McMeeking, R. M., & Kraus, T. (2015). Convective assembly of a particle monolayer. Langmuir, 31(51), 13655–13663. Gaspar, D., Fernandes, S. N., De Oliveira, A. G., Fernandes, J. G., Grey, P., Pontes, R. V., ... Fortunato, E. (2014). Nanocrystalline cellulose applied simultaneously as the gate dielectric and the substrate in flexible field effect transistors. Nanotechnology, 25(9). German, R. M. (2010). 1 - Thermodynamics of sintering A2 - Fang, Zhigang Zak. Sintering of advanced materials. Woodhead Publishing3–32. Gillman, P. K. (2006). Methylene blue implicated in potentially fatal serotonin toxicity. Anaesthesia, 61(10), 1013–1014. Gürses, A., Karaca, S., Doğar, Ç., Bayrak, R., Açıkyıldız, M., & Yalçın, M. (2004). Determination of adsorptive properties of clay/water system: Methylene blue sorption. Journal of Colloid and Interface Science, 269(2), 310–314. Habibi, Y., Lucia, L. A., & Rojas, O. J. (2010). Cellulose nanocrystals: Chemistry, selfassembly, and applications. Chemical Reviews, 110(6), 3479–3500. Hay, B. P., Dixon, D. A., Lumetta, G. J., Vargas, R., & Garza, J. (2004). Structural aspects of hydrogen bonding with nitrate and sulfate. In B. A. Moyer, & R. P. Singh (Eds.). Fundamentals and applications of anion separations (pp. 43–57). Boston, MA: Springer US. Huang, J., & Rodrigue, D. (2015). Comparison of the mechanical properties between carbon nanotube and nanocrystalline cellulose polypropylene based nano-composites. Materials & Design, 65(1980-2015), 974–982. Ibrahim, M. B., & Sani, S. (2014). Comparative isotherms studies on adsorptive removal of Congo Red from wastewater by watermelon rinds and neem-tree leaves. Open Journal of Physical Chemistry, 4, 139–146. Iram, M., Guo, C., Guan, Y., Ishfaq, A., & Liu, H. (2010). Adsorption and magnetic removal of neutral red dye from aqueous solution using Fe3O4 hollow nanospheres. Journal of Hazardous Materials, 181(1–3), 1039–1050. Itodo, A., & Itodo, H. (2010). Sorption energies estimation using Dubinin-Radushkevich and Temkin adsorption isotherms. Life Science Journal- Acta Zhengzhou University Overseas Edition, 7(4), 31–39. Jabli, M., Saleh, T. A., Sebeia, N., Tka, N., & Khiari, R. (2017). Dimethyl diallyl ammonium chloride and diallylamin Co-polymer modified bio-film derived from palm dates for the adsorption of dyes. Scientific Reports, 7(1), 14448. Jabli, M., Tka, N., Ramzi, K., & Saleh, T. A. (2018). Physicochemical characteristics and dyeing properties of lignin-cellulosic fibers derived from Nerium oleander. Journal of Molecular Liquids, 249, 1138–1144. Kang, S. J. L. (2010). 5 - Liquid phase sintering A2 - Fang, Zhigang Zak. Sintering of advanced materials. Woodhead Publishing110–129. Kralchevsky, P. A., & Denkov, N. D. (2001). Capillary forces and structuring in layers of colloid particles. Current Opinion in Colloid & Interface Science, 6(4), 383–401. Landsmann, S., Luka, M., & Polarz, S. (2012). Bolaform surfactants with polyoxometalate head groups and their assembly into ultra-small monolayer membrane vesicles. Nature Communications, 3, 1299. Lemke, T., Williams, D., Roche, V., & Zito, W. (2013). Foye’s principles of medicinal chemistry. Philadelphia: Wolters Kluwer, Lippincott Williams & Wilkins. Machulek, A., Oliveira, S. C., Osugi, M. E., Ferreira, V. S., Quina, F. H., Dantas, R. F., ... Nogueira, J. A. (2013). Application of different advanced oxidation processes for the degradation of organic pollutants. Mahmoodi, N. M., Hayati, B., Arami, M., & Lan, C. (2011). Adsorption of textile dyes on Pine Cone from colored wastewater: Kinetic, equilibrium and thermodynamic studies. Desalination, 268(1-3), 117–125. Mallick, A., Mohapatra, B. C., & Sarangi, N. (2014). Acute toxicity of calcium chloride on different stages (egg, spawn, fry and fingerling) of Rohu (Labeo Rohita, Hamilton). Research Journal of Animal, Veterinary and Fishery Sciences, 2(8), 11–16. Mancell-Egala, W. A. S. K., Su, C., Takacs, I., Novak, J. T., Kinnear, D. J., Murthy, S. N., ... De Clippeleir, H. (2017). Settling regimen transitions quantify solid separation