Intensification of Biodiesel Synthesis Reactor Using Biphasic

5 days ago - Watson shifts from drug discovery to the clinic. Responding to reports that it will no longer sell its Watson artificial intelligence sys...
2 downloads 0 Views 1MB Size
Download PDF
Research Note pubs.acs.org/IECR

Cite This: Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Intensification of Biodiesel Synthesis Reactor Using Biphasic Homogenous Catalytic Reaction: Parametric Study† Nur Adiba Mohd Noor,‡ Said Nurdin,‡ Zahira Yaakob,§ and Mohd Sabri Mahmud*,‡ ‡

Downloaded via UNIV PARIS-SUD on May 15, 2019 at 20:16:18 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

Faculty of Chemical and Natural Resources Engineering, Universiti Malaysia Pahang, Tun Razak Highway, Gambang, 26300 Kuantan, Pahang, Malaysia § Department of Chemical Engineering, Faculty of Engineering and Built Environment, Universiti Kebangsaan Malaysia, 43600 Bangi, Selangor, Malaysia ABSTRACT: Parameters of extraction and reaction of biodiesel synthesis under partially immiscible liquid−liquid phase conditions were studied in a laboratoryscale, multiple stage, stirred, counter-current tubular reactor called a multistage stirred column (MSC). Ethanol was used as the excess reactant with refinedbleached-deodorized palm oil (RBDPO) feedstock and simultaneously acted as an extractant to biodiesel and glycerol in transesterification reaction catalyzed by 1.5 wt % potassium ethoxide. The molar ratios of RBDPO to ethanol of 1:6 and 1:12 obtained from their feed flows appeared as two liquid phases and the reactor was agitated at 200 and 300 rpm. Initial study of liquid mixing revealed nonlinear equilibria of ethyl oleate concentration with dynamic distribution coefficient between the two organic liquid phases. The batch reactor study optimized stirring speed and was set as a benchmarking. MSC study showed better efficiency, having a Damköhler number (Da) greater than unity (27 ≤ Da ≤ 1688.9).

■

INTRODUCTION

because of the content of saturated fatty acids, where 20 wt % ethanol in palm oil can form two liquid phases at 45 °C.4 Intensification of the biodiesel synthesis reaction with separation targets low product or high reactant concentrations in the reactive medium and thus attainment of high rate and yield of reaction. For the case of liquid−liquid system, impurities of reactant can be separated from the products as well.5 The intensification study is normally performed in either a liquid−liquid system or a gas−liquid system. Nevertheless, the latter is not suitable for biodiesel, because the reaction temperature windows is not at separation temperature. The boiling point of methanol or ethanol is far lower than that of fatty acid acyl esters, while the reaction occurs in the liquid phase that usually contains a catalyst. An extractive reaction using a counter-current-flow column, such as the Kuhni column or the Oldshue−Rushton column, requires high catalyst activity so that the reaction can only be limited by mass transfer and the Damköhler number (Da) is high regardless of the time of mixing, as long as it is longer than the time of reaction. Using immobilized lipase is hindered by deactivation and leaching of high concentration of ethanol and limits the reaction to the use of diluted alcohol. The highest Da value, using an optimum hydrous ethanol of 20 vol %, is 0.25.6 This study attempted to oversee the use of a liquid−liquid system in transesterification reaction using pure reactants, with

Biodiesel, which is an alternative fuel to diesel from renewable resources, is important for the sustainability of energy for automotive and power generators. In particular, commercial biodiesel is widely produced from methanol derived from petroleum downstream. Ethanol also is of interest, because it can be produced from the fermentation of sugar-contained feedstock and is an engine-friendly fuel, but it creates many process issues in biodiesel production and is not economical. Biodiesel synthesis is a reversible reaction and involved with the dynamic phase of liquids. The alcohols, reactants or glycerol, are partially immiscible in vegetable oil. The formation of a second liquid phase leads to high equilibrium conversion if the actual concentration of reacting alcohols is at least stoichiometric in the oily phase. This was clearly evidenced in biphasic reaction of esterification between oleic acid and ethanol, where the free fatty acid is insoluble in water.1,2 The effect of the immiscibility of reactants and products to the equilibrium conversion can be clearly seen if the aliphatic chain of alcohol reactant is getting shorter. In the case of ethanol, two liquid phase equilibrium will appear at 20 wt % ethanol at ambient temperature and this composition will be higher if ester is present in the mixture and/or at higher temperature. Triolein-ethyl oleate-ethanol equilibrium will secure this biphasic condition after 70 wt % ethanol and any consumption of ethanol in the transesterification reaction will not be affected much since the amount of ethanol is very much in excess.3 In palm oil, triolein represents only 35 wt % of the overall mass, while tripalmitin dominates the composition. Thus, two-phase equilibrium would be relatively different, © XXXX American Chemical Society

Received: Revised: Accepted: Published: A

February 3, 2019 May 5, 2019 May 13, 2019 May 13, 2019 DOI: 10.1021/acs.iecr.9b00693 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Research Note

Industrial & Engineering Chemistry Research

normalize the area and concentration of the samples or standards in the analysis, respectively. Liquid−Liquid Equilibrium. Alcohol-enriched phase would extract more polar species from oil-enriched phase. Hence, two-liquid phase condition must be secured in order to create an extractive system. Mixtures of FAEEs, ethanol, palm oil and glycerol simulating possible equilibria inside the multistage stirred column (MSC) were prepared under the temperature of the reaction, 60 °C and various conversions to delineate the distribution of FAEE and ethanol. The analytes were mixed in 50-mL conical flasks by using magnetic bar on a hot plate stirrer for 20 min. RBDPO:ethanol (molar) ratios that were examined included 1:3, 1:6, 1:9, 1:12, and 1:24, where five samples of different mass percentages of fatty ethyl esters were prepared for each ratio. The times elapsed to form stable two layers of liquid were recorded while the sample in an oven was set at 60 °C. Batch Reactor Study. The transesterification reaction was performed in a 500-mL three-neck-round-bottom glass flask fitted with a reflux condenser, thermometer probe, and a sampling tubing. The reactor system was heated by using a digital hot plate stirrer under a temperature controller. Various stirring speeds were employed to ensure the reaction occurred under free mass-transfer limitations. 1.5 wt % KOH was used in the reaction substrate. MSC Study. The MSC setup having 9 stainless-steel stages and 2 clear-polyvinyl-chloride decanters was fabricated similar to the design of Mahmud’s doctorate work5 and was also reported by Chesterfield et al.6 Nevertheless, as shown in Figure 1, a settler was added after the top decanter to further separate the liquids at room temperature. Besides heating of silicone-oil preheater to both feeds of oil and ethanol, heating tape of a temperature controller and insulation was also installed to secure the desired temperature of the column. The temperature of the stages/column was maintained at 60 °C. As shown in Figure 1, the stirrer rod was attached with 6-bladed impeller, 2 cm disk diameter welded perpendicularly with equally spaced 3 mm × 4 mm blades arranged around the center. The impeller was located at the middle of each stage. Dual flow plates made of stainless steel with 42% opening5 area were clamped between two stages and between the last stage and decanters to allow liquid flow in both directions. Peristaltic pumps (Milton Roy Electronic Metering, PSeries) were employed to deliver liquids to and from the decanters. Calibration of the rotameter, stroke, and speed of pump on the actual flow of liquid at each outlet was performed to avoid slight variations in the hydrodynamic property of the liquids, because of different materials. The MSC was washed twice with ethanol to remove dirt. The initial liquid drop study without agitation using a graduated cylinder, graph paper, and a 5-megapixel mobile phone camera revealed that the speed of RBDPO drop was 73 cm s−1 in the ethanol phase, which is 100 times higher than the rising ethanol droplet in the RBDPO phase. Although the flow rate of ethanol (solvent) feed is slightly higher than that of oil, flooding of opposite liquid phase did not happen. Droplet size of both liquid was identical under various stirring conditions. Therefore, ethanol was made to be a continuous liquid phase filling up the column until stage 9 (counting from the bottom), as performed by Chesterfield et al.6 and original work by the inventor for this liquid−liquid technique,8 after being mixed with potassium hydroxide at 1.5 wt %, forming ethoxide and

ethanol as the coextractant, and a commercial catalyst of potassium hydroxide, as well as the highest rate possible for the biodiesel synthesis in an extractive reactor column. Although the use of absolute ethanol and refined-bleached-deodorized palm oil (RBDPO) seems unrealistic, because of their cost, the Da value is usually determined to delineate the performance of the ideal reactor7 before pursuing to the use of technical grade ethanol and waste cooking oil or crude palm oil. The crude biodiesel was synthesized using in-site liquid extraction from counter-current flow of feed and the coextractant as per Olshue-Rushton column design.8 Preliminary works of liquid− liquid equilibrium and batch reaction were carried out to determine the condition of initial startup and benchmarking, respectively.

■

EXPERIMENTAL SECTION Materials. Palm oil, which is a refined-bleached-deodorized grade and branded Saji, was obtained from local food stores. Absolute ethanol (100% pure) supplied by EMSURE was used as extractant and reactant. The viscosity of RBDPO and ethanol measured by using a Brookfield viscometer equipped with cylindrical spindles was 0.096 and 1.875 × 10−3 Pa s, respectively. Potassium hydroxide (KOH), which was used as the homogeneous catalyst, was obtained from R&M Chemicals. Heptane was supplied by Riendemann CHMIDT used as solvent for preparing GC samples. Analytical-grade glycerol, triolein, and ethyl-oleate standards were procured from Sigma−Aldrich to calibrate the components in the chromatographic analysis. Helium, air, and nitrogen that were used for the gas chromatography (GC) analysis were supplied by Linde. Fatty acid ethyl esters (FAEE) used in liquid−liquid equilibrium study was prepared by reacting RBDPO with excess ethanol under 1 wt % KOH catalyst for 2 h. The product was washed with ultrapure deionized water and was separated using separating funnels several times until the organic phase became clear. No peak of glyceride in the final product of FAEE was eventually detectable in the GC analysis checking. Analysis Method. A flame ionization detection (FID) GC system (Agilent 7693 (GC)) that was equipped with an autosampler and an autoinjector was employed to measure the concentration of reaction species, either reactant or products. The stainless steel column MXT-Biodiesel TG 15 m with 0.32 mm ID and 0.1 μm df, and a guard column with 2 and 0.53 mm inner diameter (ID) from Restek were employed in the analysis of triolein and ethyl oleate. The analysis of RBDPO revealed that triolein represented ∼34% of the oil. DB5-HighTemperature column from Agilent was also used to measure the concentration of alcohol in the same GC. The methods of analysis were similar to those reported by Mahmud5 and Chesterfield et al.6 Calibration curve was developed based on the external standard method, where the area of a particular component in a prepared standard sample were measured at several concentrations. At the same time, ethyl oleate (EO) solution in separate vials with known concentration, from a bulk quantity preliminarily prepared for all runs, was also used as another external standard of analysis to monitor the sensitivity of the GC throughout all runs including the calibration.9,10 The EO-standard vials were located at the beginning, middle, and end of the series of samples or standards for a particular run in the autosampler tray. The result of area and concentration from the EO standard will B

DOI: 10.1021/acs.iecr.9b00693 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Research Note

Industrial & Engineering Chemistry Research

Figure 2. Mass fraction of ethyl oleate in saturated-ethanol phase (yEO) and in saturated-RBDPO phase (xEO).

Kd =

RBDPO as a dispersed phase. The time of reaction was started as RBDPO was fed into the top decanter.

(2)

FEO0 − FEO + rEOV =

■

RESULTS AND DISCUSSION Liquid−Liquid Equilibrium. The samples with the RBDPO:ethanol molar ratio of 1:3 appeared as a single liquid phase while most of the other samples of ratios formed two layers for various conversions: upper layer of saturated-ethanol phase and lower layer of saturated-RBDPO phase. Nevertheless, the data were not yet sufficient to draw a complete ternary diagram. Figure 2 is a McCabe−Thiele diagram that shows equilibrium line of mass fraction of ethyl oleate in saturated-ethanol phase (yEO), as a function of ethyl oleate mass fraction in saturated-RBDPO phase (xEO) at 60 °C for the samples with molar ratios of 1:6, 1:9, 1:12, and 1:24. The data are best fitted (R2 = 98.4%) with a rational expression, as shown in eq 1, (0.0029 + 3.8177x EO) (1 + 7.8267x EO)(1 + 6.6281x EO)

x EO

decreases as xEO increases until yEO = 0.138 and Kd goes to zero due to infinite xEO, indicating no more extraction and the liquid existed as a single liquid. This result justifies the appearance of single-phase liquid in the mixture for conversions of >0.48 for the 1:3 ratio and >0.85 for the 1:6 ratio. Higher ethanol content secured two-liquid equilibria at all conversions and Kd to be higher than unity. The equilibrium line on the McCabe−Thiele diagram can be a reference to get an extractive environment by having an operating line of mass ratio between coextractant ethanol and RBPDO below the equilibrium line. All initial ratios are below the equilibrium line. Nevertheless, unlike a diluted, immiscible liquid−liquid phase, the operating line for a variableconcentration, partially immiscible liquid phase, reactive system can probably be computed from the mole balance of a flow reactor,11 as defined by eq 3:

Figure 1. Schematic diagram of the column using the same design similar to Mahmud5 and Chesterfield et al.6 Additional parts are shown. Adapted in part with permission from corresponding author for both works [Mahmud5 and Chesterfield et al.6]. Reproduced with permission from the work of Mahmud5 (Copyright 2010, The University of New South Wales, Sydney, Australia) and Chesterfield et al.6 (Copyright 2013, American Chemical Society, Washington, DC).

yEO =

yEO

dNEO dt

(3)

where FEO0 is the molar flow rate of ethyl oleate at the inlet of a reactor, FEO the molar flow rate of ethyl oleate at the outlet of a reactor, rEO the rate of ethyl oleate formation, V the volume of the reactor, and dNEO/dt the change in the number of moles of ethyl oleate. At the initial ratio, for fresh RBDPO feed without ethyl ester at steady state enters the reactor at a rate of FEO0 = dNEO/dt = 0. Hence, FEO =

Rx EO1 + EyEO9 MEO

= rEOV

(4)

where MEO is the molecular mass of ethyl oleate, xEO9 the ethyl oleate mass fraction of raffinate flows, and yEO0 the ethyl oleate mass fraction of extract flows. Hence, ethyl oleate is only a function of reaction rate law and the volume of the reactor. Batch Reactor Study. RBDPO contains triglycerides of more than 100 fatty acids.12 Stepwise transesterification of them with ethanol can be expressed as described by eq 5.13

(1)

by nonlinear regression using the Sigma Plot 10 software, is a curve. From the equation, yE0 = xEO is 0.133 and has an asymptote at 0.138. The distribution coefficient (Kd), as defined in eq 2,

TG + EtOH F DG + FAEE + EtOH F MG + 2FAEE + EtOH F G + 3FAEE C

(5)

DOI: 10.1021/acs.iecr.9b00693 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Research Note

Industrial & Engineering Chemistry Research Triglycerides will first react with ethanol to form FAEE and diglyceride. The next series reaction will similarly react ethanol and diglyceride to form the second FAEE and monoglyceride. In the third step, glycerol will form besides FAEE. However, this study was not intended to delineate the reaction rate of all glycerides, assuming that they would collectively behave in the same pattern and triolein was preferred due to its second dominance in fatty acid contents and ease of analysis handling using GC. Focus was given to the first reaction as the limiting step for the transesterification, as reported by ref 13, and triolein was preferred, because it was the limiting reactant in the reaction tests. Two RBDPO:ethanol molar ratios were tested, i.e., 1:6 and 1:12. Figure 3 shows the concentration of triolein for 1:6 for

Table 1. Rate Constants of Transesterification Using a Batch Reactor Rate Constant

seven stirring speeds in the transesterification using 1.5 wt % KOH catalyst in a batch reactor at 60 °C. From the plot, the fitted trend line indicates the equilibria were achieved at almost zero reactant (∼100% conversion). Thus, the first step of the transesterification reaction could be considered to be irreversible, as expressed by using exponential decay model. The reaction data in Figure 3 was fitted (correlation coefficient of R2 > 95%) by using the exponential decay model, as expressed in eq 6: or

C TO0 exp( −kt ) + C TO1 (6)

where CTO is the concentration, CTO0 the initial concentration of triolein, CTO1 the virtual initial concentration of triolein, and k a rate constant. Table 1 lists the value of rate constant for both molar ratios. Some stirring speeds are not in the plot of Figure 3, for the sake of clarity. The rate law of the reaction can be subsequently derived to determine the rate via differentiation as defined by eq 7: −

dC TO = k[C TO0 exp( −kt )] dt

(7)

Transforming the differential equation to a power law equation from the mole balance of the batch reactor, −rTO = kC TO

RBDPO:ethanol molar ratio = 1:6

RBDPO:ethanol molar ratio = 1:12

400 450 500 550 600 650 700 750 800 900

0.036 0.166 0.1896 0.4831 0.6442 1.201 1.9976 2.12 2.18 2.2

2.01 2.22 2.21 2.23

biodiesel synthesis using excess methanol. However, the rate constant changed with mixing speeds, indicating that the reactions were still running under mass-transfer hindrance, although the reactants and catalyst dissolved homogeneously to form a single liquid phase. The optimum stirring speed extrapolated from the data of the ratios 1:6 and 1:12 in Table 1 are 900 and 500 rpm, respectively. This phenomenon was comparable with the mass transfer of droplet (without stirring) of saponification reaction at different droplet size run by Pawelski et al.15 When the molar ratio of RBDPO-ethanol was 1:12, the rate constants were maintained at 2.2/min, starting at 500 rpm, implying that the mixing was in the steady-state/ turbulence regime and mass transfer was not limiting the reaction anymore. The viscosity of the mixture was 0.0185 Pa s, which is three times lower than that observed at a ratio of 1:6. The result indicates that a high excess amount of ethanol in creating two-liquid phase system eliminated the mass-transfer effect at low stirring speed. The data were useful to determine the efficiency of the MSC result at various stirring speeds. Multistage Stirred Column. The flows of ethanol and RBDPO are 10 mL min−1 and 26.72 mL min−1, and 15 mL min−1 and 20.04 mL min−1, respectively, which were equivalent to RBDPO:ethanol molar ratios of 1:6 or 1:12, were pumped to the top decanter and bottom decanter, respectively, after MSC was fully filled with ethanol. The extract and raffinate were drained and pumped out accordingly by maintaining the level of liquid interface at each decanter, since both flows created liquid−liquid phase condition. The reaction tests were run under 200 and 300 rpm stirring speeds to see the effect of low agitation and to avoid stable emulsion hold-up6 since the beginning. The steady state was achieved as indicated by constant torque displayed at the stirrer motor after ∼158 and 165 min each. The molar ratios of reactants were ensured by checking their liquid holdup using graduated syringes. The samples from each stage was sucked and injected by using the same syringe back and forth several times to ensure that homogeneous samples were taken before being diluted with heptane for the chromatographic analysis. Figure 4 shows the concentration profile of triolein in the MSC stages under 200 rpm after 5 min RBDPO being fed to the column−unsteady-state condition. Initial fluctuation that occurred from stage 1 to stage 5 was probably due to the dynamic equilibrium of triolein between two liquid phases throughout the column, which is symptomatic to two interacting stirred tanks.6 The concentration of triolein

Figure 3. Concentration of triolein versus reaction time for an RBDPO:ethanol molar ratio of 1:6.

C TO = C TO0 exp( −kt )

stirring speed (rpm)

(8)

The reaction followed a pseudo-first order equation, as similarly reported by Sivasamy et al.13 and Zheng et al.14 in D

DOI: 10.1021/acs.iecr.9b00693 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Research Note

Industrial & Engineering Chemistry Research Notes †

This paper was originally intended to be part of the I&ECR special issue regarding the 2018 International Conference of Chemical Engineering & Industrial Biotechnology, published earlier this year (Ind. Eng. Chem. Res. 2019, Vol. 58, Issue 2). The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The author would like to express gratitude to Ministry of Higher Education, Malaysia and Universiti Malaysia Pahang for funding this study, under Grant Nos. RDU120111 and RDU120395, and for all facilities provided at the lab of Faculty of Chemical and Natural Resources Engineering.

Figure 4. Triolein concentration profile in MSC for an RBDPO:ethanol ratio of 1:6 at 200 rpm.

■

ABBREVIATIONS EO = ethyl oleate FAEE = fatty acid ethyl ester GC = gas chromatography MSC = multistage stirred column RBDPO = refined bleached deodorized palm oil

decreased drastically from stage 6 to stage 9, although getting closer to the feed of RBDPO. This phenomenon conforms to the McCabe−Thiele equilibrium line in Figure 2, where ethyl oleate (representing ethyl esters) was dissolved into an ethanol-enriched phase at higher content. After 3 h, triolein concentration was not detected in all stages, including raffinate and extract. This also occurred at the higher stirring speeds and at the ratio of 1:12 in all stirring speeds. The reaction was too fast, probably within 5 min, to convert all triolein even though the agitation was below optimum stirring speed determined in the batch reactor study. As the reaction followed the first order, the performance of the reactor in terms of Da, can be expressed as eq 9:11 Da =

−rTO0V kC TO0V = = τk FTO0 ν0C TO0

■

(1) Mahmud, M. S.; Safinski, T.; Nelson, M. I.; Sidhu, H. S.; Adesina, A. A. Kinetic Analysis of Oleic Acid Esterification Using Lipase as Catalyst in a Microaqueous Environment. Ind. Eng. Chem. Res. 2010, 49, 1071. (2) Foresti, M. L.; Pedernera, M.; Ferreira, M. L.; Bucala, V. Kinetic modeling of enzymatic ethyl oleate synthesis carried out in biphasic systems. Appl. Catal., A 2008, 334, 65. (3) Dussan, K. J.; Cardona, C. A.; Giraldo, O. H.; Gutiérrez, L. F.; Pérez, V. H. Analysis of a reactive extraction process for biodiesel production using a lipase immobilized on magnetic nanostructures. Bioresour. Technol. 2010, 101, 9542. (4) Dias, T. P. V. B.; Mielke Neto, P.; Ansolin, M.; FollegattiRomero, L. A.; Batista, E. A. C.; Meirelles, A. J. A. Liquid-liquid equilibrium for ternar systems containing ethylic biodiesel + anhydrous ethanol + refined vegetable oil (Sunflower oil, canola oil and palm oil): Experimental data and thermodynamic modelling. Braz. J. Chem. Eng. 2015, 32, 699. (5) Mahmud, M. S. A novel extractive reactor technology for biodiesel production from waste cooking oil. The University of New South Wales, Chemical Sciences & Engineering, Faculty of Engineering: Sydney, Australia, 2011. (6) Chesterfield, D. M.; Trung, T. C.; Lucien, F. P.; Rogers, P. L.; Adesina, A. A. Basket Impeller Extractive Reactor Column for Biodiesel Production: An Experimental Study. Ind. Eng. Chem. Res. 2013, 52, 15298. (7) Samant, K. D.; Singh, D. J.; Ng, K. M. Design of liquid-liquid phase transfer catalytic processes. AIChE J. 2001, 47, 1832. (8) Oldshue, J. Y.; Rushton, J. H. Continuous extraction in a multistage mixer column. Chem. Eng. Prog. 1952, 48, 297. (9) Adesina, A. A. Application of Periodic Operation to the Fischer− Tropsch Synthesis of Hydrocarbons; University of Waterloo: Waterloo, Ontario, Canada, 1986. (10) Hardiman, K. M. Propane Reforming Under Carbon-Induced Deactivation: Catalyst Design and Reactor Operation. The University of New South Wales: Sydney, Australia, 2007. (11) Fogler, H. S. Elements of Chemical Reaction Engineering, 4th Edition; Prentice−Hall: Upper Saddle River, NJ, 2013; p 256. (12) Rocha, E. G. d. A.; Follegatti-Romero, L. A.; Duvoisin, S., Jr; Aznar, M. Liquid−liquid equilibria for ternary systems containing ethylic palm oil biodiesel+ethanol+glycerol/water: Experimental data at 298.15 and 323.15K and thermodynamic modeling. Fuel 2014, 128, 356.

(9)

where rTO0 is the rate of reaction, V the volume of the reactor, FTO0 the flow rate of triolein, CTO0 the initial concentration of triolein, ν0 the volumetric flow rate, and τ the space time. From the results, 27 ≤ Da ≤ 1688.9. The application of in situ extraction on biodiesel reactor in continuous flow is practical because Da is greater than unity.7

■

CONCLUSION Liquid−liquid extraction using excess ethanol as the solvent increased the rate of transesterification reaction with pure palm oil in a multistage stirred column under counter-current flow. The rate of triolein conversion in the MSC was comparable with its conversion in the stirred batch reactor at higher stirring speed. The performance of the MSC reactor was promising, as indicated by the high Da value.

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Mohd Sabri Mahmud: 0000-0002-7806-7454 Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Funding

The funding was provided by Ministry of Higher Education, Malaysia registered as Fundamental Research Grant Scheme (No. RDU120111) and internal grant of UMP (No. RDU120395). E

DOI: 10.1021/acs.iecr.9b00693 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Research Note

Industrial & Engineering Chemistry Research (13) Sivasamy, A.; Cheah, K. Y.; Fornasiero, P.; Kemausuor, F.; Zinoviev, S.; Miertus, S. Catalytic applications in the production of biodiesel from vegetable oils. ChemSusChem 2009, 2, 278. (14) Zheng, S.; Kates, M.; Dubé, M. A.; McLean, D. D. Acidcatalyzed production of biodiesel from waste frying oil. Biomass Bioenergy 2006, 30, 267. (15) Pawelski, A.; Jeon, S. J.; Hong, W. H.; Paschedag, A. R.; Kraume, M. Interaction of a homogeneous chemical reaction and mass transfer in a single moving droplet. Chem. Eng. Sci. 2013, 104, 260.

F

DOI: 10.1021/acs.iecr.9b00693 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX