Kinetics of ethylic esterification of lauric acid on acid activated montmorillonite (STx1-b) as catalyst

Kinetics of ethylic esterification of lauric acid on acid activated montmorillonite (STx1-b) as catalyst

Fuel 181 (2016) 600–609 Contents lists available at ScienceDirect Fuel journal homepage: www.elsevier.com/locate/fuel Full Length Article Kinetics...

651KB Sizes 0 Downloads 15 Views

Fuel 181 (2016) 600–609

Contents lists available at ScienceDirect

Fuel journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Kinetics of ethylic esterification of lauric acid on acid activated montmorillonite (STx1-b) as catalyst Paulo Ricardo Schizaki dos Santos a, Fernando Wypych b, Fernando Augusto Pedersen Voll a, Fabiane Hamerski a, Marcos L. Corazza a,⇑ a b

Department of Chemical Engineering, Federal University of Paraná, 81531-980 Curitiba, PR, Brazil Department of Chemistry, Federal University of Paraná, PO Box 19032, 81531-980 Curitiba, PR, Brazil

h i g h l i g h t s  We used an acid activated montmorillonite as catalyst to fatty acid esterification with ethanol.  High fatty acid conversions were observed.  Eley–Rideal mechanism was successfully used to describe the kinetic reaction.  The catalyst showed activity after reuse cycles.

a r t i c l e

i n f o

Article history: Received 24 February 2016 Received in revised form 2 May 2016 Accepted 3 May 2016

Keywords: Biodiesel Fatty acid esterification Montmorillonite Heterogeneous catalyst Eley–Rideal

a b s t r a c t This work reports a kinetic study of catalyzed esterification of lauric acid with anhydrous ethanol related to biodiesel production. The catalyst consists of a clay mineral (montmorillonite STx1-b) prepared according to procedure previously described in literature. The acid activation was confirmed by X-ray diffraction (XRD) and Fourier transform infrared spectroscopy (FTIR). Operating parameters: temperature (T = 140 and 180 °C), molar ratio ethanol to lauric acid (MR = 6:1 and 12:1) and catalyst loading (CAT = 2 and 10 wt%, in relation to the fatty acid mass) were evaluated as a preliminary study to the kinetics. Temperature is clearly the factor that most contributes to higher conversions of fatty acid, however, its effect was less pronounced with the use of the catalyst. Moreover, reaction kinetic experiments were performed at different conditions (T = 140, 160 and 180 °C; MR = 3:1, 6:1 and 9:1, CAT = 0, 10 and 20 wt%) as well as kinetics modeling are presented. The Eley–Rideal mechanism with surface reaction between adsorbed ethanol and lauric acid in the bulk as the limiting step was proposed to represent the catalyzed reactions, in which the global reaction rate was expressed as the sum of non-catalyzed (thermal conversion) and catalyzed reactions. From the results presented, it can be seen that the catalyst montmorillonite STx1-b was able to lead the system to high conversions in shorter time when compared to the noncatalyzed reaction. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction The depleting reserves of fossil fuel and the increasing environmental concerns are considered an important factor to drive several studies toward the search for alternatives energy sources to supplement or replace fossil fuels, including diesel and gasoline fuels [1,2]. In this context, biodiesel has been receiving increased attention as a valuable alternative to petroleum-derived fuels. It is non-toxic, biodegradable and renewable. Moreover, the use of ⇑ Corresponding author at: Department of Chemical Engineering, Federal University of Paraná, PO Box 19011, Polytechnic Center, Curitiba 81531-980, PR, Brazil. E-mail address: [email protected] (M.L. Corazza). http://dx.doi.org/10.1016/j.fuel.2016.05.026 0016-2361/Ó 2016 Elsevier Ltd. All rights reserved.

biodiesel presents substantially decreasing in the emissions of carbon monoxide, hydrocarbons, particulate material, aromatics and sulfur [3]. It can replace diesel fuel in different applications such as in internal combustion engines without need of modifications and presents small decrease in the performance [4]. In addition, biodiesel increases lubricity, which can prolong engines’ life [5]. However, some disadvantages of biodiesel use are slightly higher NOx emissions, cold start problems and lower energy content [6]. This alternative fuel consists of monoalkyl esters of long chain fatty acids derived from renewable biological feed stocks such refined, edible or nonedible vegetable oils and animal fats [7,8] and most of biodiesel is produced by alkaline-transesterification of triacylglycerol, which is the major component of these raw

Paulo Ricardo Schizaki dos Santos et al. / Fuel 181 (2016) 600–609

materials [9]. This production route is advantageous from an energy point of view, since it is normally carried out at 60–80 °C [9], on the other hand, especially attention must be paid on the quality of the raw materials because this technology is not compatible with oils and fats with high content of free fatty acids (FFA) and water. The alkali catalyst can react with the FFA to form soaps, which reduce the biodiesel yield and inhibits the separation of the esters from glycerol. Additionally, water either contained in the feedstocks or formed during the saponification reaction can hydrolyze the triacylglycerols to diacylglycerols and monoacylglycerols to form more FFA. It is worth mentioning that the maximum amount of FFA acceptable in the mentioned route is below than 2.5 wt% [2]. The high cost of biodiesel is the major obstacle for its commercialization, usually from 10% to 50% more expensive than petroleum-based diesel fuel [10]. The two aspects that influence the cost of biodiesel are the costs of processing and the costs of raw material, which corresponds to the costly of biodiesel production chain, about 60–75% of the total [1]. Therefore, in order to overcome the process difficulties and decrease the process costs and, consequently, the final price of biodiesel, several researchers have sought alternatives catalysts systems and processes [5]. Biodiesel can also be produced by esterification reactions. It is an important production route for feedstock with high content of FFA (acid oils) due to their lower cost than refined feedstock [11,12]. In addition, it is important to mention that the esterification route can also be used to feedstock with moderate free acidity as pre-treatment step in the framework of a conventional transesterification process [11]. Besides the biodiesel production, esterification of short and long chain carboxylic acids is an important route to produce including alkyl esters that are a very important class of chemicals covering a vast variety of applications such as flavors, perfumes, plasticizers, pharmaceuticals, solvents and chemical intermediates [13–18]. For esterification of carboxylic acids with alcohols in presence of acid catalysts, both homogeneous and heterogeneous catalysis have been used [19]. In homogeneous acid catalysis, such as using concentrated sulfuric acid [2], have the potential for application since the performance is not strongly affected by the presence of FFA in the feedstock, moreover, the acid can simultaneously catalyze both esterification and transesterification reactions. However, this process requires high temperatures, high alcohol to oil molar ratios, reaction rate is about 4000 times slower than alkaline homogeneous catalysis and its usage and application presents serious environmental and corrosion problems [5,20]. An alternative to drawbacks above mentioned is the use of solid catalysts, which present several advantages as simpler downstream separation process, since it does not require any washing and further neutralization and the catalyst is recyclable [2,21]. Therefore, several studies involving the use of heterogeneous catalysis have been presented in the literature, which it has been used in the different esterification reactions, such as ion exchange resins [11,19,22–26], zeolites [27,28], different clay minerals [26,29– 34,12] and layered compounds [18,35,36]. Among the options of materials for heterogeneous acid catalysis, clay minerals represent a promising catalyst variety because of their low cost, thermal stability, reusability and mainly because their desirable catalytic properties such as high selectivity, high surface area, high pores dimension and presence of acid sites [32,37,38]. Furthermore, these properties can be improved by acid activation process, which involves treatment of the clay minerals with inorganic acids, leaching part of octahedral coordinated metals from the clay minerals layers, as well as eliminating acid soluble mineral impurities [38,39]. In addition, it is worth mentioning that clay minerals are cheap materials, non-toxics, it

601

can be acid activated by a simple and low cost process, and it also can be used as raw material in the ceramic industries. Regarding the use of mineral clay as catalyst in the esterification of fatty acids some authors have reported promising results of esterification of fatty acid including stearic acid, oleic acid, palmitic acid and lauric acid mainly with methanol and/or ethanol [17,29,33,34,40–44]. All of these studies cited were focused the catalyst synthesis, characterization and catalytic activity for fatty acid esterification with short chain alcohols without measuring and properly exploring the kinetics of these mineral clay catalyzed reactions. Since the kinetics are the fundamental stone for the industrial reactors design and optimization and esterification has acquired further improvement from engineering side its mainly depends on the research of esterification kinetics and studies dedicated to the experimental and kinetics modeling of fatty acid esterification with methanol and ethanol are needed. Indeed, in a previous work [35] we have demonstrated that the noncatalyzed conversion of lauric acid with ethanol is significant and it must be considered during the catalyzed-kinetics of these reactions. The catalyzed reactions of fatty acid esterification provide a gain of conversion over the thermal conversion, up to 30 percentage points (p.p.) in some cases. For kinetic modeling, process simulation and optimization is it extremely important that the fitted kinetic model can deal and predict either catalyzed and noncatalyzed reactions. However, from our best knowledge, there is a lack of information about the kinetics of acid activated montmorillonite-catalyzed of fatty acid esterification with ethanol. In this context, the present work aims the kinetic study of the acid activated montmorillonite STx1-b as a solid catalyst for the esterification of lauric acid with ethanol. The esterification of lauric acid was set as a target reaction due to the interest of ethyl laurate in the biodiesel industry, as well as its importance and demand in food and pharmaceutical industries. Additionally, it is worthy to mention that lauric acid was chosen as model to be a pretty stable and reliable source of saturated fatty acid, which is very important in kinetic modeling. Additionally, babassú oil and macaúba kernel oil with about 44% [45] and 24% [46] of lauric acid have being studied in Brazil to biodiesel production purposes. Likewise, the ethanol use and interest lies in the fact that it is a greener and renewable reactant compared to methanol [18], it can presents better fuel properties and lubricity [47] and it can allows the process intensification of extraction or oil pre-treatment and reaction section in bio-refineries [48]. It was studied the influence of the operating parameters and it was obtained the kinetic of the studied systems. Furthermore, the kinetic experimental data were modeled using the three-steps Eley–Rideal mechanism, with surface reaction between adsorbed ethanol and fatty acid in the bulk phase as the limiting step in addition to the thermal mechanism. Thus, a kinetic model approach that is able to predict the esterification under different lauric acid to ethanol molar ratio, ethanol assays (hydrated and anhydrous), catalyst amount, auto-thermal (non-catalyzed) conversions, and temperature is proposed. 2. Materials and methods 2.1. Experimental section Synthesis of the catalyst used in this work consisted of an acid activation process performed to samples of montmorillonite STx1b, with the chemical formula (Ca0.27Na0.04K0.01)[Al2.41Fe(III)0.09Mg0.71Ti0.03][Si8.00]O20(OH)4. The clay sample was supplied by Clay Minerals Society and they are original to Gonzales County, Texas, USA. Shortly, the montmorillonite sample as received was mixed with a solution of phosphoric acid 0.5 mol L1 (Vetec – 85%) in a

602

Paulo Ricardo Schizaki dos Santos et al. / Fuel 181 (2016) 600–609

1:4 ratio (g mL1). The reaction was carried out in a roundbottomed flask, at the condition of under vigorous stirring and at the temperature of 100 °C during 2 h. The system was heated in a glycerin bath and it was connected to a reflux condenser. After the activation, the material was washed with distilled water until pH close to 7, centrifuged, dried at 110 °C for 24 h and finally grounded into a fine powder. The X-ray diffraction measurements were obtained in a Shimadzu diffractometer model XRD-7000 with Cu Ka X-ray source (40 kV, 30 mA, k = 1.5418 Å), interval of 2h = 3–40°, at a speed of 2° min1 and scanning pace of 0.02°. The infrared spectra (FTIR) were obtained in transmission mode with a Bio-Rad FTS 3500GX spectrometer, using KBr pellets with accumulation of 32 scans in the range of 4000–400 cm1 and resolution of 4 cm1. The esterification reactions of lauric acid (C12H24O2, P98%, Sigma–Aldrich) were performed with ethanol (99.8%, Neon) using a ParrÒ reactor (model 4597), which consists of a stainless steel vessel with 50 mL of capacity equipped with both heating and stirring systems, including a PID control. The total reactants volume in the vessel corresponded about 70% of its volume capacity. The pressure inside the reactor was observed to reach values around the vapor pressure of ethanol at the reaction temperature. The reaction time was counted from the moment that the desired temperature (reaction temperature set) was reached inside the reactor, which was about 20 min. It is worth mentioning that even for the kinetics study each reaction (experiments) was independently carried out. At the end of each time of the kinetics, the reactor was cooled at low temperatures (around 50 °C) and the whole reactant mixture was analyzed. After the reaction, the catalyst was separated by filtration, and the excess of ethanol was recovered by rotary evaporation (450 mmHg, 80 °C, 100 rpm and 30 min). In order to assess the catalyst reusability, the clay recovered in filtration step was washed three times with acetone (J.T. Baker – 99.7%), centrifuged and dried at 110 °C for 24 h, and then reused to study its activity for the esterification reaction. The conversion of lauric acid to fatty esters (ethyl laurate) was measured by the American Oil Chemist’s Society (AOCS) official method Ca-5a-40, which presents good correlation with other analytical methods such Nuclear Magnetic Resonance [29,35]. 2.2. Kinetics A factorial experimental design was applied in this work in order to study the effect of the variables and the interaction among them on the lauric acid esterification catalyzed by acid activated montmorillonite STx1-b. Temperature (T), ethanol to lauric acid molar ratio (MR) and the amount of catalyst in relation to the fatty acid amount (CAT, wt%) have been evaluated to optimize the reaction conditions. These variables were assessed in two levels (MR values of 6:1 and 12:1; T = 140 and 180 °C; 2 and CAT of 10 wt%) with three replicates at the central point. For these experiments, the reaction time as well as the stirring speed were kept at 2 h and 500 rpm, respectively. A systematic study was performed in order to quantify the influence of external mass transfer limiting. The catalytic tests were performed at 180 °C, MR of 6:1, catalyst amount of 10 wt% and different stirrer speeds of 0, 50, 100, 200, 300, 400, 500 and 600 rpm. The set of kinetic data was obtained at different experimental conditions of temperatures (140, 160 and 180 °C), different ethanol to lauric acid (MR) (3:1, 6:1 and 9:1) and at catalyst amount of 10 wt% and 20 wt%. The experimental conversion data were obtained at reaction time varying from 0, 15, 30, 45, 60, 90, 120 until at least 240 min.

2.3. Reaction mechanism and kinetics modeling As showed in a previous work [35], fatty acid esterification reactions with short chain alcohols present considerable reaction rates and conversion without catalyst presence, i.e., this type of reaction is thermal catalyzed (thermal conversion). Therefore, in order to study the influence of the activated montmorillonite STx1-b on the lauric acid esterification with ethanol the kinetic modeling was taking into account both non-catalyzed (or thermal conversion) and montmorillonite STx1-b catalyzed reaction. 2.3.1. Non-catalyzed reaction An elementary kinetic model was proposed to correlate the kinetic experimental data of non-catalyzed reactions. The main reaction between a fatty acid (A) and an alcohol (as ethanol, Et) to produce fatty acid ethyl ester (E) and water (W) is represented by the following equation: k1

A þ Et $ E þ W

ð1Þ

k2

where A, Et, E and W represent one mol of fatty acid (lauric acid), ethanol, ethyl laurate and water, respectively. The reaction rate expression can be expressed as:

ðrAÞNC ¼ k1 aA aEt  k2 aE aW

ð2Þ

where ai represents the activity of each component that can be written by:

ai ¼ xi ci ¼ ci

Ci Ct

ð3Þ

where xi and ci are the molar fraction and activity coefficients of each component ‘‘i” in the mixture. The reaction equilibrium constant (Ke) can be written as:

Ke ¼

k1 ¼ KcKx ¼ KcKc k2

ð4Þ

where at the chemical equilibrium condition:



cE cw cA cEt

Kc ¼ 



CE Cw C A C Et

Kc ¼

ð5Þ e

 ð6Þ e

Therefore, using Eq. (3) into Eq. (2), we derive the following kinetic rate expression:

ðrAÞNC ¼

k1

cA cEt C 2t

C A C Et  k2

cE cW C 2t

! CECW

ð7Þ

In this work, we have assumed that the volume of the reaction is constant and the product between the activity coefficients of both reactants and both products do not vary significantly, then the kinetic rate parameters of toward and backwards reactions of Eq. (7) can be grouped as effectives reaction rate parameters assume constant we the composition variation: k1 cA cEt =C 2t ffi k1 and k2 cE cW =C 2t ffi k2 . Therefore, we rewritten Eq. (7) as follow: eff

  eff eff ðrAÞNC ¼ k1 C A C Et  k2 C E C W

eff

ð8Þ

where eff

KC ¼

k1

eff

k2

ð9Þ

603

Paulo Ricardo Schizaki dos Santos et al. / Fuel 181 (2016) 600–609

Et þ S $ Et  S adsorption of ethanol

ð10aÞ

Et  S þ A $ E  S þ W surface reaction; rate determining step ð10bÞ E  S $ E þ S desorption of ester

ð10cÞ

W þ S $ W  S adsorption=desorption of water A þ Et $ E þ W overall reaction

ð10dÞ ð10eÞ

where S is representing the solid site. Thus, the reaction rate expressions corresponding to the mechanism in Eqs. (10a–e) is be given by:

ðrAÞC ¼

qcat ðk1;cat a a  k2;cat aE aW Þ  A Et  1þ

P

i K ia ai

ð11Þ

Once more combining Eqs. (3)–(6) with Eq. (11), the complete expression for the catalyzed reaction rate can be written as:



ðrAÞC ¼

CE CW qcat keff cat C A C Et  K c



  P 1 þ iK iCi

ð12Þ

  eff where qcat ¼ mcat =V, kcat ¼ kcat;1 cA cEt =C 2t , and K i ¼ K ia ci =C t . Additionally, since that the chemical equilibrium constant must be the same for both catalyzed and non-catalyzed reaction K c is given by Eq. (9). Reaction volume of reactant mixture for all experimental conditions (temperature and pressure) by using the density of pure components at each reaction condition. For pure lauric acid, ethanol and water equations obtained from the DIPPR data bank were used [49], and for ethyl laurate the PC-SAFT equation was used [50]. The reactant mixture was considered an ideal solution (VEx = 0) for all experimental conditions carried out in this study. Finally, the global reaction rate is expressed as the sum of the non-catalyzed and catalyzed reactions.

ðrAÞ ¼ ðrAÞNC þ ðrAÞC

ð13Þ

Since the reaction mixture mainly consist of solvent, C Et can be considered constant following the reaction path. As mentioned in the literature the rate expression in term of concentration is, rather than activities, more convenient for reactor design [22]. In this work, the Arrhenius equation was used to calculate the kinetic constants, considering both direct (j ¼ 1) and reversible (j ¼ 2) non-catalyzed reactions, and for the toward catalyzed reaction, as follow: eff eff

k1;cat



Eaj RT



ðj ¼ 1; 2Þ ðnon-catalyzed reactionÞ Eacat  ¼ k0;cat exp RT ðcatalyzed reactionÞ

kj ¼ k0;j exp

Therefore, considering the non-catalyzed and catalyzed reaction at same time, the kinetic parameters to be fitted are k0;1 , k0;2 , EaR1 EaR2 (pre-exponential factor in the reaction rate constants for the noncatalyzed reaction and the activation energy); k0;cat and EaRcat , which are the pre-exponential factor and the activation energy for catalyzed reaction. Finally, K Et represents the adsorption/desorption coefficient of ethanol in the reactant mixture on the catalyst surface. The objective function (OF) presented in Eq. (15) was minimized using the stochastic Particle Swarm Optimization (PSO) algorithm, for fitting the parameters of the kinetic model. Parameters values obtained from PSO optimization were then used as initial guesses for the parameters and then the kinetic parameters were optimized using the fminsearch subroutine from MatlabÒ.

FO ¼

Nj  M X 2 X Calc X Exp A;ji  X A;ji j

Calc where X Exp A;ji and X A;ji are the experimental and calculated values of

lauric acid conversion (%) of each experimental point ‘‘i” for each kinetic curve ‘‘j” used. M is the kinetic curve set and Nj is the number of experimental data for each kinetic curve. In this work, the standard deviation (rP) in the kinetic parameters were estimated following the procedure presented by [51,52]. Furthermore, the root mean square deviation (rmsd(%)) was used to evaluate the correlation between the model predictions and the experimental data.

vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi uP PN  2 u M Exp Calc j t j i¼1 X A;ji  X A;ji rmsdð%Þ ¼ 100 NOBS

ð16Þ

NOBS is the number of total experimental runs (M  Nj). It is worth to mention that the kinetic parameters were obtained from a global estimation by regressing all experimental kinetic data for all conditions along, including non-catalyzed and catalyzed kinetics. 3. Results and discussion 3.1. Catalyst characterization The activated montmorillonite STx1-b was characterized and compared with its raw material and to results presented in the lit-

b

a 5

ð14Þ

ð15Þ

i

Intensity (a.u.)

2.3.2. Activated montmorillonite STx1-b catalytic reaction The heterogeneous catalyzed-esterification kinetic data experimentally obtained were modeled following the Eley–Rideal model, which is the reaction that is taking place between an adsorbed molecule of one reactant and another one from the bulk phase. In this work, it was assumed the reaction between the ethanol adsorbed on acid sites of the catalyst and the lauric acid in the bulk solution, as proposed by Merchant et al. [23]. These authors presented different adsorption approaches, in which the best fit was obtained when the alcohol assumed adsorbing on the catalysis surface and the carboxylic acid in the bulk phase for the esterification of acetic, propanoic and pentanoic acids catalyzed by Amberlite IR-120. In the original proposal by Merchant et al. [23], the adsorption status of products consists of ester adsorbed and water in bulk. However, as the montmorillonite is a hydroscopic material and it can adsorb water, therefore in the present study was considered the adsorption/desorption of water on this solid material. The reaction mechanism can be given as follows:

10

15

20

25

30

35

40

2θ (°) Fig. 1. X-ray patterns of the raw STx1-b before (a) and after (b) activation with phosphoric acid (0.5 mol L1 and 2 h).

604

Paulo Ricardo Schizaki dos Santos et al. / Fuel 181 (2016) 600–609

erature that used the same activation procedure [30,53,54]. Fig. 1 shows the X-ray diffraction patterns of the raw STx1-b, before and after activation with phosphoric acid (0.5 mol L1) during 2 h. The acid activation process led to the same modifications in the structure of the montmorillonite, according to the literature [30]. The observed diffraction peaks are in agreement with that observed by Zatta et al. [30] and Viani et al. [55], verifying the preservation of the clay bulk structure after acid activation. The basal spacing calculated for the acid activated montmorillonite was 15.7 Å and, after the activation process, the first basal peak (5.6° in 2h) turned broader and less intense after acid activation process, which led the material to lower crystallinity, compared to its precursor. The non-basal reflections (between 18° and 24° in 2h) were less affected by activation process possibly due to the preferential leaching of cations from the octahedral sheet of the montmorillonite structure. Fig. 2 shows the FTIR spectra of the raw STx1-b, before and after activation with phosphoric acid (0.5 mol L1) during 2 h. Comparing the raw montmorillonite (Fig. 2a) to the acid leached sample (Fig. 2b) the bands did not show any significant decrease in intensity, suggesting that the structural changes attributed to the acid treatment are small. According to the reported by Zatta et al. [30], the FTIR bands at 915, 840 and 524 cm1 are attributed to Al–OH–Al, Al–OH–Mg and Al–OH–O–Si vibrations respectively. These bands decreased in intensity after the acid activation process due to the leaching of octahedral cations (Al3+ and Mg2+) from the montmorillonite structure. The presence of free silica is corresponded to the Si–O–Si deformation (466 cm1) and in-the-plane stretching of Si–O (1035 cm1). The results of both analyses are able to confirm the correct catalyst synthesis procedure, as previously described [30].

fore, all experiments were performed at constant stirring speed of 500 rpm. Compared to the methyl esterification, this is one of the parameters that least affect the ethyl ester yield, due to the better miscibility of the reactants [56]. These results show that this type of clay presents a similar behavior with some esterification reactions catalyzed by Amberlyst type resin [22,24,25] and (trans)esterification reactions catalyzed by layered zinc carboxylates [36]. Usually, in these systems, the external mass transfer resistance can be neglected.

3.2. External diffusive phenomena

3.3. Experimental design

The effect of external diffusion limitation on the esterification reaction is directly related to the stirring speed [24]. In order to assess this effect, in the present study, experiments were carried out under the same reactions parameters of 180 °C, ethanol to lauric acid molar ratio of 6:1, catalyst amount of 10 wt% and 2 h of reaction time. The stirring speed was varied from 0, 50, 100, 200, 300, 400, 500 to 600 rpm, as presented in Fig. 3. As it can be seen from Fig. 3 that the conversion of lauric acid does not present significant increase with the increase of the stirring speed, mainly from 200 rpm. The observed behavior indicates that the external diffusion is not the rate-controlling step. There-

Initially, it was evaluated the effect of the temperature (T) and the ethanol to lauric acid molar ratio (MR) in esterification reactions of lauric acid with ethanol in non-catalytic reaction (named blanks). The thermal conversion of lauric acid was measured and each reaction was performed once and the acidity analysis in triplicate. Table 1 presents the acidity results with the standard deviation of the triplicate analysis as well as the lauric acid conversion. The temperature is clearly the factor that contributes the most to the higher conversion and its estimated effect was 39.69 points percent (p.p.) and the percent contribution of 97.20% in the lauric acid conversion values. The effect of molar ratio was 4.86 p.p., and its percent contribution was 1.41%. Moreover, the simultaneous increase of both variables (T  MR) had a positive effect of 3.83 p.p. and a percent contribution of 0.91% in the lauric acid conversion. Both factors as well as the interaction between them were statistically significant, with a 95% confidence level (p-values <0.05), as observed in Fig. 4(A). It can be observed in Table 2 that the effect of the catalyst amount (CAT) shifted the lauric acid conversion to a gain in relation to the non-catalyzed esterification (blanks). Higher amounts

Transmitance (a.u.)

b

a

100

Conversion (%)

95 90 85 80 75 70 0

100

200

300

400

500

600

Stirring Speed (RPM) Fig. 3. Effect of stirring speed on the conversion of lauric acid at 180 °C, MR 6:1, catalyst amount of 10 wt% and 2 h reaction time.

Table 1 Conversion of non-catalyzed reactions after 2 h. Experimental conditions

4000

3500

3000

2000

1500

Wavenumber

1000

Blank Blank Blank Blank Blank

500

(cm-1)

Fig. 2. FTIR spectra of STx1-b before (a) and after (b) acid activation.

a

1 2 3 4 5

a

Results

T

MR

CAT

Acidity (%)

Conversion (%)

140 180 160 140 180

6:1 6:1 9:1 12:1 12:1

0 0 0 0 0

63.00 ± 1.99 27.17 ± 1.81 39.48 ± 0.89 71.75 ± 1.81 28.19 ± 0.41

37.00 72.83 60.52 28.25 71.81

Ethanol used with 0.27 wt% of water measured by Karl Fisher titration.

605

Paulo Ricardo Schizaki dos Santos et al. / Fuel 181 (2016) 600–609

(A)

100

T

39.69

Conversion (%)

RM

-4.86

TxRM

90

80

70

60

3.83

50 0

p=0.05

2

4

6

8

10

12

Ethanol to lauric acid molar ratio (RM)

Effect Estimate

Fig. 5. Lauric acid esterification with ethanol at 180 °C, 10 wt% of catalyst and different ethanol to lauric acid molar ratio (2 h of reaction).

(B) T

39.10

CAT

13.85

MR

-4.67

TxMR

3.90

MRxCAT

1.67

TxCAT

-1.34

TxMRxCAT

-0.79 p=0.05

Effect Estimate Fig. 4. Effects estimate for 22 (a) and 23 (b) factorial design.

of catalyst (10 wt%) provide higher conversion to esters. However, these results presented both lower conversions and conversion gain than the methyl esterification of lauric acid using the same catalyst, as reported by Zatta et al. [12]. The described behavior was expected, whereas ethanol is lesser reactive in comparison to methanol [57]. Better conversions were obtained at high temperatures (180 °C), like the non-catalyzed reactions, at ethanol to lauric acid molar ratio of 6:1 and 10 wt% of catalyst amount. Fig. 4(B) shows that for the reaction conditions catalyzed by acid activated montmorillonite, the factors and the variables that contribute the most

to the highest conversion of lauric acid to ethyl laurate were temperature (effect of 39.10 p.p., percent contribution of 86.61% and the lowest p-value) and amount of catalyst (effect of 13.85 p.p. and percent contribution of 10.87%). Molar ratio and all interactions among the variables were also statistically significant at 95% confidence level (p-values < 0.05). The experimental standard deviation observed at central point (0) triplicates was 1.92% in conversion of lauric acid, and then this value was assumed as the error in all experiments. In order to better understand the ethanol to lauric acid molar ratio (MR) on the conversion, some reaction were carried out at different MR conditions, and these results are presented in Fig. 5. It can be seen that at higher MR (>6:1) the lauric acid conversion is practically constant with the increase in the ethanol to lauric acid ratio. This behavior can be due the water content in the ethanol (0.27 wt%). Increasing the ethanol to lauric acid relation the water to acid relation is also increased. Even smalls amount of water (in terms of mole number) can lead a decreasing in the lauric acid conversion. From this experimental preliminary study, kinetics were measured for this reaction of lauric acid esterification with ethanol catalyzed by activated montmorillonite STx1-b, considering all factors that were observed as significant. 3.4. Kinetic of lauric acid esterification The kinetic data of non-catalyzed (thermal conversion) esterification of lauric acid with ethanol were taken from our previous

Table 2 Lauric acid conversions obtained for reactions catalyzed by acid activated montmorillonite STx1-b after 2 h. Experimental conditions T Run Run Run Run Run Run Run Run Run Run Run a

1 2 3 4 5 6 7 8 9 10 11

140 180 140 180 140 180 140 180 160 160 160

() (+) () (+) () (+) () (+) (0) (0) (0)

Acidity (%)

MR

CAT

6:1 () 6:1() 12:1 (+) 12:1 (+) 6:1 () 6:1 () 12:1 (+) 12:1 (+) 9:1 (0) 9:1 (0) 9:1 (0)

2 () 2 () 2 () 2 () 10 (+) 10 (+) 10 (+) 10 (+) 6 (0) 6 (0) 6 (0)

Ethanol used with 0.27 wt% of water measured by Karl Fisher titration.

Conversion (%)

Results Conversion gain (%)

59.06 ± 1.93 23.31 ± 0.44 70.09 ± 0.50 24.95 ± 1.40 46.32 ± 0.18 11.68 ± 0.20 52.44 ± 0.56 11.56 ± 0.67 29.83 ± 1.90 33.44 ± 1.01 30.50 ± 1.12

40.94 76.69 29.91 75.05 53.68 88.32 47.56 88.44 70.17 66.56 69.50

3.94 3.86 1.66 3.24 16.68 15.49 19.31 16.63 9.65 6.04 8.97

Paulo Ricardo Schizaki dos Santos et al. / Fuel 181 (2016) 600–609

100 90 80

Conversion (%)

work [35]. As mentioned before, for the estimation of kinetic the parameters, the non-catalyzed data of lauric acid esterification with ethanol were used with the activated montmorillonite STx1-b catalyzed kinetic data obtained in the presented work. All experimental kinetic data of catalyzed reaction as well as the fitted curves are presented in Figs. 6–8. The kinetics parameters obtained, their respective standard deviation and the root mean square deviation (rmsd) are showed in Table 3. It is observed from the rmsd value presented in Table 3 that the proposed model was capable of correlate the experimental kinetics obtained. Another important consideration is that the sorption/ desorption equilibrium constant is around 100 times higher than the ethanol constant, showing that the energy related to water adsorption on the with the activated solid catalyst is higher for water. Furthermore, these equilibrium constants (KEt and KW) are temperature independent parameters for this process. In a general way, the kinetic model proposed can be used to simulate the lauric acid esterification with ethanol for both non-catalyzed and montmorillonite STx1-b catalyzed reactions. It is worth mentioning that better fitting was obtained when the water sorption/desorption on the montmorillonite surface was considered, showing the adsorption on this catalyst type is an important step. Fig. 6 depicts the kinetics obtained for the non-catalyzed reaction (thermal conversion), using the raw montmorillonite and two different catalyst amount (10 and 20 wt%). It can be seen from this figure (Fig. 6) that raw montmorillonite does not show any catalytic activity, presenting the same behavior as the non-catalyzed esterification. On the other hand, as expected, the initial reaction rate increases by using the catalyst amount (acid activated montmorillonite) leading to high conversions (approximately 90%) in shorter time (90 min). Although this variable is significant (from the statistical analysis), from the catalyst amount of 10 until 20 wt%, it was not observed any significant difference in both reaction rates, presented by the same kinetic behavior, which were certified by experimental observations and model prediction. Additionally, the use of the catalyst does not affect the equilibrium conversion of the system indicating that this catalyst type does not permanently adsorb water from the bulk phase. The water that is adsorbed/desorbed on the catalyst surface affects just the kinetic rate of the reaction without displacing the equilibrium of the reaction. Such behavior is observed by the same equilibrium conversion that is reach in both systems, approximately 90%.

70 60 50 40 30 20 10 0

50

100

150

200

250

300

350

Time (min) Fig. 7. Experimental and calculated (kinetic model) conversions of lauric acid esterification at MR 6:1 and 10 wt% of catalyst and different temperatures 180 °C (d, exp; —, model), 160 °C (j, exp; - - - -, model) and 140 °C (N, exp; – - –, model).

100

(A)

90 80

Conversion (%)

606

70 60 50 40 30 20 0

50

100

150

200

250

300

350

400

250

300

350

400

Time (min) 100

(B)

90

Conversion (%)

80

100 90

Conversion (%)

80

70 60 50 40

70 30 60 20 50

0

50

100

150

200

Time (min)

40

Fig. 8. Experimental and calculated (kinetic model) conversions of lauric acid esterification at 180 °C, 10 wt% of catalyst and (A) MR of 3:1 (+, exp; – - –, model), 6:1 (s, exp; —, model) and 9:1 (h, exp; - - - -, model) and (B) 6:1 using hydrated ethanol 93.1 wt% (}, exp; —, model).

30 20 10 0

50

100

150

200

250

300

350

Time (min) Fig. 6. Kinetics of lauric acid esterification with ethanol at 180 °C and MR of 6:1: non-catalyzed (h, exp; – - –, model), obtained from Paiva et al. [12]; catalyzed by 10 wt% of raw STx1-b (N, exp), 10 wt% of acid activated STx1-b (d, exp; —, model) and 20 wt% of acid activated STx1-b (+, exp; - - - -, model).

As it can be observed in Fig. 7, an increase in the temperature leads to an increase in the reaction rates as well as the equilibrium conversions. This behavior is expected once these esterification reactions are endothermic and are strongly affected by temperature, as seen in factorial design results, where this factor was the

607

Paulo Ricardo Schizaki dos Santos et al. / Fuel 181 (2016) 600–609 Table 3 Kinetic parameters obtained from the non-catalyzed and acid activated STx1-b catalyzed esterification of lauric acid with ethanol.

f

e d

min1 Þ

Ea=R ðKÞ

k0 ðL mol

Non-catalyzed k0;1 k0;2

209.04 ± 0.14 144.41 ± 0.11

5451.18 ± 2.31 5447.49 ± 7.90

12.892  1010 ± 0.014  1010 Þ

15140.32 ± 6.58 0.0345 ± 0.0038

min1 Þ

3.1392 ± 0.0019

Catalyzed kcat K Et ðL mol

1

K W ðL2 mol rmsd(%)

1

min

1

c b

3630

3.11

most significant to lead the system to highest conversions, according to presented in Section 3.3. At the 140 °C, the achieved conversion in 240 min, approximately 65%, was achieved in only 15 min at 180 °C. In Fig. 8, the results regarding the kinetics obtained at different ethanol to lauric acid molar ratio are shown. The kinetic experimental observations are in agreement with the results presented in Fig. 5, which shows that the lauric acid conversion is almost constant for molar ratio higher that around 5:1. The initial rates of this reaction are also similar for this different ethanol to lauric acid esterification catalyzed by STx1-b activated montmorillonite STx1-b. In addition, the kinetic model fitted is in agreement with the experimental data, however indicating to some extent that higher ethanol amount in the system, even in low concentrations, leads to a slightly higher equilibrium conversion. Real values (corrected considering the water content in the ethanol) used in this work were: ethanol to lauric acid molar ratio 3:1 = 2.98:1, 6:1 = 5.96:1 and 9:1 = 8.94:1 and water to lauric acid were at 3:1 (0.02:1), 6:1 (0.04:1) and 9:1 (0.06:1), respectively. It is important to mention that the water to lauric acid molar ratio is also increased for higher ethanol to acid ratios reaching lower conversions. It is better viewed in Fig. 8(B), where hydrated ethanol was used (ethanol to lauric acid and water to lauric acid are 5.38 and 0.62, respectively) a much lower equilibrium conversion was found. Small deviations observed in all these kinetics (Figure A and B) might be due the experimental error attributed to this type of reaction measurements and there is no significant difference between 6:1 and 9:1 M ratio kinetics. The behavior of lauric acid esterification with ethanol reaching the chemical equilibrium conditions are in agreement with the simulated values using UNIFAC-LV model, as presented by de Paiva et al. [35]. In a general way, from the kinetic experiments and the modeling approach proposed it is seen that the fitted model is adequate to the most of the experimental sets, the low value of rmsd(%) and the agreement between the fitted model and experimental point suggest the selected mechanism explain the experimentally reaction kinetics in an adequate way, higher temperatures lead to higher conversions. In addition, the acid activated montmorillonite shows good catalytic properties, leading the system to higher conversions in shorter time. 3.5. Catalyst reuse in ethylic esterification of lauric acid Catalytic stability of montmorillonite STx1-b activated by phosphoric acid was evaluated in four use cycles, in which were performed consecutive reaction runs with the same catalyst sample. The reactions were carried out at conditions where the highest conversions of lauric acid were measured (180 °C, MR of 6:1 and CAT of 10%). The reaction time, as well as the stirring speed were kept in 2 h and 500 rpm, respectively.

**

a

845 916

628 521

4000 3500 3000 2000

1500

1000

500

Wavenumber (cm-1) Fig. 9. FTIR spectra of STx1-b before (a), after activation with phosphoric acid (b) and after (c–f) four consecutive use cycles.

Fig. 9 shows the FTIR spectra of the catalyst before (a), after (b) acid activation and after four consecutive reuse cycles (c–f). After performing the ethylic esterification of lauric acid esterification reactions, some important changes in the FTIR spectra are observed (Fig. 9c–f). The changes can be summarized as the reduction of the band at 3630 cm1, suggesting the reduction of the hydroxyl groups bonded to the octahedral structural metals. The same drastic reduction of the bands was observed at lower wavenumber, attributed to the OH deformation (Al–Al–OH – 916 cm1; Al–Mg– OH – 845 cm1; M–O (M = Mg, Al or Fe) – 628 cm1 and Al–O–Si – 521 cm1), suggesting the progressive dihydroxylation of the montmorillonite structure and leading to the formation of mixture of oxides, especially silica. The bands indicated with an asterisk (⁄) in the range of 2850–2960 cm1, arise from small organic contaminants. The increase of the band at 798 cm1 is attributed to the increase of trydimite amount based on the montmorillonite structure collapse, as also shown by the broad band around 1100 cm1, attributed to the Si–O stretching vibration [30,54]. In spite of the structural changes, the catalyst kept its activity after four reuse cycle, as Fig. 10 shows. A similar behavior was reported by Zatta et al. [30] in the esterification of lauric acid with methanol using acid activated STx-1 montmorillonite as heterogeneous catalyst. Therefore, despite using a fatty acid that is a model molecule representing biodiesel related fatty acids and ethanol, these results showed the application of this material presented desirable catalytic characteristics such as resistance to the deactivation, 90 85 80

Conversion (%)

1

Kinetic constants

Transmitance (a.u.)

Parameters

2

798

75 70 65 60 55 50 1

2

3

4

Use cycle Fig. 10. Reuse of the catalyst in the esterification of lauric acid with ethanol (T = 180 °C, MR = 6:1 and CAT = 10 wt%).

608

Paulo Ricardo Schizaki dos Santos et al. / Fuel 181 (2016) 600–609

mechanically and thermal stability under the reaction conditions. In a general way, there are many advantages in use of these clays: they are natural, inexpensive and widely available, the acid activation process is simple and the materials are environmental friendly. 4. Conclusions Sample of montmorillonite STx1-b activated by phosphoric acid (0.5 mol L1) was prepared, and their correct synthesis procedure was confirmed by X-ray diffraction and FTIR methods. The acid activated montmorillonite presented good catalytic properties to the esterification of lauric acid with ethanol, leading to equilibrium conversions of approximately 90%, at 180 °C and in shorter time (90 min). The presence of catalyst in the reactant mixture did not affect the equilibrium conversion. This reaction was favored by the highest temperature (180 °C), catalysts amount (10 wt%) and ethanol to lauric acid molar ratio (6:1) was found to be enough to reach the higher conversion observed. The heterogeneous reaction follows the Eley–Rideal model, in which the reaction occurs between adsorbed molecules of ethanol on the acid sites of the catalyst and the lauric acid molecules in the bulk phase. The mechanism adopted to model the kinetics of this reaction showed satisfactory agreement with experimental data and the root means square deviation obtained was 3.11% in terms of lauric acid conversion. The model used to represent the system as well as the kinetic parameters calculated can be useful to better understanding the reaction kinetics and it also can be suitable to further investigations of this promising route for biodiesel production from fatty acid esterification. The acid activated STx1-b showed to easily recovered from the reaction system and kept its properties after four consecutive reuse cycles. These characteristics can represent important points to application of these materials for free fatty acids esterification route for biodiesel production in industrial scale. Acknowledgments The authors are grateful to ANP/PRH-24, CAPES, CNPq (Proc. Num. 406737/2013-4 and 303573/2012-0) and Fundação Araucária for the financial support and scholarships. References [1] Singh SP, Singh D. Biodiesel production through the use of different sources and characterization of oils and their esters as the substitute of diesel: a review. Renew Sustain Energy Rev 2010;14:200–16. [2] Leung DYC, Wu X, Leung MKH. A review on biodiesel production using catalyzed transesterification. Appl Energy 2010;87:1083–95. [3] Canakci M, van Gerpen J. Biodiesel production from oils and fats with high free fatty acids. T ASAE 2001;44:1429–36. [4] Atabani AE, Silitonga AS, Badruddin IA, Mahlia TMI, Masjuki HH, Mekhilef S. A comprehensive review on biodiesel as an alternative energy resource and its characteristics. Renew Sustain Energy Rev 2012;16:2070–93. [5] Lotero E, Liu Y, Lopez DE, Suwannakarn K, Bruce DA, Goodwin JG. Synthesis of biodiesel via acid catalysis. Ind Eng Chem Res 2005;44:5353–63. [6] Saxena P, Jawale S, Joshipura MH. A review on prediction of properties of biodiesel and blends of biodiesel. Proc Eng 2013;51:395–402. [7] Ma F, Hanna Ma. Biodiesel production: a review 1. Sci Technol 1999;70:1–15. [8] Meher LC, Vidya Sagar D, Naik SN. Technical aspects of biodiesel production by transesterification – a review. Renew Sustain Energy Rev 2006;10:248–68. [9] Di Serio M, Tesser R, Pengmei L, Santacesaria E. Heterogeneous catalysts for biodiesel production. Energy Fuels 2008;22:207–17. [10] Leung DYC, Guo Y. Transesterification of neat and used frying oil: optimization for biodiesel production. Fuel Process Technol 2006;87:883–90. [11] Tesser R, Casale L, Verde D, Di Serio M, Santacesaria E. Kinetics and modeling of fatty acids esterification on acid exchange resins. Chem Eng J 2010;157:539–50. [12] Zatta L, Paiva EJM, Corazza ML, Wypych F, Ramos LP. The use of acid activated montmorillonite as a solid catalyst for the production of fatty acid methyl esters. Energy Fuels 2014;28:5834–40.

[13] Lilja J, Murzin DY, Salmi T, Aumo J, Mäki-Arvela P, Sundell M. Esterification of different acids over heterogeneous and homogeneous catalysts and correlation with the Taft equation. J Mol Catal A: Chem 2002;182–183:555–63. [14] Aranda DAG, De Goncalves JA, Peres JS, Ramos ALD, De Melo CAR, Antunes OAC, et al. The use of acids, niobium oxide, and zeolite catalysts for esterification reactions. J Phys Org Chem 2009;22:709–16. [15] Srilatha K, Lingaiah N, Devi BLAP, Prasad RBN, Venkateswar S, Prasad PSS. Esterification of free fatty acids for biodiesel production over heteropoly tungstate supported on niobia catalysts. Appl Catal A 2009;365:28–33. [16] Bosco SMD, Gonçalves M, Figueiredo FCA, Galhardo T, Carvalho WA. Sulfated pillared clay as catalyst in glycerol esterification with caprylic acid. Waste Biomass Valorization 2016;1:1–10. [17] Rezende MJC, Pinto AC. Esterification of fatty acids using acid-activated Brazilian smectite natural clay as a catalyst. Renew Energy 2016;92:171–7. [18] Paiva EJM, Sterchele S, Corazza ML, Murzin DY, Wypych F, Salmi T. Esterification of fatty acids with ethanol over layered zinc laurate and zinc stearate – kinetic modeling. Fuel 2015;153:445–54. [19] Altiokka MR, Çitak A. Kinetics study of esterification of acetic acid with isobutanol in the presence of amberlite catalyst. Appl Catal A 2003;239:141–8. [20] Kulkarni MG, Gopinath R, Meher LC, Dalai AK. Solid acid catalyzed biodiesel production by simultaneous esterification and transesterification. Green Chem 2006;8:1056. [21] Marchetti JM, Miguel VU, Errazu AF. Heterogeneous esterification of oil with high amount of free fatty acids. Fuel 2007;86:906–10. [22] Akyalçin S, Altiokka MR. Kinetics of esterification of acetic acid with 1-octanol in the presence of Amberlyst 36. Appl Catal A 2012;429–430:79–84. [23] Merchant SQ, Almohammad Ka, Al Bassam AaM, Ali SH. Biofuels and additives: comparative kinetic study of Amberlite IR 120-catalyzed esterification of ethanol with acetic, propanoic and pentanoic acids to produce eco-ethylesters. Fuel 2013;111:140–7. [24] Ali SH, Tarakmah A, Merchant SQ, Al-Sahhaf T. Synthesis of esters: development of the rate expression for the Dowex 50 W8–400 catalyzed esterification of propionic acid with 1-propanol. Chem Eng Sci 2007;62:3197–217. [25] Ilgen O. Investigation of reaction parameters, kinetics and mechanism of oleic acid esterification with methanol by using Amberlyst 46 as a catalyst. Fuel Process Technol 2014;124:134–9. [26] Calgaroto C, Calgaroto S, Mazutti MA, de Oliveira D, Pergher S, de Oliveira JV. Production of biodiesel from soybean and Jatropha curcas oils with KSF and amberlyst 15 catalysts in the presence of co-solvents. Sustain Chem Process 2013;1:17. [27] Purova R, Narasimharao K, Ahmed NSI, Al-Thabaiti S, Al-Shehri A, Mokhtar M, et al. Pillared HMCM-36 zeolite catalyst for biodiesel production by esterification of palmitic acid. J Mol Catal A: Chem 2015;406:159–67. [28] Baroi C, Dalai AK. Esterification of free fatty acids (FFA) of Green Seed Canola (GSC) oil using H-Y zeolite supported 12-Tungstophosphoric acid (TPA). Appl Catal A 2014;485:99–107. [29] Zatta L, Nepel A, Barison A, Wypych F. Montmorilonita modificada como catalisador heterogêneo em reações de esterificação (m)etílica de ácido láurico. Quim Nova 2012;35:1711–8. [30] Zatta L, Ramos LP, Wypych F. Acid-activated montmorillonites as heterogeneous catalysts for the esterification of lauric acid acid with methanol. Appl Clay Sci 2013;80–81:236–44. [31] Bhorodwaj SK, Dutta DK. Heteropoly acid supported modified Montmorillonite clay: an effective catalyst for the esterification of acetic acid with sec-butanol. Appl Catal A 2010;378:221–6. [32] Neji SB, Trabelsi M, Frikha MH. Esterification of fatty acid over Tunisian acid activated clay: kinetic study. J Oleo Sci 2011;60:293–9. [33] do Nascimento LAS, Tito LMZ, Angélica RS, da Costa CEF, Zamian JR, da Rocha Filho GN. Esterification of oleic acid over solid acid catalysts prepared from Amazon flint kaolin. Appl Catal B 2011;101:495–503. [34] Jeenpadiphat S, Tungasmita DN. Acid-activated pillar bentonite as a novel catalyst for the esterification of high FFA oil. Powder Technol 2013;237:634–40. [35] Paiva EJM, Graeser V, Wypych F, Corazza ML. Kinetics of non-catalytic and ZnL2-catalyzed esterification of lauric acid with ethanol. Fuel 2014;117:125–32. [36] Reinoso DM, Damiani DE, Tonetto GM. Zinc carboxylic salts used as catalyst in the biodiesel synthesis by esterification and transesterification: study of the stability in the reaction medium. Appl Catal A 2012;449:88–95. [37] Christidis GE, Scott PW, Dunham AC. Acid activation and bleaching capacity of bentonites from the islands of Milos and Chios, Aegean, Greece. Appl Clay Sci 1997;12:329–47. [38] Panda AK, Mishra BG, Mishra DK, Singh RK. Effect of sulphuric acid treatment on the physico-chemical characteristics of kaolin clay. Colloids Surf, A 2010;363:98–104. [39] Wang TH, Liu TY, Wu DC, Li MH, Chen JR, Teng SP. Performance of phosphoric acid activated montmorillonite as buffer materials for radioactive waste repository. J Hazard Mater 2010;173:335–42. [40] Ghiaci M, Aghabarari B, Gil A. Production of biodiesel by esterification of natural fatty acids over modified organoclay catalysts. Fuel 2011;90:3382–9. [41] Do Nascimento LAS, Angélica RS, da Costa CEF, Zamian JR, da Rocha Filho GN. Comparative study between catalysts for esterification prepared from kaolins. Appl Clay Sci 2011;51:267–73. [42] Bouguerra Neji S, Trabelsi M, Frikha MH. Esterification of fatty acids with short-chain alcohols over commercial acid clays in a semi-continuous reactor. Energies 2009;2:1107–17.

Paulo Ricardo Schizaki dos Santos et al. / Fuel 181 (2016) 600–609 [43] Kantam ML, Bhaskar V, Choudary BM. Direct condensation of carboxylic acids with alcohols: the atom economic protocol catalysed by Fe3+-montmorillonite. Catal Lett 2002;78:185–8. [44] Zatta L, da Costa Gardolinski JEF, Wypych F. Raw halloysite as reusable heterogeneous catalyst for esterification of lauric acid. Appl Clay Sci 2011;51:165–9. [45] Paiva EJM, da Silva MLCP, Barboza JCS, de Oliveira PC, de Castro HF, Giordani DS. Non-edible babassu oil as a new source for energy production-a feasibility transesterification survey assisted by ultrasound. Ultrason Sonochem 2013;20:833–8. [46] Navarro-Díaz HJ, Gonzalez SL, Irigaray B, Vieitez I, Jachmanián I, Hense H, et al. Macauba oil as an alternative feedstock for biodiesel: characterization and ester conversion by the supercritical method. J Supercrit Fluids 2014;93:130–7. [47] Hoekman SK, Broch A, Robbins C, Ceniceros E, Natarajan M. Review of biodiesel composition, properties, and specifications. Renew Sustain Energy Rev 2012;16:143–69. [48] Dagostin JLA, Carpiné D, Corazza ML. Extraction of soybean oil using ethanol and mixtures with alkyl esters (biodiesel) as co-solvent: kinetics and thermodynamics. Ind Crops Prod 2015;74:69–75.

609

[49] Rowley RL, Wilding WV, Oscarson JL, Yang Y, Zundel NA, Daubert TE, et al. DIPPR data compilation of pure chemical properties. Design Institute for Physical Properties; 2006. [50] Corazza ML, Fouad WA, Chapman WG. Application of molecular modeling to the vapor–liquid equilibrium of alkyl esters (biodiesel) and alcohols systems. Fuel 2015;161:34–42. [51] Gomes EM. Desenvolvimento de isoterma de sorção de metal pesado baseada no cálculo MSc Thesis. Brazil: Universidade Federal do Paraná; 2012. [52] Bard Y. Nonlinear parameter estimation. New York: Academic Press; 1974. [53] Zatta L, Paiva EJM, Corazza ML, Wypych F, Ramos LP. The use of acid-activated montmorillonite as a solid catalyst for the production of fatty acid methyl esters. Energy Fuels 2014;28:5834–40. [54] Zatta L, Ramos LP, Wypych F. Acid activated montmorillonite as catalysts in methyl esterification reactions of lauric acid. J Oleo Sci 2012;61:497–504. [55] Viani A, Gualtieri AF, Artioli G. The nature of disorder in montmorillonite by simulation of X-ray powder patterns. Am Mineral 2002;87:966–75. [56] Brunschwig C, Moussavou W, Blin J. Use of bioethanol for biodiesel production. Prog Energy Combust Sci 2012;38:283–301. [57] Kaur N, Ali A. Kinetics and reusability of Zr/CaO as heterogeneous catalyst for the ethanolysis and methanolysis of Jatropha crucas oil. Fuel Process Technol 2014;119:173–84.