Esterification of rapeseed oil fatty acids using a carbon-based heterogeneous acid catalyst derived from cellulose

Accepted Manuscript Title: ESTERIFICATION OF RAPESEED OIL FATTY ACIDS USING A CARBON-BASED HETEROGENEOUS ACID CATALYST DERIVED FROM CELLULOSE Author: Kristaps Malins Janis Brinks Valdis Kampars Ilze Malina PII: DOI: Reference:

S0926-860X(16)30143-0 http://dx.doi.org/doi:10.1016/j.apcata.2016.03.020 APCATA 15812

To appear in:

Applied Catalysis A: General

Received date: Revised date: Accepted date:

15-2-2016 17-3-2016 18-3-2016

Please cite this article as: Kristaps Malins, Janis Brinks, Valdis Kampars, Ilze Malina, ESTERIFICATION OF RAPESEED OIL FATTY ACIDS USING A CARBONBASED HETEROGENEOUS ACID CATALYST DERIVED FROM CELLULOSE, Applied Catalysis A, General http://dx.doi.org/10.1016/j.apcata.2016.03.020 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ESTERIFICATION OF RAPESEED OIL FATTY ACIDS USING A CARBONBASED

HETEROGENEOUS

ACID

CATALYST

DERIVED

CELLULOSE

Kristaps Malins*, Janis Brinks, Valdis Kampars, Ilze Malina

Institute of Applied Chemistry, Paula Valdena Str. 3, LV-1048, Riga, Latvia

FROM

Graphical abstract

Highlights 1. The conversion of cellulose into catalyst in presence of conc. H2SO4 reached ~61%; 2. High amounts of COOH, OH and SO3H sites have attached on obtained catalyst surface; 3. Acidic sites increase hydrophilicity of catalyst surface and its catalytic performance; 4. The conversion of rapeseed oil fatty acids over obtained catalyst reached 96.7%; 5. The catalyst initial activity can be partially recovered by simple regeneration method.

Abstract The present work demonstrates a single-step preparation of sulfo carbon-based solid acid catalyst (CSO3H) obtained from cellulose by direct carbonization– sulfonation with conc. H2SO4. The variation of reaction conditions, such as, reaction temperature (50-180 oC), weight ratio of H2SO4 to cellulose (5/1–10/1) and reaction time (1-10 h) provided CSO3H with relatively abundant density of acid sites attached on catalyst surface. The CSO3H with the highest density of SO3H groups (0.81 mmol H+/g) was obtained at 100 oC using weight ratio of H2SO4 to cellulose (20/1) after 3 h of reaction. The catalyst was characterized by means of N2 sorption analysis (average pore diameter 12.8 nm, specific surface area 1.38 m2/g, pore volume 0.004 cm3/g), FT-IR, TG, XRD, FE-SEM and S elemental analysis. The conversion of rapeseed oil fatty acids (RFA) in esterification reactions with CSO3H as catalyst was quite similar to the commercial Amberlyst-15 and the biodiesel with ester content of ≥96.5% was successfully obtained under optimized reaction conditions.

Keywords: Cellulose; Carbon-based solid acid catalyst; Sulfonation; Esterification; Amberlyst-15; Biodiesel 1.

Introduction

Biodiesel is renewable, clean biofuel and it can be manufactured from vegetable oils, animal fats and micro-algal oil [1] in transesterification reaction with lower alcohols methanol, ethanol etc. [2]. Typically the feedstock of commercial biodiesel is a food grade vegetable oil such as soya bean, palm and rapeseed oil (RO) [3]. Furthermore, the biodiesel also can be produced from any kind of raw material containing fatty acids or their glycerides. The properties of biodiesel are close to fossil diesel fuel; thus it becomes a promising alternative fuel for the employment in common diesel engines [4]. NaOH, KOH, NaOCH3 are the commonly used homogeneous alkaline catalysts for biodiesel production from glycerides with low content of water and free fatty acids (FFA) [5]. Alkaline catalysts have a high catalytic activity in

transesterification reactions, but even low amounts of the water (≥0.1 wt.%) and FFA in reaction mixture can significantly affect the biodiesel yield [1]. The alkaline catalysts react with FFA and also glycerides in presence of water forming soaps. This leads to irreversible reduction of catalyst concentration in reaction mixture. The susceptibility of acid catalysts to these contaminants is significantly lower [6]. Strong mineral acids as H2SO4 can be successfully used as catalyst for transesterification [4] and esterification [7-8] reactions of glycerides and FFA, respectively. However the homogeneous catalysts are difficult to reuse, due to its separation from reaction mixture which commonly are technologically complicated and energy consuming processes [9]. The preparation of biodiesel using reusable heterogonous acid catalysts is one of the solutions for development of low-waste technologies and cost minimization of biodiesel production process. High strong acid density and suitable textural properties of carbon-based solid acid catalysts (Csac) usually provides promising catalytic activity in esterification reactions of FFA [10-12]. Furthermore, the carbon materials or their precursors are low cost feedstock for Csac preparation even more increasing its potential for practical application in biodiesel production processes [9]. There are two general groups of Csac described in literature – supported [13] and functionalized catalysts. Functionalized Csac can be synthesized in two main ways – direct sulfonation and sulfonation by alkylation or arylation [10, 14-15]. Direct sulfonation of carbon material is the most extensively studied Csac preparation method. Csac have been prepared from different carbon sources such as microalgae residue [16], vegetable oil asphalt [17], lignin [18], glucose [19], glucose/starch mixture [20], cellulose [21] etc. Typically these precursors of carbon material were partially carbonized in N2 or Ar atmosphere at 400 – 700 oC and then sulfonated with conc. H2SO4 at ~150 – 210 oC or fuming H2SO4 [22]. In the report [21] sulfonation of carbon material was conducted at lower temperature (~80 oC) using H2SO4/SO3 (15 wt.%). Moreover, in the paper [23] the preparation of Csac was performed by partial carbonization–sulfonation of glycerol in single-step at 220 oC. In similar manner a Csac was prepared from residual lignin (derived from Xanthoceras sorbifolia Bunge hulls) [18] and oilseed cake [24] at 150 and 120 oC, respectively. Textural properties and overall acid densities of Csac depend on utilized raw materials and reaction conditions [25-26]. The impact of textural properties such as average pore diameter, specific surface area and pore volume on catalytic activity in biodiesel production processes is directly related to the diffusion ability of methanol, FFA and glycerides through catalyst framework and interaction with strong acid sites during the reaction [10, 15]. Thereby Csac with mesoporous or larger pore size framework is more suitable for biodiesel production processes. There are two types of acidic surface functional groups of Csac: weak – phenolic hydroxyl (OH), carboxyl (COOH) and strong - sulfo (SO3H) acid sites. These oxygen containing acidic sites were formed in partial carbonization of feedstock and sulfonation of obtained carbon material [27]. The influence of SO3H groups on catalytic activity of Csac in esterification reactions is significantly higher in comparison to weak acid sites [10]. Hence, SO3H density is one of the key factors which determine the catalytic activity of Csac in esterification reactions of FFA. The total and strong acid densities of synthesized Csac varied in wide range [9]. However, the Csac with higher SO3H densities than ≥2 mmol H+/g were obtained only in limited number of scientific reports [17]. Furthermore, most of the studies in literature are focused on the Csac preparation using two-step carbonization–sulfonation method without investigation the impact of reaction conditions on SO3H density of catalysts. The present study is conducted to investigate the effect of different experimental conditions such as weight

ratio of H2SO4 to cellulose (5/1 – 30/1), reaction temperature (50 – 180 oC) and reaction time (1 – 10 h) on SO3H density of sulfo carbon-based solid acid catalyst (CSO3H) prepared in single-step synthesis by direct carbonization–sulfonation of cellulose with conc. H2SO4. In many scientific reports pure oleic [10, 18, 20], palmitic [13, 28-29] and stearic acid [13] were used as the main reactants in catalytic tests of Csac. The use of FFA consisting of various long chain fatty acids or its mixture with glycerides is more applicable model providing reaction environment closer to industrial biodiesel production processes when FFA content in vegetable oil is elevated. In the present study the effects such as stirring rate (120-1200 rpm), reaction time (20 - 360 min), reaction temperature (50 - 65 oC), molar ratio of rapeseed oil fatty acids (RFA) to methanol (1/1 – 1/40), catalyst amount (1-40 wt.%) on esterification reactions of RFA over synthesized CSO3H were investigated. The industrial Amberlyst-15, utilized as reference, was also tested under similar experimental conditions. The reusability tests of CSO3H were conducted using pure RFA and RFA (10wt.%)/RO mixture.

2.

Materials and methods 2.1. Materials

Methanol, ethanol, ethyl acetate, cyclohexane, NaHCO3, NaCl, NaOH (purity ≥98 wt.%), conc. mineral acids HCl (35.8 wt.%), H2SO4 (96.3 wt.%) and Amberlyst-15 hydrogen form - dry were supplied from Sigma–Aldrich. The concentrations of mineral acids were determined using volumetric acid - base titration. Medical cotton wool "Premium" (moisture content 8.4 wt.%) was used as cellulosic material and purchased from Mediteks Ltd. C, H and ash content of cellulose was 43.09, 6.21 and 0.06 wt.% (on a dry basis), respectively. N and S content was lower than method detection limit (≤0.3 wt.%). RFA was derived from RO by hydrolysis reaction with water in the presence of H2SO4. The main characteristics of RFA and RO are given in Table 1. Table 1 2.2. Preparation and treatment of catalysts Before the carbonization-sulfonation cellulose was dried at 110 oC for 4 h. Effect of different weight ratios of conc. H2SO4 to cellulose (5/1, 10/1, 15/1, 20/1, 30/1) and reaction time (1, 2, 3, 6, and 10 h) on S content was investigated. The cellulose (10g) was gradually immersed in single-neck round-bottom 500 ml flask filled with heated conc. H2SO4 (50, 100, 150, 180 oC) for ~5-10 min (depends on the pre-determined temperature, to avoid spontaneous rise of reaction temperature) and stirred (magnetic stirrer, 500 rpm). From this point of reference the reaction time of the carbonization–sulfonation of cellulose was measured. After the reaction, mixture was diluted with warm deionized water to reduce concentration of H2SO4 and then filtered. Obtained CSO3H was washed several times with deionized water to pH ~7 and then with methanol. Washed catalyst was refluxed with water/ethanol/ethyl acetate solution (volume ratio 0.1/1.0/1.0) in Soxhlet extractor for 3 h. After

separation and treatment CSO3H was dried at 65 oC for 6 h in vacuum oven (0.9 kPa) and then grinded and sieved to mesh 200. The Amberlyst–15 was submerged in methanol for 12 h at room temperature before catalytic tests. Every experiment was repeated twice and S content, acid densities (SO3H, COOH and phenolic OH) of CSO3H, conversion of cellulose to COS3H and conversion of RFA to rapeseed oil methyl esters (RME) were determined as average arithmetic from two independently repeated experiments. A third independent experiment was performed in cases when previous two experimental values differed more than 5 wt.%, then the average result of experiment was determined using two closest experimental values. In the similar manner results of all quantitative analysis data were obtained and the maximum deviation more than 3.0% was not allowed. 2.3. Characterization of catalyst Ash content of cellulose and CSO3H was determined according to ASTM E1755–01 (2007) standard method. Thermogravimetric (TG) analysis of CSO3H and cellulose was carried out by Simultaneous Thermal Analyzer (STA) 6000 (PerkinElmer). Experiments were conducted using N2 flow rate 10 ml/min, heating rate 10 oC/min at maximum temperature 900 oC. Fourier transform infrared (FT-IR) absorbance spectra of CSO3H in the range of 500–4000 cm−1 were recorded on a Nicolet 5700 FT-IR Spectrometer (Thermo Scientific) using KBr pellets. C, H, N, S elemental analysis of cellulose and CSO3H was determined by EA3000 elemental analyzer (Euro Vector). N2 sorption analysis was performed on a Quadrasorb SI surface area and pore size analyzer at -195.85 oC (Quantachrome Instruments). The CSO3H and Amberlyst15 were dried and degassed at 65 oC and 100 oC for 12 h in vacuum oven (10 Pa) using P2O5 as water adsorbent before N2 sorption analysis. The specific surface area was determined using multipoint Brunauer–Emmett–Teller (BET) method based on the adsorption data in the relative pressure (P/P0) range 0.05–0.30. The total pore volume was estimated from the amount of N2 adsorbed at P/P0 of 0.99. The surface morphology of the CSO3H was investigated by a Field Emission Scanning Electron Microscope (FE-SEM) MIRA LMU (Tescan). FE-SEM was operated at 30.0 kV of an accelerating voltage. Powder X-ray diffraction (XRD) patterns were recorded on a D8 Advance (Bruker) diffractometer using Cu Kα radiation at 40 kV and 40 mA. The samples were scanned in the range of 2θ of 5-60o at a scanning speed of 1o/min. The SO3H density of Amberlyst–15 was calculated from the S content. The densities of SO3H, COOH and phenolic OH groups of CSO3H were determined by method, based on the Boehm acid–base back titration [30]. NaCl (acid density of SO3H groups), NaHCO3 (acid density of SO3H/COOH groups) and NaOH (total acid density), were used as reactants. The acid densities of phenolic OH and COOH groups were calculated from obtained titration analysis data. The 0.2 g of sample and aqueous solution of reactant (30 ml, 0.1 M) were mixed (magnetic stirrer, 300 rpm) at 50 oC for 1 h in 50 ml Erlenmeyer flask. After adding all components the flask was purged with argon. Then the sample was filtered and washed with deionized water (100 ml) several times. Obtained filtrate was diluted to 150 ml with deionized water. Afterwards 15 ml of diluted filtrate was titrated with aqueous NaOH or HCl (0.01 M)

using phenolphthalein as an indicator. NaOH was used as titrant only when acid density of SO3H groups was determined. H2SO4 adsorbed onto catalyst surface can significantly affect or mislead the catalytic test results of RFA esterification reactions. Therefore in order to evaluate the adsorption capacity of H2SO4 onto catalyst matrix the 50 ml of aqueous solution of H2SO4 (molar ratio 1/1, 0.1 M) was mixed (magnetic stirrer 300 rpm) with 0.1g of CSO3H at 65 oC for 1 h. Then CSO3H mixture was separated, washed, refluxed and dried in similar manner as in catalyst preparation. S content of CSO3H mixture before and after treatment with H2SO4 was determined. The behavior of CSO3H (0.4 g) in contact with water, methanol and cyclohexane was visually determined and performed in flat-bottom tubes (2.85 ml). After intensive stirring (30 min) of catalyst with water or solvents the mixtures were centrifuged at 3000 rpm for 10 min in the same flat-bottom tubes. Then the mixtures were allowed to stay at room temperature for 6 h. Described procedure was carried out to catalyst sample without water or solvent for comparison. 2.4. Catalytic tests The effects of stirring rate (120, 600, 1200 rpm), reaction temperature (50, 60, 65 C), catalyst amount (1, 5, 10, 20, 40 wt.% of the RFA amount) and molar ratio of methanol to RFA (1/1, 1/3, 1/5, 1/10, 1/20, 1/40) on esterification reactions of RFA were investigated over CSO3H. Utilized amounts of methanol for all esterification reactions were calculated from acid value (AV) of RFA (Table 1). Esterification reactions were carried out in 500 ml three-neck round-bottom flask equipped with reflux condenser. For all experiments the methanol was mixed with CSO3H and heated to pre-determined temperature. Then a 50 g of RFA or RFA/ RO mixture was added to reaction mixture for ≤0.05-0.10 min. From this point of reference the reaction time of esterification reactions of RFA was measured. During the esterification reaction 1.0 ml of the sample after 20, 40, 60, 90, 120, 180, 240, 300, 360 min was taken from reaction mixture using 1.0 ml plastic pipette with filter to separate CSO3H from the sample. Then a sample was dried in vacuum oven (0.9 kPa) at 90 oC for 30 min to remove residual methanol and water. The ester content and AV of dried samples were determined. Under similar conditions esterification tests of RFA were conducted using Amberlyst-15. Likewise the influence of CSO3H on transesterification reaction of pure RO also was determined. o

2.5. Reusability of catalyst To investigate the reusability of CSO3H the esterification reactions were carried out ten times with the same catalyst 10.0 wt.% (5g) using pure RFA and RFA (10 wt.%)/RO (50 g). The molar ratio of RFA to methanol (1/20), reaction temperature (65 oC), reaction time (4 h) and stirring rate (600 rpm) was used in esterification experiments. After the each reaction cycle the spent catalyst was separated from reaction mixture by centrifugation and washed with RFA/methanol mixture (molar ratio 1/20) to remove adsorbed RME and water formed in previous reaction cycle. Washed catalyst was used in next esterification reaction cycles without further treatment. The experiments were also conducted over CSO3H which was intensively and sequentially washed with ethanol/ethyl acetate solution (volume ratio 1/1), aqueous HCl (0.1 M) and deionized water. Then CSO3H was dried at 65 oC for 3 h in vacuum

oven (0.9 kPa). After separation of catalyst the AV and RME content of biodiesel containing samples (treated in similar manner as in catalytic tests) were determined. 2.6. Characterization of feedstock and biodiesel containing sample Monoglycerides, diglycerides and triglycerides content in RO and RFA was determined by gas chromatography (GC) according to a standard method EN14105. GC system 7890A (Agilent Technologies) with a capillary column HT DB-5 (15 m x 320 µm x 0.1 µm) and a flame ionization detector (FID) was used. The column temperature program: 50 °C hold for 1 min; 15 °C/min up to 180 °C; 7 °C/min up to 230 °C; 10 °C/min up to 370 °C; hold for 5 min at final temperature. Helium was used as a carrier gas at flow rate 5 ml/min (pressure 80 kPa) and the detector temperature was set 390 oC. Injection volume of sample was 1.0 µl. Saponification value of RO was determined according to the EN ISO 3657 standard method. AV of main feedstock – RO and RFA used in experiments was determined according to the standard method EN 14104. The AV of RFA/RME and RFA/RO/RME mixtures in catalytic tests were determined in similar manner using reduced amount of the sample (~ 0.2 - 0.4 g). Fatty acid composition of RO was determined by GC analysis using rapeseed oil fatty acid methyl esters (RMEs) obtained from RO in accordance to the standard method ISO 5509. The fatty acid composition of RMEs and ester content of RFA/RO/RME mixtures were determined using standard method EN14103. The 7890A GC system (Agilent Technologies) equipped with a FID and a HP Innowax capillary column (30 m x 250 µm x 0.25 µm) were used. The column temperature program was set at 200 °C for 25 min and detector temperature at 390 °C. Helium was used as a carrier gas at a flow rate 2 ml/min (pressure 215.5 kPa). Injection volume of sample was 1.0 µl. Fatty acid composition of RMEs was calculated on the basis of peak area ratio of alkyl esters. RMEs peaks were identified by comparison to reference standards. We assumed that RO and RFA have the same fatty acid composition as RMEs. Ester content of RME containing samples were calculated based on the sum of existing fatty acid alkyl esters peak area towards the area of internal standard peak (17–methyl heptadecanoate). 3. Results and discussion 3.1. Effects of reaction conditions on carbonization–sulfonation of cellulose with conc. H2SO4 The S content of CSO3H is the appropriate parameter to estimate density of SO3H groups [10, 14]. The carbonization–sulfonation of cellulose with conc. H2SO4 was optimized using S content as control parameter. Fig. 1 shows the effect of weight ratio of H2SO4 to cellulose on S content of CSO3H. The lowest S content (2.36 wt.%) was obtained using weight ratio of H2SO4 to cellulose 5/1. The increase of weight ratio of H2SO4 to cellulose elevates S content. The maximal S content (2.87 wt.%) was reached using weight ratio of H2SO4 to cellulose 20/1. Further increase of weight ratio of H2SO4 to cellulose does not affect the S content of CSO3H. Fig. 1

Fig. 2 illustrates the effect of reaction temperature on S content of CSO 3H. The lowest S content (1.32 wt.%) was obtained at reaction temperature 50 oC. This can be explained by reduced rate of carbonization–sulfonation related reactions at lower temperature. The maximum S content reached 3.28 wt.% at 100 oC which is more than two times higher than CSO3H obtained at 50 oC. Further increase of reaction temperature reduced the S content of catalyst. The CSO3H with 1.78 wt.% of S content was obtained at maximal studied temperature (180 oC). The negative impact on S content of catalyst at higher reaction temperature than 100 oC can be attributed to low thermal stability of SO3H groups under studied experimental conditions. Fig. 2 Fig. 3 depicts the effect of reaction time on S content of CSO3H. Reaction time has a relatively lower impact on carbonization–sulfonation of cellulose in comparison with weight ratio of H2SO4 to cellulose and reaction temperature. The lowest S content (2.92 wt.%) of COS3H was obtained after 1 h. Increase of reaction time has a positive impact on the catalyst preparation process increasing its S content. The conversion of cellulose to COS3H reached 60-61% (calculated from initial mass of cellulose and obtained catalyst amount) under optimal catalyst preparation conditions when weight ratio of H2SO4 to cellulose 20/1, reaction time 3 h, reaction temperature 100 oC and stirring rate 500 rpm was utilized. Longer reaction time than ~3 h do not increase the S content of COS3H. The main characteristics of COS3H obtained under optimal experimental conditions are shown in Table 2. Fig. 3 3.2. Characterization of catalyst The main characteristics of CSO3H and Amberlyst–15 are shown in Table 2. Table 2 The elemental composition of obtained CSO3H showed significant increase in C 61.78 and decrease in H 4.27 wt.% content (by ~43 and ~31%, respectively) in comparison with cellulose. These changes and increase of S content fundamentally attribute the dehydration of cellulose molecule in presence of high amount of conc. H2SO4 and conversion of partially dehydrated catalyst precursor to sulfonated carbonaceous product. In the FT-IR spectra of CSO3H, the O=S=O vibration bands at 1026 cm−1 (symmetric stretching) and 1172 cm−1 (asymmetric stretching) were identified, indicating the presence of SO3H groups on the catalyst surface. These specific vibration bands were not observed in cellulose FT-IR spectra. Appearance of bending vibration band at 636 cm−1 is characteristic to the –OH groups hydrogen bonded with -SO3H. Another stretching vibration bands at 3426, 1619 and 1707cm−1 are likely related to the presence of –OH, C=C and C=O (carboxyl) groups in catalyst framework. Similar FT-IR absorbance spectra of Csac were described in the report [10, 24].

The catalyst adsorption capacity test of H2SO4 showed insignificant increase of S content (≤0.2 wt.%) of CSO3H after treatment with aqueous H2SO4 (0.1 M) and following purification steps with water, methanol and water/ethanol/ethyl acetate solution (volume ratio 0.1/1.0/1.0). This test showed that adsorbed H2SO4 can be successfully removed from catalyst matrix after intensive treatment with water and polar solvents. Obtained results suggest that SO3H groups after carbonization– sulfonation of cellulose were introduced onto catalyst framework, but not adsorbed. The TG curves of cellulose and CSO3H are shown in Fig. 4. There are evident differences of the form of TG curves for both samples. The TG curve of cellulose showed significantly higher total weight loss (~92%) in comparison with CSO3H at temperature range from 30 to 900 ºC. Furthermore, rapid decomposition and carbonization of cellulose at temperature range from ~270 to ~390 oC was observed. CSO3H showed more moderate thermal decomposition (total weight loss ~60%) due to the obtained catalyst contains high amount of carbon fraction formed in the catalyst preparation process. The temperature peaks of derivative thermogravimetric (DTG) curve were observed at ~90, ~220 and ~405 oC. The weight loss of CSO3H (~15 wt.%) at temperature below ~ 150 oC indicates the presence of adsorbed water in the sample. Further weight loss at temperatures above ~150 oC, refers to the thermal decomposition of various oxygen containing groups (COOH, OH and SO3H) from the framework of catalyst. The authors [24] also observed that the thermal decomposition of COOH, OH and SO3H groups of Csac obtained from lignocellulosic material occurs at temperatures above ~150 oC. Steady weight loss of CSO3H at higher temperature than ~400 oC showed further elimination of oxygen containing groups. Fig. 4

The form of TG curve and XRD pattern (Fig. 5) of obtained catalyst is rather similar to carbonaceous materials such as activated carbon-based catalysts functionalized with 4-sulfophenyl groups (10, 14]. The appearance of broad and asymmetric diffraction peaks of 2θ ~23º (002) and ~42º (101) attributes to randomly arranged amorphous carbon-based structures with poor crystallinity [31]. Fig. 5 The acid densities of COOH, phenolic OH and SO3H groups attached on catalyst surface are shown in Table 2. SO3H density determined using method based on the Boehm acid–base back titration was by 21% lower than strong acid density calculated from the S content. This indicates that part of the SO3H groups is locked in the catalyst matrix and interaction of reactants with these acidic sites is blocked. CSO3H also contains relatively high densities of the weak acid sites, including COOH and phenolic OH from incomplete carbonization of cellulose. The surface of carbon materials has a hydrophobic character which is more important for the interaction with vegetable oil and FFA in biodiesel preparation processes [9, 24]. On the other hand, the oxygen containing acidic groups on the surface of catalyst decrease surface hydrophobicity of carbonaceous material. The hydrophilic surface of catalyst is important for interaction with the polar methanol. The behavior of CSO3H with water, methanol and cyclohexane are depicted in Fig. 6. Obvious increase of CSO3H volume in water, polar and nonpolar solvent was observed. This indicates that CSO3H has an amphiphilic character, but its interaction with water and polar solvents are more

pronounced. Similar observations were described in the paper [24] using biphasic mixture of water and hexane for test. Fig. 6 Fig. 7 illustrates the sorption isotherm of CSO3H with evident hysteresis loop in the P/P0 range of ~0.05 - 0.96. The textural properties CSO3H and commercial catalyst Amberlyst-15 are summarized in Table 2. CSO3H has significantly lower pore volume and specific surface area than commercial Amberlyst-15. Fig. 7 Likewise, the FE-SEM image (Fig. 8) of the CSO3H shows that catalyst structure has a low porosity. Obtained results suggest that CSO3H is partially carbonized cellulosic material containing high amount of carbon fraction and COOH, phenolic OH, SO3H groups attached on catalyst framework. Furthermore, CSO3H has a low porosity, but high interaction and swelling ability in contact with water, polar and nonpolar solvents. Fig. 8

3.3. The catalytic performance of sulfo carbon-based solid acid catalyst (CSO3H) on conversion of rapeseed oil fatty acids. The effect of stirring rate on conversion of RFA over CSO3H is depicted in Fig. 9. Lowest conversion of RFA at the beginning of reaction to 240 min was obtained at stirring rate of 120 rpm. The conversions of RFA after 240 min of reaction at stirring rate from 120 to 1200 rpm almost coincide. Likewise all conversion data of RFA obtained at stirring rate ≥600 rpm were almost identical and after 360 min of reaction reached 90.0-92.3%. Obtained results showed that under studied experimental conditions stirring rate has a low influence on esterification reaction rate and conversion of RFA, notably when stirring rate higher than 600 rpm was used. Fig. 9

The temperature has a significant impact on esterification reaction rate and conversion of RFA (Fig 10.). The lowest conversion of RFA was obtained at 50 oC. The rise of reaction temperature significantly increased conversion of RFA. After 360 min of reaction the conversion of RFA at 65 oC reached 92.3%. Fig. 10 Likewise to the reaction temperature the catalyst amount and molar ratio of RFA to methanol has a similar effect on the esterification reaction rate and conversion of RFA (Fig. 11-12). Higher esterification reaction rate and conversion of RFA were observed when increased catalyst amount or molar ratio of RFA to methanol was utilized. Highest conversions of RFA were obtained using 20-40 wt.% of catalyst and after 360 min of reaction reached 96.7-97.2%. Furthermore, the highest conversions of RFA were obtained using molar ratios of RFA to methanol in range from 1/10 to

1/40. Obtained conversions of RFA were quite similar using these amounts of methanol and after 360 min of esterification reaction reached 89.8-92.9%. Increased amount of CSO3H in reaction mixture has a positive effect on interaction ability of catalyst surface with RFA and methanol. Biodiesel with ester content ≥96.5% (requirements of European biodiesel standard EN 14214) can be obtained from pure RFA in single-step using molar ratio of RFA to methanol 1/20, CSO3H amount 20 wt.% reaction temperature 65 oC, reaction time 360 min and stirring rate 600 rpm. There is a high feasibility that obtained CSO3H and explored optimal conditions can be successfully used for the esterification of another kind of FFA. Fig. 11 Fig. 12 The catalytic tests under similar experimental conditions also were performed with RO. It was observed that under studied experimental conditions the CSO3H has a weak impact on transesterification reaction of RO and the conversion of RO to RME achieved only 1.4%. Fig. 13 The conversion of RFA to RME in presence of CSO3H and commercial Amberlyst-15 are shown in Fig. 13. The higher reaction rate at the beginning of the esterification reaction was observed over Amberlyst–15 using identical catalyst amounts by weight. Despite the fact that Amberlyst–15 has a significantly higher SO3H density, specific surface area, pore volume and average pore diameter than CSO3H (Table 2) the conversion of RFA was quite similar for both catalysts when esterification reaction time longer than ~180 min was used. The water that forms in large quantities during the esterification reaction has an adverse effect on the equilibrium of esterification reaction and conversion of RFA. Therefore the esterification reaction curves showed decrease of reaction rate and rather similar conversions of RFA at the end of reaction were observed. The conversion of RFA was by ~44% higher in presence of CSO3H than Amberlyst–15 when identical amounts of SO3H sites in reaction mixture were utilized in catalytic tests. Under these experimental conditions total acid density of CSO3H is significantly higher than Amberlyst-15. High amounts of weak (COOH and phenolic OH) acidic sites increase hydrophilicity of catalyst surface [9, 24] and its interaction ability with polar compounds in reaction mixture, thereby the conversion of RFA over CSO3H was higher in comparison with commercial Amberlyst-15. 3.4. Reusability of catalyst The reusability of the CSO3H through ten esterification reaction cycles of RFA and RFA/RO mixture is shown in Fig. 14. Rapid deactivation of the catalyst was observed when it was separated by centrifugation from reaction mixture and then washed with RFA/methanol mixture. After ninth reaction cycle the conversion of RFA decreased to 16.3 %. This suggests to the adsorption of esterification reaction products such as RME, water and different rapeseed oil containing impurities on catalyst surface, hinders the interaction of RFA and methanol to acidic sites of CSO3H. The catalytic performance of the spent CSO3H can be partially recovered by intensive and sequential washing with ethanol/ethyl acetate solution (volume ratio

1/1), aqueous HCl (0.1 M), deionized water and drying. The conversion of RFA utilizing this treatment methodology of spent CSO3H was significantly higher and reached 80.2 % after ninth esterification reaction cycle. It was also observed that the gradual decrease of RFA conversion after each subsequent reaction cycle was more robust when pure RFA was used in experiments. While the conversion of RFA using RFA (10 wt.%)/RO mixture was higher and reached 92.0-95.1 % in all esterification reaction cycles. This indicates to the more intensive and irreversible elimination of SO3H groups from catalyst framework in presence of water formed in larger quantities during the esterification reaction of pure RFA in comparison to RFA (10 wt.%)/RO. The presence of water also has an adverse effect on the equilibrium of esterification reactions. Obtained results suggest that CSO3H has a suitable and reusable catalyst for successful catalytic esterification of FFA. Fig. 14 4. Conclusions The cellulose can be converted into a CSO3H by a single-step process of carbonization-sulfonation with conc. H2SO4. The weight ratio of H2SO4 to cellulose and reaction temperature has a stronger impact on SO3H density of obtained CSO3H than reaction time. Under optimal catalyst preparation conditions the conversion of cellulose to COS3H reached 60-61% when weight ratio of H2SO4 to cellulose 20/1, reaction time 3 h, reaction temperature 100 oC and stirring rate 500 rpm was utilized. Despite the fact that CSO3H has a significantly lower average pore diameter (12.8 nm), specific surface area (1.38 m2/g), pore volume (0.004 cm3/g), and SO3H density (0.81 mmol H+/g) than commercial Amberlsyt-15 the conversions of RFA under studied experimental conditions were close for both catalysts and reached 92.3 and 93.4%, respectively. The CSO3H also contains abundant amount of weak (COOH and phenolic OH) acidic groups attached on catalyst surface. Acidic sites increase hydrophilicity of catalyst surface and its interaction ability with polar compounds in reaction mixture. The biodiesel with ester content ≥96.5% (requirements of European biodiesel standard EN 14214) can be successfully obtained from pure RFA over under explored optimal esterification reaction conditions. Inexpensive, widely available raw materials, simple single-step preparation method and high catalytic performance of CSO3H increases its application potential for biodiesel production from feedstock with elevated content of FFA. Furthermore, the catalyst initial activity of spent CSO3H can be partially recovered by intensive and sequential washing with ethanol/ethyl acetate solution (volume ratio 1/1), aqueous HCl (0.1 M), deionized water and drying.

Acknowledgments This work was supported by the National Research Program of Latvia ‘‘LATENERGI’’

References 1. A. Abbaszaadeh, B. Ghobadian, M.R. Omidkhah, G. Najafi, Energ. Convers. Manage. 63 (2012) 138–148. 2. M.K. Lam, K.T. Lee, A.R. Mohamed, Biotechnol. Adv. 28 (2010) 500–518. 3. M. Balat, Energ. Convers. Manage. 52 (2011) 1479–1492. 4. K. Malins, V. Kampars, R. Kampare, J. Prilucka, J. Brinks, R. Murnieks, L. Apseniece, Fuel 137 (2014) 28–35. 5. M. Fangrui, M.A. Hanna, Bioresource. Technol. 70 (1999) 1–15. 6. A. Kawashima, K. Matsubara, K. Honda, Bioresource. Technol.100 (2009) 696–700. 7. J.M. Marchetti, A.F. Errazu, Biomass Bioenerg. 32 (2008) 892–895. 8. J. Brinks, K. Malins, V. Kampars, J. Prilucka, L. Apseniece, Pol. J. Chem. Technol. 15 (2013) 54–59. 9. L.J. Konwar, J. Boro, D. Deka, Renew. Sust. Energ. Rev. 29 (2014) 546–564. 10. L. Geng, G. Yu, Y. Wang, Y. Zhu, Appl. Catal. A-Gen. 427-428 (2012) 137– 144. 11. X.Y. Liu, M. Huang, H.L. Ma, Z.Q. Zhang, J.M. Gao, Y.L. Zhu, X.J. Han, X.Y. Guo, Molecules 15 (2010) 7188–7196. 12. M. Toda, A. Takagaki, M. Okamura, J. N. Kondo, S. Hayashi, K. Domen, M. Hara, Nature 438 (2005) 178. 13. J.R. Kastner, J. Miller, D.P. Geller, J. Locklin, L.H. Keith, T. Johnson, Catal. Today. 190 (2012) 122–132. 14. K. Malins, V. Kampars, J. Brinks, I. Neibolte, R. Murnieks, Appl. Catal. BEnviron. 176 (2015) 553–558. 15. R. Liu, X. Wang, X. Zhao, P. Feng, Carbon 46 (2008) 1764 –1669. 16. X. Fu, D. Li, J. Chen, Y. Zhang, W. Huang, Y. Zhu, J. Yang, C. Zhang, Bioresource. Technol. 146 (2013) 767–770. 17. Q. Shu, Q. Zhang, G.H. Xu, Z. Nawaz, D.Z. Wang, J.F. Wang, Fuel Process. Technol. 90 (2009) 1002–1008. 18. F. Guo, Z.L. Xiu, Z.X. Liang, Appl. Energ. 98 (2012) 47–52. 19. F.A. Dawodu, O. Ayodele, J. Xin, S. Zhang, D. Yan, Appl. Energ. 114 (2014) 819–826. 20. G. Chen, B. Fang, Bioresource. Technol. 102 (2011) 2635-2640. 21. M. Hara, Top. Catal. 53 (2010) 805–810. 22. A.M. Dehkhoda, A.H. West, N. Ellis, Appl. Catal. A-Gen. 382 (2010) 197– 204. 23. B.L.A. Prabhavathi Devi, T. Vijai Kumar Reddy, K. Vijaya Lakshmi, R.B.N. Prasad, Bioresource Technol. 153 (2014) 370–373. 24. E.M. Santos, A.P.D.C. Teixeira, F.G. da Silva, T.E. Cibaka, M.H. Araújo, W.X.C. Oliveira, F. Medeiros, A.N. Brasil, L.S. de Oliveira, R.M. Lago, Fuel 150 (2015) 408–414. 25. M. Zong, Z. Duan, W. Lou, T.J. Smith, H. Wu, Green Chem. 9 (2007) 434– 437. 26. M. Okamura, A. Takagaki, M. Toda, J.N. Kondo, K. Domen, T. Tatsumi, M. Hara, S. Hayashi, Chem. Mater. 18 (2006) 3039–3045. 27. W. Lou, M. Zong, Z. Duan, Bioresource Technol. 99 (2008) 8752–8758. 28. C. Poonjarernsilp, N. Sano, H. Tamon, Appl. Catal. B-Environ. 147 (2014) 726–732.

29. J.M. Fraile, E. García-Bordejé, E. Pires, L. Roldán, J. Catal. 324 (2015) 107118. 30. H.P. Boehm, Carbon 40 (2002) 145–149. 31. J. Zhao, L. Yang, F. Li, R. Yu, C. Jin, Carbon 47 (2009) 744–751.

Fig. 1 The effect of weight ratio of H 2SO4 to cellulose on S content of sulfo carbon-based solid acid catalyst (CSO3H). Carbonization–sulfonation conditions: reaction time 3 h, reaction temperature 150 o C, stirring rate 500 rpm.

Fig. 2 The effect of reaction temperature on S content of sulfo carbon-based solid acid catalyst (CSO3H). Carbonization–sulfonation conditions: weight ratio of H2SO4 to cellulose 20/1, reaction time 3 h, stirring rate 500 rpm.

Fig. 3 The effect of reaction time on S content of sulfo carbon-based solid acid catalyst (CSO3H). Carbonization–sulfonation conditions: weight ratio of H2SO4 to cellulose 20/1, reaction temperature 100 oC, stirring rate 500 rpm.

Fig. 4. Thermogravimetric (TG) and derivative thermogravimetric (DTG) curves of sulfo carbon-based solid acid catalyst (CSO3H) and TG of cellulose.

Fig. 5. X-ray diffraction pattern of sulfo carbon-based solid acid catalyst (CSO3H).

Fig. 6. The behavior of sulfo carbon-based solid acid catalyst (CSO3H) in contact with water, methanol and cyclohexane.

Fig. 7. N2 sorption isotherm of sulfo carbon-based solid acid catalyst (CSO3H).

Fig. 8. FE-SEM image of sulfo carbon-based solid acid catalyst (CSO3H).

Fig. 9. The effect of stirring rate on conversion of rapeseed oil fatty acids (RFA) in presence of sulfo carbon-based solid acid catalyst (CSO3H). Esterification reaction conditions: molar ratio of RFA to methanol 1/20, catalyst amount 10 wt.%, reaction temperature 65 oC.

Fig. 10. The effect of reaction temperature on conversion of rapeseed oil fatty acids (RFA) in presence of sulfo carbon-based solid acid catalyst (CSO3H). Esterification reaction conditions: molar ratio of RFA to methanol 1/20, catalyst amount 10 wt.%, stirring rate 600 rpm.

Fig. 11. The effect of catalyst amount on conversion of rapeseed oil fatty acids (RFA) in presence of sulfo carbon-based solid acid catalyst (CSO3H). Esterification reaction conditions: molar ratio of RFA to methanol 1/20, reaction temperature 65 oC, stirring rate 600 rpm.

Fig. 12. The effect of molar ratio of rapeseed oil fatty acids (RFA) to methanol on conversion of RFA in presence of sulfo carbon-based solid acid catalyst (CSO3H). Esterification reaction conditions, catalyst amount 10 wt.%, reaction temperature 65 oC, stirring rate 600 rpm.

Fig. 13. Esterification of rapeseed oil fatty acids (RFA) in presence of sulfo carbon-based solid acid catalyst (CSO3H) and Amberlyst-15. Esterification reaction conditions: molar ratio of RFA to methanol 1/20, reaction temperature 65 oC, stirring rate 600 rpm.

Fig. 14. Reusability of sulfo carbon-based solid acid catalyst (CSO3H) after esterification reaction cycles of rapeseed oil fatty acids (RFA) and RFA (10 wt.%)/rapeseed oil (RO) mixture. Esterification reaction conditions: molar ratio of RFA to methanol 1/20, amount of catalysts 10 wt.%, reaction time of cycle 4 h, reaction temperature 65 oC, stirring rate 600 rpm.

Table 1 The main characteristics of rapeseed oil fatty acids (RFA) and rapeseed oil (RO). Property Value RFA RO Monoglycerides, wt.% 0.8 0.3 Diglycerides,wt.% 0.9 0.7 Triglycerides, wt.% 0.0 97.9 Saponification value, mg KOH/g N.A.a 191.71 Acid value, mg KOH/g 199.91 0.01 Fatty acid composition of RO, wt.% Palmitic acid (C16:0) 4.1 Stearic acid (C18:0) 1.4 Oleic acid (C18:1) 62.5 Linoleic acid (C18:2) 21.7 α–Linolenic acid (C18:3) 8.7 Arachidic acid (C20:0) 0.4 Other fatty acids 1.2 a Not analyzed

Table 2 Textural properties, acidity and S content of sulfo carbon-based solid acid catalyst (CSO3H) and Amberlyst–15. SBETb, V c, Dd, S Total acid SO3H COOH Phenolic m2/g cm3/g nm content, density density density OH mmol/g mmol H+/g mmol mmol H+/g density H+/g mmol H+/g CSO3H 1.38 0.004 12.8 1.03 3.12 0.81 0.95 1.36 Amberlyst-15 40 0.329 32.9 4.64 4.73 4.64 N.A.a a Not analyzed b Specific surface area c Pore volume d Average pore diameter