Synthesis and properties of the rapeseed meal-grafted-poly(methyl methacrylate-co-butyl acrylate) oil-absorbents

Abstract

Owing to the unique properties such as oil body (OB) membranes and cellulose components within the structures of rapeseed meals (RSMs), the novel rapeseed meal-grafted-poly(methyl methacrylate-co-butyl acrylate) (RSMs-g-P(MMA-co-BA)) oil-absorbents were prepared and used for oil/water separation in the present study. Specifically, the RSMs-g-P(MMA-co-BA) were successfully synthesized through free radical graft copolymerization from RSMs, methyl methacrylate (MMA) and butyl acrylate (BA) with benzoyl peroxide (BPO) as initiator and N,N′-methylene-bisacrylamide (MBA) as crosslinker. The synthetic mechanism and 3D network structures were confirmed by FTIR and SEM, and the effects of reaction conditions, swelling kinetics, thermodynamics as well as reusability were investigated in detail. The RSMs-g-P(MMA-co-BA) demonstrated good absorption capacity for both organic solvents and oils, and the change of oil absorbency after eight repeated cycles of swelling/deswelling was marginal. Furthermore, the RSMs-g-P(MMA-co-BA) could selectively remove oils from water rapidly, which is promising for their potential applications in oil spill cleanup.

---

Introduction

In the past few years, water pollution resulting from oil spills and the discharge of organic pollutants have caused a severe crisis in marine ecosystems and human health.^1–4 To overcome this problem, considerable efforts have been made to remove the oil contaminants from the water surface: mechanical methods, chemical dispersion, in situ burning, bioremediation and oil-absorbents.⁵ Among them, the most economical and efficient methods should be granted by absorption by oil-absorbents due to their easy operation and high efficiency.⁶ So far, three types of oil-absorbent have been frequently used to handle oil contamination including inorganic mineral products (like activated carbon, zeolite, grapheme, etc.),^7–9 organic natural products (like wood, poplar, cotton, kapok fiber, etc.)^10–13 and synthetic organic products (like polyacrylate, polystyrene, polypropylene, polyurethane, etc.).^14–18 In recent years, efficient utilization of abundant agricultural residues as organic natural products has been investigated extensively to solve the intractable problem of oil spills.¹⁹ It is verified that natural products are low-cost, sustainable and nontoxic, but suffer from inferior separation for oil and water, slow oil absorbency rate and inconvenient recovery drawbacks.²⁰ In comparison with the inherent limitations of natural products, the synthetic organic oil-absorbents have obvious advantages for disposing of oil contaminants owing to their excellent oleophilic and hydrophobic characteristics.²¹ For example, high oil-absorption resins polymerized by lipophilic monomers through appropriate crosslinking have been ascertained to possess brilliant merits in absorption processes, such as good buoyancy, high oil absorbency, rapid absorption rate, heat and cold resistance, good selectivity in oil/water separation and so on,^22–24 and consequently represent a kind of functional synthetic oil-absorbents with unique 3D network structures.²⁵ Nevertheless, the relative high cost of synthetic oil-absorption resins makes it a limited process for final industrial production, especially in developing countries. In an effort to reduce the cost of the synthetic resins, another novel alternative method, i.e., synchronous utilization of the combined properties originating from the natural products and synthetic polymers, has been developed and applied to overcome the existing defects.^25,26 Typically, Wang J. T. et al.²⁷ obtained the functionalized oil-absorption resins by polymerization from butyl methacrylate and kapok fiber. Song C. and coworkers²⁸ reported a new class of oil-absorbents by incorporating β-cyclodextrin moieties into octadecyl acrylate and butyl acrylate co-monomers. Shan G. R. et al.²⁹ also investigated the oil-absorption resins filled with polybutadiene with a better swelling capability. By these means, the polymer network crosslinked tightly through chemical crosslinking could be loosen by the introduction of physical fillers, and thereby improved the oil absorbency compared with their original components.³⁰ Moreover, the incorporation of bulky and hard reinforcing fillers as frameworks could enhance the mechanical stability and led to a remarkable reusability. More importantly, adding appropriate natural products into resins, i.e., partially substitute for conventional petroleum-based polymers in the construction of hybrid oil-absorbents, has actually reduced the production costs due to the low-cost, abundant and renewable traits of natural products.

Rapeseeds, as an important oil-bearing crop, are widely planted all over the world. From the aspect of their microstructures, oils are mainly stored in spherical organelles (0.5–2.0 μm diameter) consisting in rapeseeds called oil bodies (OBs).³¹ Differing from other biological organelles with bilayer membranes, the OBs are composed of an oil core surrounded by a phospholipid monolayer membrane and proteins.³² Owing to the coexistence of oleophilic non-polar groups and hydrophilic polar groups in the phospholipid molecules of OB membranes, the OBs have been verified as novel micro-capsules and used in the extraction of organic compounds from aqueous solutions.³³ RSMs are the by-products of rapeseeds after oil expression. Traditionally, massive RSMs are usually handled as agricultural wastes, which cause a dissipation of resources and contaminate the environment. Actually, rapeseeds are rich in cellulose, OBs, and proteins.³⁴ Based on this background, we judge that plenty of cellulose and OB membranes would be inevitably left in RSMs after oil expression. In terms of their molecular structures, the primary functional groups in cellulose are hydroxyl,³⁵ which can give rise to an interlinkage between RSMs and oil-absorption resins through grafting reaction. Additionally, the OB membranes possess various functional groups such as acyl, long-chain alkyl, phosphate, amidogen groups and so on.³¹ Thereinto, long-chain alkyls belong to hydrophobic groups, which endow RSMs with partial lipophilicity and capture oils voluntarily by hydrophobic interaction. Moreover, the inherent void environments within RSMs can further help retain oils through capillary movements. Hereby, such concurrence of the hydrophilic reactive groups, hydrophobic functional groups, as well as void fraction makes good reasons for RSMs to be employed in the preparation of hybrid oil-absorbents, which have not been reported up to now.

Based on the above considerations, here we firstly prepared the RSMs-g-P(MMA-co-BA) by free radical graft copolymerization of RSMs, MMA and BA using BPO as initiator and MBA as crosslinker. The detailed mechanism for the construction and absorption processes were put forward. In addition, the performances of the RSMs-g-P(MMA-co-BA) were studied in the whole scope of oil absorption, including oil absorbency, kinetics, thermodynamics, regeneration and circulation. Benefiting from their excellent hydrophobicity and lipophilicity, good oil absorbency, rapid absorption rate and good buoyancy, the RSMs-g-P(MMA-co-BA) could effectively remove oils from water rapidly, which is promising for the potential applications in the remediation of oil spills.

Experimental

Materials

RSMs were obtained from Delingha region (Qinghai, China). MMA and BA were purchased from Tianjin Kemiou Chemical Reagent Co., Ltd (Tianjin, China). BPO, MBA and sodium hydroxide (NaOH) were provided by Tianjin Fuchen Chemical Reagent Factory (Tianjin, China). All above agents were of analytical grade and used without further purification. Absolute ethanol, chloroform (CHCl₃), dichloromethane (CH₂Cl₂), toluene and ethyl acetate were of chemical pure, and were supplied by Xi'an Chemical Reagent Factory (Shaanxi, China). Gasoline, diesel, engine oil and rapeseed oil were collected from the local service station. Nitrogen was obtained from Kunming Messer gas products Co., Ltd. (Yunnan, China). Distilled water was used throughout the work.

Pretreatment of RSMs

The original RSMs were washed with distilled water to remove the impurities and dried. Then the desiccated samples were transferred into a micro plant grinding machine, ground and screened through a 110-mesh sieve. Subsequently, a certain amount of the RSM powders were dispersed into a dilute aqueous solution of NaOH (2 wt%) under stirring at 90 °C. After 3.0 h, the precipitates were collected by centrifugation and washed with water and ethanol for several times. The resulting products were dried absolutely, milled into powders, and the pretreated RSMs were obtained.

Preparation of the RSMs-g-P(MMA-co-BA) oil-absorbents

A series of samples with different reaction conditions were prepared by the following procedures. First, appropriate amounts of pretreated RSMs, MMA, BA, BPO and MBA were successively added in a beaker under stirring. With the completely dissolving of BPO and MBA, all of the mixture was then transferred into a 200 mL three-necked flask equipped with a mechanical stirrer, a thermometer and a nitrogen pipe. The slurry was heated by a water bath at 70 °C under nitrogen atmosphere. After being purged with oxygen-free nitrogen for 15 min, the reaction was preceded under stirring and 1.0 h was employed to complete the whole free radical graft copolymerization. The resultants were taken out from the flask and washed by absolute ethanol and water for several times, then dried at 60 °C to get the composites used for the later tests.

The blank control sample P(MMA-co-BA) was synthesized according to the similar procedures described above except for the addition of RSMs.

The RSM intermediates were prepared by dispersing moderate pretreated RSMs in the mixture of MMA, BA and MBA at 70 °C under vigorously stirring for 1.0 h, and then dried at 60 °C for the subsequent characterization of FTIR.

Characterizations

Two techniques were applied to characterize the related products. The functional groups and chemical structures were confirmed using a Nicolet FTIR spectrometer in the range of 4000–450 cm⁻¹ via potassium bromide (KBr, optical grade) pellet. The surface morphologies were observed through a Hitachi S-4800 scanning electron microscope using an accelerated voltage of 5.0 kV. The specimens applied for SEM visualization were ground to a fine powder, fixed on aluminum stubs and treated by gold-sputtering before determination.

Measurement of the properties

Oil absorbency. In pure organic solvent or pure oil systems: the oil absorbency of the RSMs-g-P(MMA-co-BA) was determined by a weighing method.³⁶ First, the dried oil-absorbents (about 0.1 g) with known weight were put in a filter bag and immersed in a pure organic solvent (chloroform, dichloromethane, toluene and ethyl acetate) or a pure oil (gasoline, diesel, engine oil and rapeseed oil) at 25 °C. After a predetermined time (1.0 h was needed to ensure full saturation), the filter bag was lifted, drained for 1 min, wiped with a filter paper and immediately weighted. The oil absorbency of RSMs and P(MMA-co-BA) was determined following the same procedure as mentioned above and was calculated according to the following equation:where Q (g g⁻¹) is the oil absorbency defined as grams of oil per gram of dried samples, m₁ (g) and m₂ (g) are the weights of samples before and after oil absorption, respectively.

In oil/water systems: the tested oils (gasoline, diesel, engine oil and rapeseed oil) were separately mixed with water in a beaker under constant stirring through a magnetic stirrer. Then a quantity of about 0.1 g of dried oil-absorbents with known weight were immersed in the mixture. After 1.0 h, the samples were removed, drained, wiped and weighed. The total absorbency of both oil and water was calculated following eqn (1). To further determine the individual oil absorbency and water absorbency, the residual oils left in the mixture was extracted by n-hexane for an hour, and then the supernatant mixture was withdrawn by suction using a syringe. The water absorption capacity was determined by weighing the mass decrement of water before and after absorption processes. The oil absorbency could be obtained through a difference method between the above two numerical values (total absorbency and water absorbency).

In order to investigate the effect of the temperature on oil absorbency, studies were severally undertaken at 283.15, 293.15, 303.15, 313.15 K in pure organic solvent systems and at 278.15, 298.15, 318.15 K in pure oil and oil/water systems.

Swelling kinetics. The measurement to investigate the swelling kinetics of the RSMs-g-P(MMA–BA) in four organic solvents was substantially the same as the method for testing the oil absorbency mentioned above but repeated several times at predetermined time intervals. The oil absorbency at each given absorption time was determined following eqn (1).

Reusability. The reusability of the RSMs-g-P(MMA-co-BA) and P(MMA-co-BA) were investigated by the following steps. First, the two samples with known weights were separately put into filter bags and immersed in an pure oil (gasoline was tested in this part) at 25 °C for an hour, the oil absorbency of the composites was determined following eqn (1). Afterward, the saturated samples were immersed into 200 mL of n-hexane for another hour to collect the absorbed gasoline, and then lifted from n-hexane, drained, wiped and weighted. The oil absorbency after desorption was also calculated by eqn (1). Then, the samples were dried thoroughly in a forced air oven at 60 °C and were taken for the second swelling process. The same procedures were repeated eight times in gasoline and n-hexane and the oil absorbency in both absorption and desorption during each cycle was obtained.

In all cases above, three parallel experiments were measured and average value was reported in this study.

Results and discussion

Synthesis and characterization of the RSMs-g-P(MMA-co-BA) oil-absorbents

The RSMs-g-P(MMA-co-BA) were prepared by grafting co-monomers (MMA and BA) onto RSMs in the presence of initiator (BPO) and crosslinker (MBA). The proposed mechanism was depicted in Scheme 1.

Scheme 1 Proposed mechanisms for the formation of the RSMs-g-P(MMA-co-BA) oil-absorbents.

Since the surface of RSMs was covered with waxes, rap oils and other impurities, the alkali pretreatment was necessary to make hydroxyl groups exposed for the subsequent graft polymerization reaction (step 1).³⁷ In the succedent formation procedure of the RSMs-g-P(MMA-co-BA), co-monomers and crosslinker could be firstly captured on RSMs through hydrogen bond interaction. Afterwards, the initiator in the mixture was decomposed and changed into BPO radicals under heating conditions. With the help of BPO radicals, the hydrogen of hydroxyl groups on the pretreated RSMs could be taken away to produce some active alkoxy groups and consequently the RSM macro-radicals were generated.³⁸ Meanwhile, the co-monomer molecules were synchronously transformed into free co-monomer radicals by BPO radicals, that is, a chain with the fracture of carbon–carbon double bonds was initiated (step 2). Thereafter, the free co-monomer radicals turned into radical donors to the neighboring co-monomer molecules and the addition polymerization between MMA and BA took place to produce polymer chain (denoted here as P(MMA-co-BA) chain) radicals, resulting in the growth of P(MMA-co-BA) chains. Moreover, the procreant P(MMA-co-BA) chain radicals in close vicinity of the reaction sites were able to become accepters of RSM macro-radicals, which made P(MMA-co-BA) chains and RSMs closely interconnected (step 3).³⁹ During the aforementioned course of chain propagation, the grafted P(MMA-co-BA) chains could also react with the end of vinyl groups of MBA (step 4).⁴⁰ In such multiple ways, an interpenetrating polymer network structure was formed with substantial ester groups in it,⁴¹ and hereby the distinctive crosslinked 3D network structures with good hydrophobicity were eventually formed.⁴²

From the above analysis, it was obvious that RSMs were significant for the RSMs-g-P(MMA-co-BA). Generally, RSMs have played three pivotal roles during the assembling of RSMs-g-P(MMA-co-BA). First, substantial hydroxyl groups on RSMs had excellent affinity to seize MMA, BA and MBA through hydrogen bond interaction and RSM macro-radicals could be generated through BPO initiator under heating conditions. In this way, the graft copolymerization could take place on the surface of the RSMs to form the structures of the composites. Second, the mechanical stability is greatly strengthened because of the addition of the tough and hard RSMs. Third, partially substituted RSMs for the polymers in the construction of the hybrid oil-absorbents could inevitably reduce the production costs owing to the utilization of the abundant and renewable agriculture residues. From these points of view, the obtained RSMs-g-P(MMA-co-BA) are demonstrated to be new oil-absorbents with good economy and practical applicability.

In order to illustrate the above-mentioned formation procedures, FTIR spectroscopy was firstly employed to characterize the chemical bonding structures of the intermediates and final products. The FTIR spectra of primitive RSMs (a), pretreated RSMs (b), RSM intermediates (c), RSMs-g-P(MMA-co-BA) (d) and P(MMA-co-BA) oil-absorbents (e) are shown in Fig. 1.

Fig. 1 FTIR spectra of primitive RSMs (a), pretreated RSMs (b), intermediates (c), RSMs-g-P(MMA-co-BA) (d) and P(MMA-co-BA) oil-absorbents (e).

In the spectrum of primitive RSMs (Fig. 1(a)), the broad and strong peak observed at 3314 cm⁻¹ is due to the stretching vibration of hydroxyl groups.⁴³ The stretching vibration of methylene groups is proved by the appearance of absorption peak at around 2927 cm⁻¹ and the sharp band at 1044 cm⁻¹ is attributed to C–O stretching vibration.^27,44 These above peaks are assigned to the primary characteristic absorption peaks of cellulose in RSMs. Besides, according to literatures,^4,35,45,46 the absorption peaks at 1649, 1535 and 1385 cm⁻¹ are ascribed to the stretching vibration of the skeletal C=C in aromatic rings bands of lignin in RSMs. Fig. 1(b) is the FTIR spectrum of the pretreated RSMs, it is obvious that the characteristic absorption peak at 3389 cm⁻¹ becomes stronger compared with primitive RSMs, suggesting that more hydroxyl groups are exposed after alkali treatment. Analogously, the spectrum of the RSM intermediates (Fig. 1(c)) is basically consistent with the pretreated RSMs except for the red shift of the hydroxyl groups at 3343 cm⁻¹. Such deviation from 3389 to 3343 cm⁻¹ implies the intermolecular forces among RSMs, co-monomers and crosslinker have been aroused through hydrogen bond interaction. Following the reaction between RSMs and co-monomers, the obtained RSMs-g-P(MMA-co-BA) exhibit some changes in characteristic absorption peaks as shown in Fig. 1(d). Compared with the aforementioned spectrums of RSMs, the characteristic peaks at 3443 and 1200–1000 cm⁻¹ are distinctly weakened after the graft copolymerization with MMA and BA, demonstrating hydroxyl groups in RSMs have participated in the copolymerization reaction.^27,39 Moreover, the as-prepared composites have some new absorption peaks at 2958 and 2876 cm⁻¹ (CH₃ asymmetric stretching vibration and symmetric stretching vibration), 1733 cm⁻¹ (C=O stretching vibration in ester groups), 1452 and 1391 cm⁻¹ (CH₂ and CH₃ asymmetric bending vibration and symmetric bending vibration), 1243 and 1163 cm⁻¹ (C–O–C asymmetric stretching vibration and symmetric stretching vibration in ester) and 992 cm⁻¹ (vinyl groups),^{14,16,47–51} which are similar to the spectrum of pure P(MMA-co-BA) in Fig. 1(e). Those evidences provide direct confirmation of the presence of P(MMA-co-BA) in RSMs-g-P(MMA-co-BA). On the whole, with holding the primary characteristics peaks of both RSMs and P(MMA-co-BA), it can be concluded that the graft copolymerization between RSMs and co-monomers has indeed taken place through the active hydroxyl groups in the condition of BPO and MBA.

The grafting of P(MMA-co-BA) on RSMs can be further verified by SEM micrographs. Fig. 2 displays the surface morphologies of primitive RSMs (a), pretreated RSMs (b), and the RSMs-g-P(MMA-co-BA) oil-absorbents (c and d), respectively.

Fig. 2 SEM images of RSMs before (a) and after (b) pretreatment and the RSMs-g-P(MMA-co-BA) oil-absorbents (c and d).

As shown in Fig. 2(a), the primitive RSMs disperse unequally as blocky particles of varying sizes. From the enlarged drawing, the appearances of them are like cottons with a relatively rough surface, which endow RSMs with oil-absorbing ability. Fig. 2(b) displays the RSMs after alkali treatment. Unlike the rough surface of primitive RSMs, pretreated RSMs exhibit a smoother surface in the inset. This morphology change provides assertive evidence that the waxes and impurities on RSMs are removed, which are conducive to the succedent graft copolymerization. The surface profiles of the RSMs-g-P(MMA-co-BA) are demonstrated in Fig. 2(c) and (d). Fig. 2(c) shows the surface morphology of the attrite RSMs after grafting P(MMA-co-BA). Owing to the continuous accumulation and interconnection of P(MMA-co-BA) resins, the sizes of grafted RSMs become larger compared with the pretreated RSMs. Besides, it can be intuitively seen from Fig. 2(d) that RSM particles have been united into a bulky aggregation through P(MMA-co-BA) and act as skeletons in the composites. Both of the two images demonstrates that the RSMs-g-P(MMA-co-BA) possess a highly heterogeneous and coarse surface with a lot of cavities, folds, pores and loose structures. Such unique hybrid constructions of the final products have combined the advantages of both RSMs and P(MMA-co-BA), which are beneficial to increase the absorbing channels and boost the penetration speed. Meanwhile, the grafting of P(MMA-co-BA) onto RSMs has inescapably led to a transform of the surface wettability. The embedded images in Fig. 2(d) displays the morphology of a water and a diesel droplet (about 6 μL) on the RSMs-g-P(MMA-co-BA) surface at 25 °C. The water contact angle is as high as about 120°. In contrast, when a diesel droplet is dropped on the surface, it immediately spreads and permeates into the composites thoroughly, and the oil contact angle is measured to be 0°, definitely implying the excellent hydrophobicity and lipophilicity of the RSMs-g-P(MMA-co-BA).

Oil absorbency properties of the RSMs-g-P(MMA-co-BA) oil-absorbents

Effect of the mass ratio of MMA to BA on oil absorbency. The effect of the mass ratio of MMA to BA (1/3, 1/2, 1/1, 2/1, 3/1) on oil absorbency are illustrated in Fig. 3. The optimal oil absorbency of 30.72 g g⁻¹ in chloroform, 25.19 g g⁻¹ in dichloromethane, 18.94 g g⁻¹ in toluene and 14.98 g g⁻¹ in ethyl acetate all appears at a 1/2 mass ratio of MMA to BA. As we know, the co-monomer mass ratio can not only affect the affinity to oils, but also influence the effective volume of the 3D network structures.⁵² Longer hydrophobic alkyl chains in BA units can bring in better hydrophobicity,⁵³ so a suitable increase in BA concentration can enhance the oil affinity. Nevertheless, a further increase in BA content will lead to superabundant long alkyl groups, which in turn block the channels and decrease the effective volume of the oil-absorbents.⁴¹ A proper addition of MMA can introduce a number of short methyl branches into the composites which is good for the formation of 3D network structures and increases the effective volume, and thus increases the oil absorbency.

Fig. 3 Effect of the mass ratio of MMA to BA on oil absorbency of the RSMs-g-P(MMA-co-BA) (using 0.7 wt% BPO, 0.19 wt% MBA and 3.2 wt% RSMs).

Effect of initiator (BPO) concentration on oil absorbency. The relationship between oil absorbency of the RSMs-g-P(MMA-co-BA) (using 0.04 wt% MBA, MMA/BA = 1/2, and 3.2 wt% RSMs) and BPO content (from 0.3 to 1.1 wt% of total co-monomers weight) is demonstrated in Fig. 4. As Fig. 4 shows, with increasing BPO content, oil absorbency initially increases to a maximum and then decreases. The maximal oil absorbency in chloroform, dichloromethane, toluene and ethyl acetate are obtained: 40.95, 34.90, 25.21 and 20.50 g g⁻¹ at 0.7 wt% of BPO. The initiator has significant impacts on the polymer chain length and effective network volume.⁵³ Specifically, when a low BPO concentration is employed, the generated radicals are not enough to initiate more active sites in the propagation of P(MMA-co-BA) chains and the crosslinking network will become too loose. In addition, the co-monomers that are not initiated will remain in the network and decrease the effective volume.⁴¹ Conversely, a high BPO dosage will produce too many free radical centers, which will increase the rate of the termination reaction of free radicals.²⁶ As a result, the chain length will be shortened and the crosslinking density will be tightened,^29,52 which lead to a limited oil absorbency.

Fig. 4 Effect of initiator (BPO) or crosslinker (MBA) concentration on oil absorbency of the RSMs-g-P(MMA-co-BA).

Effect of crosslinker (MBA) concentration on oil absorbency. According to Flory's swelling theory,⁵⁴ swelling capacity is mainly affected by elasticity, the affinity to oils and crosslinking density. Here, oil absorbency of the RSMs-g-P(MMA-co-BA) (using 0.7 wt% BPO, MMA/BA = 1/2, and 3.2 wt% RSMs) with different amounts of MBA (from 0.01 to 0.13 wt% of total co-monomers weight) was studied in Fig. 4. It can be observed that oil absorbency initially increases with raising MBA content, and then decreases when the MBA dosage ranges from 0.04 to 0.13 wt%. It's well known that appropriate crosslinker is responsible for the formation of network structures.⁵⁵ A low MBA concentration will make the network too loose, and many soluble segments will exist in the oil-absorbents including liner polymers and polymer chains insufficiently crosslinked,⁵³ resulting in a low oil absorbency. However, as MBA content exceeds 0.04 wt%, excessive crosslinker will tighten the network and shorten the chain length, making it hard to stretch in the process of oil absorption.⁵⁶ These bad elasticity are unfavorable for oil molecules to penetrate into the network and accordingly the oil absorbency reduces.

Effect of RSMs content on oil absorbency. In this part, primitive RSMs, P(MMA-co-BA) and several RSMs-g-P(MMA-co-BA) with varied amounts of RSMs (from 1.6 to 12.8 wt% of total co-monomers weight) were used for analyzing the effect of RSMs content on oil absorbency, and the results are demonstrated in Fig. 5. The maximum value is achieved when RSMs dosage is 3.2 wt%, then decreases when more RSMs are incorporated into the composites. Oil absorption behavior of the oil-absorbents is an expansion process, in which oil molecules penetrate into a 3D network continually to give it rise to swelling.²⁶ Relying on this rule to speculate, the inordinate tightness of the network caused by chemical crosslinking is adverse to the extension of polymer chains and refrains the invasion behavior of oil molecules. Hereby, with an appropriate addition of RSMs, an improvement of oil absorbency can be achieved by decreasing the crosslinking density. More importantly, the pure polymer network originated from co-monomers is easy to collapse after swelling.⁵³ As an additional skeleton to support the network, the introduction of RSMs can strengthen the mechanical stability and mitigate the collapse, resulting in an extra oil absorbency.²⁷ Nevertheless, excessive RSMs in oil-absorbents bring about a sharp decrease of oil absorbency, and primitive RSMs even show a minimum value. The detailed causes can be ascribed to the following facts: (1) the absorption mechanism of RSMs is mainly on account of the capillary forces, which are quite weaker compared with the van der Waals forces provided by the pure P(MMA-co-BA) resins. (2) Excessive RSMs will cause a deep loss to the percentage of hydrophobic groups on P(MMA-co-BA) and thereby decrease the oil affinity. (3) More RSMs serve as physical fillers will obstruct the porous channels and weaken the elasticity and stretch properties. As a consequence, oil absorbency will drastically decrease with the introduction of excessive RSMs.

Fig. 5 Effect of RSMs content on oil absorbency of the RSMs-g-P(MMA-co-BA) (using 0.7 wt% BPO, 0.04 wt% MBA and MMA/BA = 1/2).

Swelling kinetics. Analysis of swelling kinetics is beneficial to clarify the mechanism of the swelling process and evaluate the oil absorption efficiency. Fig. 6 illustrates the oil absorbency of RSMs-g-P(MMA-co-BA) in four organic solvents as a function of immersion time. Judging from the curves in Fig. 6, the observed trends of oil-absorbents in the four solvents are quite similar. Oil absorbency initially increases rapidly with immersion time and 10 min later increases slowly and levels off after about 1 hour, then remains almost constant. Similar trends had been reported in previous literatures.^57,58

Fig. 6 Oil absorbency of the RSMs–P(MMA-co-BA) (using 0.7 wt% BPO, 0.04 wt% MBA, MMA/BA = 1/2, and 3.2 wt% RSMs) in four organic solvents as a function of immersion time.

The absorption processes can be divided into three steps: firstly, oil molecules are absorbed on the surface of the composites by hydrophobic interactions and van der Waals forces,^4,59 which can be accomplished in a very short time. Secondly, oil molecules start penetrating into the 3D network, making it expand and swell through porous channels. As time goes on, the lipophilic groups on the inner channels decrease with the increment of oil absorption, so a slower oil movement is observed. Thirdly, oil molecules can also enter RSM voids through internal capillary movements.⁴⁶ Nevertheless, the absorption is quite weaker compared with the first and second step. Therefore, there is a slight increase in the later period of the absorption processes.

According to swelling kinetics for the high water absorption resins with 3D network,^60,61 the pseudo-first-order and pseudo-second-order kinetic models were adopted to fit the experimental data, which had already been employed by early studies.^29,62

Pseudo-first-order kinetic equation: ln(Q_e − Q_t) = ln Q_e − k₁t (2)

where Q_e (g g⁻¹) and Q_t (g g⁻¹) are the oil absorbency defined as grams of oil per gram of dried sample at equilibrium and time t (min), respectively; k₁ (min⁻¹) is the absorption rate constant of pseudo-first-order absorption and k₂ (g g⁻¹ min⁻¹) is the absorption rate constant of pseudo-second-order absorption.

The straight lines obtained from a linear fitting technique are displayed in Fig. 7 and the relevant parameters are summarized in Table 1. Clearly, the linearity fitted by pseudo-second-order model is better, and the correlation coefficients R² of the pseudo-second-order model are greater than those of the pseudo-first-order model. In addition, the RSMs-g-P(MMA-co-BA) present an oil absorbency, at equilibrium, of 40.95 g g⁻¹ in chloroform, 34.90 g g⁻¹ in dichloromethane, 25.21 g g⁻¹ in toluene and 20.50 g g⁻¹ in ethyl acetate. Analogously, theoretical values Q_e,calc obtained from the pseudo-second-order model are 41.49, 35.26, 25.48 and 21.11 g g⁻¹, which are quite consistent with the above-mentioned experimental values Q_e,exp. These results suggest that the overall rate of absorption processes is controlled by the pseudo-second-order kinetic model rather than the pseudo-first-order model.

Fig. 7 Swelling kinetics of the RSMs–P(MMA-co-BA) oil-absorbents in four organic solvents. (a) Pseudo-first-order. (b) Pseudo-second-order.

Table 1 Kinetic parameters for oil absorbency of the RSMs-g-P(MMA-co-BA) oil-absorbents in four organic solvent

Organic solvent	Qe,exp (g g^−1)	Pseudo-first-order			Pseudo-second-order
		k1 (min^−1)	Qe,calc (g g^−1)	R^2	k2 (g g^−1 min^−1)	Qe,calc (g g^−1)	R^2
Chloroform	40.95	0.0542	13.44	0.8599	0.0162	41.49	0.9998
Dichloromethane	34.90	0.0472	11.39	0.8746	0.0187	35.26	0.9999
Toluene	25.21	0.0509	8.49	0.8794	0.0279	25.48	0.9998
Ethyl acetate	20.50	0.0855	11.87	0.9183	0.0167	21.11	0.9997

---

By comparing the values of k₂, the order of the absorption rate is: toluene > dichloromethane > ethyl acetate > chloroform. This phenomenon can be explained by the different polarities among these solvents. As presented in Fig. 6, the polarities are arranged according to the following orders: toluene < dichloromethane < ethyl acetate < chloroform. It can be distinctly concluded that the organic solvent with a lower polarity shows a faster absorption rate. As already discussed above, the primary driving factors in oil absorption are van der Waals forces. The smaller polarity of the solvent is, the stronger interaction between the solvents and oil-absorbents generates and consequently leads to a quicker absorption rate.

Absorption thermodynamics. To investigate the effect of temperature on oil absorbency, experiments are firstly carried out at 283.15, 293.15, 303.15, 313.15 K in organic solvents, and the relevant results are displayed in Fig. 8(a).

Fig. 8 Effect of temperature on oil absorbency. (a) In pure organic solvents. (b) In pure oil and oil/water systems.

It can be easily drawn from Fig. 8(a) that the oil absorbency changes slightly with the increase of temperature, demonstrating temperature has little effects on these solvents. To further evaluate the specific effects of temperature on oil absorbency, four different types of oils (gasoline, diesel, engine oil and rapeseed oil) that are widely used in daily life are severally investigated in pure oil and oil/water systems at 278.15, 298.15 and 318.15 K. Fig. 8(b) displays the experimental findings.

As Fig. 8(b) shows, oil absorbency in pure oils decreases in the following order: gasoline > diesel > engine oil > rapeseed oil, which is consistent with the ranks in oil/water systems. Obviously, among the four oils, gasoline has the lowest viscosity and the highest oil absorbency, superadding the similar viscosity of the organic solvents and the slight change in oil absorbency at different temperatures from Fig. 8(a), we can hereby draw a conclusion that viscosity is an important factor for oil absorbency.^63,64 Specifically, although a high viscosity can improve the adherence between oils and absorbents, it's quite difficult for oils to enter the network. Conversely, lower viscosity makes it easier for oils to penetrate into the network and thereout generates a higher oil absorbency. Furthermore, viscosity often changes with temperature. At 278.15 K, the viscosity is high, inhibiting the penetration of oils into the network and leading to a limited oil absorbency. A continuing increase in test temperature will accompany with a decreased viscosity which is beneficial to oil absorbency. In addition, making a comparison between the pure oils and oil/water systems, it can be observed that there is no evident differences in oil absorbency, and the values of water absorbency in the mixture of oil and water are very low (in the range of 1.18–1.86 g g⁻¹), which implies the RSMs-g-P(MMA-co-BA) have an excellent selectivity for oil and water.

In order to further ascertain the inherent energetic changes in oil absorption process, gasoline (the initial concentration was 16.5 g L⁻¹ of 100 mL gasoline/water mixture) was taken for example to determine the thermodynamic parameters such as the changes in Gibbs free energy (ΔG⁰), enthalpy (ΔH⁰) and entropy (ΔS⁰) using the following equations:^19,51

ΔG⁰ = −RT ln K_c (4)

where R (=8.314 J mol⁻¹ K⁻¹) is the universal gas constant, T (K) is the absolute temperature, K_c is the equilibrium constant of the absorption, C_e (absorbent) is the concentration of the absorbed oils in RSMs-g-P(MMA-co-BA) at equilibrium, and C_e (mixture) is the concentration of oils in the mixture at equilibrium. The slopes (−ΔH⁰/R) and intercepts (ΔS⁰/R) of the plots of ln K_c versus 1/T gives the values of ΔH⁰ and ΔS⁰.

The values of these thermodynamic parameters were listed in Table 2. The negative ΔG⁰ values for all the three disquisitive temperatures (−1.56, −4.07 and −5.49 kJ mol⁻¹ at 278.15, 298.15 and 318.15 K, respectively) manifest the absorption process is spontaneous and thermodynamically favorable.⁶⁵ With the increase of temperature, the absolute values of ΔG⁰ increase, indicating that a higher temperature is beneficial to oil absorption.⁶⁶ Moreover, it has been reported that ΔG⁰ for physisorption are between −20 and 0 kJ mol⁻¹.⁶⁰ Here, the values of ΔG⁰ are within the above range, demonstrating physisorption is the main process. The positive ΔH⁰ value indicates that the absorption process is endothermic; besides, the magnitude of ΔH⁰ can be also applied to distinguish the physisorption and chemisorption. Normally, when ΔH⁰ is smaller than 40 kJ mol⁻¹, the absorption process is described as physisorption, and when it is in range of 80–400 kJ mol⁻¹, the absorption is viewed as chemisorption.⁶⁷ In this study, the ΔH⁰ (25.57 kJ mol⁻¹) is in the physisorption range, further implying the absorption process is dominated by physisorption, which holds the similar conclusions with the lauric acid treated oil palm leaves.¹⁹ The relative results obtained from ΔH⁰ are also in line with the elevated oil absorbency with the increase of temperature. Meanwhile, the ΔS⁰ (98.16 J mol⁻¹ K⁻¹) is observed to be positive, revealing that an increasing randomness occurs at the solid/mixture interface during the absorption of gasoline from the mixture.⁶⁸ All the thermodynamic parameters testify that the RSMs-g-P(MMA-co-BA) can be selected as promising oil-absorbing materials to remove oils from oil/water mixture spontaneously.

Table 2 Thermodynamic parameters for oil absorbency of the RSMs-g-P(MMA-co-BA) oil-absorbents in gasoline/water system

Oil	T (K)	Qe (g g^−1)	Kc	ΔG^0 (kJ mol^−1)	ΔH^0 (kJ mol^−1)	ΔS^0 (J mol^−1 K^−1)	R^2
Gasoline	278.15	10.92	1.96	−1.56	25.57	98.16	0.9455
	298.15	13.82	5.16	−4.07
	318.15	14.66	7.97	−5.49

---

Reusability

Reusability is a significant property for oil-absorbents in practical application. In early studies, the methods for reutilization were diverse, such as vacuum filtration,⁵⁸ squeezing,² drying in an oven²⁷ and so on. These methods could indeed desorb oils from absorbents but might make an irreversible deformation of the network and produce harmful gases. To conquer this dilemma, reutilization through solvent elution appears to be a better solution. As a good solvent for gasoline, n-hexane was selected as an extractant to achieve the procedure of recycling. The change in oil absorbency of the oil-absorbents with and without RSMs for gasoline after eight swelling/deswelling cycles is displayed in Fig. 9.

Fig. 9 Reusability of the RSMs–P(MMA-co-BA) and P(MMA-co-BA) oil-absorbents in gasoline: O₁–O₈ were the oil absorption cycle conditions; D₁–D₇ were the oil desorption cycle conditions.

It is obvious in Fig. 9 that the overall trends in oil absorbency of the both two samples are downward with the increase of the absorption cycles and the serious decreases are mainly observed after the first cycle. The reduction can be ascribed to the fact that a handful of soluble fraction remained in the composites is dissolved in repeated swelling, and a fraction of network might collapse by extraction and drying.²⁷ After eight cycles of swelling/deswelling, oil absorbency of the RSMs-g-P(MMA-co-BA) still can reach 12.26 g g⁻¹, which is around 83.42% of their initial value. By contrast, P(MMA-co-BA) only retains approximately 64.59%. This phenomenon demonstrates that stronger mechanical stability can be generated by the introduction of RSMs, which is effective to improve the reusability of the RSMs-g-P(MMA-co-BA).

Selectively removal of diesel from water surface

On the basis of the above results, a more immediate application for the RSMs-g-P(MMA-co-BA) is the possibility of separating oils from water surface completely and rapidly. As a proof-of-concept experiment, diesel with the moderate viscosity and oil absorbency is chosen in this part to investigate the general selectively in the mixture of oil and water. The absorption processes for removing diesel form water surface by RSMs-g-P(MMA-co-BA) at different time intervals are vividly displayed in Fig. 10.

Fig. 10 The optical images of removing diesel from water.

In the first step, about 6 mL diesel is placed on 30 mL water in a culture dish under static condition to form the oil layer. Then the oil-absorbents are put into the diesel/water system. Once they are added, the oil-absorbents float on the mixture surface and immediately absorb diesel and swell. Interestingly, the oil-absorbent particles tend to aggregate spontaneously and most of the composites are gathered together 2 min later. After about 20 min, all of the particles aggregate together and almost all diesels are taken up by them. The swollen composites still float on the cleaned water surface after absorbing diesel, so they can be easily skimmed by a filter bag or a stainless steel mesh. As a consequence, there is no obvious residual diesel in the dish. This test indicates that the RSMs-g-P(MMA-co-BA) oil-absorbents could be utilized in large-scale cleanup of the spilled oils on water surface due to their good buoyancy, high oil absorbency, rapid absorption rate, and excellent selectivity for oil and water.

Conclusions

In summary, the RSMs-g-P(MMA-co-BA) oil-absorbents were successfully prepared through free radical graft copolymerization from RSMs, MMA and BA with BPO as initiator and MBA as crosslinker. Synthesis procedures of the composites with unique 3D network structures were separately characterized by FTIR and SEM. Moreover, the effects of reaction conditions, swelling kinetics, absorption thermodynamics as well as reusability were all investigated in detail. The performance tests demonstrated that the RSMs-g-P(MMA-co-BA) possess high oil absorbency and fast swelling rate for a wide range of organic solvents, and was found to follow the pseudo-second-order model. The thermodynamics analysis indicated that the gasoline absorption was a spontaneous, endothermic and physisorption process. In addition, great reusability for gasoline was further proved and could be reusable up to 8 cycles holding 83.42% of their initial uptake capacity, which exerted more excellent properties than P(MMA-co-BA). The outstanding selectivity of the RSMs-g-P(MMA-co-BA) in the mixture of diesel and water was also visually verified in the end. Furthermore, the utilization of natural agricultural residues (RSMs) for the preparation of the RSMs-g-P(MMA-co-BA) oil-absorbents could not only reduce the production costs significantly, but also improve the composites' mechanical stability and oil absorbency. These findings of the study may provide a reference for the fabrication of other natural materials as oil-absorbents for oil/water treatment and purification.

Acknowledgements

This work was supported by Shaanxi Provincial Natural Science Foundation of China (No. 2015JM2071), National Natural Science Foundation of China (No. 21176031) and Fundamental Research Funds for the Central Universities (No. 310829162014).

References

1. Y. Gao, Y. S. Zhou, W. Xiong, M. M. Wang, L. S. Fan, H. Rabiee-Golgir, L. J. Jiang, W. J. Hou, X. Huang, L. Jiang, J.-F. Silvain and Y. F. Lu, ACS Appl. Mater. Interfaces, 2014, 6, 5924–5929.
2. S. T. Nguyen, J. D. Feng, N. T. Le, A. T. T. Le, N. Hoang, V. B. C. Tan and H. M. Duong, Ind. Eng. Chem. Res., 2013, 52, 18386–18391.
3. V. Singh, R. J. Kendall, K. Hake and S. Ramkumar, Ind. Eng. Chem. Res., 2013, 52, 6277–6281.
4. D. Li, F. Z. Zhu, J. Y. Li, P. Na and N. Wang, Ind. Eng. Chem. Res., 2013, 52, 516–524.
5. A. A. Al-Majed, A. R. Adebayo and M. E. Hossain, J. Environ. Manage., 2012, 113, 213–227.
6. X. F. Sun, R. C. Sun and J. X. Sun, J. Agric. Food Chem., 2002, 50, 6428–6433.
7. A. Bayat, S. F. Aghamiri, A. Moheb and G. R. Vakili-Nezhaad, Chem. Eng. Technol., 2005, 28, 1525–1528.
8. M. O. Adebajo, R. L. Frost, J. T. Kloprogge, O. Carmody and S. Kokot, J. Porous Mater., 2003, 10, 159–170.
9. H. C. Bi, X. Xie, K. B. Yin, Y. L. Zhou, S. Wan, L. B. He, F. Xu, F. Banhart, L. T. Sun and R. S. Ruoff, Adv. Funct. Mater., 2012, 22, 4421–4425.
10. M. Likon, M. Remškar, V. Ducman and F. Švegl, J. Environ. Manage., 2013, 114, 158–167.
11. G. Deschamps, H. Caruel, M.-E. Borredon, C. Bonnin and C. Vignoles, Environ. Sci. Technol., 2003, 37, 1013–1015.
12. X. F. Huang and T.-T. Lim, Desalination, 2006, 190, 295–307.
13. H.-M. Choi and J. P. Moreau, Microsc. Res. Tech., 1993, 25, 447–455.
14. J. Z. Gao, X. F. Li, Q. F. Lu, Y. Li, D. L. Ma and W. Yang, Polym. Bull., 2012, 68, 37–51.
15. J. Wu, N. Wang, L. Wang, H. Dong, Y. Zhao and L. Jiang, ACS Appl. Mater. Interfaces, 2012, 4, 3207–3212.
16. Y. Feng and C. F. Xiao, J. Appl. Polym. Sci., 2006, 101, 1248–1251.
17. Q. F. Wei, R. R. Mather, A. F. Fotheringham and R. D. Yang, Mar. Pollut. Bull., 2003, 46, 780–783.
18. D. X. Wu, W. J. Wu, Z. Y. Yu, C. Y. Zhang and H. T. Zhu, Ind. Eng. Chem. Res., 2014, 53, 20139–20144.
19. S. M. Sidik, A. A. Jalil, S. Triwahyono, S. H. Adam, M. A. H. Satar and B. H. Hameed, Chem. Eng. J., 2012, 203, 9–18.
20. S. Suni, A.-L. Kosunen, M. Hautala, A. Pasila and M. Romantschuk, Mar. Pollut. Bull., 2004, 49, 916–921.
21. N. Y. Ji, H. Chen, M. M. Yu, R. J. Qu and C. H. Wang, Polym. Adv. Technol., 2011, 22, 1898–1904.
22. X. M. Zhou and C. Z. Chuai, J. Appl. Polym. Sci., 2010, 115, 3321–3325.
23. A. M. Atta, S. H. El-Hamouly, A. M. AlSabagh and M. M. Gabr, J. Appl. Polym. Sci., 2007, 105, 2113–2120.
24. B. Wu and M. H. Zhou, J. Environ. Manage., 2009, 90, 217–221.
25. H. Maimaiti, K. Arken, M. Wumaier and L. Wushuer, Iran. Polym. J., 2014, 23, 671–678.
26. G. R. Shan, P. Y. Xu, Z. X. Weng and Z. M. Huang, J. Appl. Polym. Sci., 2003, 90, 3945–3950.
27. J. T. Wang, Y. A. Zheng and A. Q. Wang, J. Appl. Polym. Sci., 2013, 127, 2184–2191.
28. C. Song, L. Ding, F. Yao, J. P. Deng and W. T. Yang, Carbohydr. Polym., 2013, 91, 217–223.
29. G. R. Shan, P. Y. Xu, Z. X. Weng and Z. M. Huang, J. Appl. Polym. Sci., 2003, 89, 3309–3314.
30. X. M. Zhou and C. Z. Chuai, Polym. Eng. Sci., 2013, 53, 540–545.
31. Z. Purkrtova, P. Jolivet, M. Miquel and T. Chardot, C. R. Biol., 2008, 331, 746–754.
32. Y. S. Ying, L. P. Zhao, L. Z. Kong, X. Z. Kong, Y. F. Hua and Y. M. Chen, Food Chem., 2015, 181, 179–185.
33. J. Boucher, F. Cengelli, D. Trumbic and I. W. Marison, Chemosphere, 2008, 70, 1452–1458.
34. A. M. Pustjens, H. A. Schols, M. A. Kabel and H. Gruppen, Carbohydr. Polym., 2013, 98, 1650–1656.
35. X. H. Xu, B. Bai, H. L. Wang and Y. R. Suo, J. Phys. Chem. Solids, 2015, 87, 23–31.
36. H. T. Zhu, S. S. Qiu, W. Jiang, D. X. Wu and C. Y. Zhang, Environ. Sci. Technol., 2011, 45, 4527–4531.
37. Y. A. Zheng, J. T. Wang, Y. F. Zhu and A. Q. Wang, J. Environ. Sci., 2015, 27, 21–32.
38. Y. Bao, J. Z. Ma and N. Li, Carbohydr. Polym., 2011, 84, 76–82.
39. Y. M. Zhou, S. Y. Fu, L. L. Zhang and H. Y. Zhan, Carbohydr. Polym., 2013, 97, 429–435.
40. A. Pourjavadi, P. E. Jahromi, F. Seidi and H. Salimi, Carbohydr. Polym., 2010, 79, 933–940.
41. L. Ding, Y. Li, D. Jia, J. P. Deng and W. T. Yang, Carbohydr. Polym., 2011, 83, 1990–1996.
42. W. B. Wang and A. Q. Wang, J. Compos. Mater., 2009, 43, 2805–2819.
43. R. Gnanasambandam and A. Proctor, Food Chem., 2000, 68, 327–332.
44. A. K. Rana, R. K. Basak, B. C. Mitra, M. Lawther and A. N. Banerjee, J. Appl. Polym. Sci., 1997, 64, 1517–1523.
45. T. Wan, R. Q. Huang, Q. H. Zhao, L. Xiong, L. L. Qin, X. M. Tan and G. J. Cai, J. Appl. Polym. Sci., 2013, 130, 3404–3410.
46. M. A. Abdullah, A. U. Rahmah and Z. Man, J. Hazard. Mater., 2010, 177, 683–691.
47. G. V. R. Reddy, V. S. Joseph and K. C. Mani, J. Appl. Polym. Sci., 2000, 77, 398–408.
48. A. Kleinová and E. Borsig, Macromol. Chem. Phys., 1996, 197, 2289–2296.
49. M. N. Siddiqui, H. H. Redhwi, D. Charitopoulou and D. S. Achilias, Polym. Int., 2014, 63, 766–777.
50. J. Mou, X. R. Li, H. H. Wang, G. Q. Fei and Q. Liu, Starch-Stärke, 2012, 64, 826–834.
51. J. T. Wang, Y. A. Zheng and A. Q. Wang, Mar. Pollut. Bull., 2013, 69, 91–96.
52. M. H. Zhou and W.-J. Cho, J. Appl. Polym. Sci., 2002, 85, 2119–2129.
53. J. Jang and B.-S. Kim, J. Appl. Polym. Sci., 2000, 77, 914–920.
54. P. J. Flory, Principles of Polymer Chemistry, Cornell University Press, Ithaca, New York, 1953, ch. 13.
55. A. M. Atta, R. A. M. El-Ghazawy, R. K. Farag and A.-A. A. Abdel-Azim, React. Funct. Polym., 2006, 66, 931–943.
56. A. Pourjavadi and M. Doulabi, Adv. Polym. Technol., 2013, 32, 21332.
57. J. T. Wang, Y. A. Zheng, Y. R. Kang and A. Q. Wang, Chem. Eng. J., 2013, 223, 632–637.
58. N. Y. Ji, H. Chen, G. X. Zong and D. J. Wang, Polym. Int., 2012, 61, 1786–1791.
59. P. Fang, P. P. Mao, J. Chen, Y. Du and X. Hou, J. Appl. Polym. Sci., 2014, 131, 40180.
60. T. S. Anirudhan and S. R. Rejeena, Sep. Purif. Technol., 2013, 119, 82–93.
61. Y. W. Huang, M. Zeng, J. Ren, J. Wang, L. R. Fan and Q. Y. Xu, Colloids Surf., A, 2012, 401, 97–106.
62. M. H. Zhou, C.-S. Ha and W.-J. Cho, J. Appl. Polym. Sci., 2001, 81, 1277–1285.
63. X. C. Gui, H. B. Li, K. L. Wang, J. Q. Wei, Y. Jia, Z. Li, L. L. Fan, A. Y. Cao, H. W. Zhu and D. H. Wu, Acta Mater., 2011, 59, 4798–4804.
64. J. Y. Lin, Y. W. Shang, B. Ding, J. M. Yang, J. Y. Yu and S. S. Al-Deyab, Mar. Pollut. Bull., 2012, 64, 347–352.
65. L. L. Ding, B. Zou, W. Gao, Q. Liu, Z. C. Wang, Y. P. Guo, X. F. Wang and Y. H. Liu, Colloids Surf., A, 2014, 446, 1–7.
66. Q. Li, Q. Y. Yue, Y. Su, B. Y. Gao and H. J. Sun, Chem. Eng. J., 2010, 158, 489–497.
67. M. A. Z. Abidin, A. A. Jalil, S. Triwahyono, S. H. Adam and N. H. N. Kamarudin, Biochem. Eng. J., 2011, 54, 124–131.
68. A. L. Ahmad, C. Y. Chan, S. R. Abd Shukor and M. D. Mashitah, Chem. Eng. J., 2009, 148, 378–384.

---