Recycling of Waste Cotton Sheets into Three-Dimensional Biodegradable Carriers for Removal of Methylene Blue.

Waste cotton sheets (WCS) are promising cellulose sources due to their high content of cellulose and large amount of disposal every year, which could be recycled and employed as low-cost structural materials. The present work aims at investigating the efficacy of hydrogel adsorbents prepared from regenerated WCS as the carriers of activated carbon (AC) for treating the dye-contaminated water. Activated WCS was directly dissolved in lithium chloride/N,N-dimethylacetamide (LiCl/DMAc) solvent and then regenerated into cellulose hydrogels, which were employed as three-dimensional biodegradable matrices for loading an extremely high content of AC (up to 5000%). The morphology and properties of resultant adsorbents were studied in detail. The results showed that different washing methods and contents of AC and cellulose had obvious effects on water contents, mechanical properties, and adsorption capacities of AC/WCS hydrogels. Especially, the hydrogels containing high AC content washed by gradient ethanol solvent exhibited outstanding compressive strengths of up to 3.0 MPa at 60% strain, while the adsorption capacity of 5000%AC/0.3CS toward a model dye methylene blue (MB, initial concentration of 200 mg/L) reached 174.71 mg/g at pH 6.9 and 35 °C. This was comparable to the adsorption capacity of original AC powders, while no AC powders were released from hydrogels to water. The adsorption of MB followed the Dubinin-Astakhov model and pseudo-first-order mechanism. Thermodynamic studies showed the spontaneous and endothermic nature of the overall physical adsorption process. Therefore, this work demonstrates the feasibility to recycle WCS into biodegradable carriers of functional compounds, and the AC/regenerated cellulose hydrogels have a high potential as a promising adsorbent with low-cost and convenient separation for dye removal from wastewater.

Introduction

About 25 million tons of cotton cloth is produced and used throughout the world every year, while a large amount of textile waste is generated simultaneously.[1] Only a small portion of waste cotton textile is used to produce mop, plush toys, and lower-grade cotton textile, and the rest is disposed via landfill or incineration, which cause serious environmental problems and are a great waste of cotton resources. Therefore, the high value-added applications of waste cotton textile have already become one of the hot topics in the scientific field.[2] It is an interesting objective to make use of cellulosic materials from such textile waste to help reduce environmental problems. Because the content of cellulose in cotton fibers is very high (upward 90%),[3−5] the recycle of such cellulosic waste for the preparation of hydrogel adsorbents is a suitable, low cost, and attractive substitute to conventional disposal methods.[6]

Every year, large quantities of contaminated wastewater are released into the environment by various industries, estimated to be around 70–200 tons per year.[7] Dyes are an important source of pollution for the environment and aqueous solutions for that they are used in a variety of industries such as textiles, cosmetics, leather, food, plastic, rubber, and so forth. Discharged dyes in wastewater pose significant threats to the aqueous environment and human health and have an economic impact. Dyes, including their degradation products, have complex structures that can be toxic to humans and other living things and cause mutations and cancers.[8] Therefore, the removal of toxic dyes in wastewater is an extremely urgent issue. Adsorption is considered a simple, low-cost, and eco-friendly technology to remove such dyes.[9] Activated carbon (AC) is a porous carbon material produced from raw vegetal or mineral materials by chemical or physical activation. Because of its very high specific area (up to 2000 m2/g), AC can be used as an adsorbent for various applications such as water and gas treatments. Molecules belonging to different classes, such as heavy metals and organics, can be removed by AC.[10] AC prepared from various precursors presented promising adsorption performance in dye removal processes. For example, AC made from rubber wood, oil palm waste, rubber wood sawdust, and sugar cane bagasse showed a maximum adsorption capacity of 1176 mg/g for acid blue 264,[11] 1845 mg/g for basic blue 1,[12] 1456 mg/g for Bismark Brown,[13] and 1060 mg/g for acid blue 25.[14] However, in practical application, AC powders cannot be easily handled and separated from an aqueous solution, making subsequent operations difficult and/or tedious.[15,16] When AC spreads into the water, it could eventually turn into the secondary wastes. Adsorbents with a suitable shape for easy operation are greatly required in industrial applications. Therefore, embedding AC powders in a proper matrix could maintain the high adsorption capacity of AC and restrict the mobility of free powders.[9,15] Hydrogels may be good candidates as matrices to fix AC by utilizing their characteristic features such as good formability, good permeability, and ability of incorporation of particles in their matrices.[9,17−19] In past decades, many researchers studied the combinations of polymers with AC for water treatment, such as gelatin and AC (256.41 mg g–1 for Rhodamine B),[20] polyacrylonitrile (PAN) and AC [187.79 mg g–1 for Cr(vi)],[21] chitosan and AC (1428 and 1370 mg g–1 for food blue 2 and food red 17, respectively),[22] and so forth. The AC powder can be immobilized in the hydrogels without dispersing in the water. Hydrogels are crosslinked polymers with a three-dimensional structure that can hold water in their structure. The AC powder can be immobilized in the hydrogels without dispersing in the water or impeding the adsorption of AC to the dyes.[23,24]

For the purpose of sustainable development and environmental protection, the demand of renewable and biosustainable resources is increasing.[25,26] Hydrogels prepared from natural polymers have captured extensive attention over the past decades because of their exceptional biocompatibility, nontoxicity, ease of gelation, and low-cost.[17,27−29] Cellulose, the most abundant natural polymer on earth, will become the main chemical resource in the future.[30] Cellulose having abundant hydroxyl groups can be used to prepare hydrogels easily with fascinating structures and properties. Therefore, cellulose hydrogels have been widely used in many applications such as tissue engineering, agriculture, sensor, and water purification.[31] Compared with natural cellulose, waste cotton fabrics are interestingly now on the rise due to the very high content of cellulose in cotton fibers,[5] and it is economical and eco-friendly to recycle such cellulosic wastes into value-added products for a sustainable future.[2,32,33]

In the past few decades, many researchers have successfully prepared hydrogel/additive composites for water purification; however, only low levels of additives could be embedded in the hydrogels, probably due to the poor network forming capacities. For example, AC powders were 50% of total weight of polyacrylamide/AC hydrogel (PAAm-FAc);[34] the maximum weight ratio of powdered AC(PAC) and carboxymethyl sago pulp (CMSP) in CMSP-PAC hydrogel was 10:1;[18] in cellulose acetate (CA) and AC composite membranes, the highest AC content was 20% of CA content;[35] the AC content in cellulose/AC composite monolith was 50% of the total composite;[9] in gelatin/AC composite beads, the GC/AC weight ratio was 27:10;[20] and so forth. Regenerated cellulose can act as a reliable matrix of AC.[15] To the best of our knowledge, the research on recycling waste cotton sheets (WCS) as the carrier of high amount of AC for dye removal has been seldom reported, which will allow the improvement of adsorption capacity per unit volume of adsorbents. This work aimed to develop cellulose hydrogels incorporated with high AC content and evaluate their performances as adsorbents for a model dye methylene blue (MB) adsorption from water, which could promote the recycle of waste cotton textile and help reduce water contaminants.

In this work, WCS were selected as the raw material, which was dissolved in N,N-dimethylacetamide/lithium chloride (DMAc/LiCl) solution and then regenerated into cellulose hydrogels. This hydrogel matrix was used to fix AC powders for dye adsorption from wastewater. Different regeneration baths were adopted to investigate their effects on the morphology of hydrogels.[36] The obtained composite hydrogels were characterized in relation to their structural characteristics and mechanical properties. Adsorption experiments were conducted to examine the effects of adsorbent dosage, pH, time, temperature, and initial dye concentration on the adsorption of MB. Adsorption isotherms and kinetic parameters were also evaluated in detail. Finally, the potential recyclability of the hydrogels was also performed.

Results and Discussion

Structure and Properties of WCS/AC Hydrogels

Both 0.3 and 0.652 g of activated WCS could be dissolved by 9% DMAc/LiCl (30 mL), giving highly transparent solutions with good fluidity (Figure c). The solution formed with 0.3 g of WCS had a lower viscosity, and the fluidity decreased when AC was added. The maximum amounts of AC that could be loaded in the solutions were 2000% of 0.652 g of WCS and 5000% of 0.3 g of WCS; otherwise, it was difficult to disperse the mixtures evenly by a magnetic stirrer. Gels formed after regeneration (Figure e), and AC was fixed tightly in the network. No AC powder was seen coming out from the gels when washing or using the samples, which meant that the regenerated cellulose hydrogels were a promising carrier. As far as we know, no other literature has reported a matrix with such high content of additives.[15,23]

Figure 1

Photographs of WCS (a), grinded WCS (b), transparent solution of WCS dissolved in DMAc/LiCl (c), WCS/AC suspension (d), and WCS/AC hydrogel (e).

Figure shows the SEM images of hydrogel surfaces and cross section. AC powder with various sizes and shapes was observed on the surfaces and interior of hydrogels, which were glued together by regenerated cellulose. The clump cavities resulted from rough surfaces could provide more reactive sites and let the dye molecules penetrate to reach to internal reactive sites during the adsorption process. The cross-sectional images of AC2000%/WCS0.3 and AC1000%/WCS0.652 with low AC contents are displayed in Figure a,e. The surface of AC particles was coated with WCS gel, and the coating became thinner as the AC content increased or the WCS amount reduced. The dye molecules are supposed to penetrate the thinner gel coating more easily than the thick one to approach the internal reactive sites.

Figure 2

SEM images of WCS/AC hydrogels (left – cross sections and right - surfaces): AC2000%/WCS0.3 (a,b), AC5000%/WCS0.3 (c,d), AC1000%/WCS0.652 (e,f), and AC2000%/WCS0.652 (g,h).

The X-ray diffraction patterns of AC, WCS, AC1000%/WCS0.652, and AC5000%/WCS0.3 are shown in Figure . For AC powder, typical peaks appeared (one sharp diffraction peak at 26° and one broad peak 43°), indicating the turbostratic structure of disordered carbon.[37] It was noticed that WCS showed a sharp crystalline peak at 22.7° (200) and two broad peaks at 14.8° (1 1(_) 0) and 16.4° (110), indicating a cellulose I crystalline structure.[25,38] The regenerated cellulose showed the characteristic diffraction pattern of cellulose II at 2θ around 20–24° with comparatively lower intensity and wider shape, which were attributed to the (002) and (10 1(_)) lattice planes of cellulose II.[39] It was evident that the cellulose I structure was completely lost during dissolution and regeneration. In AC5000%/WCS0.3 and AC1000%/WCS0.652, only diffraction pattern of AC was observed, which may be because the characteristic peak of cellulose II was masked by the high amount of AC embedded in the hydrogels. This was beneficial for the adsorption because liquid was more likely to penetrate into the cellulose II phase matrix.[40]

Figure 3

X-ray diffraction patterns of AC, WCS, regenerated cellulose, and AC1000%/WCS0.652 and AC5000%/WCS0.3 hydrogels.

During the cellulose regeneration process, DMAc and LiCl were replaced by a cellulose nonsolvent that was miscible with the LiCl/DMAc solution to trigger a phase separation. It was reported that polymer gels generally shrink in the solvents with lower solubility parameters, and the use of a solvent with high mutual affinity to the initial solvent can retain the initial volume better in comparison with solvent with low mutual affinity.[41] The utilization of water and ethanol at different ratios can cause different degrees of shrinkage, which will affect the structure and mechanical properties of hydrogels. Therefore, the effect of three washing methods on water content and mechanical properties was discussed. The water content defined here can be seen as a measure for the efficiency of volume preservation during the solvent washing. As shown in Figure a, all the samples treated with DI water had the lowest water contents, while the water contents of hydrogels treated by gradient ethanol washing were higher than those regenerated in 20% ethanol solution. It was because of that ethanol has a higher affinity to LiCl/DMAc solvent than water. In addition, regenerated cellulose without AC [0.3WCS and 0.652WCS in Figure a (0.3 g or 0.652 g WCS dissolved in 60 mL DMAc/LiCl solution)] had much higher water content than hydrogels with AC, and hydrogels with low AC content had higher water content than that with high AC content (AC1000%/WCS0.652 > AC2000%/WCS0.652, AC2000%/WCS0.3 > AC5000%/WCS0.3). This may be because the AC powder occupied more free volume of the gel matrix.

Figure 4

Water content (a) and compressive strength (b) of regenerated cellulose hydrogels prepared by different washing methods. Different asterisks on the top of columns represent significant differences among different washing methods (p < 0.05).

Good mechanical properties (such as compressive strength) of a hydrogel are the prerequisites of an adsorbent with excellent recyclability and wet stability.[42] The compressive strength at 60% strain of the hydrogels prepared by three different washing solvents is shown in Figure b. No fracture was observed when the samples were subjected to 60% strain, which indicated that the regenerated cellulose matrix could not only fix powder form additives but also provide necessary flexibility for handling. The hydrogels treated with DI water had the highest compressive strength, followed by the samples washed with 20% ethanol and gradient ethanol solutions. For hydrogels undergoing the same washing method, the compressive strength of regenerated cellulose without AC (0.3WCS and 0.652WCS) were much lower than hydrogels with AC and it increased remarkably with the increase in AC content. For example, the compressive strength of AC2000%/WCS0.652 (2.76 MPa) was approximately 3 times as high as that of AC1000%/WCS0.652 (0.97 MPa), while the compressive strength of AC5000%/WCS0.3 (3.01 MPa) was about 10 times as high as that of AC2000%/WCS0.3 (0.31 MPa). In addition, when no AC or AC content was relatively low (0.3WCS, 0.652WCS, AC1000%/WCS0.652, and AC2000%/WCS0.3), the hydrogels with higher WCS content had higher compressive strength than those with lower WCS content. The regenerated cellulose formed a three-dimensional network in the hydrogels that could resist the compressive deformation, and the higher WCS content resulted in a denser and stronger network. Furthermore, increasing the AC content in the hydrogels also improved their compressive strength.[43] It could be explained that adding AC into cellulose matrix improved the stress transfer properties between the hydrogels and AC. Additionally, the resistance created by AC was also favorable for reducing the hydrogel deformability.[17,44,45] However, the compressive strength of AC2000%/WCS0.652 and AC5000%/WCS0.3 had no significant difference, which indicated that a large amount of AC played a crucial role in enhancing the strength of hydrogels.[46]

Adsorption of MB

Based on the consideration of mechanical strength and amounts of cellulose and active adsorption sites, AC5000%/WCS0.3 washed by gradient ethanol solutions was selected for the dye adsorption tests. The effects of various parameters, such as contact time, adsorption temperature, solution pH, and initial dye concentration, on the removal efficiency of hydrogel toward MB were investigated by batch experiments. An example is shown in Figure that a piece of hydrogel could convert the dark blue solution into colorless.

Figure 5

Photograph of dye solutions before (left) and after (right) the adsorption process (0.16 g AC5000%/WCS0.3, 25 mL MB solution with initial concentrate of 200 mg L–1, 38 h, 35 °C).

Effect of pH on MB Adsorption

The effect of pH on MB adsorption is shown in Figure . It was reported that the transition of pH could change the net charge on the gel surface and affect the adsorption capacity.[15] The qe values of hydrogels increased from 157.6 to 168.5 mg/g when the pH value of solution changed from 4.0 to 7.0, and the pHzpc of cellulose hydrogels with AC was reported to be approximately 6.7.[15] It implied that the surface charges of hydrogels became positive when the solution pH was lower than pHzpc, resulting in stronger repulsion toward cationic MB molecules and lower MB removal.[15,47] However, the adsorption capacity at pH 4 was still high, indicating that the electrostatic interaction did not dominate the sorption mechanism of MB on the hydrogels.

Figure 6

Effect of pH on MB adsorption with AC5000%/WCS0.3. Different asterisks on the top of the columns indicate the significant difference (p < 0.05).

Because the effect of pH values above 7.0 on the adsorption capacity of hydrogel was negligible and the pH of the original MB solution was around 7.0, no pH adjustment was performed in the subsequent experiments. The removal rates were around 70% at different pH values, indicating that the obtained qe values were the saturation adsorption capacities of AC/WCS hydrogel.

Effect of Temperature on MB Adsorption

The effect of temperature on adsorption capacity is displayed in Figure . It was found that both the qe values and dye removal rates obviously increased when the temperature changed from 15 to 35 °C, which suggested that dye adsorption became more favorable at higher temperatures. It is widely known that the increment of temperature may accelerate the mobility of dye molecules and enhance the interactions between adsorbate and adsorbent by giving sufficient energy.[48] Moreover, at a higher temperature, the internal structure of hydrogels might swell to enable the passage of dye molecules, resulting in an increase of MB adsorption.[49] However, there was no significant difference when the temperature further increased to 45 °C.

Figure 7

Effect of temperature on MB adsorption with AC5000%/WCS0.3. Different asterisks on the top of the columns indicate the significant difference (p < 0.05).

Thermodynamic Study

To further understand the effect of temperature on the adsorption process, thermodynamic parameters, such as standard free energy (ΔG°), enthalpy (ΔH°), and entropy (ΔS°), were determined using the following equationswhere R (J/mol K) is the gas constant, T (K) is the absolute temperature, ΔG° (kJ/mol) is Gibbs free energy changes, ΔS° (J/mol K) is entropy changes, and ΔH° (kJ/mol) is enthalpy changes. k (L mg–1) is the ratio of qe to Ce(50)

As reported in Table , the negative values of ΔG° implied that the adsorption process of hydrogels happened spontaneously and was thermodynamically favorable. In addition, the changes in free energy could predict the adsorption behavior. The physisorption takes place when the ΔG° value is within the range between −20 and 0 kJ/mol, while chemisorption occurs in the range of −400 to −80 kJ/mol.[50,51] Therefore, physical interactions were mainly involved in the adsorption of MB on AC/WCS hydrogels. With the rise in temperature, ΔG° decreased accordingly, illustrating more driving forces and thereby resulting in higher adsorption capacity.[52] Typically, ΔH° for physical adsorption ranges from 4 to 40 kJ/mol, while that of chemical adsorption varies from 40 to 800 kJ/mol.[53] The positive ΔH° value also dictated that the adsorption was physical and endothermic.[54] The value of ΔS° was positive, revealing an increment of randomness at the adsorbent/adsorbate interface during the adsorption of MB on hydrogels. All these results were consistent with the other reported dye adsorption on AC,[52,55−59] which indicated that the incorporation of high amount of AC in regenerated cellulose matrix did not affect the adsorption behavior.

Table 1

Thermodynamic Parameters of MB Adsorption on AC5000%/WCS0.3

ΔG° (kJ mol–1)			ΔH° (kJ mol–1)	ΔS° (J mol–1 K–1)
288 K	298 K	333 K
–2.257	–3.027	–3.738	19.081	74.080

Effect of Contact Time and Kinetic Studies

The plots of adsorption capacity and removal rate of AC1000%/WCS0.652, AC2000%/WCS0.652, AC2000%/WCS0.3, and AC5000%/WCS0.3 against contact time are displayed in Figure . The adsorption of MB on four hydrogels was all fast in the beginning and then gradually reached equilibrium after 15 h. The initial adsorption rates of AC5000%/WCS0.3 and AC2000%/WCS0.652 were faster than those of AC2000%/WCS0.3 and/WCSAC1000%/WCS0.652, respectively, suggesting that the adsorption rate increased with the AC content. In addition, the adsorption rate of AC5000%/WCS0.3 was also faster than that of AC2000%/WCS0.652, showing that the higher ratio of cellulose in the hydrogels slowed down the adsorption of MB. It might be because the AC particles were wrapped with the cellulose gel matrix, which affected the efficient contact of dye molecules to the adsorption sites of AC, and the higher cellulose ratio in the hydrogels enlarged this effect.

Figure 8

Effect of contact time on MB adsorption with AC/WCS hydrogels.

In order to elucidate the adsorption mechanism, three kinetic models, pseudo-first order (PFO), pseudo-second order (PSO), and intraparticle diffusion (IPD) models were used to analyze the MB adsorption behavior on AC/WCS hydrogels. The fitting curves of experimental data are shown in Figure a,b, and the kinetic models’ parameters of MB adsorption are summarized in Table . Although the experimental data were fitted well with both PFO and PSD models (R2 > 0.98), the PFO model showed higher R2 value than the PSO model, and the qe values calculated by the PFO model were closer to the measured qe,exp values, indicating that the PFO model better explained the adsorption process. The IPD constants kid1, kid2, and kid3 describe the diffusion rates at different stages of the adsorption process (Table ), which followed the trend kid1 > kid2 > kid3. The first stage corresponded to the rapid diffusion (kid1) of large amounts of dye molecules (>50%) from the bulk dye solution to the surface of the adsorbent. The second subdued portion was the slow adsorption stage, where IPD was rate-controlled.[60] However, it could be found that none of the plots gave a linear straight line passing through the origin (C≠0). Therefore, the IPD was involved but not the only rate-controlling step for the adsorption of MB onto AC/WCS hydrogels. As discussed above, the adsorption process was also affected by the transport of dye molecules through the regenerated cellulose gel matrix, where the higher ratio of cellulose in the hydrogels resulted in a slower adsorption rate.

Figure 9

Adsorption kinetic of MB on AC/WCS hydrogels: (a–d) pseudo-first-order and pseudo-second-order, and (e) intraparticle diffusion.

Table 2

Kinetic Parameters of MB Adsorption on AC/WCS Hydrogels

samples	pseudo-first-order			pseudo-second-order			intraparticle diffusion model			experimental
	qe (mg g–1)	k1 (h–1)	R2	qe (mg g–1)	k2 (g mg–1 h1)	R2	kip	cip	R2	qe,exp(mg g–1)
AC5000%/WCS0.3	145.2	0.2007	0.9903	173.1	0.0013	0.9922	25.32	25.30	0.8585	148.6
AC2000%/WCS0.3	141.5	0.1611	0.9946	173.8	0.0010	0.9893	26.12	13.07	0.8769	141.9
AC2000%/WCS0.625	135.6	0.1792	0.9953	164.3	0.0012	0.9897	24.60	17.20	0.8580	137.1
AC1000%/WCS0.625	102.1	0.2072	0.9924	121.6	0.0019	0.9835	17.74	18.40	0.8145	101.9

Table 3

Kinetic Parameters of the Intraparticle Diffusion Model

samples	intraparticle diffusion model
	C1 (mg g–1)	kip1 (mg g–1 h–1)	R12	C1 (mg g–1)	kip2 (mg g–1 h–1)	R22	C3 (mg g–1 h–1)	kip3 (mg g–1 h–1)	R32
AC5000%/WCS0.3	–1.54	38.91	0.9823	116.85	5.73	0.9875	-	-	-
AC2000%/WCS0.3	–12.17	38.86	0.9843	104.55	6.73	0.9852	-	-	-
AC2000%/WCS0.625	–16.15	42.34	0.9928	52.63	19.33	0.9966	106.51	5.15	0.9576
AC1000%/WCS0.625	–10.42	33.21	0.9614	17.99	21.87	0.9604	96.58	0.88	0.9208

Adsorption Isotherm

Generally, at fixed dye concentrations, the removal rate (%) of adsorption increased with adsorbent dose (Figure ), and mainly hinged on the availability of adsorption sites. A dramatic positive impact of adsorbent dosage was indicated on the MB removal with approximately a linear relationship. However, when the dosage increased, the adsorption capacity (mg/g) deceased significantly, consequence of the unsaturation of adsorption sites during the adsorption process.[54] Cellulose has interactions with ionic dyes,[61,62] but the adsorption capacity of regenerated cellulose toward MB (Figure f) was much lower compared to WCS/AC hydrogels (Figure a–d). The maximum qe within the range of experimental conditions was 8.35 mg/g and the removal rate was 6.66% at an initial concentration of 10 mg/L (Figure f). It was worth noting that, although the existence of regenerated cellulose matrix affected the efficient contact of MB molecules to the adsorption sites, the adsorption capacity of AC5000/WCS0.3 was close to that of original AC powders (Figure b,e). This small difference indicated that the regenerated cellulose hydrogels from WCS could be used as an effective matrix to immobilize AC to treat dye-contaminated water.

Figure 10

Effect of adsorbent dose on qe (mg/g) and removal rate of MB adsorption on hydrogels, AC, and regenerated cellulose.

Least-square plots of the (a) Langmuir, (b) Freundlich, (c) Redlich–Peterson, (d) Sips, (e) Dubinin–Radushkevich, and (f) Dubinin–Astakhov isotherm for the adsorption of MB on AC/WCS hydrogels.

The values of maximum adsorption capacities (qm), correlation coefficients (R2), and other constant parameters for the six isotherm model equations for the adsorption process at 35 °C are summarized in Table . In general, R2 obtained by the adsorption data fitting of Redlich–Peterson (R–P), Sips, and Dubinin–Astakhov (D–A) models were superior to the other three models [Langmuir, Freundlich, and Dubinin–Radushkevich (D–R)]. Sips isotherm is a combined form of Langmuir and Freundlich expressions and their high values of R2 (>0.94) indicated the heterogeneous adsorption systems of MB-hydrogels.[63] The R–P model integrates elements from the Langmuir and Freundlich equations and their good fitting inferred that the adsorption did not follow an ideal monolayer mechanism.[64,65] The D–A isotherm model has the same basis as the D–R model and is more flexible than the DR equation. The excellent fitting of D–A indicated that the hydrogels had a high degree of heterogeneity due to their main adsorption component of carbonaceous solids.[66] The value of E obtained in the D–R isotherm was found to be lower than 3 kJ mol–1 for all the hydrogels. It suggested that the adsorption mechanism was physical in nature.[67] The qm values of AC5000%/WCS0.3 obtained from Sips and D–A models were very close to that of original AC powders and were the highest among four hydrogels. It indicated that the regenerated cellulose matrix was capable to hold a high amount of AC powders while not affecting their adsorption capacity.

Table 4

Isotherm Values for the Adsorption of MB onto AC/WCS Hydrogels

samples	Langmuir			Freundlich			Redlich–Peterson
	qm (mg g–1)	KL (L mg–1)	R2	KF (mg g–1)	n	R2	Q0 (L mg–1)	KRP (L g–1)	β	R2
AC5000%/WCS0.3	167.61	0.740	0.9607	96.72	7.92	0.9322	196.63	1.497	0.942	0.9878
AC2000%/WCS0.3	157.53	1.339	0.8794	113.19	13.96	0.8077	331.23	2.407	0.968	0.9202
AC2000%/WCS0.625	151.76	0.8672	0.9268	96.18	9.59	0.8893	230.16	1.891	0.948	0.9674
AC1000%/WCS0.625	100.42	0.708	0.9111	80.70	22.57	0.8812	122.72	1.353	0.980	0.9381

Comparison with Other Adsorbents

The adsorption capacity of MB by AC5000%/WCS 0.3 and other reported composite adsorbents are listed in Table . The hydrogel prepared in this work was found to be competitive with other forms of adsorbents. This clearly indicated that the coupling of regenerated cellulose from WCS with a high amount of AC could produce a promising adsorbent for MB removal from aqueous solution, because it possessed a dominant adsorption performance and was easy to recover and reuse.

Table 5

Maximum Adsorption Capacities of MB with Various Composite Adsorbents

adsorbents	adsorption capacity (mg g–1)	reference
AC/cellulose biocomposite films	103.66	(15)
hierarchical porous cellulose/AC composite monolith	159	(9)
crosslinked carboxymethyl sago pulp-powdered AC hydrogel	250	(18)
pineapple peel cellulose/magnetic diatomite hydrogels	101.94	(68)
polyvinyl alcohol/carboxymethyl cellulose hydrogels reinforced with graphene oxide and bentonite	172.14	(69)
alginate/compact discs waste-derived AC composite beads	11.78	(16)
AC/Cobalt Ferrite/Alginate Composite Beads	33.58	(70)
cellulose-derived carbon/montmorillonite nanocomposites	138.1	(71)
AC5000%/WCS 0.3	174.71	This work

Conclusions

WCS was successfully recycled into a three-dimensional matrix to fix a high amount of AC by a simple and facile method. The obtained hydrogel was an efficient adsorbent with a good handling property for wastewater treatment. The water content and compressive strength of AC/WCS hydrogels with relatively low AC content were affected obviously by regeneration methods, where the regeneration bath with higher affinity to LiCl/DMAc resulted in higher water content and lower strength. However, these properties of AC/WCS hydrogels with high AC content were mainly determined by the incorporation of AC powder because a large amount of AC was filled in the free volume of cellulose hydrogel and supported the matrix. The regenerated cellulose hydrogel matrix could hold and prevent the release of AC powder during adsorption process but also slightly affected the efficient contact of MB molecules to the adsorption sites. The maximum MB adsorption capacity of AC5000%/WCS 0.3 hydrogel reached 174.71 mg/g with the dye concentration of 200 mg/L, original solution pH of 6.9, at 35 °C for 24 h, which was comparable to the adsorption capacity of free AC powder. Adsorption of MB was best described by the Dubinin–Astakhov model and pseudo-first-order mechanism. Thermodynamic studies showed the spontaneous and endothermic nature of the overall physical adsorption process.

Materials and Methods

Reagents and Materials

WCS were collected from a second-hand market located in Montreal, Canada. The white old bed sheet contains 90–95% cellulose and a few percent of hemicellulose.[3,4] AC, MB trihydrate, LiCl, DMAc, acetone, and ethanol were purchased from Fisher Scientific (Mississauga, Canada). LiCl was dried at 200 °C for 1 h and DMAc was heated at 110 °C for 10 min to remove water before use.

Preparation of WCS/AC Hydrogels

LiCl (5.2 g) was added into 60 mL of water-free DMAc and stirred at 110 °C until dissolved, and then, the solution was cooled down to 80 °C. WCS was smashed into short floss without further purification or bleaching. For a better dissolution in DMAc/LiCl, WCS was activated by solvent exchange (H2O, acetone, and DMAc). After that, activated WCS (0.652 and 0.3 g) was added into a 60 mL DMAc/LiCl solution and stirred overnight until transparent solutions were formed. Desired amounts of AC were added into WCS/DMAc/LiCl solutions, stirred evenly, and casted into cylindrical molds to form hydrogels by replacing DMAc/LiCl with water and ethanol at different ratios. The gels were regenerated in three different baths: deionized water, 20% ethanol aqueous solution, and gradient ethanol solutions (a multistep soaking in ethanol/water mixtures with increasing concentrations of ethanol: water = 10:0, 8:2, 6:4, 4:6, 2:8, and 0:10, v/v). The hydrogels regenerated with ethanol were further soaked in water to remove ethanol. The hydrogels were coded as AC2000%/WCS0.3, AC5000%/WCS0.3, AC1000%/WCS0.652, and AC2000%/WCS0.652. For example, AC2000%/WCS0.3 was prepared by adding 6 g of AC (2000% of WCS dry weight) and 0.3 g of WCS.

Characterization

The morphologies of hydrogels were observed by scanning electron microscopy (SEM, Hitachi TM-1000, Japan) and the samples were gold/platinum coated (Leica EM ACE200, Germany). The crystal phases of AC, AC1000%/WCS0.652 and AC5000%/WCS0.3, were studied and compared using an X-ray diffractometer (D8 Discovery, Bruker, Germany) equipped with Cu Kα radiation (λ = 0.1542 nm). The compressive test was performed on cellulose hydrogels (cylinder with diameter of 4.0 mm and thickness of 4.0 mm, three replicates per each sample) at a rate of 0.4 mm/min by a compact tabletop universal tester (Instron 5965, USA) with a 10 N load cell. The water contents of hydrogels were determined gravimetrically by freeze-drying, which were defined as the ratios of the differences in weights of original hydrogel mw and dry hydrogel md relative to the original hydrogel weight mwwhere mw (g) is the weight of the wet gel, and md (g) is the weight of the dried gel.

Dye Adsorption

The adsorption efficiency of AC/WCS hydrogels toward methylene blue (MB) was investigated by batch methods. Kinetic experiments were carried out with 0.35 g of hydrogels and 300 mL of MB solutions (200 mg L–1) in a magnetically stirring-beaker at room temperature. At desired time intervals (0–38 h), the remaining amounts of MB in the aqueous solutions were determined by using a UV–vis spectrometer (GE Ultrospec 2100 pro, USA) at the wavelength of 663 nm. In the adsorption experiments of regenerated cellulose to MB, the original MB concentration was set to 10 mg/L. To evaluate the thermodynamic properties, adsorption isotherms were determined at 288, 298, 308, and 318 K, respectively, in a thermostatic orbital shaker for 38 h. The effect of pH (pH 4–9) on the adsorption of MB was investigated with the initial MB concentration of 200 mg L–1. The pH values of MB aqueous solutions were adjusted by 0.1 M HNO3 or 0.1 M NaOH solutions. The adsorption isotherm was measured by adding different amounts of hydrogels (0.02–0.16 g) in 25 mL of MB (200 mg L–1) solutions. Both tests were performed at 35 °C in a thermostatic orbital shaker for 38 h. All the adsorption tests were conducted in triplicates.

The amounts of MB absorbed were calculated by the following equationwhere qe (mg g–1) is the amount of MB adsorbed per gram of the dry adsorbent, V (mL) is the MB solution volume, C0 (mg L–1) and Ce (mg L–1) are the initial and equilibrium concentrations of MB in the solution, respectively, and m (g) is the weight of the dry adsorbent.

Removal rate of MB was calculated by the following equation

The adsorption isotherms of MB were fitted by nonlinear equations of Langmuir (eq ), Freundlich (eq ), Redlich–Peterson (eq ), Sips (eq ), Dubinin–Radushkevich (eq ), and Dubinin–Astakhov (eq ) models which are presented as follows:where qm (mg g–1) is the maximum amount of the adsorption, kL (L mg–1) is the Langmuir adsorption constant, and n and kF (mg g–1) are the Freundlich adsorption constants. KRP (L mg–1), αRP, and β are the Redlich–Peterson parameters. ks is the Sips isotherm model constant and ms is the Sips isotherm model exponent. βD is a constant related to the mean free energy of adsorption per mole of the adsorbate (mol2 J–2), ε is the Polanyi potential given by the relation ε = RTln(1 + 1/Ce), R (J mol–1 K–1) is the gas constant, and T (K) is the absolute temperature. The constant βD gives an idea about the mean free energy E (kJ mol–1) of adsorption per molecule of the adsorbate when it is transferred to the surface of the solid from the relationshipED (kJ/mol) is the characteristic energy associated to the given working pair. b is a heterogeneity parameter, considering the surface structure of the sorbent material. When b = 2, the Dubinin–Astakhov equation is referred to as the Dubinin–Radushkevich equation.

Kinetics data were represented by the PFO model, PSO model, and IPD model, given as followswhere k (h–1) is the PFO model rate constant, k2 (g mg–1 h–1) is the PSO model rate constant, kid (mg g–1 h–0.5) is the IPD rate constant, and C (mg g–1) is the intercept and related to the thickness of the boundary layer.

Statistical Analysis

Statistical evaluation was carried out by analysis of variance (ANOVA), followed by multiple-comparison tests using Duncan’s multiple-range test at the 95% confidence level. All the analyses were conducted using SPSS statistical software (version 22.0, IBM, New York, USA) with a probability of p < 0.05 considered to be significant.