Nanocellulose and Graphene Oxide Aerogels for Adsorption and Removal Methylene Blue from an Aqueous Environment.

The characteristics of aerogel materials such as the low density and large surface area enable them to adsorb large amounts of substances, so they show great potential for application in industrial wastewater treatment. Herein, using a combination of completely environmentally friendly materials such as cellulose nanofibers (CNFs) extracted from the petioles of the nipa palm tree and graphene oxide (GO) fabricated by simple solvent evaporation, a composite aerogel was prepared by a freeze-drying method. The obtained aerogel possessed a light density of 0.0264 g/cm3 and a porosity of more than 98.2%. It was able to withstand a weight as much as 2500 times with the maximum force (1479.5 N) to break up 0.2 g of an aerogel by compression strength testing and was stable in the aquatic environment, enabling it to be reused five times with an adsorption capacity over 90%. The CNF/GO aerogel can recover higher than 85% after 30 consecutive compression recovery cycles, which is convenient for the reusability of this material in wastewater treatments. The obtained aerogel also showed a good interaction between the component phases, a high thermal stability, a 3D network structure combined with thin walls and pores with a large specific surface area. In addition, the aerogel also exhibited a fast adsorption rate for methylene blue (MB) adsorption, a type of waste from the textile industry that pollutes water sources, and it can adsorb more than 99% MB in water in less than 20 min. The excellent adsorption of MB onto the CNF/GO aerogel was driven by electrostatic interactions, which agreed with the pseudo-second-order kinetic model with a correlation coefficient R 2 = 0.9978. The initial results show that the CNF/GO aerogel is a highly durable "green" light material that might be applied in the treatment of domestic organic waste water and is completely recoverable and reusable.

Introduction

There is a higher demand for esthetic materials; more pigments and dyes have been used in textile and dye industries. In particular, organic dyes are always highly appreciated for their esthetics, gloss and diffusing ability in fabrics compared to inorganic pigments.[1,2] In the process of using these organic dyes, there is always an amount of organic dyes that are lost in wastewater and discharged into the environment through rivers. Organic dyes are mostly chemicals that are considered toxic to the ecosystems around rivers. The wastewater needs to be treated to effectively remove the dye color in order to protect the environment. Methylene blue (MB, C16H18N3SCl) is a kind of organic dye; it is a blue cationic thiazine dye used widely in the textile industry with a heterocyclic aromatic structure. MB has high solubility in water, and it is difficult to decompose naturally in water due to its complex structure.[2] If MB is not removed from wastewater, it will go into the water source and cause significant harm to the environment. MB is known as a popular organic dye used in textiles, dyeing, printing, homestead, and coating for paper stock.[3,4] MB also acts as a mild antiseptic when diluted in water and applied on the skin. When MB exists in high concentrations in water, it can interfere with oxygen solubility, affecting the habitat of aquatic species and the self-cleaning process of water.[5,6] Consequently, wastewater from factories has to extrude MB before discharging.

Many methods exist for dye removal, including adsorption,[7,8] encapsulation,[9] degradation,[10] and photodegradation.[11,12] Using adsorbents with a porous structure and a large surface area are one of the common methods commonly used for the treatment of MB wastewater problems.[13,14] This method is quite simple, inexpensive, and sensitive to harmful dyes. The traditional adsorbents are often used as activated carbon derived from plants,[15−17] zeolite,[18,19] montmorillonite,[20−22] biochar from bagasse, rice husks,[23,24] or the aerogels from cellulose.[25−27] However, these materials are hydrophilic materials and well dispersed in aquatic environments, so they could be released into water partially during adsorption. This reduces their strength and pieces of adsorbents, which are potential risks of water pollution. The reuse of these adsorbent materials for subsequent cycles with good stability adsorption capacity is also a challenge.

The cellulose aerogel is an adsorbent of most recent attention in the adsorption of polar organic dyes due to the interactions between the functional groups on cellulose and these dyes. Cellulose is known as a natural polymer that is renewable, low-cost, biodegradable, and naturally abundant.[28] The structure of cellulose contains many hydroxyl groups. Consequently, it is easy to create internal and intermolecular hydrogen bonds with water, creating favorable conditions for this material to exist in the form of hydrogels, especially when they are at a nanometer size. The nanocellulose aerogels can be fabricated for many applications by many different methods, of which the vacuum freeze-drying technique is the simplest and most popular technique. The nanocellulose aerogel has a highly porous structure, light weight, and a large surface area, making it very suitable for adsorption applications.[29] The surface of the nanocellulose aerogel has good polarity, so it has good interactions with polar organic dyes. However, the nanocellulose aerogels have a limited sub-resolution[30] and often have to be latticed or reinforced with other materials in adsorption applications. In the water environment, the research to improve the adsorption capacity of dye wastewater for this type of adsorbent is very interesting to expand its applicability.

In addition to activated carbon, other carbon materials almost all have high adsorption capacity due to the characteristics of the graphite lattice structure and high surface areas, such as graphene,[31,32] carbon nanotubes,[33,34] and graphene oxide (GO).[35−37] Graphene/matrix composites have been shown to have excellent mechanical properties, including high compressive and flexural strength,[38] Young’s modulus and compressive strength,[39] and controllable pores and mechanical flexibility.[40] GO nanosheets have a special 2D structure with a high surface area, and when in the aerogel form, the GO aerogel has a highly porous 3D structure with a low density. Also, these two structures of GO should be applied as direct adsorbents for heavy metals and MB.[35,36,41] The GO aerogel material shows a high adsorption efficiency for MB with a maximum adsorption capacity of 416.6667 mg/g.[35] However, GO disperses very well in the natural aquatic environment due to the abundant oxygen surface that reduced the reusability of the GO aerogel in wastewater treatment. Moreover, this process has a hidden risk of secondary pollution[42] if the adsorbents are not recovered completely after adsorption. When GO was reinforced by agar in the agar/GO aerogel through hydrogen bonds combined with dehydration in the interaction between oxygen-containing functional groups on agar and GO during hydrogel preparation, the adsorption efficiency of the agar/GO composite aerogel (AGO) was increased to 578 mg/g according to Chen et al.[43] The porous GO/polyacrylic acid aerogel was polymerized to remove MB, crystal violet (CV), methyl orange (MO), and rhodamine B from wastewater. It exhibited significant adsorption capacities for CV and MB of about 851.31 and 771.14 mg g–1, respectively.[44] On the other hand, regenerated cellulose combined with GO (0.5 %wt) was initially shown to be effective in improving the adsorption on MB at concentrations of 5–40 mg/L. This aerogel has a good stable adsorption rate at 90.5% after being used five times under oscillatory adsorption. In recent years, there have been some publications on aerogels based on nanocellulose and GO sheets applied in the wastewater treatment.[45−48] However, in this study, this is the first time an aerogel is prepared based on nanocellulose fibers (CNFs) extracted from the petioles of the nipa palm tree, which is a byproduct. The extraction process includes a simple mechanical process and bleaching without hydrolysis to limit the use of chemicals and obtain cellulose fibers at the nanometer scale. On the other hand, GO was added to CNFs with a rather low content but still showed significant improvements in the strength and adsorption capacity of the CNF/GO aerogel. The resulting aerogel is a promising good adsorption capacity material for the removal of MB through electrostatic interactions between the aerogel and the cationic dye in environmental engineering.

Results and Discussion

Characterizations of GO, CNF, and CNF/GO Aerogels

The change between the amorphous phase and the crystalline phase in the structure of raw cellulose cells and CNFs has been evaluated by X-ray diffraction (XRD). Figure shows the main peaks for the crystal lattice (002) of cells and CNFs in the range of 2θ = 22.2 and 22.4°, respectively. This peak signal of CNFs is sharper and clearer than that of cells. This indicates an increase in the lattice surface (002) when the yarn is chemically and mechanically treated. On the other hand, the XRD patterns also show a slower diffraction peak at 2θ = 16° for the plane (110) of a crystal structure of the cellulose-I structure.[49] The crystallinity indexes calculated according to the Segal method are 55.74 for cells and 89.04% for CNFs. The increase in the crystallinity index occurs during the mechanical and chemical treatment of CNFs. The spacing for the signal peak at 2θ = 26.4° in the XRD pattern of graphite is 3.37 Å. In the XRD pattern of GO, there is a high-intensity peak at 2θ = 11°, which is typically evident in the (001) crystal plane of GO with a spacing of 8.06 Å. This result shows an increase in the number of oxygen-containing groups on the surface and edge of each GO sheet after being oxidized from graphite, which increased the distance between the layers. The XRD pattern of CNF/GO shows the peaks at 2θ = 22.4 and 16°, which characterizes the structure of cellulose I in CNFs with a much-reduced intensity compared to those of the XRD pattern of CNFs. In addition, the characteristic peak at 2θ = 11° corresponding to d001 of GO is no longer observable. This result indicates that there is a reciprocal interaction between CNFs and GO. The cellulose chains of CNFs creep in the structure of GO sheets, so this peak does not appear in the XRD diagram of CNF/GO.

Figure 1

XRD patterns for graphite, GO, cells, CNFs, and CNF/GO.

The chemical structures of cells and CNFs from the nipa palm tree were obtained by Fourier transform infrared (FTIR) spectra in Figure . It is almost similar in the shapes of the spectra of these samples. The peaks at 1100 cm–1 were linked to the C–O–C stretching vibration of the pyranose ring skeleton in cellulose molecules. The bands around 1640 cm–1 indicated the O–H bending vibration of absorbed water due to absorbed water in the cellulose molecules.[50] There was a different wavelength at 1700 cm–1 in the CNF spectrum compared to the cell spectrum. This is attributed to the reduction of the stretching vibrations of the C=O bond on CNFs after the treatment. The peak at 1700 cm–1 in the CNF spectrum has a weaker and not sharper signal than that of cells in FTIR due to removal of lignin-containing ester bonds in the structure of the raw nipa palm fiber. There was a broader band around 3400 cm–1 in the FTIR spectrum of cells than the CNF due to the increase of the free stretching vibration of OH groups in the structure of the cellulose during the extracting process.

Figure 2

FTIR spectra of cells, CNFs, GO, and CNF/GO.

Moreover, all the spectra in Figure presented a peak at 2900 cm–1, which resulted from the C–H stretching vibration of the alkyl or aliphatic in the structure of GO and cellulose. In the GO spectrum, the characteristic peaks of GO located at 1714, 1586, and 1031 cm–1 are due to C=O in carboxylic acid and carbonyl moieties, C–OH, and C–O vibrations, respectively. The FTIR spectrum of the CNF/GO aerogel is similar to that of CNF and GO. Simultaneously, the peak around 3400 cm–1 decreased remarkably and shifted to a high wavenumber. These FTIR and XRD results indicate that there is a strong interaction between cellulose chains and GO sheets.

Figure a shows the scanning electron microscopy (SEM) image of cells with the surface covered with a lot of components of plant fibers such as hemicellulose, lignin, and other impurities. The surface of the cell was full with fewer defects, and its diameter was about 50–100 μm. Meanwhile, the field emission scanning electron microscopy (FE-SEM) image of CNFs shows that the rod-like morphology of CNFs after being treated from cells had a cleaner and purer surface, and the diameter of CNFs significantly reduced to around 20–80 nm, with more than 82% of the fibers in the 20–50 nm scale (Figure b). However, cellulose microfibrils of CNFs were still agglomerated in some places which have a larger diameter due to the interaction force between cellulose fibers after treatment to remove impurities covering the fiber surface as hydrogen bonds and van der Waals force. With this result, a good extraction of the cellulose microfibers from the raw fibers of the nipa palm trees is shown.

Figure 3

SEM image of cells (a) and FE-SEM image of CNFs (b).

Figure shows that all aerogels are porous and have a light structure. The CNF aerogel is white (Figure a), and the GO/CNF aerogel (Figure g) is brown because of the color of the GO aerogel (Figure d). However, the resulting aerogels have different pore structures in SEM and OM images. These differences depend on their respective compositions and the interaction between these compositions in the aerogel. The CNF aerogel has a heterogeneous porous structure with cycle pores with an average diameter of about 200 μm (Figure b,c), and this structure is similar to the structure of CNFs in other publications.[51−53] The GO aerogel has a 2D porous structure with parallel sheets which have a large surface area (Figure e,f). The interaction between GO and CNFs in the CNF/GO aerogel creates a plate-like unidirectional porous structure in the GO/CNF aerogel. Figure h,i shows that GO nanosheets arranged in parallel combine to some small “bridges” of CNF chains to constitute disordered small pores of the CNF/GO aerogel. This structural difference played an important role in influencing the strength and density of the aerogel. Figure a,b shows images of a GO/CNF aerogel placed on flowers showing the super lightweight aerogel and high compressive strength of the GO/CNF aerogel when it can withstand a weight more than 1000 times that of the aerogel without breaking it. The densities of the aerogel of CNFs, GO, and GO/CNF are shown in Table . The CNF/GO aerogel has the lowest density due to the mixed porous structure. Besides, the porous structure of the CNF/GO aerogel also gives the highest porosity with 98.2%. The interaction of GO and CNFs in the CNF/GO aerogel not only creates a material with a lighter and more porous structure but also improves the maximum compressive force and reduces the strain of the sample (Table ) compared with the monoporous structure of the GO aerogel or CNF aerogel. The resulting CNF/GO aerogel reinforced with GO showed the same density and higher maximum compressive force, but the porosity was slightly lower than those of CNF aerogels cross-linked with citric acid.[54,55]

Figure 4

Photographs (a,d,g), SEM images (b,e,h), and cross-sectional OM images (c,f,i) of CNF, GO, and CNF/GO aerogels.

Figure 5

Lightness (a) and compressive strength (b) of the CNF/GO aerogel (0.2 g).

Table 1

Physical Characteristics of CNF, GO, and CNF/GO Aerogels

samples	density (g/cm3)	porosity (%)	max forcea (N)	elastica (N/mm2)
CNF	0.042 ± 0.02	96.8	2146.2	0.107
GO	0.039 ± 0.01	97.0	314.2	0.088
CNF/GO	0.0264 ± 0.02	98.2	1479.5	0.189

Obtained from compression strength testing.

Mechanical and Thermal Properties of GO, CNF, and CNF/GO Aerogels

The result of the compressive recovery test of the CNF/GO aerogel with strain 70% after 30 compression-recovery cycles is shown in Figure a. The recovery of the aerogel was quite stable after every five cycles with the height decreasing on average about 2.5%. After 30 compression-recovery cycles, this has only decreased about 15% of the height of the CNF/GO aerogel. This shows the durability and the reusability of the CNF/GO aerogel, even though this material is very porous and light (Figure ).

Figure 6

(a) Compress recovery of the CNF/GO aerogel at 70% strain after 30 cycles and (b) typical compress stress and strain curves of GO, CNF, and CNF/GO aerogels.

A compression test at room temperature was used in order to evaluate the durability of the aerogels. Figure b shows the stress and strain curves of the three aerogels of CNF, GO, and CNF/GO at room temperature performed on the universal testing machine for comparison. All curves can be divided into two different regions: a linear elastic region where the stress increases slowly when the strain increases and a densification region with a strong increase in stress and a high slope. The CNF/GO aerogel showed higher stress than other aerogels in both these regions, in which at the strain 50% region, the CNF/GO aerogel gave stress higher than that of the CNF aerogel and GO aerogel by 3 and 7 times, respectively. This value is almost stable in the remaining strain regions. The aerogel of CNFs and GO have almost the same stress in the low-strain region below 20%, and the CNF aerogel has stress higher than that of the GO aerogel in the remaining strain regions. This result once again shows the effectiveness of reinforcement the CNF aerogel by GO to create a material having a porous, lightweight, and high-strength structure. The −OH groups and hydrophobic −CH moieties of CNFs make CNFs behave as an amphiphilic compound,[45] in which the −OH groups on cellulose will interact with the functional groups on the edges of GO, while the hydrophobic part of CNFs will interact with the plane of GO sheets during the preparation of the CNF/GO aerogel. They adhere to the CNF and the GO sheets. This increases the dispersion of GO on CNFs as well as helps to stabilize the structure of CNFs in the porous form of aerogel materials. This interaction is also observed in the XRD, FTIR, and SEM analyses.

Thermogravimetric analysis (TGA) is used to investigate the thermal stability of the precursors during carbonization and represents a partial interaction between these components in the final CNF/GO aerogel. In Figure , the TGA curves of CNF and CNF/GO are similar to the TGA curve of raw cellulose, and the major weight loss occurred in the temperature range of 220–450 °C. This is a typical behavior of cellulosic materials,[56,57] and this thermal decomposition relates to the depolymerization of cellulose, which is carried out simultaneously with dehydration and release of the volatile compounds such as carbon dioxide, methanol, and acetic acid. The cell had lost the maximum weight at 332 °C, and the weight yield at 700 °C is about 34.4%. After going through the treatment, the maximum decomposition temperature of CNFs was lower by about 10% than that of cells and the residue at 700 °C was much lower than the raw fibers, around 22.4%. This may be explained by the presence of impure components in the raw nipa palm fiber as lignin. Lignin is considered a biopolymer having the ability to protect cellulose chains in the cell wall of woody plants under external heat effects due to its high capability of char formation.[58,59] Meanwhile, CNFs were removed with most of the lignin and hemicellulose throughout the treatment, so the thermal stability and final ash content of CNFs are lower than those of cells. In addition, the defects existing on CNFs during the mechanical and chemical treatment also contribute to reducing the thermal stability of CNFs. On the other hand, the TGA curve of GO shows an early weight loss below 250 °C with an initial 20% weight loss due to the thermal decomposition of oxygen-containing functional groups simultaneously with dehydration and release of volatile compounds.[60,61] The remaining functional groups on GO continue to undergo thermal decomposition with more steepness in the next range from 200 to 250 °C by more than 20% weight loss.

Figure 7

TGA curves of cells, CNFs, GO, and the CNF/GO aerogel.

The TGA curve of GO was quite smooth and less steep after 250 °C because of the thermal stability of the carbon skeleton remaining in the structure of GO. The TGA curve of CNF/GO was similar to that of CNFs, with a maximum decomposition temperature at 292 °C and the yield weight at 700 °C about 25%. This result shows that GO sheets are surrounded by CNFs (SEM images), so the thermal decomposition of oxygen-containing functional groups on GO does not affect the thermal stability of CNF/GO in the temperature region below 200 °C. This result also partly reflects the interaction between GO and CNFs in the resulting CNF/GO aerogel.

Study of MB Adsorption

The dynamic adsorption capacity of MB in water onto the aerogels shown in Figure a was calculated. The CNF/GO aerogel removed almost the entire amount of MB in the shortest contact time, adsorbed more than 99% of MB at the initial MB concentration of 20 mg/L within less than 40 min. Meanwhile, the adsorption capacity of MB in water at 40 min was only more than 25 and 70% onto the CNF aerogel and GO aerogel, respectively. The adsorption capacity of CNFs and the GO aerogel steadily increased until MB saturation reached after 90 min to contact the adsorbed substance. After 90 min, the amount of MB in water was removed by more than 95 and 64% for the GO aerogel and CNF aerogel, respectively. Thus, the dynamic adsorption capacity and the adsorption rate of MB in water onto GO and the CNF aerogel were much lower than that of the CNF/GO aerogel. This result shows that the combination of the two components of CNFs and GO showed significant effects in removing the organic dye MB in water within a short time.

Figure 8

MB adsorption capacity in water vs time plot (a); MB solutions after adsorption for 90 min onto aerogels of CNFs, GO, and CNF/GO (b); MB adsorption capacity in water (c) and MB solutions after 1, 4, and 6 recycling times onto aerogel CNF/GO (d).

All aerogels float on water because they have a much lower density than water. The density gradually increased with the time aerogels contact to MB in water. However, it was not much, and the aerogels remained floating near the water surface during the adsorption time until the aerogels reached the MB saturation state after 90 min in Figure b. After 90 min, the amount of MB in water reduced sharply compared to the original through the fading of the blue color of MB in the water (Figure b). Here, the water adsorbed by the CNF/GO aerogel was almost transparent with complete disappearance of the blue color of MB. Water after adsorption by the GO aerogel also showed a significant removal of MB through the blue color of MB fading with contact time. However, the GO aerogel was disintegrated in water when the adsorption time was longer than 60 min due to the hydrophilic structure of GO and the poor strength in water of the resulting aerogel. Although the GO aerogel only adsorbed about 90% of MB in water (Figure a) at 60 min, this leads to a subsequent reduction in the adsorption capacity of the GO aerogel. The CNF aerogel had the lowest adsorption efficiency with MB in water when the part of water after 90 min still clearly showed the MB color and the water containing a lot of debris of the CNF aerogel is separated from the initial CNF aerogel during the adsorption process. This significantly affects the ability to remove the adsorbents. These results also indicate that the reinforcement of GO into the CNF aerogel significantly contributes to increasing the durability and adsorption capacity of the CNF/GO aerogel and overcomes the secondary pollution from the adsorbent during the treatment water process.

The CNF/GO aerogel after the saturated adsorption of MB was desorbed by ethanol and then dried to re-adsorb MB in water in the next cycles. Figure c shows the relatively stable adsorption capacity of MB in water onto the CNF/GO aerogel after six cycles. The MB adsorption capacity decreased about 2% after each cycle and decreased sharply at cycle 6 with about 6%. This result also shows the strength of the CNF/GO aerogel after many treatments as this aerogel can be completely reused about five times with the ability to remove over 90% of MB in water. The decrease in MB adsorption capacity of the CNF/GO aerogel was shown by the darkening of the blue color of MB solution as the number of re-adsorption times was increased (Figure d).

The main adsorption mechanism of CNF, GO, and CNF/GO aerogels is the electrostatic interaction as mentioned in many previous literature studies.[45,46] In this study, the zeta potential of CNFs is −1.4 mV, while this value of CNF/GO is −111.3 mV. This result indicates that the adsorbed mechanism of MB onto the surface of the CNF/GO aerogel is via the electrostatic interaction. On the other hand, the presence of GO sheets further increased the structural strength of CNFs through the interactions between functional groups on CNF and GO surfaces. GO also plays a role as an electron acceptor through the π–π stacking interactions with the conjugated structures of MB (unsaturated double bonds).[45] All the above results explain why MB adsorbs well on the surface of CNF/GO.

Adsorption Kinetics (Pseudo-First Order and Pseudo-Second Order)

The values of qe, k1, and k2 in Table were assessed from the linear plots of Figure a,b. The adsorption kinetic models can be used to predict the equilibrium adsorption capacity and clarify the adsorption mechanism of MB onto the surface of the CNF/GO aerogel. The MB dye is known as a cationic dye dispersed in water, and the electrostatic interactions between MB and the adsorbent were formed when MB was in close contact with adsorption sites such as functional groups (−OH, −COOH) on the surface of CNFs and GO in the adsorbent. The amount of MB adsorbed on this adsorbent will gradually increase in contact time until it reaches equilibrium. MB adsorption kinetics onto the CNF/GO aerogel was obtained by two pseudo-first-order kinetic models and pseudo-second-order models after the experimental values q and qe were calculated. The experimental value of qe, 9.94 mg·g–1, did not agree well with the calculated one as obtained from the linear plots (Figure a), and the value of the correlation coefficient (R2) was relatively low (0.899 in Table ). This indicates that the adsorption process does not comply with the pseudo-first-order rate model. The pseudo-second-order model showed a better fit to describe the adsorption of MB onto CNF/GO aerogels with the calculated value qe closer to the experimental value qe (9.94 mg·g–1) and a higher correlation coefficient (R2 = 0.9978 in Table ) than the first-order kinetics model.

Table 2

Estimated Kinetic Parameters of the Two Adsorption Models for MB at a Concentration of 20 mg/L on the CNF/GO Aerogel

pseudo-first order	qe (mg g–1)	4.455
	k1 (min–1)	0.0842
	R2	0.899
pseudo-second order	qe (mg g–1)	10.48
	k2 (g mg–1 min–1)	0.0954
	v0 (mg g–1 min–1)	10.48
	R2	0.9978

Figure 9

Pseudo-first-order kinetics (a) and pseudo-second-order kinetics (b) for adsorption of MB by the CNF/GO aerogel.

This result was consistent with previous results onto adsorbents from the CNF aerogel[62,63] and nanocellulose/GO aerogel.[45,46] However, the CNF/GO aerogel obtained values of calculated qe and v0 from the pseudo-second-order model much lower than those of other studies; it may be due to the difference between the methods of the preparation aerogel, especially the orientation or no orientation of the porous structure inside the aerogels. This result initially shows the adsorption capacity of a kind of aerogel made from CNFs from the nipa palm tree with the reinforcement of GO.

The CNF/GO aerogel in this study is considered a new adsorbent based on nanocellulose extracted from the petioles of the nipa palm tree, which is a byproduct (Table ). The CNF/GO aerogel showed an efficient adsorption capacity of the cationic MB dye in an aqueous environment with a high adsorption rate. In addition, the CNF/GO aerogel was stable after more than 30 compression cycles, and the ability to effectively reuse after five cycles with the removal efficiency of MB from wastewater still over 90%.

Table 3

Studies on the Adsorbent of Arogel Materials from Different Sources

dye	aerogel materials	results	refs
MB	GO/nanocellulose (kraft pulp)	removal efficiency after three cycles: 98%	(45)
	25 wt % GO
MB	GO/carboxymethyl cellulose (CMC)	removal efficiency after 180 min: 97.2%	(46)
	GO/CMC: 2:1 (w/w)
MB	GO/cellulose nanofibril (kenaf core)	removal efficiency after five cycles: 52.2%	(47)
	20 wt % GO and 1 wt % of Fe(III) ion
MB	GO/nanocellulose (Nypa palm petioles)	removal efficiency after five cycles: 91%	this work
	13 wt % GO	compress recovery of the aerogel at 70% strain after 30 cycles: 85%

Selective Adsorption of the CNF/GO Aerogel

The selective adsorption of the CNF/GO aerogel is shown in Figure .

Figure 10

Adsorption of the MB/MO mixture on the CNF/GO aerogel.

This aerogel material showed a rapid adsorption of MB in water with the intensity of the peak at 660 nm in UV–vis. This peak remarkably reduced after 20 min contact. The solution only retained the bright-yellow color of MO, while the specific blue color of MB was no longer observed after 20 min. On the other hand, there was little change in the adsorption signal at 463 nm associated with MO even after 60 min exposure. These results indicated that there was an effective removal of MB from the MB/MO mixture. The CNF/GO aerogel showed good selective adsorption behavior with the cationic dye MB.

Conclusions

The mechanical pretreatment combined with alkaline treatment and double bleaching created cellulose microfibrils with high crystallinity more than 89.81% and the main fiber diameters in the range of 20–50 nm, with aspect ratio hundreds of times from the petioles of the nipa palm tree. GO was successfully synthesized from graphite by the modified Hummer method. The CNF/GO aerogel obtained by the freeze-drying technique was ultralight with a porous structure, a low density, high porosity, and high strength with the compressive stress at a deformation of 80% reaching 383 kPa. The CNF/GO aerogel still retains good structural recovery after 30 consecutive compression and recovery cycles, showing durability in structure despite high porosity. After investigating the ability to remove MB from water, the CNF/GO aerogel not only showed a good MB adsorption capacity (99%) in a short contact time (less than 20 min) but also still retained the stable structure during the adsorption compared to CNF and GO aerogels. This adsorbent also showed an efficient adsorption capacity (91%) after five regeneration cycles. In this study, we also performed the adsorption mechanism through electrostatic interactions between the surface of the CNF/GO aerogel and the MB cationic dye by demonstrating the adsorption model of CNF/GO suitable for the pseudo-second-order kinetic adsorption model and zeta potential result. The CNF/GO aerogel showed a rather good selective adsorption behavior with the cationic dye MB than the anionic dye MO. Therefore, it is considered as a material with potential applications in the field of adsorption and treatment of water contaminated with organic wastes such as cationic dyes.

Experimental Section

Materials

CNFs are extracted from powder grounded from the petioles of the nipa palm tree, Ho Chi Minh City, Vietnam. Graphite powder is supplied by Sigma-Aldrich, Singapore. Concentrated sulfuric acid (≥98%), phosphoric acid (≥85%), concentrated acetic acid (≥99%), solution hydrogen peroxide (∼33% in H2O), MB (≥85%), sodium hydroxide, sodium hypochlorite, MO, and potassium permanganate were purchased from Guangdong Guanghua Sci-Tech Company, China. All materials and chemicals were analytical and used without further purification. Distilled water was used in all experiments.

Synthesis of CNFs

Purification of the Nypa fruticans powder was performed with small chemical modifications. First, in order to remove pectin and residual starch, the Nypa palm powder was dispersed in 500 mL of distilled water and stirred for 1 h at 90 °C. Second, the suspension was treated with 5 wt % aqueous NaOH at 90 °C for 2 h to remove hemicellulose. Third, the samples were treated with a solution of NaClO, CH3COOH, and distilled water at 75 °C for 2 h to remove the lignin. The third steps were repeated twice to get highly purified cellulose. Finally, the cellulose was centrifuged with distilled water until pH = 7 and ultrasonication treatment in an ultrasonic generator at 30 min to obtain the nanocellulose suspension. The prepared cellulose suspensions were stored at 4 °C before future utilization.

Synthesis of CNF/GO

GO was synthesized from graphite powder using the modified Hummers method.[64] The CNF suspension (4 g, 2.5 wt %) and GO solution (1 g, 1.5 wt %) were mixed, vigorously stirred in 1 h, and then sonicated in an ultrasonic generator for 30 min. Then, the mixture was poured into a cylindrical mold (2.5 cm × 5 cm) and was stored overnight in a refrigerator at 4 °C for pre-cooling to avoid macroscopic cracking during the dry-freezing step. Then, the mold was transferred to a freeze dryer and lyophilized for 48 h at −40 °C to obtain the CNF/GO aerogel. Similarly, CNF and GO aerogels were prepared by the freeze-drying method.

Sample Characterization

All samples after preparation were stored in a desiccator with humidity <30% before testing. The microstructure and elemental analyses of CNF, GO, and CNF/GO aerogels were observed using SEM (JSM-6480LV, JEOL) and FE-SEM (S-4800, Hitachi, Japan). ImageJ software was used to calculate the diameter of CNFs in the FE-SEM image. FTIR spectroscopy (Tenser 27, Bruker, Germany) was performed in the range of 400–4000 cm–1, with a 4 cm–1 resolution at room temperature. The XRD pattern was obtained on a D8 ADVANCE under analytical conditions at room temperature with 30–35% relative humidity; the tube voltage is 40 kV, and the tube current is 40 mA, with Cu Kα radiation (plate filter Ni). The thermal properties of all samples were analyzed on a LABSYS evo TG–DSC, SETARAM. Samples were analyzed in nitrogen and heated from room temperature to 800 °C at a heating rate of 10 °C/min. The compression test was carried out using a Shimadzu-Autograph ASH-X. The test sample had a diameter of 25 mm and a height of 50 mm, and the loading rate was set to 1 mm/min. The adsorption capacity of MB was investigated using a UV–vis spectrophotometer (JASCO V-730, Japan) at wavelengths from 300 to 800 nm. The zeta potential (ζ) of CNFs and CNF/GO was measured with the dynamic light scattering technique using a HORIBA, Zetasizer Nano series.

The densities of the aerogels were calculated by measuring the mass and volume of the aerogels. The actual density (ρ) of the aerogel was calculated by eq where ρ is the density of the aerogel, m is the mass of the aerogel, and v is the volume of the aerogel [obtained by v = (π × d2 × h)/4, where d and h represent the diameter and height of aerogel, respectively].

Equation is used to calculate the density (ρs) of the solid materialwhere WCNF and WGO are the weight percentages of CNFs and GO in the aerogel and ρCNF and ρGO are the solid densities of CNFs and GO, respectively. The densities of CNFs and GO used for this research are 1692 and 1585 kg/m3, respectively. The porosity of aerogels (P) was determined according to eq where P is the porosity of the aerogel, ρ is the density of the aerogel, and ρs is the density of the solid material.

For the kinetic research, the aerogels (0.1 g) were immersed into 50 mL of the aqueous solution of MB (neutral pH, C0 = 20 mg/L) at room temperature and with stirring at 300 rpm. The current concentration of MB solution was measured using a UV–visible spectrophotometer (JASCO V-730, Japan) at 664 nm after the preset time interval, t. The elimination efficiency ( H%) of the dye, the adsorption capacity at time t (q, mg/g), and the equilibrium adsorption capacity (qe, mg/g) of aerogels were obtained using eqs –6, respectively, as follows:where m is the weight (g) of the dry aerogel, v is the volume (L) of the MB solution, co is the initial dye concentration (mg/L), c is the residual dye concentration at a given time (mg/L), and ce (mg/L) is the equilibrium concentration of MB solution.

Adsorption Kinetics (Pseudo-First Order and Pseudo-Second Order)

In the adsorption mechanism, kinetic prediction of the adsorption capacity and adsorption time is crucial. Kinetic data were evaluated using pseudo-first-order and pseudo-second-order models to describe the relationship of MB adsorption on time. The pseudo-first-order kinetic model and the second-order kinetic model can be expressed by eq and 8(65)where qe (mg/g) represents the amount of adsorbed MB at equilibrium, q (mg/g) represents the amount of adsorbed MB at time t, and k1 (min–1) and k2 (g·mg–1·min–1) represent the pseudo-first-order and pseudo-second-order rate constants, respectively.

The initial adsorption rate v0 at t = 0 could be calculated using eq

Selective Adsorption of the CNF/GO Aerogel on the Dye Solution Mixture

The adsorption selectivity of the CNF/GO aerogel was determined by the adsorption analysis of the MB and MO mixture on the CNF/GO aerogel in an aqueous environment. Typically, 0.1 g of the CNF/GO aerogel was added to 100 mL of the MO/MB mixture (1:1 v/v) at room temperature and pH neutral in 60 min. The residual dye concentration in the solution was determined by UV–vis after certain time points.