The Effect of Operating Parameters of Alkali Catalyzed Transesterication of Sunower Oil With Methanol in the Presence of the Cosolvent Assisted by Hydrodynamic Cavitation on the Degree of Triglyceride Conversion

Method of independent variation of the value of one operating parameter has been used to investigate the effect of operating parameters on alkali-catalyzed transesterification of sunflower oil with methanol in presence of tetrahydrofuran (THF) as cosolvent, assisted by hydrodynamic cavitation (ACTC) on the value of the degree of triglyceride conversion (DTC). ACTC was performed by a venturi-type hydrodynamic cavitation reactor (VCR) of our construction. To determine the effect of ACTC on DTC following operating parameters were varied: reaction mixture inlet pressure (p 1 ) from 202.3 kPa to 1013.2 kPa; methanol to oil molar ratio (M 1 ) from M 1 =3 to M 1 =12; concentration of catalyst (C c ) from 0.3wt% to 1.5wt%; methanol to THF molar ratio (M 2 ) from M 2 =1.25 to M 2 =2.0; temperature (T) from 20°C to 55°C, number of passes through the VCR ( n ) from n =1 to n =10. It was found, based on the obtained results, that: a) the values of DTC increase with the increase in p 1 , M 1 , C c, and n , b) the values of the DTC decrease with the increase in T and c) maximum values of the DTC are obtained at C c =1.0~1.1wt% and M 2 =1.5.


Introduction
Biodiesel is an environmentally friendly alternative fuel, as it is renewable, biodegradable, non-toxic, and has substantially reduced sulfur oxides and carbon monoxide emissions. Most typically it is produced by alkaline or acid-catalyzed transesterification reaction of various vegetable oils with short-chain alcohols, mainly methyl alcohol [1]. The reaction rate in transesterification reactions is limited by mass transfer because reaction takes place in a liquid-liquid two-phase system since oils and methyl alcohol are immiscible. Therefore, achieving a higher reaction rate by increasing the efficiency of the oil, methanol, and catalyst contact is crucial to perform the reaction in the reactor of economic size. Regarding this problem, several technologies of process intensification can be applied for increasing contact of two liquid phases and mass transfer rate, consequently [2].
Alkaline catalysis and acid catalysis are conventional methods for biodiesel production.
Since transesterification is an equilibrium reaction, alcohol is used in excess to shift the reaction towards the formation of fatty acid methyl esters (FAME) [1]. Alkaline catalyzed process, the most commonly worldwide used industrial process, is strongly affected by mixing and stirring, both of which increase the contact area between the reactants. Having limitations of classical procedures in mind, there has been a strong urge to evolve a novel, time and cost-saving, efficient, and environmentally friendly biodiesel process of production, superior to now existing classical procedures [3].
Considering whether the catalyst remains or not in the same, liquid, phase as the reactants, the catalysts can be either homogenous or heterogeneous. Alkaline homogenous catalyzed process: a) operates on low temperatures and pressures; b) gives higher levels of triglyceride conversion in shorter periods, 60 minutes mostly reported; c) has higher catalytic activity at low catalyst concentrations: d) is easily available and cheap. The downsides of the process are: a) requires high purity oil; b) sensitivity to higher free fatty acid and water content; c) catalyst cannot be reused and d) costly separation of the reaction products. Acidic homogenous catalyzed process: a) uses low-cost oils; b) alcohol is used as a solvent and esterification reagent, leading to separation and transesterification in one step; c) has a lower sensibility to free fatty acid content in the oils. On the other hand, process: a) is more sensitive to the content of water in the oil; b) there is significant equipment corrosion; c) produces more waste and unintended byproducts; d) requires higher temperatures; e) requires longer reaction times; f) has weaker catalytic activity. Heterogeneous catalysis has several upsides: a) catalyst separation and reuse are easy; b) lower cost of catalyst; c) process has a lower impact on the environment; d) no soap formation. The downside of heterogeneous catalysis is that process requires high temperatures, up to 200°C, making the process considerably expensive [4].
One way to intensify transesterification reaction is enhancing the miscibility of oil and methanol, thus turning the reaction system from two-phase to one-phase, which can be done by adding a cosolvent. This way, the reaction system doesn`t need additional stirring, and separation of glycerin phase, after the completion of transesterification, is several times faster.
Various cosolvents have been reported: different light ethers, acetone, hexane, heptane, CO2, and several ionic liquids. The most frequently used cosolvent is tetrahydrofuran (THF), due to its low price, non-toxicity, and non-reactivity [2,5,6,7]. Boocock et al. [8] performed alkali catalyzed transesterification of soybean oil by methanol and with THF as cosolvent, at ambient temperature, at M1=6 and M2=1,25. They found that with THF as cosolvent, the system shifted to one phase, thus shortening reaction time significantly. Banković-Ilić et al. [9] found out that although the addition of THF as cosolvent has positive influences on hydrodynamic properties of the reaction mixture, the addition of cosolvent in excess leads to change in reaction kinetics from models with mass transfer limitations towards irreversible and/or reversible second-order kinetics. Mao et al. [5], investigating transesterification of soybean oil by methanol, with cosolvent THF, at ambient temperatures, M1=6, with NaOH, KOH, and MeO-Na + as catalysts, found out that the reaction rate is very fast at the beginning, while the system is one-phase, but slows down rapidly after formation of the two-phase system due to glycerin formation. Ataya et al. [6] examined canola oil transesterification using acid or base catalyst in a two-phase system and one-phase system with THF as cosolvent. The authors stated that a one-phase reaction has first-order chemical reaction kinetics and that in a one-phase reaction medium there is no mass transfer resistance. Roosta and Sabzpooshan [10] used mathematical modeling to predict the use of cosolvent and confirmed that adding cosolvent does not affect the final yield of FAME, only speeds up the reaction. Encinar et al. [7] examined the effect of various cosolvents on rapeseed oil alkali catalyzed transesterification, also varying M1 and methanol to cosolvent ratios, catalyst concentrations, temperature, and agitation speed, observing high FAME yields in short reaction times for all the different system setups. The authors also concluded that the kinetics of the reaction can be described with the pseudo-first-order kinetic model.
Intensification in alkali catalyzed transesterification of oil can be also obtained by alternating energy sources for the reaction process. High gravity [11,12], ultrasound [13,14,15], hydrodynamic cavitation [3, 13, 16~25], and microwaves [26~31] were investigated, both in a laboratory and commercial scale. Beljić Durković et al. [26] examined comparative kinetics of the alkali catalyzed sunflower oil methanolysis with THF as cosolvent under conventional and microwave heating. Authors concluded that microwave heating does not affect the kinetic model of alkaline catalyzed sunflower oil methanolysis with THF as cosolvent, the isothermal value of the rate of the transesterification under microwave heating is 2.5-3.5 larger than the rate of the transesterification under conventional heating, and that microwave heating causes a significant increase in the value of the preexponential factor.
Alkali catalyzed transesterification of oil with methanol by venturi-type hydrodynamic cavitation reactor was investigated by several researchers. Maddikeri et al. [22] examined biodiesel synthesis by interesterification of waste cooking oil intensified by the plate with orifices and two different types of venturi cavitation reactors by varying reaction mixture inlet pressure, the molar ratio of oil to methyl acetate, and concentration of catalyst. Ladino et al.
[32] did a numerical study of geometrical properties of venturi cavitation reactors used for biodiesel production to find the best possible venturi configuration. Chitsaz et al. [24] used response surface methodology to optimize biodiesel production from sunflower waste frying oil by VCR. Bargole et al. [33] studied the intensification of biodiesel synthesis from waste cooking oil by use of circular venturi and several different plates with orifices. Simpson et al. [34] developed several computational fluid dynamic models to simulate various venturi-type cavitation flows to provide data to be used for designing and optimizing venturi-type hydrodynamic cavitation reactors. In their review on advances and perspectives in controlled hydrodynamic cavitation Panda et al. [35] compared venturi and plate of orifices cavitation reactors and suggested that VCRs produce more stable cavitation, more cavities in number and size thus leading to better cavitation yields.
As far as the best knowledge of the authors, there is no literature data about the effect of the operating parameters of the hydrodynamic cavitation assisted alkaline catalyzed transesterification of refined sunflower oil with methanol, in presence of cosolvent, on the degree of triglyceride conversion. Considering that, the main goal of here presented work was the determination of the effects of operating parameters: reaction mixture inlet pressure, methanol to oil molar ratio, the concentration of catalyst, temperature, methanol to THF molar ratio, and the number of passes through the VCR on the degree of triglyceride conversion, in ACTC conditions, performed by here presented VCR of own construction and make.

Hydrodynamic cavitation-assisted alkaline catalyzed transesterification (ACTC)
Alkaline catalyzed transesterification of sunflower oil in presence of cosolvent was performed by the VCR of our construction. The schematic diagram of hydrodynamic cavitation equipment and the VCR geometric construction are respectively shown in Figure 1   Geometric characteristics of VCR are given in Table 2: Length of the convergent section 15mm Length of the divergent section 65mm Half angle of the convergent section 23.2° Half angle of the divergent section 6.4° The experimental procedure of transesterification was performed as follows: refined sunflower oil (300 mL) was dosed into the storage tank and the predetermined mass of cosolvent THF was added to achieve the required methanol to THF molar ratio (1.25 -2.0). A mixture of oil and THF was heated to previously designated T (25°C -50°C). The previously calculated mass of catalyst KOH was dissolved in a predetermined mass of methanol to achieve a certain value of methanol to oil ratio (3 -12) and concentration of catalyst (0.3 -1.5wt%).
The resulting solution was heated to a predetermined temperature, as same as for oil and THF mixture, and dosed into the storage tank. The one-phase reaction mixture thus formed was loaded into the VCR by an electrical pump at the predetermined reaction mixture inlet pressure (202.3 kPa -1013.2 kPa) for a determined time, to achieve a previously determined number of passes through the VCR (1 to 10). Since hydrodynamic cavitation increases the temperature of the reaction mixture, the temperature was maintained by the external cooling of the storage tank and regularly measured by the thermometer.
During the ACTC, predefined samples were taken from the reaction mixture, to be analyzed, at the sampling valve (SV). To stop the reaction the 1wt% acetic acid (corresponding to the amount of KOH used) was added until the neutral pH was reached. The mixture was transferred into a separating funnel and left for two hours to allow the gravitational separation of glycerin. Upon removal of glycerin from the funnel, FAME was washed for 15 minutes with warm redistilled water heated to 40 °C (volume ratio 1:1 to FAME), at 600 rpm. Then, the liquid was poured into a second separation funnel and left for several minutes to separate the water from FAME. Upon the separation and removal of the water layer, traces of residual water and residual methanol were removed by heating FAME at 105 °C.

Determination of concentration of methyl esters in transesterification product
The determination of methyl esters concentration was performed by a method of gas chromatography following EN 14103:2003 [51]. The method of determining concentrations is thoroughly described in the work of Beljić Durković et al. [26].
The methyl ester concentration ( ) given in percentage (wt%) was calculated using the Eq. 1: ΣA being the total peak area from the methyl ester C14 to that in C24:1, Aei the peak area corresponding to the methyl heptadecanoate, Aer the peak corresponding to the methyl heptadecanoate of the referent sample, Cei the concentration of the methyl heptadecanoate solution being used, Vei the volume of the solution of methyl heptadecanoate used, and m the weight of the sample.

Determination of the converted triglycerides weight
Determination of the weight of the converted triglycerides (Wtg), was calculated by the Eq.2: Mtg is the molar mass of triglycerides, Wi is the initial weight of triglycerides in the reaction mixture, Cme is the concentration of methyl esters, and Mme is the molar mass of methyl esters.

Determination of the degree of triglyceride conversion
The DTC was calculated by the following Eq. 3:   A similar effect of the reaction mixture inlet pressure on the yield of biodiesel was showed by Maddikeri et al. [22] in the study of alkali catalyzed interesterification of waste cooking oil with methyl-acetate using the VCR, where it was shown that an increase in inlet pressure from 200 kPa to 300 kPa, increases the biodiesel yield up to 89%. Also, Ghayal et al.
[ 16] in a study of the alkali catalyzed transesterification of waste frying oil assisted by hydrodynamic cavitation showed that the reaction mixture inlet pressure of 300 kPa was required to achieve the 94% degree of the conversion of triglycerides to methyl esters.
It was established that an increase in the value of DTC with reaction mixture inlet pressure is caused by the increase in turbulent flow velocity of the reaction mixture through the venturi-type hydrodynamic cavitation reactor. This leads to the decrease in hydrodynamic cavitation reactor`s cavitation number and increases in the reactor`s cavitation effectivity due to an increase in the number of the bubbles that collapsed; the amplitudes of pressure and temperature impulses; the energy of collapsing bubble; the generated shock wave intensity. As a consequence, the values of DTC increase. The experimentally obtained results in this work, that transesterification does not occur at p1≤304.0 kPa under given conditions, implies that ACTC reaction does not happen inside the core of the cavity or at the interphase region between the cavity and bulk reaction mixture, which explains why there are no small intermediary molecules or radicals in the reaction products. Most likely, an extremely high rate of ACTC reaction is a consequence of a significant increase in reactant mass transfer which, in response, significantly accelerates the reaction.

The effect of methanol to oil molar ratio on the degree of triglyceride conversion
Eckey and Am [37], Sridharan and Mathai [38], proposed a mechanism of conventional alkaline catalyzed transesterification of oil with various types of alcohols. Following that mechanism, the transesterification of oil with alcohol is a complex reaction that consists of three elemental consecutive reverse reactions. Due to this, to achieve higher conversion and FAME yield, conventional transesterification is performed with methanol to oil molar ratio M1 much higher than the stoichiometric molar ratio of methanol and oil M1=3.  Figure   3.

The effect of catalyst concentration on the degree of triglyceride conversion
Catalyst concentration in the reaction mixture, as well as p1 and M1, has a significant impact on transesterification rate, the DTC, and FAME yield.     The obtained results reveal that with the increase of temperature within the investigated range, the value of DTC shows a clear linear decrease from DTC=95 % at T=20 °C to DTC=88 % at T=55 °C. The dependence can be mathematically described by Exp. 1: The influence of temperature on the rate and yield of a chemical reaction that takes place under conditions of hydrodynamic cavitation is complex and it is defined by the simultaneous influence of temperature on the rate of a chemical reaction and on physicochemical properties of the reaction mixture: vapor pressure (pv); coefficient of viscosity (ή) and coefficient of surface tension (γ), which affect the dynamics of forming, growth and implosion of the   The degree of triglyceride conversion increases linearly with an increase in the value of n, from DTC=94%, at n=1(5s) to DTC=99%, at n=10(1min). The dependence can be mathematically described by Exp 2: The increase of DTC with an increase of n is the consequence of the increase in the number of collapsing bubbles with the increase in the number of reaction mixture's passes through the cavitation device.

Cavitation yield
The application of hydrodynamic cavitation leads to a significant intensification of the chemical reaction due to: reaction time reduction, reaction yield increase, use of less forcing conditions (T, p), changes in reaction pathways resulting in increased selectivity [22]. To quantify the extent of intensification of the examined reaction/process due to the application of hydrodynamic cavitation relative to other procedure performances, a new quantity is introduced: cavitation yield (CY) [16]. Cavitation yield is defined as the yield of the product per unit of energy supplied to the system. Table 3. presents the change of cavitation yield with changes in the values of the DTC and the number of passes through the VCR.  [16] and waste cooking oil [22] with methanol assisted by hydrodynamic cavitation (1.28-1.22·10 -3 g/J) which indicates that ACTC represents an energy-efficient process for biodiesel production.

Conclusions
The

Author contributions
The manuscript was written through the contributions of all authors. All authors have approved the final version of the manuscript.

Funding sources
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

ABBREVIATIONS
ACTCalkaline catalyzed transesterification of refined sunflower oil with methanol in presence of tetrahydrofuran as cosolvent, assisted by hydrodynamic cavitation DTCthe degree of triglyceride conversion FAMEfatty acid methyl esters CYcavitation yield VCRventuri type hydrodynamic cavitation reactor