Catalytic synthesis of fatty acid methyl esters from Madhuca indica oil in supercritical methanol

Energy Conversion and Management 173 (2018) 412–425

Contents lists available at ScienceDirect

Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman

Catalytic synthesis of fatty acid methyl esters from Madhuca indica oil in supercritical methanol Neha Lamba, Sangeeta Adhikari, Jayant M. Modak, Giridhar Madras

T

⁎

Department of Chemical Engineering, Indian Institute of Science, Bangalore 560012, India

A R T I C LE I N FO

A B S T R A C T

Keywords: Supercritical methanol Mahua oil Biodiesel Metal oxides Combustion synthesis Kinetics

Fatty acid methyl esters (FAMEs) that are used as biodiesel can be synthesized in supercritical methylating agents such as supercritical methanol. While the synthesis can be conducted both non-catalytically and catalytically, the synthesis in the presence of oxides is significantly faster. In this study, FAMEs were synthesized by transesterification of a non-edible oil (Mahua, Madhuca indica) in supercritical methanol with a wide variety of oxides. The reaction was extensively studied with eleven different oxides, synthesized using the solution combustion method, as catalysts. In addition, the best two catalysts, namely MgO and Mn3O4, were synthesized using four different fuels in the combustion synthesis. The catalytic effect of all these oxides was investigated and conversions ranging from 5% to 100% were obtained over the investigated range of temperature from 503 K to 583 K, and with reaction time varying between 2 min and 80 min. Among all the catalysts, MgO synthesized with ascorbic acid as the fuel for the solution combustion gave the best results. Therefore, this catalyst was chosen and the influence of operating temperature for the transesterification reaction (503–583 K) on the rate of the reaction was studied. A pseudo first order kinetic model was obtained based on the proposed Eley-Rideal reaction mechanism and, the rate constants were obtained. The rate constants varied between 1.61 × 10−3 s−1 to 4.93 × 10−3 s−1 with an activation energy of 36 kJ/mol and a pre-exponential factor of 9.33 s−1. The rate constant obtained for the non-catalytic supercritical transesterification with oxide as catalyst was significantly higher than the rate constant of 9.9 × 10−5 s−1 obtained for the non-catalytic reaction at 523 K. The activation energy for the catalyzed reaction (36 kJ/mol) was notably lower than the activation energy (75 kJ/mol) for the uncatalyzed reaction indicating the efficacy of the catalyst.

1. Introduction The energy demands are increasing throughout the world with the industrial and transportation sectors having the highest share in energy consumption [1]. The energy requirements (∼80%) of the transportation sector are mainly fulfilled by the fossil fuels. However, it has been predicted that the global fossil oil reserves would deplete within the next 45 years [2]. Further, the greenhouse gas (CO2) emissions has increased by 92% in the last thirty years [1]. Thus, there is a need to find alternate ecofriendly energy sources that are socially acceptable and also economically viable. Biodiesel, which is a mixture of fatty acid alkyl esters, results in an average reduction of 40% in the emissions when used as the fuel in the diesel engine [3]. Thus, biodiesel is a nontoxic, renewable, biodegradable and low sulfur alternate fuel and has been considered as a cleaner substitute for the diesel [4,5]. The raw materials that are mostly employed for the biodiesel synthesis are edible and non-edible vegetable oils. However, non-edible

⁎

oils are preferred as these crops can be grown on barn lands, avoiding competition for the land between edible and non-edible crops [6,7]. Transesterification is the preferred method of biodiesel synthesis in comparison to other methods such as dilution, microemulsion and pyrolysis [8]. In transesterification, the triglycerides are converted to fatty acid methyl esters (FAMEs) using a methylating agent through an exchange of the alkoxy moiety either in the presence or absence of a catalyst [9]. Homogeneous catalysts involve either acid or base catalysts but they are sensitive to the impurities present in the feedstock such as water and free fatty acids [2,10]. Further, when the homogeneous catalysts are employed in the reaction the downstream processing for product separation becomes complex, tedious, time consuming and further leads to a large amount of waste water. Moreover, the homogeneous catalysts cannot be reused. Biological catalysts such as immobilized enzymes can also be used but these are expensive and reusability is another issue. However, these limitations associated with the homogeneous and enzymatic catalyst can be overcome by a

Corresponding author. E-mail address: [email protected] (G. Madras).

https://doi.org/10.1016/j.enconman.2018.07.067 Received 29 April 2018; Received in revised form 19 July 2018; Accepted 20 July 2018 Available online 04 August 2018 0196-8904/ © 2018 Elsevier Ltd. All rights reserved.

413

Palm Palm Castor Rapeseed Triolein Rapeseed Palm Soybean

Crude rapeseed Waste frying oil

Sunflower

Jatropha

Palm products (crude palm oil (CPO), refined (RPO), palm fatty acid distillate (PFAD)) Crude rapeseed

Soybean

Soybean Soybean Used vegetable oil

Purified palm oil

Soybean Soybean

3. 4. 5. 6. 7. 8. 9. 10.

11. 12.

13.

14.

15.

16.

17.

18. 19. 20.

21.

22. 23.

Methanol Subcritical Methanol

Near critical Methanol

Subcritical Methanol Subcritical Methanol Methanol

Subcritical Methanol

C based catalyst (synthesized by incomplete carbonization of naphthalene in H2SO4) CH3ONa Na2SiO3

K3PO4 KOH Cs doped heteropolyacid

Liquid organic amine (ethylene diamine, diethyl amine, triethylamine) + co-catalyst (propylene oxide) MnCO3/ZnO

Near critical Methanol

Methanol

SO4-ZrO2, WO3-ZrO2, TiO2-ZrO2

Near critical Methanol

CaO ZnO CaO CaO/Al2O3 CaO/KI/ɤ-Al2O3 NaOH SrO, CaO, ZnO, TiO2, ZrO2 Acid exchange resin, Nafion + SCCO2 Zinc nitrate (metal precursor for ZnO) SO4-ZrO2 Nano-MgO

Catalyst

NaOH Zeolite Y solid acid catalyst (molar ratio of SiO2/Al2O3 = 4.88) CeO2, WO3, ZnO, ZrO2, mixed metal oxides (50–50 wt %), ZrO2-SO4, WO3-ZrO2, CeO2-ZrO2, ZnO-La2O3, Al2O3 Base: Ca & La mixed oxide (CaLaO)

Ethanol Methanol Ethanol Methanol Subcritical Methanol Methanol Methanol Sub or Supercritical Methanol Methanol Sub or Supercritical Methanol CO2 + Methanol

Methanol Ethanol

SCF

Calculated at the optimum operating conditions using the pseudo first order kinetic model (Eq. (16a)).

Sunflower Rapeseed

1. 2.

a

Oil

S.N

250 °C, 23:1, 1 wt%, 20 min 220 °C, 3 MPa, 36:1, 0.5 wt%, 30 min

(Mn/Zn = 1:1 M ratio) 18:1, 448 K, 1 h, 4 wt% of catalyst 220 °C, 8 MPa, 24:1, 30 min, 1 wt% 160 °C, 24:1 0.1 wt%, 10 min 260 °C, 20 MPa, 40:1, 40 min, 3 wt% of Cs2.5PW12O40 270 °C, 12:1, 0.5 wt%, 30 min

240 °C, 8.2 MPa, 21:1, 10 min, 1 wt% (CaLa4 most active) 250 °C, 24:1 (for CPO and RPO), 18:1 (for PFAD) 10 min, 20% WO3-ZrO2 calcined @ 800 °C 250 °C, 8.5 MPa, 24:1, 10 min

28.8 18.3

10.6

95% 97.4% 95.6%

25.6 20.9 7.26

9.72

[68] [52]

[98]

[95] [96] [97]

[94]

[93]

[92]

–

14.2

[91]

[90]

[88] [89]

[81] [82] [83] [48] [84] [85] [86] [87]

[80] [66]

Reference

27.2

163

116 76.8

35.2 – – – 6.17 164 17.3 276 – – 65.2

Rate constantsa (s−1) × 104

95.6% 98% 92%

99.25%

89.5%

91% (CPO), 94% (RPO), 81% (PFAD)

93%

98%

200 °C, 20 MPa, 4 min, 25:1, 5 g of ZrO2-SO4

100% 93% 88% 96% 95% 98.9% 95% 88% 96.9% 90% 100% 100% 100%

24 MPa, 6 min, 3 wt%, 41:1 15 MPa, 60 min, 42:1, 3 wt% 15 MPa, 60 min, 42:1, 3 wt% 20 MPa, 30:1, 4.85 min 12 MPa, 24:1, 3 wt%, 60 min 9.52 MPa, 40:1, 10 min, 0.1 wt% 10.5 MPa, 40:1, 1 wt% of ZnO, 10 min 25 MPa, 25:1, 9 g of catalyst, 2 min 35 MPa, 3 wt% 10 min 0.5 wt%, 24:1, 20 min 28 MPa, 10 min, 3 wt%, 36:1

Yield

250 °C, 10.2 MPa, 24:1, 6 min, 0.8 wt% 240 °C, 12 MPa, 15 min, 5:1, 0.2 g

252 °C, 270 °C, 270 °C, 285 °C, 290 °C, 300 °C, 250 °C, 205 °C, 250 °C, 250 °C, 260 °C,

Optimum operating conditions

Table 1 Literature available on the transesterification reaction performed under sub or supercritical conditions in the presence of catalysts.

N. Lamba et al.

Energy Conversion and Management 173 (2018) 412–425

Energy Conversion and Management 173 (2018) 412–425

N. Lamba et al.

oxidizer (mostly water soluble metal nitrates) and the fuel (readily available water soluble compounds such as urea, glycine, metal acetates or hydrazides) for the synthesis of the desired oxide [26]. The exothermicity of the reaction makes the combustion synthesis reaction selfsustaining by reducing the external energy requirements. Further, the high temperature and the large evolution of gases associated with the process normally result to nanosized particles with high surface area and stability [27–30]. Recently, an engineering and economic evaluation for the feasibility of combustion synthesis route was performed by synthesizing a metal composite using the combustion synthesis [31]. The operating temperatures employed in the study were 90–300 °C and ca. 250 g of catalyst was synthesized. The process was found to be profitable with a gross profit margin of 2424.771 USD and can survive for 20 years of production. Thus, solution combustion synthesis is one of the most efficient route for materials synthesis in terms of time, energy, rate, simplicity and labour in comparison to other synthesis routes [32,33]. Further, the use of different precursors (metal salt + fuel) in solution combustion synthesis can lead to oxides with diverse properties, resulting in a wide range of catalytic activities [34,35]. These oxides have been used extensively for gas phase reactions, as summarized in reviews [36,37]. It has also been observed from various existing studies employing combustion synthesized oxides for different reactions that the combustion synthesized oxides provide better catalytic activity in comparison to the commercial oxides [26,34,38,39]. For example, a higher photocatalytic activity towards methylene blue and phenol degradation was obtained using combustion synthesized TiO2 in comparison to commercial TiO2, Degussa P-25 [40,41]. A porous spongy combustion synthesized ZnO lead to superior photoactivity towards bacterial enumeration in comparison to Degussa P-25 [28] and an improved dye uptake by combustion synthesized WO3 in comparison to commercial samples was obtained [42]. The combustion synthesized ZrO2 and Co3O4 and their composite showed higher adsorption capabilities in comparison to commercial activated carbon [43]. A better electrochemical activity and high reversible capacity was obtained using combustion synthesized Co3O4 and nanocomposite of Co3O4/CoO/graphene as anode material in lithium ion battery in comparison to the commercially used graphite [44] and, a higher hydrogen generation rates (by hydrolyzing the NaBH4) were obtained using the combustion synthesized Co3O4 in comparison to the commercial Co3O4 and Co nanopowders [45]. However, there were no studies available using solution combustion synthesized oxides for the supercritical transesterification reaction. Thus, an attempt was made to employ this route of synthesis based on its efficiency in terms of lesser time, energy, labour, simple equipment requirement and higher activity in the present study to investigate transesterification reaction using supercritical methanol and solution combustion synthesized oxides. Thus, the main objectives of the present study consist of (a) synthesizing different metal oxides using solution combustion synthesis using ascorbic acid as a fuel and studying their respective performance towards the transesterification of mahua oil with supercritical methanol, (b) proposing a mechanism for the reaction and obtaining the rate constants, activation energies and pre-exponential factors for the reaction with best oxides, (c) comparing the rate constants obtained with different catalysts with the non-catalyzed supercritical transesterification reaction at same operating conditions and (d) studying the effect of different fuels utilized in the synthesis of best oxide on the activity or performance of the catalyst for the integrated transesterification reaction.

heterogeneous catalyst which is easy to separate from the product and also can be recycled and reused. Thus, heterogeneous catalysis is preferred and acid (zeolites, heteropolyacids, sulphated zirconia) and alkali (transition metal oxides, alkali doped metal oxides, mixed metal oxides) catalysts have been reported [11]. In general, higher activity is obtained with the basic catalysts [12]. However, a limitation of slower rates and sensitivity towards the feedstock is associated with this catalytic pathway. Therefore, a non-catalytic pathway employing the usage of supercritical fluids such as supercritical methanol for the transesterification reaction has been studied. Supercritical fluids have properties that are intermediate of a liquid and gas. The liquid-like densities and gas-like viscosities of supercritical fluids reduce the mass transfer resistances associated with the reaction and increase the solubility of methanol in the oil phase leading to the enhanced reaction rate. This pathway reduces the number of process steps and also leads to a higher yield of esters within a few minutes [13,14]. A number of studies are reported for the FAMEs synthesis in supercritical methanol [15–19]. Although this route of synthesis has several advantages over the catalytic pathway, it is energy intensive because of high temperature and pressure conditions and this makes the process economically unviable. It has been observed from the literature that the heterogeneous catalytic path for FAMEs or biodiesel synthesis is the most economical pathway among the homogenous alkaline catalytic with acid pretreatment, homogenous acid catalytic, heterogeneous catalytic and supercritical pathways [20,21]. On comparing these four pathways the total operating and capital investment was highest with the supercritical pathway with lowest internal rate of return (1.80%). However, it has been shown in a study that has employed propane as a cosolvent in the supercritical FAMEs synthesis that the supercritical pathway can be made economically viable [22] upon usage of cosolvent that can reduce the operating conditions for the reaction. Similarly, there is another way to reduce the energy consumption associated with the supercritical process without compromising over the triglycerides conversion and the rates of reaction obtained using this pathway i.e., by adding a heterogeneous catalyst to the reaction mixture under supercritical conditions of methylating agent. It should be noted that this coupling of heterogeneous catalyst with non-catalytic supercritical pathway would overcome the limitations associated with both of these routes for FAMEs synthesis, individually. It can be observed from the literature that the energy required can be reduced to ca. one third (1205 kW) of the energy consumption in supercritical synthesis alone (3247 kW) for a biodiesel plant with a capacity of FAMEs production of 10,000 tons/ year [23]. Similar observations of less energy associated with the integrated process have also been discussed in a review [24]. Thus, if the reaction is performed using the supercritical alkylating agent in the presence of a heterogeneous catalyst, complete conversions can be obtained at lower temperatures compared to the non-catalytic supercritical route. Only a few such studies, however, have been performed. All of these studies have been compiled in Table 1 to understand the types of catalysts and the operating conditions employed in such an integrated process. Most of the catalysts being used in these studies have been synthesized using the co-precipitation method and the operating conditions in all the studies varied from 473 K to 573 K and 8.2 MPa to 24 MPa, respectively. Similar to the effect of different types of catalysts (acid or base), the methods of synthesis of a catalyst (co-precipitation, sol-gel, hydrothermal, self-propagating high temperature synthesis, ion-exchange, electrochemical synthesis, etc.) also affects its activity towards a particular reaction [25]. However, it is often difficult to compare the studies on different catalysts across various conditions. In this work, the reaction was conducted with eleven different oxides prepared using the solution combustion synthesis and the reactions were conducted under similar operating conditions. Such a comprehensive study of using different oxides under similar conditions is lacking in the literature. Combustion synthesis is an exothermic reaction between the

2. Experimental 2.1. Materials 2.1.1. Raw material Mahua (Madhuca indica) oil was used in the present study for synthesizing FAMEs. Mahua is a large semi-evergreen or evergreen forest 414

Energy Conversion and Management 173 (2018) 412–425

N. Lamba et al.

24 Mg (NO3)2 + 5 C12 H22 O11 → 24 MgO + 24 N2 + 55 H2 O + 60 CO2

tree that belongs to a family of Sapotaceae tree and found mainly in the subtropical region of India. The tree grows upto a height of 20 m. An average yield of 5–200 kg of mahua seeds per tree is obtained depending on the age and size of the tree. The seed contain ca. 50–61% of oil leading to a yield of ca. 18 lakhs metric tons of mahua oil per year [46]. However, the presence of aflatoxins (poisonous carcinogenic compounds) in the mahua seed samples makes it unsuitable to human or animal consumption. Aflatoxin B1 is mainly found with a concentration of 282.36 ppb in seed, 201.57 ppb in oil and 74.35 ppb in cake from mahua tree. However, the total aflatoxin present in mahua seed, oil and cake is 315.51 ppb, 220.66 ppb and 87.55 ppb, respectively. Other chemicals such as saponins and tannins are also present in the mahua seeds along with the aflatoxins, adding further to the nonedible nature of the seeds and oil [47]. Thus, the non-edible mahua oil was chosen for the FAME synthesis in the present study. The mahua oil (extracted by the cold pressing of the seeds) used in the study was obtained from Gandhi Krishi Vigyana Kendra (GKVK), Bangalore (India). The oil was heated (for melting the fat present and homogenizing it to a single phase) and vacuum filtered using the Whatman filter paper (150–200 mesh) (for removing any solid impurities present in the oil) prior to use in the reaction.

(5) Similarly, the equations can be written for the rest of the oxides for obtaining the stoichiometric amounts of metal precursor and fuel for the synthesis. 2.3. Oxides characterization The synthesized oxides were analysed using the Rigaku X-ray diffractometer equipped with the Cu-Kα radiation (1.542 Å) source. The samples were analysed by moving the detector between 10 and 80° at a scanning rate of 1°/min. The crystallite size was calculated using the Scherrer equation comprising of full width half maxima (FWHM) value and the Bragg’s angle.

D=

2.2. Oxides synthesis The metal oxides (MgO, Mn3O4, Fe2O3, MoO3, CeO2, CuO, ZrO2, Bi2O3, TiO2, Co3O4 and WO3) used in this study were synthesized using the solution combustion synthesis. The metal precursors (Mg (NO3)2·6H2O, Mn(NO3)2·4H2O, Fe(NO3)3·9H2O, (NH4)6Mo7O24·4H2O, Ce(NH4)2(NO3)6, Cu(NO3)2·3H2O, Zr(NO3)4·5H2O, Bi(NO3)3·5H2O, TiO (NO3)2, Co(NO3)2·6H2O and (NH4)6W12O39·xH2O) and fuels (ascorbic acid (C6H8O6), urea (CH4N2O), glycine (C2H5NO2), oxalyl dihydrazide (C2H6N4O2) and lactose (C12H22O11)) were procured from Sigma Aldrich, Bangalore. All the fuels employed were water soluble and have ignition temperatures less than 773 K. Stoichiometric amounts of metal precursor (used as the oxidizer) and the fuel (acts as a reducing agent) were mixed to form a solution in minimum amount of distilled water in a small Petri-dish made up of quartz glass. The calculations were performed by calculating the ratio of oxidizing to reducing valencies that can be obtained using the propellant chemistry [33]. The dish having the solution was then introduced into the preheated muffle furnace at 623 ± 10 K. The solution first boils then dehydrates and finally undergoes decomposition leading to a large evolution of gases and formation of a froth. This froth then swells and converts to a foam which ruptures because of the incandescent flame giving a voluminous, porous solid. The formed oxide was then ground and calcined for removing the deposited carbon on the material’s sites. The final product after the calcination was characterized and used for the reactions studied. A theoretical equation for the redox reaction occurring between the metal precursor (Mg(NO3)2·6H2O) and the fuel (ascorbic acid, urea, glycine, oxalyl dihydrazide, lactose) for the formation of MgO (assuming the complete combustion) can be written as follows,

2.4. Transesterification reaction and product analysis The reaction was performed in an 11 ml batch reactor made up of a SS316 tubing with one end fixed with a removable plug (to load and unload the reactor) and the other being welded. A detailed description of the reactor used has been provided earlier [50]. Lorentz-Berthelot (LB) mixing rule along with the equation of state, developed by PengRobinson (PR-EOS) was used to calculate the volume of reactants to be loaded inside the reactor as has been discussed earlier for an isochoric system [51]. An optimized amount of 1 wt% (w.r.t oil) of the solid oxide was added to the reactor. The reactor with the reactants (oil and methanol) and the oxide was put inside the preheated furnace at the required operating temperature (503–583 K). The temperature inside the furnace was controlled by the PID controller with the help of k-type thermocouple at ± 5 K. The reaction was stopped after a certain interval of time by quenching the reactor in a water bath maintained at 263 K. The catalyst from the liquid reaction mixture was separated first using a centrifuge (Remi C-854/6) by applying a relative centrifugal force (G) of 1175g for 15 min. The liquid mixture was then pipetted out

(1)

3 Mg (NO3)2 + 5 CH4 N2 O → 3 MgO + 8 N2 + 10 H2 O + 5 CO2 (2)

9 Mg (NO3)2 + 10 C2 H5 NO2 → 9 MgO + 14 N2 + 25 H2 O + 20 CO2

(3)

Mg (NO3)2 + C2 H6 N4 O2 → MgO + 3 N2 + 3 H2 O + 2 CO2

(4)

(6)

In Eq. (6), K is Scherrer’s constant having a value of 0.9, λ (1.5418 Å) is wavelength of the radiation, β (radian) is full width at half maximum (FWHM) and θ (degree) is half of the Bragg’s angle of diffraction. The surface areas of all the oxides were obtained using the Brunauer-Emmett-Teller (BET) method. The oxides were regenerated at 120 °C for 2 h prior to the measurements for removing any adsorbed molecules from the pores and surface. Liquid nitrogen at −196 °C and water at 25 °C were used for the adsorption and desorption of N2. The area under the adsorption and desorption curves was analysed by the Belsorb surface area analyser (Smart Instruments) for obtaining the surface area of a particular oxide. The basic strength (H_) of the synthesized oxides was obtained using 5 different Hammett indicators with different basic strengths (neutral red (H_ = 6.8), bromothymol (H_ = 7.2), phenolphthalein (H_ = 9.3), 2,4-dinitroaniline (H_ = 15) and 4-nitroaniline (H_ = 18.4)) [48]. A solution of 0.1 wt% in methanol of a particular Hammett indicator was prepared. 25 mg of a particular oxide was then shaken with 1 ml of this solution. The solution with oxide was then allowed for equilibration for 2 h. The color change of the solution indicates the basic strength of a particular oxide (the oxide is stronger than the Hammett indicator if color changes and weaker if no change in color is observed). The basicity for the most active oxides was determined using the Hammett indicator-benzoic acid titrations [49]. 10 mg of the oxide was shaken in an Erlenmeyer flask having 2 ml of the 0.1 wt% solution of Hammett indicator in methanol. A solution of 0.1 mol/L of benzoic acid in methanol was then added using a burette. The volume of the benzoic acid used was noted when the basic color disappeared.

2.1.2. Chemicals The reagent, methanol with purity > 99.99% and the solvent, nheptane (> 99.99% purity, determined by GC) were purchased from the Merck Chemicals, Bangalore (India). The internal standard used for the FAMEs analysis, butyl laurate with a purity of > 99% was procured from the Sigma Aldrich, Bangalore (India).

2 Mg (NO3)2 + C6 H8 O6 → 2 MgO + 2 N2 + 4 H2 O + 6 CO2

Kλ β cos θ

415

Energy Conversion and Management 173 (2018) 412–425

N. Lamba et al.

3. Results and discussion

Table 2 Physical and thermodynamic properties of Madhuca indica (Mahua) oil. Properties

Mahua oil [51]

Acid value (mg KOH/g) Saponification value (mg KOH/g) Average molecular weight (g/mol) Critical temperature, Tc (K) Critical pressure, Pc (MPa) Critical volume, Vc (l/mol) Critical compressibility factor, Zc

14.90 192.6 821.3 959.4 0.38 3.05 0.15

Fatty acid composition of oil (wt %) Palmitic acid Stearic acid Oleic acid Linoleic acid Arachidic acid Total saturated fatty acids Total unsaturated fattyacids

24.5 22.7 37.0 14.3 1.50 48.7 51.3

3.1. Characterization 3.1.1. XRD analysis Eleven different oxides were synthesized using the solution combustion synthesis with ascorbic acid (AA) as fuel and can be named as WO3-AA, Co3O4-AA, TiO2-AA, Bi2O3-AA, ZrO2-AA, CuO-AA, CeO2-AA, MoO3-AA, Fe2O3-AA, Mn3O4-AAand MgO-AA. The two most active oxides among these oxides (MgO-AA and Mn3O4-AA) were synthesized with various fuels (urea (U), glycine (G), oxalyl dihydrazide (ODH), lactose (L)) and these can be named as MgO-U, MgO-G, MgO-ODH, MgO-L and Mn3O4-U, Mn3O4-G, Mn3O4-ODH, Mn3O4-L, respectively. The synthesized oxides were analysed using X-ray diffraction for their purity and crystal structure. The diffraction pattern for all the oxides are shown in Figs. 1 and 2. The peaks obtained for a particular oxide were matched against the reference pattern (JCPDS (Joint Committee on Powder Diffraction Standards) file). The JCPDS files corresponding to the different oxides have also been mentioned in the Figs. 1 and 2. Different crystal structures for different oxides were observed with varying lattice parameters. Oxides such as MgO (all fuels), CeO2-AA, Co3O4-AA, ZrO2-AA were found to have cubic crystal structure. Mn3O4 (all fuels) and TiO2-AA have tetragonal structure, CuO-AA and Bi2O3AA have monoclinic crystal system, WO3-AA and MoO3-AA have orthorhombic and Fe2O3-AA possess rhombohedral structure, respectively. These structures match with the JCPDS reported in the literature and the crystal structure reported in previous studies for CeO2 [55], Co3O4 [43], ZrO2 [56], CuO [57], Bi2O3 [58], MgO [59], WO3 [60], MoO3 [61], Fe2O3 [62], Mn3O4 [63]. All the synthesized oxides were obtained in pure form except for Mn3O4-AA. Peaks corresponding to Mn2O3 were observed in case of Mn3O4-AA and Mn3O4-U. Peaks corresponding to MnO were observed in case of Mn3O4-ODH. However, no impurity peaks were obtained in case of Mn3O4-G and Mn3O4-L. These can be clearly obtained from Fig. 2. The crystallite size for each oxide was calculated according to the highest intensity peak using the Debye Scherrer equation. The crystallite sizes for all the oxides have been tabulated in Table 3. It was found to be highest for Co3O4-AA (71 nm) and was least for TiO2-AA (9 nm). As expected, compounds with smaller crystallite size will have higher specific surface areas and this trend was followed for these oxides (Table 3) with Co3O4-AA having the least surface area (3 m2/g) and TiO2-AA possessing the highest surface area (158 m2/g). However, this opposing trend of crystallite size and specific surface area was not observed with MgO and Mn3O4 oxides synthesized using different fuels. MgO-AA and Mn3O4 –U were found to have highest crystallite size whereas, MgO-L and Mn3O4 –G have the largest specific surface area. This may be due to the difference in the complex formation between the metal nitrate precursor and different fuels. However, this trend was consistent for compounds synthesized with a single fuel. The catalytic activity of a material is dependent on various parameters such as specific surface area, pore volume, pore size and the actual active site concentration [64]. It has been reported earlier that the catalytic activity of a material for the conventional transesterification mainly depends upon the specific surface area and the basic strength of the material [64]. Among these factors, the basicity of the catalyst is bound to play a major role and, if basicities of the catalysts are similar, then the surface area of the catalyst will play a role. Therefore, it is not necessary that the catalysts with the highest surface area will yield the highest conversion. This is apparent in this study wherein TiO2-AA was found to have the highest specific surface area but the highest catalytic activity was obtained using the MgO-AA. It should be noted that the catalytic activity discussed in the study has been referred to the effect on the performance of the reaction due to the addition of a catalyst. It has been discussed or analysed on the basis of rate constants (discussed later in the manuscript) obtained for the

and kept inside an oven at 363 K for 15 min for evaporating the unreacted methanol. After the methanol evaporation, the solution was again centrifuged to separate the glycerol. The top layer consisting of FAMEs was separated out and was used for the further analysis. The fatty acid composition of the oil was determined by GC-MS. The oil sample was derivatized prior to the analysis by a two-step process to produce FAMEs, as has been discussed in an earlier work [51]. The physical properties of the oil sample such as acid value and the saponification value were obtained by the AOCS methods [52]. Moreover, the thermodynamic properties such as critical temperature, pressure, volume and compressibility factor for oil were also obtained along with these physical properties using the Constantinou and Gani group contributions and the LB mixing rule [53,54]. All of these properties of oil have been presented in Table 2. FAMEs obtained after the reaction were analysed using the Gas chromatography (GC) equipped with a VF-5 capillary column having dimensions of 30 m × 0.25 mm × 0.25 µm. The FAMEs sample was diluted with the solvent, n-heptane and an appropriate amount of internal standard (butyl laurate) was added to it. 1 µl of the prepared sample was injected into the column through the injector port maintained at a temperature of 573 K with a split ratio of 100. The detector and the column oven temperatures were maintained at 573 K and 523 K (isothermal, 10 min), respectively. Helium gas (> 99.999% purity) was used as the carrier gas. The fuel gases were hydrogen and oxygen and nitrogen was the makeup gas for detector. The area under the peaks was analysed to obtain the triglycerides conversion and was calculated using the below expression,

% Conversion =

WFAMEs, actual × 100 WFAMEs, theory

where WFAMEs,actual and WFAMEs,theory denotes the weight of FAMEs obtained after the reaction (from GC analysis) and the theoretically obtained weight of FAMEs corresponding to the 100% conversion of triglycerides (from the stoichiometry of the reaction), respectively. However, it is important to note that if there is only one reaction is occurring (such as only transesterification) and, the glycerol and methanol after the reaction have been separated from the product mixture, the yield of FAMEs and conversion can be assumed same. Generally, this sample after the separation has been analyzed for obtaining the amount of FAMEs formed in most of the existing studies (Table 1). Thus, both can be used interchangeably. The experiments were performed in triplicate and the variation in conversions was found to be less than ± 2%.

416

Energy Conversion and Management 173 (2018) 412–425

N. Lamba et al.

40

202 -113 -311 113 311 004 80

Fe2O3-AA

130 -211

110

220 231

121 -110

JCPDS: 80-2377

Mn3O4-AA

211

321 224

105

220

JCPDS: 24-0734 * Mn2O3 * MgO-AA

200

004

*

60

2 (degree)

103

j

331

220

20

80

112

Intensity (a.u.)

110

i

k

220

JCPDS: 87-0652 311 222

111

20

MoO3-AA

002 211 112 171

221 -104 -322 -241 60

2 (degree)

CeO2-AA JCPDS: 75-0390

JCPDS: 05-0508

h

40

311

220 -202 020

111

110 040 021 101 111 060

Bi2O3-AA JCPDS: 27-0053

CuO-AA JCPDS: 80-1916

311

111

200

200

g

107

213

105

100

Intensity (a.u.)

101

440

422 511

400 200

103 002 -112 012 -120 -202 -212 -113

ZrO2-AA JCPDS: 81-1550

-110

420

140

311 220

222

TiO2-AA JCPDS: 71-1167

d

20

f

Co3O4-AA JCPDS: 43-1003

101

c

222

220 002 120 022 202 Intensity (a.u.)

b

e

WO3-AA JCPDS: 43-1035

111

a

40

60

80

2 (degree) Fig. 1. XRD plots for the oxides (a) WO3-AA, (b) Co3O4-AA, (c) TiO2-AA, (d) Bi2O3-AA, (e) ZrO2-AA, (f) CuO-AA, (g) CeO2-AA, (h) MoO3-AA, (i) Fe2O3-AA, (j) Mn3O4AA and (k) MgO-AA synthesized using solution combustion synthesis with ascorbic acid (AA) as fuel.

(Brønsted base). In a typical transesterification reaction between oil and methanol, this Brønsted base accepts the proton from the methanol and leads to the formation of methoxide ion. This methoxide ion then reacts with the triglyceride molecule and forms esters [65]. Thus, the basic strength of an oxide is an important factor that can affect the activity of a catalyst for transesterification. Hence, the basic strength of each oxide investigated in the present study was obtained using the Hammett indicators as discussed in the Section 2.2 and have been provided in the Table 3. It was observed that WO3-AA and MoO3-AA have the lowest basic strength whereas MgO-AA has the highest basic strength. The rate constants for the reaction using WO3-AA and MgO-AA also followed the same trend as the basic strength, as can be seen from the Table 3. However, oxides such as MoO3-AA and ZrO2-AA also have similar basic strength as WO3-AA (H_ < 6.8) and MgO-AA (9.3 < H_ < 1 5) but

investigated reaction using different oxides synthesized in the present study. The above discussion indicates that the basicity of the catalyst is an important factor, whose role is described in the subsequent section. 3.1.2. Basic strength or basicity of oxides: The focus of the current study was to synthesize the oxides using solution combustion synthesis for their utilization in the transesterification reaction under sub or supercritical conditions of reaction mixture and evaluate their corresponding basic strength. However, surface area of a catalyst is another important factor associated with the heterogeneous catalysts. Thus, only these two factors have been obtained in this preliminary study with combustion synthesized oxides. A metal oxide consists of two parts: a positive metal ion that accepts electron (Lewis acid) and the anionic oxygen ion that accepts proton 417

Energy Conversion and Management 173 (2018) 412–425

*

* Mn3O4-G

321 224

105

220

JCPDS: 24-0734 Mn3O4-U

Intensity (a.u.)

MgO-ODH

Intensity (a.u.)

*

103

*

*

004

MgO-L

112

101

JCPDS: 87-0652

b

311 222

220

111 200

a

211

N. Lamba et al.

MgO-G

Mn3O4-L

Mn3O4-ODH #

# MnO

#

MgO-U

MgO-AA 20

321

Mn3O4-AA * Mn2O3 *

40

60

80

20

2 (degree)

40

*

60

80

2 (degree)

Fig. 2. XRD plots for oxides (a) MgO-AA, MgO-U, MgO-G, MgO-ODH, MgO-L and (b) Mn3O4-AA, Mn3O4-ODH, Mn3O4-L, Mn3O4-G, Mn3O4-U synthesized using solution combustion synthesis with different fuels (ascorbic acid (AA), urea (U), glycine (G), oxalyl dihydrazide (ODH) and lactose (L)).

oxides (SrO, CaO, ZnO, ZrO2 and TiO2) that the basic strength followed a trend of ZrO2 < TiO2 < ZnO < CaO < SrO. However, the yield of esters obtained with ZnO in supercritical methanol at 523 K was found to be similar to SrO and CaO [27] but it was lower at temperatures below 523 K. Thus, the observations from the current and previous study suggest that basic strength is one of the important factors that affect the ester yields from the transesterification reaction. However, the operating conditions also play a key role in the activity of these oxides towards a particular reaction. Basicity (basic sites per unit mass of the catalyst) was obtained for the two best oxides (gave highest rate constants for reaction) i.e., MgOAA and Mn3O4-AA. It was found to be 1.23 mol/g and 0.77 mol/g for MgO-AA and Mn3O4-AA, respectively as can be seen from Table 4. These are higher than the earlier reported values of basicity for different oxides [49,65]. This can be attributed to the method of synthesis for oxides employed in the present study i.e., solution combustion

the rate constants obtained for the reaction with these catalysts were not similar to WO3-AA and MgO-AA. The rate constant obtained using MoO3-AA was nearly four times higher than the WO3-AA oxide and, there was an order of magnitude difference between the rate constants obtained with ZrO2-AA and MgOAA. Similar observations of catalytic activities with surface area and basicity have been made earlier with ZnO and CaO. It was noted that ZnO with lesser specific surface area and lower basic strength provided better catalytic activity than CaO [66]. Oxides such as Co3O4-AA, TiO2AA, Bi2O3-AA and Mn3O4-AA were found to have similar basic strength (6.8 < H_ < 7.2). However, the rate constant obtained using Mn3O4AA was higher in comparison to the other 3 oxides. This can be because of the presence of other oxide (Mn2O3) in case of Mn3O4-AA that might be enhancing the activity of Mn3O4-AA oxide in comparison to the other (Co3O4-AA, TiO2-AA, Bi2O3-AA) pure oxides. It has been observed from an existing study [27] with 5 different

Table 3 Conditions at which the oxides were synthesized and the characteristics of the oxides. Oxide

Calcination temperature (K) and time (h)

Specific surface area (m2/g) ( ± 1)

Crystallite size (nm)

Basic strength

Rate constants (s−1) × 104 at 523 K

WO3-AA Co3O4-AA TiO2-AA Bi2O3-AA ZrO2-AA CuO-AA CeO2-AA MoO3-AA Fe2O3-AA Mn3O4-AA MgO-AA MgO-L MgO-ODH MgO-G MgO-U Mn3O4-U Mn3O4-G Mn3O4-L Mn3O4-ODH

773 K/2 h [99] 773 K/2 h [100] Not calcined [30] 873 K/4 h [101] 773 K/24 h [102] 723 K/1 h [103] Not calcined [104] 773 K/2 h [105] 773 K/2 h [62] 673 K/2 h [106] 1073 K/1 h [107] 1073 K/1 h [107] 1073 K/1 h [107] 1073 K/1 h [107] 1073 K/1 h [107] 673 K/2 h [106] 673 K/2 h [106] 673 K/2 h [106] 673 K/2 h [106]

7 3 158 4 26 38 35 4 13 39 21 32 9 22 4 22 58 56 13

52 71 9 59 31 26 15 59 70 27 39 36 29 33 33 70 22 37 40

H_ < 6.8 6.8 < H_ 6.8 < H_ 6.8 < H_ 9.3 < H_ 7.2 < H_ 7.2 < H_ H_ < 6.8 7.2 < H_ 6.8 < H_ 9.3 < H_ 9.3 < H_ 9.3 < H_ 9.3 < H_ 9.3 < H_ 7.2 < H_ 7.2 < H_ 6.8 < H_ 6.8 < H_

0.432 0.602 0.922 0.930 1.58 2.17 4.26 4.55 8.68 13.2 19.8 4.72 6.27 7.28 11.9 7.62 8.72 9.41 10.4

418

< < < < < <

7.2 7.2 7.2 15 9.3 9.3

< < < < < < < < < < <

9.3 7.2 15 15 15 15 15 9.3 9.3 7.2 7.2

Energy Conversion and Management 173 (2018) 412–425

N. Lamba et al.

Table 4 Basicity of the two best oxides and the activation energies for the transesterification reaction using supercritical methanol with these oxides. Basicity (mol/g)

MgO-AA Mn3O4-AA Reaction systems Mahua oil + MeOH + MgO-AA Mahua oil + MeOH + Mn3O4-AA

6.8 < H_ < 7.2

7.2 < H_ < 9.3

9.3 < H_ < 15

Total basicity (mol/g)

0.87 0.77

0.08 – Activation energy (kJ/mol)

0.28 – Pre-exponential factors (s−1)

1.23 0.77

36 51

9.33 140

catalysts [72]. If all the above 13 models based on Hattori, LHHW and ER mechanisms are considered [69] then, it is clear that 5 of these models reduce to first order expressions in terms of triglyceride’s concentration. These models are ER model with surface reaction as rate limiting step, the LHHW model with triglyceride adsorption as rate limiting step and the Hattori model with triglyceride adsorption as rate limiting step, respectively. Thus, it was observed when the surface reaction or the triglyceride adsorption is the rate controlling step then the overall reaction order is unity, when appropriate assumptions are made. However, a detailed kinetic study for a wide range of data shows that LHHW with methanol adsorption and ER with surface reaction as rate limiting step leads to more appropriate models [69,72]. The data obtained in the present study shows that the experimental data followed the first order kinetics in triglyceride’s concentration. A recent study shows that the surface reaction step should be assumed to rate limiting step for the catalysts having more basic strength [73] and, a range of basic strength corresponding to different synthesized oxides have been obtained in the present study. Thus, an intrinsic kinetic model was developed in the present study for the transesterification of mahua oil in supercritical methanol in the presence of metal oxides using the Eley-Rideal (ER) reaction mechanism with surface reaction as rate limiting step. The elementary steps involved in the mechanism can be written as: adsorption of methanol on the catalyst surface; reaction of the adsorbed methanol with oil to form FAMEs and adsorbed glycerol; desorption of the products as shown below.

synthesis in comparison to the usually applied co-precipitation method for the oxides synthesis. 3.2. Transesterification reaction with supercritical methanol using metal oxides Mahua oil was reacted with supercritical methanol in the presence of different metal oxides. The physical and thermodynamic properties of oil and methanol (fatty acid composition of oil, average molecular weight and critical point) have been reported earlier [51]. 3.2.1. Mechanism of the transesterification reaction A conventional heterogeneous reaction occurs in several steps: diffusion of one liquid to the interface of other liquid, then to the bulk liquid; transport of the reactants to the catalyst surface; pore diffusion; surface reaction and product desorption. The proposed mechanism for heterogeneous base catalyzed transesterification consists of extraction of proton by the O2− on the catalyst site from CH3OH leading to the formation of methoxide ion [67]. This CH3O− ion then attacks the carbonyl carbon of the triglyceride molecule and forms a tetrahedral intermediate. The tetrahedral complex then takes the proton from the surface of catalyst or reacts with another methanol molecule to form another methoxide ion [68]. The rearrangement of this tetrahedral complex results to the formation of FAMEs [65,69]. In the present work, the interliquid mass transfer (associated with the conventional heterogeneous reaction) between the reactants was overcome because of the single phase formed in supercritical methanol [70]. However, the external and internal mass transfer limitations associated with the reactant and the catalyst phase were neglected in the present study. These resistances are generally overcome by stirring the reaction mixture (external resistance) and reducing the size of catalyst (internal resistance) [71]. In case of reactions employing supercritical methylating agents (such as supercritical methanol), due to the high diffusivity, there is no external mass transfer resistance and, therefore, no stirring of the reactor is normally required. Due to the size of the material (nanomaterials) used as catalyst, there is negligible internal mass transfer resistance. There are three different types of mechanisms available in the literature for the heterogeneous transesterification: Hattori, LangmuirHinshelwood-Watson (LHHW) and Eley-Rideal mechanism. MgO catalyzed transesterification of ethyl acetate with methanol has been studied earlier, and in order to obtain a kinetic model for the same reaction, 13 different models have been illustrated using these three different mechanisms. Hattori and LHHW, mechanisms consist of 5 steps. However, Eley-Rideal is a simple three step mechanism. Various investigators have investigated the applicability of the above models to different catalytic systems [69,72,73]. The model developed using EleyRideal mechanism by considering the adsorption of methanol on the catalyst’s surface as the rate determining step, provided the best results for the MgO catalyzed transesterification of ethyl acetate with methanol [69]. Further, the rate expression obtained using methanol adsorption as the rate determining step in LHHW mechanism was found to be the most suitable for the production of biodiesel with the hydrotalcite

k1

A + * ⇌ A*

(7)

k−1 k2

A* + B ⇌ C + D*

(8)

k−2 k3

D* ⇌ D + *

(9)

k−3

A, B, C and D denote methanol, oil, FAMEs and glycerol, respectively. The rate equation for each step are

r1 = k1 CA θS−k−1 θA

(10)

r2 = k2 θA CB−k−2 CC θD

(11)

r3 = k3 θD−k−3 CD θS

(12)

From the total site balance, (13)

θA + θD + θS = 1

The second step (surface reaction) was assumed to be rate determining step (RDS) step and the other two steps were quasi-equilibrated. Thus, equating the rate of reaction from step 1 (Eq. (10)) and 3 (Eq. (12)) to zero to obtain θA and θD .

θA =

419

k1 CA θS k−1

and θD =

k3 CD θS . k−3

Energy Conversion and Management 173 (2018) 412–425

N. Lamba et al.

Using the values of θA and θD and from Eq. (13), θs was obtained, k1 1 θs = (adsorption equilibrium constant) and CD . Where, K1 = k 1 + K1 CA +

1.0

−1

K3

0.8

k

⎛ 1 r2 = ⎜ ⎜ 1 + K1 CA + ⎝

⎞ k−2 ⎛ ⎞ ⎟ k2 K1 CA CB− K CC CD CD ⎟ 3 ⎝ ⎠ K3 ⎠ ⎜

% Conversion

K3 = k 3 (desorption equilibrium constant). Thus, the rate of reaction −3 can be obtained as,

⎟

(14)

The value of the desorption equilibrium constant, K3, is high because the rate of the reverse step of desorption i.e., k−3 is very small. Thus, Eq. (14) can be simplified as,

kKC C r2 = ⎛ 2 1 A B ⎞ ⎝ 1 + K1 CA ⎠ ⎜

(15)

0.2 0

10

20

30 40 t / min

50

60

70

⎟

(16) the oxide was employed for the reaction studied in the present work. The variation of pseudo first order rate constant with different oxides can be seen from Fig. 4 and Table 3, respectively. The slope of the regressed linear fit between −ln(1 − XB) and t gives the rate constant for the reaction (see electronic supplementary information, Fig. S1). Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.enconman.2018.07.067. MgO having the highest basic strength showed the best activity and WO3 with lowest basic strength was least active towards the transesterification reaction with supercritical methanol at 523 K and 10 MPa. The rate constants for the reaction with other oxides have been discussed earlier in Section 3.1.2. The two oxides (MgO-AA and Mn3O4AA) that gave the highest rate constants for the transesterification reaction were investigated further. These were synthesized using different fuels (urea, glycine, lactose and ODH) with various functional groups. Urea consists of 2 amide groups; glycine possesses 1 amide + 1 carboxylic group; lactose has 8 hydroxyl groups; ODH has 2 amide groups with 2 carbonyl carbons and ascorbic acid consists of 4 hydroxyl + 1 carboxylic groups. Different type of functional groups could lead to different type of combustion with the metal precursor and thus would lead to formation of an oxide with different specific surface areas and basic strength with different fuels [75]. It has been observed earlier in a study of synthesis of iron oxide powders using different fuels groups such as glycine, hydrazine and citric acid resulted to different phases of iron oxides with different specific surface areas [76]. The possible reason for this was reported to be the presence of different types of functional groups in the fuels. It was concluded from the study that fuels with eNH2 type ligand were most active towards the combustion followed by eOH and eCOOH types of ligands [76]. A similar observation of obtaining different phases for an oxide was made with Mn3O4 synthesized in the present study using different fuels (ascorbic acid, urea, glycine, lactose and ODH). It can be observed from the XRD results for Mn3O4 shown in Fig. 2(b). It can be seen from the XRD plot that along with Mn3O4, there are peaks corresponding to Mn2O3 and MnO in case of synthesis with ascorbic acid, urea (Mn2O3) and ODH (MnO), respectively. However, pure Mn3O4 was obtained with lactose and glycine, Fig. 2(b) and similarly a single phase MgO (Fig. 2(a)) was obtained with different fuels. This can be a possible reason for difference in basic strength for Mn3O4 with various fuels, which is not observed in case of MgO. Thus, inclusion of different types of fuels results to different extent of exothermicity and evolution of gases during the combustion that would impact the phase, porosity or particulate properties of synthesized oxide [75]. These oxides synthesized with different fuels results to different activity towards a particular reaction or process. For example, the

For high concentrations of methanol (CA), it is apparent that K1CA ≫ 1 and thus Eqs. (15) or (16) simplifies to,

r2 = k2 CB

0.5 % 1.0 % 2.0 % 3.0 %

Fig. 3. Variation of conversion of Mahua oil to FAMEs with time for different loadings (■, 0.5 wt%, ●, 1 wt%, ▴, 2 wt% and ▾, 3 wt%) of MgO-AA at 523 K and 10 MPa. Solid lines represent the first order fit.

Eq. (15) can be rewritten as, ⎜

0.4

0.0

⎟

1 1 ⎛ 1 = + 1⎞ r2 k2 CB ⎝ K1 CA ⎠

0.6

(16a)

The rate constants were obtained by using Eq. (16a) expressed in terms of conversion i.e., −ln (1 − XB) = k2t. Here, XB denotes the conversion of triglycerides to FAMEs that can be expressed using the initial (CBo) and final (CB) concentration of triglycerides as XB = (CBo − CB)/CBo. It is important to note that the intermediate kinetic parameters such as K3 and K1 could not be determined because of the overall kinetics was first order with parameter k2. Therefore, Eq. (16a) was fitted to the experimental data and the value of k2 was obtained. However, if the concentrations of methanol are near stoichiometric i.e., when K1CA is not significantly greater than unity, then Eq. (15) should be used for determining the rate constants, as discussed earlier [74]. In this case, the inverse of rate should be plotted versus the inverse of CA keeping CB constant and then the inverse of rate with the inverse of CB keeping CA constant. Both will be linear fits and the rate constant, k2 and the equilibrium constant, K1, can be obtained. Eq. (16a) is the simplified pseudo first order irreversible kinetic model with k2 being the forward rate constant and CB is the concentration of the reactant, oil. As the molar ratio of CA to CB is very high (= 40) in the current study, Eq. (16a) applies for this system. This equation was used to obtain the rate constants for integrated transesterification reaction with different oxides. 3.2.2. Catalytic activity The transesterification reaction in the presence of different oxides was performed at 523 K, 10 MPa, 40:1 (molar ratio of methanol to oil) and 1 wt% of a particular oxide. The operating temperature and pressure that gives the maximum conversion of triglycerides were selected based on the literature available for the integrated (catalytic and noncatalytic) transesterification (Table 1). The selected operating conditions ensure the supercritical phase of methanol (Tc = 512 K and Pc = 7.85 MPa). The amount of catalyst to be used was optimized by performing the reaction with the above operating conditions (T = 523 K, P = 10 MPa, molar ratio of oil to methanol = 40:1) at various concentrations of oxide (MgO-AA) varying from 0.5 to 3 wt% as shown in Fig. 3. It was observed that the rate constants (pseudo first order) doubled when the catalyst concentration was changed from 0.5 to 1% (k = 0.96 × 10−4 to 19.8 × 10−4 s−1). However, there was a slight decrease in rate constants was observed from 1 to 2% (18.2 × 10−4 s−1) and further to 3% (17.9 × 10−4 s−1). These observations were similar to other reported literature [48]. Thus, 1% of 420

Energy Conversion and Management 173 (2018) 412–425

N. Lamba et al.

1.0 0.8

TiO2 Bi2O3

0.6

% Conversion

% Conversion

1.0

WO3 Co3O4

0.4

0.6 MgO - L MgO - ODH MgO - G MgO - U MgO - AA

0.4 0.2

0.2

(a) 0.0

0.8

0

20

40

60

0.0

80

(a) 0

10

20

% Conversion

0.8

ZrO2 CuO CeO2

1.0

MoO3

0.8

% Conversion

1.0

0.6 0.4 0.2 0.0

0

10

20

30

40

50

60

70

50

60

Mn3O4 - L Mn3O4 - ODH Mn3O4 - G

0.4

(b) 80

0.0

Mn3O4 - U Mn3O4 - AA

(b) 0

10

20

30

40

t / min

50

60

70

Fig. 5. Variation of conversion of Mahua oil to FAMEs with time at 523 K and 10 MPa using (a) MgO-AA, MgO-U, MgO-G, MgO-ODH, MgO-L and (b) Mn3O4AA, Mn3O4-U, Mn3O4-G, Mn3O4-ODH, Mn3O4-L under sub or supercritical conditions. Solid lines represent the first order fit. The rate constants were obtained using Eq. (16a) (pseudo first order kinetics) and are shown in Fig. S2 (see electronic supplementary information).

1.0

% Conversion

40

0.6

0.2

t / min

0.8 0.6 Fe2O3 Mn3O4 MgO

0.4 0.2 0.0

30

t / min

t / min

MgO-AA, MgO-U, MgO-G, MgO-ODH, MgO-L and Mn3O4-AA, Mn3O4-U, Mn3O4-G, Mn3O4-ODH, Mn3O4-L were obtained using the pseudo first order kinetic fit (see electronic supplementary information, Fig. S2). The activity of different oxides followed a trend wherein the maximum catalytic activity was shown by the oxides prepared with AA. Generally, glycine is the most commonly used fuel in the combustion synthesis because of its ignition temperature being equal to the decomposition temperature for nitrates leading to the formation of catalysts with higher specific area. In the present study, the oxides prepared with glycine also showed good activity in both the cases. The rate constant with MgO-AA was found to be nearly thrice of MgO-G and, the rate constant for Mn3O4-AA was 1.5 times of Mn3O4-G. The trend can be discerned from the Table 3 and Fig. 5. It can be observed from the trend that in the case of an oxide synthesized using different fuels, the activity of the catalyst might depend on some other factors as well along with the basic strength (as the rate constants did not follow the trend of basic strength). One of the possible reasons can be the structure of the oxides that have been employed for the reaction. In order to verify that scanning electron microscopy (SEM) was performed for some of the oxides synthesized in the present study. It was observed from the micrographs shown in Figs. S3 and S4 (see electronic supplementary information) that nanoparticles were forming a network like structure with a wide size distribution with all the different oxides. It was observed that a big network having more porous structure was leading to higher specific surface area. The area of MgO-U and MgO-ODH was lesser than the other three MgO because of the comparatively smaller network formation in case of both of these oxides (Fig. S4 (see electronic supplementary information)). Similarly, in case

(c) 0

10

20

30

40

50

60

70

t / min Fig. 4. Variation of conversion of mahua oil to FAMEs with time at 523 K and 10 MPa using different catalysts (a) WO3-AA, Co3O4-AA, TiO2-AA, Bi2O3-AA, (b) ZrO2-AA, CuO-AA, CeO2-AA, MoO3-AA, and (c) Fe2O3-AA, Mn3O4-AA, MgOAA. Solid lines represent the pseudo first order fit. The straight lines based on Eq. (16a) (pseudo first order kinetics) to obtain rate constants are shown in Fig. S1 (see electronic supplementary information).

photocatalytic activity of the semiconductor oxide, WO3 showed significantly different photocatalytic activity for the degradation of methylene blue when it was synthesized with diverse fuels such as urea, glycine and thiourea as fuels [42]. When TiO2 was synthesized with various fuels such as glycine, hexamethylene-tetramine and ODH [40], it showed different activity for the degradation of various dyes. Similar studies of luminescence, antibacterial and photocatalytic activity of ZnO synthesized using different fuels have also been reported [28,77,78]. This clearly indicates that the choice of fuel significantly determines the catalytic activity of the oxide. In this study, MgO-AA and Mn3O4-AA gave the highest rate constants (Table 3) transesterification. Therefore, MgO and Mn3O4 were synthesized using different fuels. The rate constants for different oxides

421

Energy Conversion and Management 173 (2018) 412–425

N. Lamba et al.

of MgO-AA, TiO2-AA, Bi2O3-AA and CeO2-AA, the cluster of nanoparticles were leading to a smaller network in comparison to other three oxides (Fig. S3 (see electronic supplementary information)). It suggests that there was less evolution of gases in the combustion synthesis during the formation of these oxides which are having lower specific surface areas and smaller networks. However, the activity did not seem to follow any trend. It might vary in case when different morphological structures (shapes) such as nanorods, spheres, towers, etc. are obtained. However, no such shapes were obtained in the present study. The catalytic activity of the best oxide (as determined based on the conversions and rate constants for the reaction) was also compared with the existing catalysts that have been employed in the transesterification reaction under sub or supercritical conditions of reaction mixture (Table 1). The optimum temperature for the reactions in studies with serial number (S.N) 1 (CaO), 6 (ZnO), 11 (NaOH), 16 (triethylamine + propylene oxide) and 22 (CH3ONa) was 523 K, similar to what have been employed in the present study to obtain the rate constants for the reaction. Thus, only these studies from Table 1 were selected for the comparison with the best oxide obtained in present study. It can be observed from the Table 1 that similar order of rate constants as that with MgO-AA (synthesised in this study) were obtained with CaO, ZnO, triethylamine + propylene oxide and CH3ONa except that of the reaction using NaOH. However, it is important to note here that triethylamine + propylene oxide, CH3ONa and NaOH leads to homogeneous reaction because of their complete dissolution in reaction mixture. Thus, NaOH being having the highest dissolution power resulted to a higher rate constant in comparison to other catalysts. Other catalysts such as CaO from the first study is resulting to almost double the rate constant (35.2 × 10−4 s−1) as obtained with MgO-AA (19.8 × 10−3 s−1). However, it should be noted from the existing study (Serial number, S.N 6) with different oxides such as SrO, CaO, ZnO, TiO2 and ZrO2 synthesized using coprecipitation has resulted to ZnO as the best oxide with a rate constant of 17.3 × 10−4 s−1 which is lesser than the reaction with MgO-AA at similar operating conditions. Thus, MgOAA synthesized using solution combustion synthesis can be a potential candidate for the transesterification reaction under supercritical conditions of methanol.

1.0

% Conversion

0.8 MgO - AA 503 K 523 K 543 K 563 K 583 K

0.6 0.4 0.2

(a) 0.0

0

10

20

30

40

50

60

70

80

t / min

% Conversion

1.0 0.8 Mn3O4 - AA

0.6

503 K 523 K 543 K 563 K 583 K

0.4 0.2 0.0

(b) 0

10

20

30

40

50

60

70

t / min Fig. 6. Synthesis of FAMEs under sub or supercritical conditions using the best catalysts (a) MgO-AA and (b) Mn3O4-AA at different temperatures: ■ 503 K, ● 523 K, ▴ 543 K, ▾ 563 K and ♦ 583 K. Solid lines represent first order fit. Eq. (16a) was used to obtain the rate constants and the linear lines are shown in Fig. S5 (see electronic supplementary information).

[51]. Further, conversions of more than 99% were obtained at 698 K, 30 MPa in 10 min. Conversions of less than 40% at 523 K and 30 MPa were obtained after an hour in the non-catalytic pathway of FAMEs synthesis from mahua oil. The rate constant for the non-catalytic synthesis of FAMEs from mahua oil at 523 K and 30 MPa was observed to be 9.9 × 10−5 s−1. It was found that the rate constants obtained from the integrated (supercritical methanol + catalyst) transesterification of mahua oil with different oxides (ZrO2-AA, CuO-AA, CeO2-AA, MoO3-AA, Fe2O3-AA, Mn3O4-AA, MgO-AA) at 523 K and 10 MPa were always higher than the non-catalytic (only supercritical methanol) pathway. The rate constants (k × 10−4 s−1) for the transesterification reaction for the two most active oxides (MgO-AA and Mn3O4-AA) were found to be 19.8 and 13.2, respectively. Thus, the rate constants obtained by this catalytic integrated transesterification reaction is higher by an order of magnitude compared to the non-catalytic system. However, it can also be observed from the above comparison that complete conversions of triglycerides can be obtained at temperatures almost 100 K lower (583 K) than the non-catalytic reaction operating temperature and also at lower pressures. The activation energies and pre-exponential factors for the reaction with the most active oxides (MgO-AA and Mn3O4-AA) were obtained using the Arrhenius plot, as shown in Fig. 7. The activation energies were obtained from the slope whereas; the pre-exponential factors were obtained from the intercept of the regressed line of the logarithm of the rate constant with the inverse of temperature. These values have been provided in Table 4. The activation energies were found to be 36 kJ/mol and 51 kJ/mol for transesterification of mahua oil with supercritical methanol using MgO-AA and Mn3O4-AA, respectively. However, higher activation

3.2.3. Influence of temperature and time on transesterification The reaction with mahua oil with sub or supercritical methanol with a molar ratio of 40:1 using 1 wt% of MgO-AA and Mn3O4-AA was performed at various temperatures of 503–583 K for 2–72 min at 10 MPa. These operating conditions are in the region of near critical point (Tr = 0.9–1.5, Pr = 1–2) of pure methanol. The critical point of the reaction mixture was obtained using the LB mixing rules and was found to be 570 K and 6.5 MPa. Thus, the reactions were performed in the near critical or supercritical region of the mixture. The rate constants for the reaction at various temperatures using MgO-AA and Mn3O4-AA were obtained using the Eq. (16a) expressed in terms of conversion (see supplementary information, Fig. S3). It was observed for both the reactions (with MgO-AA and Mn3O4-AA) that conversion of oil to FAMEs increases with an increase in temperature and time of reaction as can be observed from the Fig. 6. A conversion of more than 85% was obtained within 20 min at temperatures of 503 K and 523 K and almost complete conversion to FAMEs at 543–583 K was obtained within 15 to 30 min using MgO-AA with supercritical methanol. In the reaction using Mn3O4-AA, 85% of conversion of oil was obtained at 503 K in 45 min, a conversion of more than 95% was obtained at 523–563 K in 45 min and at 583 K almost complete conversion was obtained within 30 min. The rate constants all the reactions with different oxides have been obtained using Eq. (16a) and tabulated in Table 3. In an earlier study of reaction of mahua oil with only supercritical methanol (non-catalytic), conversions of more than 80% were obtained at a temperature of 623–698 K and pressure of 30 MPa within 25 min 422

Energy Conversion and Management 173 (2018) 412–425

N. Lamba et al.

-3

-4

ln k

conversions of mahua oil to FAMEs using this integrated transesterification system were reduced by 100 K and 20 MPa. Thus, both the capital cost and the operating cost associated with the supercritical pathway can be reduced to make this route more economically feasible. In this regard, future studies can investigate different oxide catalysts with various morphologies (nanorods, nanoflowers, nanospheres, etc.) or substitutions.

MgO - AA, ln k = - 4.38/ T + 2.23 Mn3O4 - AA, ln k = - 6.16 / T + 4.97

-5

Acknowledgements

-6

The authors thank Council of Scientific and Industrial Research (CSIR), India for the financial support provided for the work. Sangeeta Adhikari would like to thank UGC-Dr D.S. Kothari Postdoctoral Fellowship and the Government of India for research funding. The corresponding author sincerely thanks the Department of Science and Technology, India for J.C Bose fellowship.

-7 1.70

1.75

1.80

1.85

1.90

(1000 / T ) / K

1.95

2.00

-1

Fig. 7. Arrhenius plot for synthesis of FAMEs under sub or supercritical conditions using MgO-AA and Mn3O4-AA as catalysts.

References [1] Mahmudul H, Hagos F, Mamat R, Adam AA, Ishak W, Alenezi R. Production, characterization and performance of biodiesel as an alternative fuel in diesel engines–a review. Renew Sust Energ Rev 2017;72:497–509. [2] Hajjari M, Tabatabaei M, Aghbashlo M, Ghanavati H. A review on the prospects of sustainable biodiesel production: a global scenario with an emphasis on waste-oil biodiesel utilization. Renew Sust Energ Rev 2017;72:445–64. [3] Gajdobranski A, Latković D, Krmpot V. Biofuels in serbia-production and basic costs. Izazovi zelene ekonomije 2018:173. [4] Liew WH, Hassim MH, Ng DK. Review of evolution, technology and sustainability assessments of biofuel production. J Clean Prod 2014;71:11–29. [5] Tamilselvan P, Nallusamy N, Rajkumar S. A comprehensive review on performance, combustion and emission characteristics of biodiesel fuelled diesel engines. Renew Sust Energ Rev 2017;79:1134–59. [6] Murugesan A, Umarani C, Chinnusamy T, Krishnan M, Subramanian R, Neduzchezhain N. Production and analysis of bio-diesel from non-edible oils—a review. Renew Sust Energ Rev 2009;13:825–34. [7] Balat M. Potential alternatives to edible oils for biodiesel production–a review of current work. Energy Convers Manage 2011;52:1479–92. [8] Lin L, Cunshan Z, Vittayapadung S, Xiangqian S, Mingdong D. Opportunities and challenges for biodiesel fuel. Appl Energy 2011;88:1020–31. [9] Atabani AE, Silitonga AS, Badruddin IA, Mahlia T, Masjuki H, Mekhilef S. A comprehensive review on biodiesel as an alternative energy resource and its characteristics. Renew Sust Energ Rev 2012;16:2070–93. [10] Aransiola E, Ojumu T, Oyekola O, Madzimbamuto T, Ikhu-Omoregbe D. A review of current technology for biodiesel production: state of the art. Biomass Bioenergy 2014;61:276–97. [11] Chouhan AS, Sarma A. Modern heterogeneous catalysts for biodiesel production: a comprehensive review. Renew Sust Energ Rev 2011;15:4378–99. [12] Islam A, Taufiq-Yap YH, Chan E-S, Moniruzzaman M, Islam S, Nabi MN. Advances in solid-catalytic and non-catalytic technologies for biodiesel production. Energy Convers Manage 2014;88:1200–18. [13] Madras G, Kolluru C, Kumar R. Synthesis of biodiesel in supercritical fluids. Fuel 2004;83:2029–33. [14] Saka S, Isayama Y. A new process for catalyst-free production of biodiesel using supercritical methyl acetate. Fuel 2009;88:1307–13. [15] García-Martínez N, Andreo-Martínez P, Quesada-Medina J, de los Ríos AP, Chica A, Beneito-Ruiz R, et al. Optimization of non-catalytic transesterification of tobacco (Nicotiana tabacum) seed oil using supercritical methanol to biodiesel production. Energy Convers Manage 2017;131:99–108. [16] Sakdasri W, Sawangkeaw R, Ngamprasertsith S. Techno-economic analysis of biodiesel production from palm oil with supercritical methanol at a low molar ratio. Energy 2018. [17] Sawangkeaw R, Bunyakiat K, Ngamprasertsith S. A review of laboratory-scale research on lipid conversion to biodiesel with supercritical methanol (2001–2009). J Supercrit Fluids 2010;55:1–13. [18] Son J, Kim B, Park J, Yang J, Lee JW. Wet in situ transesterification of spent coffee grounds with supercritical methanol for the production of biodiesel. Bioresour Technol 2018;259:465–8. [19] Jazzar S, Olivares-Carrillo P, de los Ríos AP, Marzouki MN, Acién-Fernández FG, Fernández-Sevilla JM, et al. Direct supercritical methanolysis of wet and dry unwashed marine microalgae (Nannochloropsis gaditana) to biodiesel. Appl Energy 2015;148:210–9. [20] Marchetti J, Errazu A. Technoeconomic study of supercritical biodiesel production plant. Energy Convers Manage 2008;49:2160–4. [21] Marchetti J, Miguel V, Errazu A. Techno-economic study of different alternatives for biodiesel production. Fuel Process Technol 2008;89:740–8. [22] Van Kasteren J, Nisworo A. A process model to estimate the cost of industrial scale biodiesel production from waste cooking oil by supercritical transesterification. Resour Conserv Recy 2007;50:442–58. [23] Glišić S, Lukic I, Skala D. Biodiesel synthesis at high pressure and temperature: analysis of energy consumption on industrial scale. Bioresour Technol

energy of 75 kJ/mol was obtained in case of non-catalytic transesterification of mahua oil using supercritical methanol at T = 523–698 K and P = 30 MPa [51]. Thus, the catalysts significantly enhance the reaction rates and reduce the activation energy and this study clearly shows the advantage of using oxides as catalysts. It should be noted that the reusability of the catalyst was not investigated in this study. This is because the amount of catalyst used in this reaction was very small and it was difficult to recover the same amount of catalyst (as used in first cycle) after washing it with heptane and methanol and then drying for using it for second cycle. However, when the XRD of the used catalyst was taken, there was no change in the pattern indicating no change in the structure of the catalyst even after reaction. Further, combustion synthesized catalysts have been extensively studied and reused. For example, the reusability of these catalysts has been studied for photocatalytic reactions [27,28], adsorption [43] and gas phase reactions [29,79] with the catalyst being active for several repeated cycles. Therefore, it is believed that the catalyst can be reused without loss of activity in this case also. 4. Conclusions The present study demonstrates an integrated transesterification of mahua oil using supercritical methanol in the presence of different oxides synthesized using solution combustion synthesis. Eley-Rideal type of mechanism was proposed for the reaction leading to the pseudo first order kinetics. Ascorbic acid synthesized MgO and Mn3O4 were found to be the most active catalysts (with the highest rate constants) due to their highest basic strength. The first order rate constants for the reaction with MgO-AA and Mn3O4-AA (AA-ascorbic acid as the fuel) at 523 K and 10 MPa were 19.8 × 10−4 s−1 and 13.2 × 10−4 s−1, respectively and were higher by an order of magnitude compared to the rate constant of 9.9 × 10−5 s−1 at 523 K and 30 MPa obtained for the non-catalytic supercritical transesterification reaction. The influence of temperature and time on the conversion of oil was studied and, the rate constants and activation energies were obtained. The increase in temperature and time had a significant effect on the conversions. The rate constants for the transesterification with most active oxides combusted using different fuels followed a trend, wherein, the oxides prepared with AA exhibited the highest activity and the oxides prepared with glycine showed moderate activity. The activation energies obtained were 36 kJ/mol (with MgO-AA) and 51 kJ/mol (with Mn3O4-AA) for the reaction that were lower than the non-catalytic supercritical synthesis (75 kJ/mol) of FAMEs. While the non-catalytic reaction achieved near-complete conversions (> 99%) at 698 K and 30 MPa, whereas, similar conversions were obtained at much lower temperature 523 K at 10 MPa. Thus, the operating conditions for the complete 423

Energy Conversion and Management 173 (2018) 412–425

N. Lamba et al.

soybean oil. Chem Eng J 2014;241:418–32. [55] Mokkelbost T, Kaus I, Grande T, Einarsrud M-A. Combustion synthesis and characterization of nanocrystalline CeO2-based powders. Chem Mater 2004;16:5489–94. [56] Roy S, Ghose J. Synthesis of stable nanocrystalline cubic zirconia. Mater Res Bull 2000;35:1195–203. [57] Azam A, Ahmed AS, Oves M, Khan M, Memic A. Size-dependent antimicrobial properties of CuO nanoparticles against Gram-positive and-negative bacterial strains. Int J Nanomed 2012;7:3527. [58] Jalalah M, Faisal M, Bouzid H, Park J-G, Al-Sayari S, Ismail AA. Comparative study on photocatalytic performances of crystalline α-and β-Bi2O3 nanoparticles under visible light. J Ind Eng Chem 2015;30:183–9. [59] Orante-Barrón V, Oliveira L, Kelly J, Milliken E, Denis G, Jacobsohn L, et al. Luminescence properties of MgO produced by solution combustion synthesis and doped with lanthanides and Li. J Lumin 2011;131:1058–65. [60] Adhikari S, Mandal S, Sarkar D, Kim D-H, Madras G. Kinetics and mechanism of dye adsorption on WO3 nanoparticles. Appl Surf Sci 2017;420:472–82. [61] Li X-L, Liu J-F, Li Y-D. Low-temperature synthesis of large-scale single-crystal molybdenum trioxide (MoO3) nanobelts. Appl Phys Lett 2002;81:4832–4. [62] Manikandan A, Vijaya JJ, Kennedy LJ. Structural, optical and magnetic properties of porous α-Fe2O3 nanostructures prepared by rapid combustion method. J Nanosci Nanotechnol 2013;13:2986–92. [63] Gao W, Ye S, Shao M. Solution-combusting preparation of mono-dispersed Mn3O4 nanoparticles for electrochemical applications. J Phys Chem Solids 2011;72:1027–31. [64] Refaat A. Biodiesel production using solid metal oxide catalysts. Int J Environ Sci Tech 2011;8:203–21. [65] Yan S, Kim M, Salley SO, Ng KS. Oil transesterification over calcium oxides modified with lanthanum. Appl Catal A 2009;360:163–70. [66] Hoang D, Bensaid S, Saracco G, Pirone R, Fino D. Investigation on the conversion of rapeseed oil via supercritical ethanol condition in the presence of a heterogeneous catalyst. Green Process Synth 2017;6:91–101. [67] Lee D-W, Park Y-M, Lee K-Y. Heterogeneous base catalysts for transesterification in biodiesel synthesis. Catal Surv Asia 2009;13:63–77. [68] Zeng D, Yang L, Fang T. Process optimization, kinetic and thermodynamic studies on biodiesel production by supercritical methanol transesterification with CH3ONa catalyst. Fuel 2017;203:739–48. [69] Dossin TF, Reyniers M-F, Marin GB. Kinetics of heterogeneously MgO-catalyzed transesterification. Appl Catal B 2006;62:35–45. [70] Baiker A. Supercritical fluids in heterogeneous catalysis. Chem Rev 1999;99:453–74. [71] Chantrasa A, Phlernjai N, Goodwin Jr. JG. Kinetics of hydrotalcite catalyzed transesterification of tricaprylin and methanol for biodiesel synthesis. Chem Eng J 2011;168:333–40. [72] Kapil A, Wilson K, Lee AF, Sadhukhan J. Kinetic modeling studies of heterogeneously catalyzed biodiesel synthesis reactions. Ind Eng Chem Res 2011;50:4818–30. [73] Xiao Y, Gao L, Xiao G, Lv J. Kinetics of the transesterification reaction catalyzed by solid base in a fixed-bed reactor. Energy Fuels 2010;24:5829–33. [74] Narayan RC, Madras G. Kinetics of non-catalytic synthesis of bis (2-ethylhexyl) sebacate at high pressures. React Chem Eng 2017;2:27–35. [75] Chen P, Qin M, Chen Z, Jia B, Qu X. Solution combustion synthesis of nanosized WO x: characterization, mechanism and excellent photocatalytic properties. RSC Adv 2016;6:83101–9. [76] Deshpande K, Mukasyan A, Varma A. Direct synthesis of iron oxide nanopowders by the combustion approach: reaction mechanism and properties. Chem Mater 2004;16:4896–904. [77] Pathak TK, Kumar A, Swart C, Swart H, Kroon R. Effect of fuel content on luminescence and antibacterial properties of zinc oxide nanocrystalline powders synthesized by the combustion method. RSC Adv 2016;6:97770–82. [78] Potti PR, Srivastava VC. Comparative studies on structural, optical, and textural properties of combustion derived ZnO prepared using various fuels and their photocatalytic activity. Ind Eng Chem Res 2012;51:7948–56. [79] Singh SA, Madras G. Detailed mechanism and kinetic study of CO oxidation on cobalt oxide surfaces. Appl Catal A 2015;504:463–75. [80] Demirbas A. Biodiesel from sunflower oil in supercritical methanol with calcium oxide. Energy Convers Manage 2007;48:937–41. [81] Sawangkeaw R, Tejvirat P, Ngamcharassrivichai C, Ngamprasertsith S. Supercritical transesterification of palm oil and hydrated ethanol in a fixed bed reactor with a CaO/Al2O3 catalyst. Energies 2012;5:1062–80. [82] Asri NP, Machmudah S, Budikarjono K, Roesyadi A, Goto M. Palm oil transesterification in sub-and supercritical methanol with heterogeneous base catalyst. Chem Eng Process 2013;72:63–7. [83] Rodríguez-Guerrero JK, Rosa PT. Production of biodiesel from castor oil using sub and supercritical ethanol: effect of sodium hydroxide on the ethyl ester production. J Supercrit Fluids 2013;83:124–32. [84] Maçaira J, Santana A, Recasens F, Larrayoz MA. Biodiesel production using supercritical methanol/carbon dioxide mixtures in a continuous reactor. Fuel 2011;90:2280–8. [85] Kim M, Lee H-S, Yoo SJ, Youn Y-S, Shin YH, Lee Y-W. Simultaneous synthesis of biodiesel and zinc oxide nanoparticles using supercritical methanol. Fuel 2013;109:279–84. [86] Petchmala A, Laosiripojana N, Jongsomjit B, Goto M, Panpranot J, Mekasuwandumrong O, et al. Transesterification of palm oil and esterification of palm fatty acid in near-and super-critical methanol with SO4–ZrO2 catalysts. Fuel 2010;89:2387–92.

2009;100:6347–54. [24] Glisic SB, Orlović AM. Review of biodiesel synthesis from waste oil under elevated pressure and temperature: phase equilibrium, reaction kinetics, process design and techno-economic study. Renew Sust Energ Rev 2014;31:708–25. [25] Shao Z, Zhou W, Zhu Z. Advanced synthesis of materials for intermediate-temperature solid oxide fuel cells. Prog Mater Sci 2012;57:804–74. [26] Varma A, Mukasyan AS, Rogachev AS, Manukyan KV. Solution combustion synthesis of nanoscale materials. Chem Rev 2016;116:14493–586. [27] Eswar NK, Singh SA, Madras G. Photoconductive network structured copper oxide for simultaneous photoelectrocatalytic degradation of antibiotic (tetracycline) and bacteria (E. coli). Chem. Eng J 2018;332:757–74. [28] Adhikari S, Gupta R, Surin A, Kumar TS, Chakraborty S, Sarkar D, et al. Visible light assisted improved photocatalytic activity of combustion synthesized spongyZnO towards dye degradation and bacterial inactivation. RSC Adv 2016;6:80086–98. [29] Shinde VM, Madras G. Nanostructured Pd modified Ni/CeO2 catalyst for water gas shift and catalytic hydrogen combustion reaction. Appl Catal B 2013;132:28–38. [30] Nagaveni K, Sivalingam G, Hegde M, Madras G. Solar photocatalytic degradation of dyes: high activity of combustion synthesized nano TiO2. Appl Catal B 2004;48:83–93. [31] Nandiyanto A, Hayati W, Aziz T, Ragadhita R, Abdullah A, Widiaty I. Engineering analysis and economic evaluation of the synthesis of composite CuO/ZnO/ZrO2 nanocatalyst. IOP conf ser mater sci eng. IOP Publishing; 2018. p. 012012. [32] Aruna ST. Solution combustion synthesis. Concise encyclopedia of self-propagating high-temperature synthesis. Elsevier; 2017. p. 344–6. [33] Wen W, Wu J-M. Nanomaterials via solution combustion synthesis: a step nearer to controllability. RSC Adv 2014;4:58090–100. [34] Rajeshwar K, De Tacconi NR. Solution combustion synthesis of oxide semiconductors for solar energy conversion and environmental remediation. Chem Soc Rev 2009;38:1984–98. [35] Lucilha AC, Afonso R, Silva PR, Lepre LF, Ando RA, Dall'Antonia LH. ZnO prepared by solution combustion synthesis: characterization and application as photoanode. J Braz Chem Soc 2014;25:1091–100. [36] González-Cortés SL, Imbert FE. Fundamentals, properties and applications of solid catalysts prepared by solution combustion synthesis (SCS). Appl Catal A 2013;452:117–31. [37] Hegde M, Madras G, Patil K. Noble metal ionic catalysts. Acc Chem Res 2009;42:704–12. [38] Ravishankar TN, Ramakrishnappa T, Nagaraju G, Rajanaika H. Synthesis and characterization of CeO2 nanoparticles via solution combustion method for photocatalytic and antibacterial activity studies. ChemistryOpen 2015;4:146–54. [39] Ali MA, Idris MR, Quayum ME. Fabrication of ZnO nanoparticles by solutioncombustion method for the photocatalytic degradation of organic dye. J Nanostruct Chem 2013;3:36. [40] Nagaveni K, Hegde M, Ravishankar N, Subbanna G, Madras G. Synthesis and structure of nanocrystalline TiO2 with lower band gap showing high photocatalytic activity. Langmuir 2004;20:2900–7. [41] Eswar NK, Ramamurthy PC, Madras G. High photoconductive combustion synthesized TiO2 derived nanobelts for photocatalytic water purification under solar irradiation. New J Chem 2015;39:6040–51. [42] Morales W, Cason M, Aina O, de Tacconi NR, Rajeshwar K. Combustion synthesis and characterization of nanocrystalline WO3. J Am Chem Soc 2008;130:6318–9. [43] Singh SA, Vemparala B, Madras G. Adsorption kinetics of dyes and their mixtures with Co3O4–ZrO2 composites. J Environ Chem Eng 2015;3:2684–96. [44] Rai AK, Gim J, Anh LT, Kim J. Partially reduced Co3O4/graphene nanocomposite as an anode material for secondary lithium ion battery. Electrochim Acta 2013;100:63–71. [45] Pfeil T, Pourpoint T, Groven L. Effects of crystallinity and morphology of solution combustion synthesized Co3O4 as a catalyst precursor in hydrolysis of sodium borohydride. Int J Hydrogen Energy 2014;39:2149–59. [46] Atabani A, Silitonga A, Ong H, Mahlia T, Masjuki H, Badruddin IA, et al. Nonedible vegetable oils: a critical evaluation of oil extraction, fatty acid compositions, biodiesel production, characteristics, engine performance and emissions production. Renew Sust Energ Rev 2013;18:211–45. [47] Vijay Kumar M, Veeresh Babu A, Ravi Kumar P. Experimental investigation on mahua methyl ester blended with diesel fuel in a compression ignition diesel engine. Int J Ambient Energy 2017:1–13. [48] Yoo SJ, Lee H-S, Veriansyah B, Kim J, Kim J-D, Lee Y-W. Synthesis of biodiesel from rapeseed oil using supercritical methanol with metal oxide catalysts. Bioresour Technol 2010;101:8686–9. [49] Mootabadi H, Salamatinia B, Bhatia S, Abdullah AZ. Ultrasonic-assisted biodiesel production process from palm oil using alkaline earth metal oxides as the heterogeneous catalysts. Fuel 2010;89:1818–25. [50] Narayan RC, Madras G. Esterification of sebacic acid in near-critical and supercritical methanol. Ind Eng Chem Res 2017;56:2641–9. [51] Lamba N, Modak JM, Madras G. Fatty acid methyl esters synthesis from non-edible vegetable oils using supercritical methanol and methyl tert-butyl ether. Energy Convers Manage 2017;138:77–83. [52] Yin J-Z, Ma Z, Hu D-P, Xiu Z-L, Wang T-H. Biodiesel production from subcritical methanol transesterification of soybean oil with sodium silicate. Energy Fuels 2010;24:3179–82. [53] Constantinou L, Gani R. New group contribution method for estimating properties of pure compounds. AIChE J 1994;40:1697–710. [54] Olivares-Carrillo P, Quesada-Medina J, de los Ríos AP, Hernández-Fernández FJ. Estimation of critical properties of reaction mixtures obtained in different reaction conditions during the synthesis of biodiesel with supercritical methanol from

424

Energy Conversion and Management 173 (2018) 412–425

N. Lamba et al.

[87] Wang L, Yang J. Transesterification of soybean oil with nano-MgO or not in supercritical and subcritical methanol. Fuel 2007;86:328–33. [88] Wang L, He H, Xie Z, Yang J, Zhu S. Transesterification of the crude oil of rapeseed with NaOH in supercritical and subcritical methanol. Fuel Process Technol 2007;88:477–81. [89] Medina-Valtierra J, Ramirez-Ortiz J. Biodiesel production from waste frying oil in sub-and supercritical methanol on a zeolite Y solid acid catalyst. Front Chem Sci Eng 2013;7:401–7. [90] Saez B, Santana A, Ramirez E, Maçaira J, Ledesma C, Llorca J, et al. Vegetable oil transesterification in supercritical conditions using co-solvent carbon dioxide over solid catalysts: a screening study. Energy Fuels 2014;28:6006–11. [91] Teo SH, Goto M, Taufiq-Yap YH. Biodiesel production from Jatropha curcas L. oil with Ca and La mixed oxide catalyst in near supercritical methanol conditions. J Supercrit Fluids 2015;104:243–50. [92] Laosiripojana N, Kiatkittipong W, Sutthisripok W, Assabumrungrat S. Synthesis of methyl esters from relevant palm products in near-critical methanol with modified-zirconia catalysts. Bioresour Technol 2010;101:8416–23. [93] Wang L, Tang Z, Xu W, Yang J. Catalytic transesterification of crude rapeseed oil by liquid organic amine and co-catalyst in supercritical methanol. Catal Commun 2007;8:1511–5. [94] Wan L, Liu H, Skala D. Biodiesel production from soybean oil in subcritical methanol using MnCO3/ZnO as catalyst. Appl Catal B 2014;152:352–9. [95] Yin J-Z, Ma Z, Shang Z-Y, Hu D-P, Xiu Z-L. Biodiesel production from soybean oil transesterification in subcritical methanol with K3PO4 as a catalyst. Fuel 2012;93:284–7. [96] Yin J-Z, Xiao M, Wang A-Q, Xiu Z-L. Synthesis of biodiesel from soybean oil by coupling catalysis with subcritical methanol. Energy Convers Manage 2008;49:3512–6. [97] Shin H-Y, An S-H, Sheikh R, Park YH, Bae S-Y. Transesterification of used

[98]

[99]

[100]

[101] [102]

[103] [104] [105]

[106] [107]

425

vegetable oils with a Cs-doped heteropolyacid catalyst in supercritical methanol. Fuel 2012;96:572–8. Chanchaochai P, Boonnoun P, Laosiripojana N, Goto M, Jongsomjit B, Panpranot J, et al. Transesterification of palm oil at near-critical conditions using sulfonated carbon-based acid catalyst. Chem Eng Commun 2013;200:1542–52. Singh SA, Madras G. Photocatalytic degradation with combustion synthesized WO3 and WO3TiO2 mixed oxides under UV and visible light. Sep Purif Technol 2013;105:79–89. Merino MG, Palermo M, Belda R, De Rapp MF, Lascalea G, Vázquez P. Combustion synthesis of Co3O4 nanoparticles: fuel ratio effect on the physical properties of the resulting powders. Procedia Materi Sci 2012;1:588–93. Li L, Yan B. CeO2–Bi2O3 nanocomposite: two step synthesis, microstructure and photocatalytic activity. J Non·Cryst Solids 2009;355:776–9. Reddy B, Mal I, Tewari S, Das K, Das S. Aqueous combustion synthesis and characterization of nanosized tetragonal zirconia single crystals. Metall Mater Trans A 2007;38:1786–93. Zhou K, Wang R, Xu B, Li Y. Synthesis, characterization and catalytic properties of CuO nanocrystals with various shapes. Nanotechnol 2006;17:3939. Mimani T, Patil K. Solution combustion synthesis of nanoscale oxides and their composites. Mater Phy Mech (Russia) 2001;4:134–7. Parviz D, Kazemeini M, Rashidi A, Jozani KJ. Synthesis and characterization of MoO3 nanostructures by solution combustion method employing morphology and size control. J Nanopart Res 2010;12:1509–21. Anil C, Madras G. Kinetics of CO oxidation over Cu doped Mn3O4. J Mol Catal A: Chem 2016;424:106–14. Balamurugan S, Ashna L, Parthiban P. Synthesis of nanocrystalline MgO particles by combustion followed by annealing method using hexamine as a fuel. Journal of Nanotechnol 2014;2014.