capsulated magnetite nanoparticles as a recyclable

0 downloads 6 Views 3MB Size Report
FFAs occurs, which substantially decreases the biodiesel yield, especially when a feedstock with a high FFA content is adopted. ... amount of water for washing, thus leading to environmental pollution ... while can be conveniently separated from the final product, is of ..... To determine the optimal catalyst dose for the efficient.

NJC View Article Online

Published on 10 July 2018. Downloaded by University of California - Santa Barbara on 7/13/2018 1:17:59 AM.

PAPER

Cite this: DOI: 10.1039/c7nj05075e

View Journal

Sulfonated poly(styrene-divinylbenzene-glycidyl methacrylate)-capsulated magnetite nanoparticles as a recyclable catalyst for one-step biodiesel production from high free fatty acid-containing feedstocks† Jinming Chang, Haojun Fana

ab

Xiaoyu Guan,a Siyu Pan,a Maolin Jia,a Yi Chen*a and

The decrease in fossil fuel resources and the negative environmental impact of greenhouse gas emissions from their combustion have stimulated interest in finding alternative energy resources. Of various alternatives, biodiesel is surging in popularity given its renewability and lower environmental impact. Herein, magnetite nanoparticles were prepared by a chemical co-precipitation technique, and then capsulated with a sulfonated poly(styrene-divinylbenzene-glycidyl methacrylate) copolymer, yielding a recyclable catalyst suitable for biodiesel production from high free fatty acid-containing feedstocks, such as waste or recycled oil. The morphology and structure of the resultant catalyst were characterized by multiple techniques, and the influence of catalyst dose, temperature, molar ratio of methanol to feedstock oil, and free fatty acid content on the biodiesel yield was evaluated, respectively. The results indicated that this catalyst was capable of catalyzing the transesterification of triglycerides and the esterification of free Received 21st December 2017, Accepted 28th June 2018

fatty acids simultaneously, enabling a simple, one-step procedure for biodiesel production from

DOI: 10.1039/c7nj05075e

was found to reach 91.1%. In addition, such a catalyst was proved to be easily recyclable under an external

feedstocks containing a high content of free fatty acids. Under the optimal conditions, the biodiesel yield magnetic field, and reusable, which promised its potential application in the environmentally-benign

rsc.li/njc

production of biodiesel from cheap waste or recycled oil.

Introduction Since the dawn of the industrial revolution, fossil fuels have been the driving force behind the industrialized world and its economic growth. However, their finite supply, together with the environmental pollution caused by fossil fuel combustion, has stimulated interest in renewable energy resources with lower environmental impact. Biodiesel, a renewable and biodegradable energy resource, has emerged as one of the most strategically important alternatives. It refers to vegetable oil- or animal fat-based diesel fuels, consisting of long chain alkyl (methyl, ethyl, or propyl) esters. Use of biodiesel reduces greenhouse gas emissions in that carbon dioxide released from

a

Key Laboratory of Leather Chemistry and Engineering of Ministry of Education, Sichuan University, Chengdu 610065, P. R. China. E-mail: [email protected] b Chemical Synthesis and Pollution Control Key Laboratory of Sichuan Province, China West Normal University, Nanchong 637009, P. R. China † Electronic supplementary information (ESI) available. See DOI: 10.1039/c7nj05075e

biodiesel combustion is offset by that absorbed from growing soybeans or other feedstocks used to produce the fuel.1–4 Biodiesel also reduces tailpipe particulate matter, hydrocarbon, and carbon monoxide emissions, because the presence of oxygen in biodiesel allows the fuel to burn more completely, so fewer unburned fuel emissions result.5–7 In particular, biodiesel fuel can be used in any diesel engine in the pure form or blended with fossil fuels at any level. Even a blend of 20% biodiesel and 80% fossil fuel will significantly reduce carcinogenic emissions and gases that may contribute to global warming.8,9 However, the commercialization of biodiesel still suffers from high cost compared with that of fossil fuels. This is mainly because biodiesel is mostly produced from expensive virgin vegetable oil, which accounts for 70–80% of its total production cost.10,11 Hence, using cheap feedstocks such as waste or recycled oil, which contains a high content of free fatty acids (FFAs), is considered a feasible way to make biodiesel economically competitive. The manufacture of biodiesel from feedstock oil typically involves the transesterification of the oil with methanol using an alkali as the catalyst. During this process, saponification of

This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

New J. Chem.

View Article Online

Published on 10 July 2018. Downloaded by University of California - Santa Barbara on 7/13/2018 1:17:59 AM.

Paper

NJC

FFAs occurs, which substantially decreases the biodiesel yield, especially when a feedstock with a high FFA content is adopted. Also, it has been claimed that, when the FFA level exceeds 5%, the soap formed inhibits the separation of glycerol by-product from the resultant biodiesel.12,13 In order to alleviate saponification, there are two alternatives. The first solution is to convert FFAs into esters by an acid-catalyzed pretreatment step, before the alkali-catalyzed transesterification process. Despite its efficiency, this method requires time-consuming and laborious separation/purification processes. Alternatively, acid catalysts such as sulfuric acid can be employed to catalyze the esterification and transesterification reactions simultaneously. In this approach, however, homogeneous acid catalysts are corrosive to equipment, and consume a large amount of water for washing, thus leading to environmental pollution. Therefore, developing an acid catalyst capable of catalyzing the esterification and transesterification reactions simultaneously, while can be conveniently separated from the final product, is of great importance for the environmentally benign production of biodiesel from cheap feedstocks with a high FFA content. In recent years, core–shell nanoparticles with magnetically responsive cores and functional shells have been widely investigated given their broad applications.14–18 This combination is desirable, because immobilizing functional groups on nanoparticles creates a significantly large accessible surface area, thus offering a higher capacity for adsorption, separation, or catalysis. Meanwhile, magnetite-supported functional shells can be readily separated using an external magnet, and thus, filtration, centrifugation or other tedious workup procedures can be omitted after treatment. Previously, various magneticallyseparable catalysts intended for biodiesel production have been reported.19–21 In spite of their efficiency in catalyzing the transesterification of feedstock oil, they were either corrosive, prone to induce significant saponification or inactivated in the presence of a high FFA content, preventing them from being successful in cases where cheap waste or recycled oil are employed as the feedstocks. Herein, we attempted to capsulate magnetite nanoparticles with a sulfonated poly(styrene-divinylbenzene-glycidyl methacrylate) copolymer, with the aim of designing a recyclable catalyst for onestep biodiesel production from feedstocks with a high FFA content. The structure of the core–shell-structured catalyst was systematically characterized, and the influence of catalyst dose, temperature, molar ratio of methanol to feedstock oil, and FFA content on the biodiesel yield was evaluated, respectively. It was found that this catalyst could catalyze the transesterification of triglycerides and the esterification of free fatty acids simultaneously, affording a biodiesel yield of 91.1% under the optimal conditions. According to these results, this catalyst holds great potential as a recyclable catalyst for the cost-efficient and environmentally benign production of biodiesel from cheap waste or recycled oil.

Experimental Materials Ferric chloride hexahydrate (FeCl36H2O, purity Z 99.0%), ferrous sulfate heptahydrate (FeSO47H2O, purity Z 99.0%),

New J. Chem.

ammonium hydroxide solution (NH3H2O, 28.0–30.0% NH3 basis), and anhydrous ethanol (CH3CH2OH, purity Z 99.5%) were purchased from Kelong Chemical Engineering Co. Ltd (Chengdu, China). Styrene (ST, purity Z 99.0%), divinyl benzene (DVB, purity Z 98.0%), and glycidyl methacrylate (GMA, purity Z 97.0%), potassium persulfate (KPS, purity Z 99.5%), and methanol (CH3OH, purity Z 99.5%) were supplied by Sigma-Aldrich (Shanghai, China). In addition, soybean oil with an average molecular weight of 832 g mol1 was supplied by Cheng Jie Chemical Co. Ltd (Shanghai, China). As specified by the supplier, the fatty acids constituting the soybean oil triglycerides are 31% oleic acid, 55% linoleic acid, 8% palmitic acid, 5% stearic acid, and traces of other acids. Ultrapure water with a resistivity equal to 18.2 MO cm at 25 1C was obtained using a Millipore Synergy water purification system. All reagents were used without further purification unless otherwise stated. Synthesis of the [email protected]@SO3H catalyst Magnetite nanoparticles were prepared by using a chemical co-precipitation technique according to a previously disclosed procedure.22,23 In brief, 7.37 g FeCl36H2O and 4.33 g FeSO47H2O were first dissolved in 100 mL nitrogen-purged ultrapure water, yielding a mixture in which the molar ratio of ferric and ferrous cations was equal to 1.75 : 1. Then, a NH3 aqueous solution was added dropwise under continuous stirring until the pH value reached 10.0. After incubation at 60 1C for 30 min, the black precipitate, denoted as MNPs, was magnetically separated, and washed with anhydrous ethanol three times. Finally, the product was dried in a vacuum oven at 50 1C until a constant weight was obtained. Typically, waste or recycled oil contains a high level of FFAs compared with that in virgin vegetable oil, while native MNPs are corrosive under acidic conditions, which significantly compromises their magnetism. Therefore, a polymer-coating technology was employed herein to protect the MNPs against acid corrosion by FFAs. Meanwhile, the ST monomeric units constituting the polymer shell in this study was susceptible to sulphonation by sulfuric acid, thus enabling sulfonated MNPs to act as a heterogeneous, recyclable catalyst. To this end, 2.0 g MNPs were first dispersed under ultrasonic oscillation (2000 W, 50 Hz) into 100 mL ultrapure water, followed by addition of 1.5 g oleic acid. The temperature was then elevated to 80 1C, and the mixture was ultrasonically oscillated (2000 W, 50 Hz) for another 30 min. During this process, the oleic acid molecules self-assembled on the surface of the MNPs into a two-layered sheet with the hydrophobic tails pointing toward the center of the sheet. After that, 3.01 g ST, 1.50 g DVB, and 0.52 g GMA were added, and then subjected to ultrasonic oscillation (2000 W, 50 Hz) at 30 1C for 40 min. These monomers accumulated within the center of the oleic acid double-layered sheet. Subsequently, 0.08 g KPS was added, and the mixture was stirred at 65 1C for another 20 h under a nitrogen atmosphere, when the monomers copolymerized to form a shell. In the above formula, DVB served as a crosslinker, while GMA regulated the polarity of the shell, which facilitated the wrapping of the hydrophilic magnetite core. The engineered magnetite, referred

This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online

Published on 10 July 2018. Downloaded by University of California - Santa Barbara on 7/13/2018 1:17:59 AM.

NJC

Paper

accuracy of 0.2 eV. The magnetization curves were obtained using a vibrating sample magnetometer (VSM, LakeShore 7307) in a magnetic field of 10 kOe. Thermogravimetric (TA) analysis was performed using a NETZSCH TG 209 F1 analyzer with a TASC 414/3 thermal analysis controller. The sample in a tared platinum pan was heated from 35 to 800 1C at a rate of 10 1C min1 under a nitrogen atmosphere. The elemental composition of the char residue was analyzed using an energy dispersive X-ray (EDX, X-MAX50, JEOL, Japan) spectroscope. Biodiesel production

Scheme 1

Synthesis procedure and structure of [email protected]@SO3H.

to as [email protected], was collected using a magnet, washed five times with ethanol, and dried at 50 1C until a constant weight was obtained. Subsequently, 1 g [email protected] was mixed with a 20 mL cold concentrated sulfuric acid solution (96 wt%), which was shaken at room temperature for 1 h. Afterwards, the mixture was added dropwise into 200 mL ultrapure water to obtain the sulfonated catalyst, denoted as [email protected]@SO3H. Finally, the [email protected]@SO3H catalyst was isolated by using a magnet, and washed with ultrapure water five times to remove any unreacted sulfuric acid. The synthesis procedure and the structure of [email protected]@SO3H are outlined in Scheme 1. Characterization of the [email protected]@SO3H catalyst The morphologies of the nanoparticles were examined by using a scanning electron microscope (SEM, JSM7500F, JEOL, Japan), and a high resolution transmission electron microscope (HR-TEM, Hitachi HF750, Japan) operated at an accelerating voltage of 200 kV. Low-temperature nitrogen adsorption experiments were carried out to measure the Brunauer–Emmett– Teller specific surface area of the nanoparticles using an ASAP 2020 Accelerated Surface Area and Porosimetry System (Micromeritics, United States). The testing temperature was set at the boiling temperature of liquid nitrogen. The sample was degassed under vacuum for 24 h before analysis. Fourier transform infrared (FT-IR) spectra were collected at ambient temperature using a Nicolet iS10 FT-IR spectrometer (Thermo Scientific, United States) over a wavenumber range from 400 to 4000 cm1 after 64 scans at a resolution of 2 cm1. X-ray photoelectron spectroscopy (XPS) measurements were performed on an AXIS Ultra DLD instrument (Kratos, U.K.), equipped with a standard and monochromatic Al Ka X-ray excitation source (1486.6 eV). The analysis was carried out in the constant analyzer energy mode at a pass energy of 40 eV. The pressure in the analysis chamber was 2  107 Pa. The binding energies were corrected by referencing the C 1s peak at 284.6 eV with an

The feedstock oil with different contents of FFAs was prepared by adding different amounts of oleic acid into a fixed amount of soybean oil. Biodiesel was then generated by exposing a predetermined amount of [email protected]@SO3H to a mixture containing the feedstock oil and methanol in a different molar ratio. The reaction was carried out in a four-neck flask equipped with a condenser, a thermometer, and a stirrer. The influence of catalyst dose (2%, 5% and 8%), temperature (80 1C, 100 1C and 120 1C), molar ratio of methanol to feedstock oil (3 : 1, 15 : 1 and 25 : 1), and the content of FFAs (10%, 20% and 40%) on the biodiesel yield was investigated, respectively. At a fixed time interval, the catalyst was separated temporarily using a magnet. The residual oil was centrifuged at 4000 rpm and 25 1C for 15 min, and the composition of the upper layer was analyzed by using a gas chromatography instrument (GC, Agilent, 7890B). A chromatographic column (HP-5MS, Agilent) was used for the separation. The oven temperature was programmed from 190 to 280 1C at an increasing rate of 10 1C min1, and held at 280 1C for 6 min. Both the injector and detector temperatures were 300 1C. Methyl salicylate was used as the internal standard. The biodiesel yield was calculated according to the following equation:24 Yield ¼ 

mactual Cesters  n  Vesters   100% mtheoretical moil Cester  n  Voil Cesters  n  100%   100% roil moil

where, mactual (g) and mtheoretical (g) are the actual mass and theoretical mass of the biodiesel, respectively; moil (g) represents the mass of the feedstock oil; n denotes the diluted multiple of the biodiesel; Cesters (g mL1) refers to the mass concentration of the biodiesel; roil (g mL1) is the density of the feedstock oil; Vesters (mL) and Voil (mL) are the volumes of the biodiesel and the feedstock oil, respectively.

Results and discussion The morphology of these native magnetite nanoparticles (MNPs) co-precipitated in this study is visible in the SEM and TEM images presented in Fig. 1(a1) and (a2), respectively. It could be seen that they were roughly spherical with a similar particle size. The particle size distribution of these MNPs was statistically determined by analyzing the SEM image using the Image J software developed by the National Institutes of Health

This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

New J. Chem.

View Article Online

Published on 10 July 2018. Downloaded by University of California - Santa Barbara on 7/13/2018 1:17:59 AM.

Paper

NJC

(United States). As displayed in the inset of Fig. 1(a1), the diameter of these MNPs averaged 17  5 nm with the maximum not exceeding 26 nm. According to the low-temperature nitrogen adsorption isotherm, these MNPs exhibited a specific surface area of 62.2  1.4 m2 g1. As a support for the catalyst, an exceptionally large specific surface area was beneficial for immobilizing more catalyst, which notably enhanced the catalytic efficiency. After coating with a poly(styrene-divinylbenzene-glycidyl methacrylate) copolymer, or PST, several MNPs were wrapped simultaneously by the copolymer, forming a larger multi-core–shell nanocomposite with an average particle size of about 75 nm (Fig. 1b1 and b2). This multi-core–shell morphology was formed because the native magnetite nanoparticles produced by the chemical co-precipitation technique tended to aggregate themselves so as to minimize their surface energy. Accordingly, the specific surface area of the nanocomposite decreased to 15.2  2.2 m2 g1, as measured by the low-temperature nitrogen adsorption experiment. This multi-core– shell morphology was also the case for the [email protected]@SO3H nanoparticles (Fig. 1(c)). After sulphonation, the presence of sulfonic acid moieties on the outermost layer of the nanocomposite was further qualitatively verified by FT-IR analysis. As demonstrated in Fig. 2(a), the absorption band peaking at 578 cm1 was assigned to the Fe–O bond from the core of the MNPs. The absorption band peaking at 3001 cm1 was associated with the unsaturated hydrocarbon

Fig. 1 SEM images of the (a1) MNPs and (b1) [email protected] TEM images of the (a2) MNPs, (b2) [email protected] and (c) [email protected]@SO3H nanoparticles. The insets in (a1 and b1) show the histogram of their diameter distribution, statistically obtained by using the Image J program (version 1.40, National Institutes of Health, United States), which converts the image into a binary one using a rational threshold for segmenting the spherical particles from the background.

New J. Chem.

stretching vibration from the benzene structure, derived from the ST and DVB monomeric units in the PST layer. In addition, the peaks at 1605, 1582, 1497, and 1451 cm1 were ascribed to the CQC stretching vibration from the benzene structure, further indicating the presence of the copolymer shell. Furthermore, the successful sulphonation of the copolymer shell was evidenced by the presence of absorptions at 1155, 1049, and 989 cm1, corresponding to the stretching vibration of S–O, and the symmetric and asymmetric vibrations of SQOQO from the sulfonic acid moieties.25 Moreover, as shown in Fig. 2(b), two additional XPS signals peaking at 227.4 and 168.1 eV appeared for [email protected]@SO3H, corresponding to S 2s and S 2p, respectively, which also confirmed that the sulfonic acid groups had been successfully introduced by aromatic sulphonation. The magnetic properties of the catalyst were investigated by using a vibrating sample magnetometer in the field range of 10 o H o 10 kOe. As could be seen in Fig. 3, all three samples displayed superparamagnetism, in which there was almost no coercivity once the magnet was removed. Furthermore, the saturated magnetization of the MNPs, [email protected], and

Fig. 2 (a) FT-IR spectra of the MNPs, [email protected] and [email protected]@SO3H nanoparticles. (b) XPS survey spectra of [email protected] and [email protected]@SO3H.

This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online

Published on 10 July 2018. Downloaded by University of California - Santa Barbara on 7/13/2018 1:17:59 AM.

NJC

Fig. 3 Room-temperature magnetic hysteresis loops of native the MNPs, [email protected], and [email protected]@SO3H nanoparticles. The inset shows the appearance of the 1.0 mg mL1 [email protected]@SO3H catalyst dispersed in oleic acid methyl ester, and its fast response to an externally applied magnetic field.

[email protected]@SO3H nanoparticles was found to be 32, 17, and 13 emu g1, respectively. This decrease in the saturation magnetization could be ascribed to the decreased MNP component after surface modification.22,26 In spite of such a decrease, the magnetization of the [email protected]@SO3H catalyst was still strong enough for it to be separated quickly using an external magnet. As displayed in the inset of Fig. 3, the [email protected]@SO3H catalyst could be magnetically recovered from oleic acid methyl ester within only 3 min. To determine the optimal catalyst dose for the efficient production of biodiesel, the influence of [email protected]@SO3H dose on the biodiesel yield from a feedstock with the FFA content as high as 20% was investigated. We employed this high FFA-containing feedstock to mimic waste or recycled oil, which had been suggested as an effective and cost efficient feedstock for biodiesel production as it is readily available. The molar ratio of methanol to the feedstock was set at 15 : 1, and the reaction was carried out at 100 1C. According to Fig. 4(a), the biodiesel yield, regardless of the [email protected]@SO3H catalyst dose, experienced a surge over the first 120 min, followed by a significantly decelerated increase as a function of time. It was noted that the yield almost approached the maximum after 480 min. For time efficiency, therefore, we chose 480 min as the optimal reaction time to maximize the biodiesel yield using MNPs[email protected]@SO3H as the catalyst. In addition, the final biodiesel yield increased significantly from 61.9 to 91.1% when the [email protected]@SO3H dose was increased from 2 to 5%. Accordingly, the acid value of the final product decreased from 9.3 to 0.8 mg KOH per g. As the catalyst dose was further increased to 8%, the final biodiesel yield only slightly increased to 92.8%. For cost effectiveness, therefore, we chose 5% as the optimal [email protected]@SO3H dose for further experiments. These results also indicated that [email protected]@SO3H prepared herein is capable of catalyzing the transesterification of

Paper

Fig. 4 Influence of (a) [email protected]@SO3H dose, (b) temperature, (c) molar ratio of methanol to feedstock oil, and (d) FFA content on the biodiesel yield by using [email protected]@SO3H as the nanocatalyst in a one-step procedure. The reported data are the means  standard deviation of triplicate samples for each measurement.

triglycerides and the esterification of FFAs simultaneously, enabling a simple, one-step procedure for biodiesel production from feedstocks containing a high content of FFAs. Compared with the two-step procedure, in which the esterification of FFAs and the transesterification of triglycerides were performed successively, a one-step procedure overcame the problem of repeated separation and purification. Temperature impacts both rate and extent of most chemical reactions. In our experiment, three temperatures, i.e. 80 1C, 100 1C, 120 1C were selected to run the experiment, and the biodiesel yield was measured as a function of time. The molar ratio of methanol to the feedstock was set at 15 : 1, while the catalyst dose was set at 5%. As illustrated in Fig. 4(b), the yield of the biodiesel increased rapidly during the first 120 min at three temperatures, owning to the high content of the reactants (i.e. triglycerides, free fatty acids, and methanol) at the very beginning. When the optimal reaction time, 480 min, elapsed, the final biodiesel yield at 80 1C was found to be 69.5%, while the yield at 100 and 120 1C increased to 91.1% and 95.4%, respectively. In spite of the higher biodiesel yield at 120 1C, the increase was found to be only 4.3% compared with that at 100 1C. Accordingly, the benefit of the higher yield would be offset by higher energy consumption resulting from the high reaction temperature. Therefore, we chose 100 1C as the optimal reaction temperature for further experiments. Meanwhile, the high yield of the biodiesel at 100 1C indicating that the [email protected]@SO3H catalyst prepared in this study exhibited good heat resistance. The molar ratio of methanol to the feedstock is another important factor that influences the conversion degree of the feedstock oil to biodiesel. In principle, three moles of methanol are required per mole of triglyceride for the stoichiometric transesterification reaction. However, in the production of biodiesel

This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

New J. Chem.

View Article Online

Published on 10 July 2018. Downloaded by University of California - Santa Barbara on 7/13/2018 1:17:59 AM.

Paper

NJC

from waste or recycled oil, the molar ratio of methanol to the feedstock should be much larger than 3 : 1, because FFAs also consume methanol in esterification. In this study, three different molar ratios of methanol to the feedstock, i.e. 5 : 1, 15 : 1 and 25 : 1 were adopted in the biodiesel production experiment. The reaction was run for 480 min with 20% FFAs and 5% [email protected]@SO3H at 100 1C. According to Fig. 4(c), the difference among the three reactions was obvious, and the yield of the biodiesel increased as a function of the methanol content. The yield was only 51.3% when the molar ratio of methanol to the feedstock oil was 5 : 1. However, the yield of the biodiesel reached 91.1% and 95.9%, respectively, when the molar ratio was increased from 15 : 1 to 25 : 1. Since the yield only increased slightly by 4.8% when the molar ratio of methanol to the feedstock was increased from 15 : 1 to 25 : 1, we chose 15 : 1 as the optimal methanol dose in the reaction. In general, the FFA content in waste or recycled oil varies, and therefore the influence of FFA content on the biodiesel yield must be taken into account. The yields of biodiesel from feedstock oils with 10%, 20%, 40% FFA contents are presented in Fig. 4(d). The molar ratio of methanol to the feedstock was set at 15 : 1, and the reaction was carried out at 100 1C with a 5% catalyst dose. At the very beginning of the reaction, the conversion rate to methyl ester was greatly correlated to the FFA content. This was simply because the concentration of methanol was higher at the beginning of the reaction, thus resulting in more successful collisions between the reactant particles, which pushed the reaction to proceed in the forward direction. After 120 min, the biodiesel yield gradually levelled off. The yield in the case of 10% FFAs was found to be 62.0%, while doubling the FFA content to 20% resulted in a significant increase in the yield to 91.1%. As the FFA content was further increased to 40%, the final biodiesel yield deceased to less than 79.0%. This phenomenon might be associated with the higher content of water produced by the esterification of the FFAs, which suppressed the reaction from proceeding in the forward direction. In our experiment, a reflux condenser was adopted so that the water produced would return back to the reactor, and inhibit the formation of methyl ester from FFAs. This indicated that [email protected]@SO3H prepared herein could be applied to a feedstock with 20% FFAs directly, and there was no need to separate water, which suppressed the formation of biodiesel. However, when the FFA content was increased to as high as 40%, the resultant water must be isolated if a higher yield is about to be achieved. For a catalyst, reusability is associated with its cost effectiveness for potential applicability. In this study, after carrying out the biodiesel production experiment under the optimal conditions (catalyst dose = 5%; temperature = 100 1C; reaction time = 480 min; molar ratio of methanol to feedstock oil = 15 : 1; FFA content = 20%), the [email protected]@SO3H catalyst was recycled from the biodiesel with a magnet, and washed with ethanol twice before drying. The recycled catalyst was then employed for catalyzing biodiesel production under the optimal conditions again. As could be seen in Fig. 5, the biodiesel yield decreased slightly after five continuous catalytic

New J. Chem.

Fig. 5 Biodiesel yield by using the [email protected]@SO3H as the catalyst in five cycles ([email protected]@SO3H dose = 5%; temperature = 100 1C; reaction time = 480 min; molar ratio of methanol to feedstock oil = 15 : 1; FFA content = 20%). The reported data are the means  standard deviation of triplicate samples for each measurement.

cycles, but the yield still exceeded 86%. Thus, it could be concluded that the [email protected]@SO3H catalyst was useable, with reproducible catalytic potency, which promised its cost effectiveness as a catalyst for biodiesel production from feedstocks with a high FFA content. Furthermore, the resultant biodiesel in each catalytic cycle was heated up to 800 1C under a nitrogen atmosphere by TG. It was found by EDX analysis that neither S nor Fe element was present in the char residue (see the ESI,† Fig. S1), indicating complete removal of the catalyst under magnetic fields.

Conclusions Nanoscale magnetite prepared by a chemical co-precipitation technique could be capsulated with a poly(styrene-divinylbenzeneglycidyl methacrylate) copolymer, forming a shell which could be further sulfonated, yielding a recyclable catalyst suitable for biodiesel production from high free fatty acid-containing feedstocks. The biodiesel yield by using the sulfonated magnetite catalyst was a complex function of catalyst dose, temperature, molar ratio of methanol to the feedstock oil, and the content of free fatty acids. The catalyst could be conveniently recycled from the oil, and reused without significantly compromising its catalytic potency, which promised its potential application in the environmentally-benign production of biodiesel from a cheap feedstock with a high content of free fatty acids, such as waste or recycled oil.

Conflicts of interest The authors declare no competing financial interest.

This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online

NJC

Published on 10 July 2018. Downloaded by University of California - Santa Barbara on 7/13/2018 1:17:59 AM.

Acknowledgements We gratefully acknowledge financial support of this work by the Science and Technology Planning Project of Sichuan Province (2017JY0218) and the Solid-state Fermentation Resource Utilization Key Laboratory of Sichuan Province (2015GTY012). We also thank Zhonghui Wang (College of Light Industry, Textile and Food Engineering, Sichuan University) for her help in conducting TG experiment.

Notes and references ¨o ¨k and X. Tang, Energy Policy, 2013, 52, 797–809. 1 M. Ho 2 S. H. Kim, S.-j. Kim, S. Park and H. K. Kim, J. Mol. Catal. B: Enzym., 2013, 85, 10–16. 3 M. Y. Koh and T. I. M. Ghazi, Renewable Sustainable Energy Rev., 2011, 15, 2240–2251. 4 A. B. Avagyan, J. Environ. Prot., 2010, 1, 183. 5 M. Bowman, D. Hilligoss, S. Rasmussen and R. Thomas, Hydrocarbon Process., 2006, 85, 103. 6 B. K. Mishra and R. Kumar, IOSR J. Appl. Chem., 2012, 2, 41–45. 7 A. Murugesan, C. Umarani, R. Subramanian and N. Nedunchezhian, Renewable Sustainable Energy Rev., 2009, 13, 653–662. 8 A. Demirbas, Energy Convers. Manage., 2009, 50, 14–34. 9 C. Ilkilic and R. Behcet, Energy Sources, Part A, 2010, 32, 839–850. 10 M. M. R. Talukder, J. C. Wu and L. P.-L. Chua, Energy Fuels, 2010, 24, 2016–2019.

Paper

11 C. Huang, M.-h. Zong, H. Wu and Q.-p. Liu, Bioresour. Technol., 2009, 100, 4535–4538. 12 R. Banani, S. Youssef, M. Bezzarga and M. Abderrabba, J. Mater. Environ. Sci., 2015, 6, 1178–1185. 13 J. Van Gerpen, Fuel Process. Technol., 2005, 86, 1097–1107. 14 R. Liu, Y. Guo, G. Odusote, F. Qu and R. D. Priestley, ACS Appl. Mater. Interfaces, 2013, 5, 9167–9171. 15 J. Liu, S. Z. Qiao and Q. H. Hu, Small, 2011, 7, 425–443. 16 V. Polshettiwar, R. Luque, A. Fihri, H. Zhu, M. Bouhrara and J.-M. Basset, Chem. Rev., 2011, 111, 3036–3075. ¨th, Angew. Chem., 17 A. H. Lu, E. E. L. Salabas and F. Schu Int. Ed., 2007, 46, 1222–1244. 18 X.-C. Yin, X. Liu, J.-C. Fan, J.-J. Wu, J.-L. Men and G.-S. Zheng, J. Appl. Polym. Sci., 2017, 134, DOI: 10.1002/app.44466. 19 Y. Fan, F. Su, K. Li, C. Ke and Y. Yan, Sci. Rep., 2017, 7, 45643. 20 M. Rengasamy, K. Anbalagan, S. Kodhaiyolii and V. Pugalenthi, RSC Adv., 2016, 6, 9261–9269. 21 E. C. S. Santos, T. C. dos Santos, R. B. Guimaraes, L. Ishida, R. S. Freitas and C. M. Ronconi, RSC Adv., 2015, 5, 48031–48038. 22 J. Chang, Y. Chen, S. Zhao, X. Guan and H. Fan, Polym. Chem., 2015, 6, 8150–8160. 23 X. Guan, S. Yan, Q. Zeng, Z. Xu, Y. Chen and H. Fan, Fibers Polym., 2016, 17, 1131–1139. 24 S. Hu, Y. Guan, Y. Wang and H. Han, Appl. Energy, 2011, 88, 2685–2690. 25 B. K. Nath, A. Khan and J. Chutia, Mater. Res. Bull., 2015, 70, 887–895. 26 A. S. Bhatt, D. K. Bhat and M. S. Santosh, J. Appl. Polym. Sci., 2011, 119, 968–972.

This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

New J. Chem.

Suggest Documents