Ethanolysis of fish oil via optimized protocol and purification by dry washing of crude ethyl esters

ARTICLE IN PRESS

JID: JTICE

[m5G;July 23, 2015;21:45]

Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–13

Contents lists available at ScienceDirect

Journal of the Taiwan Institute of Chemical Engineers journal homepage: www.elsevier.com/locate/jtice

Ethanolysis of fish oil via optimized protocol and purification by dry washing of crude ethyl esters Abdelrahman B. Fadhil∗, Adnan I. Ahmed Industrial Chemistry Research Laboratory, Department of Chemistry, College of Science, Mosul University, Mosul, Iraq.

a r t i c l e

i n f o

Article history: Received 12 December 2014 Revised 31 May 2015 Accepted 7 June 2015 Available online xxx Keywords: Fish oil Ethanolysis De-oiled fish waste Activated carbon Dry washing purification Fuel properties

a b s t r a c t In this research work, fatty acid ethyl ester (biodiesel) was successfully developed from fish oil. The acid value of fish oil used is 1.23 mg KOH/g. As a result, transesterification of fish oil with ethanol was performed via one-step transesterification, namely alkaline-catalyzed transesterification using potassium hydroxide (KOH) as a catalyst. The influence of transesterification variables including amount of KOH, ethanol to oil molar ratio, reaction temperature, reaction time and type of the alkali catalyst on yield of fish oil ethyl esters (FOEE) were investigated. The dry washing method which used the activated carbon produced from de-oiled fish waste was used to purify the crude ethyl esters. The best yield of FOEE (98.04% ∼ 97.11% w/w ester content) was obtained at 0.75% wt. KOH, 9:1 ethanol to oil molar ratio, 70 °C reaction temperature and 60 min of reaction. The fuel properties of FOEE were complied with the limits prescribed in the ASTM D6751 standards and EN 14214, where applicable. The viscosity of the produced ethyl ester was found much lower than those reported for the ethyl esters produced from various feedstocks. The transesterification of fish oil with ethanol followed first order kinetics and the activation energy was found to be 14.45 kJ/mol. © 2015 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

1. Introduction Biodiesel (BD) was used as a successful alternative to petro diesel due to its renewability, non-toxicity and bio-degradability. Moreover, no poisonous pollutants are emitted from it on combustion [1–3]. Production of BD from vegetable oils was widely investigated [4–8]. However, there are some restrictions on the use of vegetable oils in the production of BD, because lands available to cultivate various seed oils are limited. Furthermore, most of vegetable oils are important source of food for humankind. Thus, it is very important to find nonedible feedstocks for BD production. Animal fats such as chicken, beef, duck and lard fats were widely used in the production of BD, because their availability are more guaranteed when compared to vegetable oils. Furthermore, most of animal fats are disposed as wastes. In addition, animal fats are not recommended for human as a source of food due to its negative impact on the health [9–15]. The BD produced from animal fats has higher oxidation stability and higher cetane number than vegetable oils due to their higher content of saturated fatty acids [1]. Iraq is a fish producer country and is one of the major fish producer countries in the world. Besides petroleum, Iraq is also rich in different types of fish found in Tigris and Euphrates rivers as well as Hawizah marsh. Grypus Barbus, Barbus luteus Heckel,

∗

Corresponding author. Tel.: +9647701642855. E-mail address: [email protected] (A.B. Fadhil).

Hypophthalmichthys molitrix and Silurus triostegus Heckel are the main types of fish that can be caught in fresh water surfaces in Iraq. Fish slaughter houses produce various fish wastes such as backbone, skin, fatty layers, heads, tail and stomach which are either disposed as waste or may be used as a manure. Due to the environmental and health risks associated with accumulation of such waste, it is either used as fertilizers or in the production of protein concentrates which in turn is used as feed for fish. Therefore, such waste represents a real threat to the environment unless it is handled. Recycling oil from this waste and using this raw material for BD production may be one of the real solutions [14,15]. Fatty acid ethyl ester (FAEE) emits less greenhouse gases (CO2 and NOX ) as well as particulate matters than fatty acid methyl ester (FAME). Moreover, it is more biodegradable in water than FAME and has higher cetane number and heating value but lower cloud and pour points than FAME. However, production of FAEE through alkali-catalyzed transesterification is difficult compared to FAME due to the formation of stable emulsion during ethanolysis process which makes separation of glycerin from BD very hard [16]. Water washing method is widely used in the purification of BD after separation of glycerin. However, a part of the ester is lost in the form of soaps which happens more in the case of ethanol than methanol. Moreover, wet washing method produces large amounts of the contaminated water which contains various pollutants such as glycerin, soap, free fatty acids, methanol and oil. As a result, treatment of the wastewater is necessary and thereby, the cost of production

http://dx.doi.org/10.1016/j.jtice.2015.06.010 1876-1070/© 2015 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

Please cite this article as: A.B. Fadhil, A.I. Ahmed, Ethanolysis of fish oil via optimized protocol and purification by dry washing of crude ethyl esters, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.06.010

ARTICLE IN PRESS

JID: JTICE 2

[m5G;July 23, 2015;21:45]

A.B. Fadhil, A.I. Ahmed / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–13

increases. Consequently, the dry washing method using various adsorbents such as ion-exchanger, magnesol, activated carbon and rice husk ash was proposed to avoid loss of ester in the form of soaps. Furthermore, the use of adsorbents for purifying the crude BD reduces amounts of the polluted water originated during the production of BD [17–20]. Activated carbon has higher surface area and various pore size distribution which help in removing pollutants of various molecular sizes. Besides, the use of adsorbents in the purification of BD results in bleaching and de-coloring of the produced BD [20]. Few works were reviewed in the literature on the production of FAEE from animal fats [21,22]. The only published research on the production of FAEE from fish oil was used ultrasonic energy [23]. To the best of the author’s knowledge, the optimized ethanolysis of fish oil with the purification of the crude ethyl esters through the dry washing method using the activated carbon produced from its deoiled solid waste was not reported in the literature. Herein, FAEE was produced from fish oil via KOH-catalyzed transesterification with ethanol. The crude ethyl ester was purified by using the activated carbon produced from de-oiled fish waste. Ethanolysis variables such as amount of KOH, ethanol to oil molar ratio, reaction temperature, reaction time and type of the alkali catalyst were investigated. Properties of the produced ethyl esters were evaluated in accordance to the ASTM standards. Kinetics of transesterification of fish oil with ethanol was also investigated, and finally different blends of FAEE and petro diesel were prepared and studied.

After separating the glycerin (lower layer), excess ethanol was recovered from the ethyl esters layer by using a rotary evaporator [3,4]. The unpurified ethyl ester was mixed with the AC (3 wt.%) for purification. Finally, the ethyl esters yield was determined as follows [20,24].

Product yield (wt.%) =

Weight of FOEE produced × 100 Weight of oil used

Biodiesel yield (wt.%) =

Weight of purified FOEE × 100 Weight of oil used

2.4. Analysis of fish oil ethyl esters

2. Experimental

The fatty acids composition of FOEE was analyzed using gas chromatography (GC, Perkin Elmer, Auto system GLX, Shelton, U.S.A.). A silica capillary column (Supelco SPTM -2380, 30 m, 0.25 mm i.d., 0.25 μm film thickness) provided with a flame ionization detector (FID) was used for separation. The sample was dissolved in hexane and helium was used as the carrier gas at a flow rate of 0.5 mL/min. The injector and detector temperature were 280 and 260 °C, respectively. The method proposed by Bindhu et al. [25] was used to determine the ester content on the BD samples. The thin layer chromatography (TLC) was used as a rapid and an easy means to monitor ethanolysis of fish oil by using silica gel plates and the iodine vapor to visualize the spots after fractionation [15]. The key functional groups on the BD and the parent oil were determined using the Fourier transform Infra-Red (FTIR) spectrum (Biotechnology, UK).

2.1. Materials

2.5. Testing and evaluation of fuel properties

Fish wastes were collected from the fish Slaughterhouse located in the city of Mosul, north of Iraq. Chemicals and solvents used in the present study were of analytical reagent grade and were used as received without any further purification. Absolute ethanol (99.90%), potassium and sodium hydroxides (KOH and NaOH, pellets), sodium methoxide (CH3 ONa), sodium ethoxide (CH3 CH2 ONa), potassium iodide, formic acid, sodium periodate and sodium sulfate (Na2 SO4 ) were purchased from Merck (Darmstadt, Germany). Chloroform, nhexane, phenolphthalein indicator, and hydrochloric acid were provided from Fluka (Swiss).

Fish oil ethyl ester produced was evaluated for several interesting properties according to ASTM procedures. Density was measured at 15.6 °C according to ASTM D 4052-91 using a calibrated pycnometer. A kinematic viscometer (Canon F F24 U-tube glass viscometer) was used to determine viscosity of FOEE at 40 °C according to (ASTM D 455). The refractive index (D1747–09) was measured at 20 ± 0.1 °C using the Abbe refractometer connected to a thermostatically controlled water bath that maintained the temperature of the refractometer. The cloud point and pour points were determined according to ASTM D 2500. Conradson carbon residue (ASTM D4530) was used to determine the amount of carbon remained after complete burning of FOEE. The acid value (ASTM D664) was determined using titration method. The flash point (ASTM D93) was determined using a Pensky–Martens closed-cup tester. The iodine value (IV) was measured according to Hanus method. The cetane number was determined using a digital cetane number meter (Shatox, Russian Federation). The total glycerin of the produced fuel was determined using method proposed by Pisarello et al. [26]. The glycerin titration was based on its oxidation to formic acid using sodium periodate, followed by a titration with sodium hydroxide [26]. Soap content was determined in accordance with AOCS Cc 17-95.

2.2. Extraction of fish oil The fish oil was extracted from the discarded parts of fish (fatty layers, backbone, tails, heads and skin) without any chemical treatment as shown in Fig. 1. The fish waste was placed into a 1 L conical flask. The flask was then heated by a boiled water bath in order to melt the oil. The fish oil was then separated from the solid impurities such as meat and particles of bones by filtration. The obtained oil was then transferred to a separating funnel and left overnight to separate water (if any). The oil was dried over freshly activated sodium sulfate (Na2 SO4 ), placed in a dark container and kept at 5 °C for further use. The yield of fish oil (FO) was calculated on weight basis. 2.3. Transesterification of fish oil with ethanol Ethanolysis of FO was performed using a 500 mL round-bottom flask provided with thermostat, mechanical stirrer, sampling outlet, and a condenser. The FO (100 g) was fed to the reactor and then a freshly prepared ethanolic potassium hydroxide solution (a preestablished amount of KOH dissolved in ethanol at 6:1 ethanol to oil molar ratio) was added to the oil. The mixture was heated by using a digital water bath at 70 °C for 60 min with simultaneous stirring at 600 rpm. After completion of the reaction, the products were transferred to a separating funnel and left overnight to obtain two layers.

2.6. Preparation of activated carbon from de-oiled fish waste After extraction of the oil from fish waste, the de-oiled residue was used as a feedstock to produce the activated carbon (AC) via onestep process, i.e. carbonization and activation method as depicted in Fig. 1. The de-oiled waste was mixed with a solution of phosphoric acid (50% H3 PO4 v/v) and left for a day. The mixture was then filtered and oven dried until all water was evaporated. The product was then activated in a muffle furnace at 600 °C for 1 h. The produced carbon was washed with (0.1 M solution of HCl) and then by distillated water until neutral water was obtained [27]. The carbon was then dried, crushed and allowed to pass through a (100 μm) sieve. The micro pore surface area of the produced AC was determined using ethylene glycol mono ethyl ester (EGME) retention method

Please cite this article as: A.B. Fadhil, A.I. Ahmed, Ethanolysis of fish oil via optimized protocol and purification by dry washing of crude ethyl esters, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.06.010

JID: JTICE

ARTICLE IN PRESS

[m5G;July 23, 2015;21:45]

A.B. Fadhil, A.I. Ahmed / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–13

3

Fig. 1. The experimental steps for FOEE production from fish waste.

[28]. Adsorption of methylene blue was used to determine the meso pore surface area of the produced carbon [29]. For the infrared spectroscopy analysis, the produced AC carbon was macerated and supported in KBr pellets. The analysis was performed using a FTIR spectrophotometer (Bio-Technology, UK) in the range of 4000–400 cm−1 with resolution of 4 cm−1 and 32 scans. Scanning electron microscopy (SEM) image for the AC was obtained with 20-kV accelerating voltage with a field emission scanning electron microscope (FEI QUANTA 200, Holland).

2.7. Blending evaluation The FOEE sample which was produced under the optimal conditions was blended with conventional diesel fuel with various volume percentages (10–50% v/v) at room temperature. The blends were placed in a beaker and mixed well by means of a magnetic stirrer at 1200 rpm to ensure homogeneity of the mixture. Then, several interesting properties of the blends such as the density, viscosity, flash

point, refractive index, acid value and pour point were evaluated according to the ASTM standard methods. 3. Results and discussions 3.1. Properties of fish oil It was found that the oil content in fish waste is 76.0% w/w of waste. This value is appreciably higher than the oil content of the soap stock of the marine fish (37.0% w/w) [30] and the fat content of pork waste (70%w/w) [31]. Besides, this percentage yield of oil is much higher than that observed for other vegetable seeds such as rubber seeds (49%) [32]. As a result, fish oil can be a promising source of oil for BD production. Table 1 lists properties of FO used in the present study compared to other animal fats. The acid value (AV) of FO is 1.23 mg KOH/g oil which corresponds to 0.651% FFA content. Dias et al. [31] reported the AV of waste lard to be 14.57 mg KOH/g, whereas Encinar et al. [33] reported the AV of various animal fats to be 21.2, 9.7, and 26.9 mg KOH/g. As a result, no pre-treatment such

Please cite this article as: A.B. Fadhil, A.I. Ahmed, Ethanolysis of fish oil via optimized protocol and purification by dry washing of crude ethyl esters, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.06.010

ARTICLE IN PRESS

JID: JTICE 4

[m5G;July 23, 2015;21:45]

A.B. Fadhil, A.I. Ahmed / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–13 Table 1 Properties of fish oil compared to other animal fats.

Table 2 Fatty acid composition of FOEE compared to other fish oils methyl esters.

Property

FO

Lard fata

Beef tallowb

Fatty acid

FOEE

MFOMEc

SFOMEd

BTMEe

LTMEd

Density @ 15.6 (D) Kinematic viscosity @ 40° C (KV) Flash point, ° C (TF ) Acid value, mg KOH/g oil (AV) Sapon. value mg KOH/g oil (SV) Iodine value, 100 mg I2 /g oil (IV) Cloud point, ° C (CP) Pour point, ° C (PP) Conradson carbon residue % (CCR) Refractive index @20 ° C (RI)

0.9120 18.44 272 1.23 183.54 100.25 12 0 0.120 1.4708

– 47.87 206 – 198 61.22 – – – –

0.9216 37.39 165 4.66 172 33 18 14 0.230 1.4780

C14 C14:1 C15 C16 C16:1 C17 C18 C18:1 C18:2 C18:3 C20 C20:1 C20:2 C20:4 C20:5 C21 C22 C22:4 C22:5 C22:6 Others Total saturated FA Total unsaturated FA Total polyunsaturated FA Total monounsaturated FA

3.83 0.21 2.50 20.20 12.46 3.70 6.34 21.70 5.89 3.25 0.57 1.98 1.85 2.48 7.77 0.35 – – – 4.20 1.07 37.14 61.79 25.44 36.35

3.16 – – 19.61 5.61 1.82 5.24 20.94 2.69 0.90 4.75 – 0.81 2.54 3.70 – 1.55 3.86 2.44 15.91 3.01 37.06 59.93 33.58 26.35

5.08 – – 15.39 7.55 0.46 4.00 20.76 3.78 0.99 0.15 – 0.30 2.08 9.49 – 0.09 0.30 4.94 13.99 2.47 25.70 64.18 30.83 33.35

1.3 0.0 0.0 23.5 2.6 0.4 13.5 41.7 10.7 0.0 – – – – – – – – – – 6.30 38.70 55.0 10.70 44.30

3.10 1.30 0.5 23.8 4.7 1.1 12.7 47.2 2.6 0.8 – – – – – – – – – – 2.20 42.30 55.80 3.40 52.40

a b

from Ref. [13] from Ref. [10]

as an acid treatment is required to produce BD from FO used in the present study. The density and viscosity of FO were lower than those observed for beef and lard fats. The cloud and pour points as well as the iodine value of an oil or fat relate greatly to its content of saturated and unsaturated fatty acids. It was found that FO has lower cloud and pour points but higher iodine value than lard and beef fats which might be attributed to its higher content of unsaturated fatty acids [10,13]. In comparison with other fish oils, FO used in the present study has lower AV than oils obtained from the marine fish oil, salmon fish, the acidified salmon fish oil and waste fish oil from canning industry which required a two-steps process viz. acid–base catalyzed transesterification to produce the BD [30,34,35]. Thus, the production cost of BD from FO used in the present study will be less. The kinematic viscosity (KV) of FO used in the present study was lower than that reported for fish oil which contains 90 wt.% of salmon oil and the remaining 10 wt.% as a mixture of other fish oil in season (45.34 mm2 /s) [36]. The presence of larger percentages of the saturated fatty acids with larger carbon chains increases the KV of the oil. The iodine value (IV) of FO used in the present study was much lower than that reported for waste fish oil from canning industry [35]. This variation could be attributed to the high content of the unsaturated fatty acids content of the latter compared to the former. As shown in Table 2, the principal fatty acids in FOEE determined by GC are oleic (21.70%), palmitic (20.20%), palmitoleic acid (12.46%), stearic (6.34%), linoleic (5.89%) and linolenic (3.25%) acids. The level of total monounsaturated, polyunsaturated and saturated fatty acid of the extracted oil was 36.40%, 25.4% and 61.80%, respectively. The total content of saturated fatty acids in FO used in the present study was close to that reported for the marine fish oil and lower than that observed for salmon fish oil [30,34]. On the other hand, the total content of the unsaturated fatty acids of FO used in the present study was close to that reported for salmon fish oil and lower than that observed for the marine fish oil. On the other hand, FO used in the present study has lower content of the polyunsaturated fatty acids and higher content of the monounsaturated fatty acids than the marine fish oil and salmon fish oil [30,35]. Lin and Li [30] reported that the presence of fatty acids of multiple double bonds (greater than three) in an oil results in deterioration of its oxidation stability [30]. Moreover, the weight fractions of the long chain fatty acids in the range of (C20 –C22 ) in FO used in the presence was lower than those observed in the marine fish and salmon fish oils [30,35].

f

from Ref. [31] c from Ref. [30] d from Ref. [35] e from Ref. [12]

AC determined by the adsorption from aqueous solution method using methylene blue as a solute was found to be 122.72 m2 /g. This indicates that the produced AC has mixed micro and meso porous structure. The SEM can be used to have information about the structure and morphological characteristics of the AC. The SEM image of the produced AC which is depicted in Fig. 2 shows that its surface is rough and contains many pores with different sizes. The presence of various pores increases the surface area and adsorptive capacity of the AC. As a result, pollutants of various molecular sizes such as soaps, free fatty acids, methanol, mono, di and triglyceride and glycerin could be removed by the AC. The FTIR spectrum of the produced AC which is depicted in Fig. 3 shows that the AC surface contains various functional groups. The band at 1618 cm−1 attributes to the stretching mode of vibration for (C═C) bond in aromatic ring. The absorption peak at 2932 cm−1 belongs to stretching mode of vibration of (C–H) bond of methyl and methylene groups. The presence of (–OH) stretching vibration of phenol, alcohol, or carboxylic acid could be attributed to the absorption broad band at 3400–3300 cm−1 . The band at 1735 cm−1 is attributed to stretching vibration of (C═O) in ketones, aldehydes or carboxyl, while that observed at 1451 cm−1 is attributed to stretching vibrations of (C–O) group as well as bending modes of (C–O–C, O–H) in carboxylic acids, carboxyl, lactones, esters, and ethers. The peak at 874 cm−1 belongs to stretching mode of vibration of (O–O) group. The presence of various oxygen groups on the surface of the AC helps removal of impurities from the CEE [20]. 3.3. Biodiesel production from fish oil

3.2. Activated carbon production from the extraction cake The micro pore surface area of the AC produced from de-oiled fish waste was determined using EGME retention method. This method was used by several researchers for determination of micro pore surface area of clays and activated carbons [28,29,37]. The micro pores surface area of the AC determined by EGME retention method was found to be 340 m2 /g, whereas the meso pore surface area of the

The FFA content of an oil or fat should be between (3 and 5%) so as to produce high yield of BD via alkali-catalyzed transesterification [38]. Kafuku and Mbarawa [39] produced BD from Croton megalocarpus oil which has an acid value of 3.34 mg KOH/g oil (1.76% FFA content) via one step alkali-catalyzed transesterification, while Fei et al. [40] produced BD from Xanthium sibiricum seed oil with its acid value 1.38 mg KOH/g oil (0.69% FFA content) via the same

Please cite this article as: A.B. Fadhil, A.I. Ahmed, Ethanolysis of fish oil via optimized protocol and purification by dry washing of crude ethyl esters, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.06.010

JID: JTICE

ARTICLE IN PRESS

[m5G;July 23, 2015;21:45]

A.B. Fadhil, A.I. Ahmed / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–13

Fig. 2. Scanning electron micrograph of the AC.

Fig. 3. The FTIR spectra of the AC.

Please cite this article as: A.B. Fadhil, A.I. Ahmed, Ethanolysis of fish oil via optimized protocol and purification by dry washing of crude ethyl esters, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.06.010

5

ARTICLE IN PRESS

JID: JTICE 6

[m5G;July 23, 2015;21:45]

A.B. Fadhil, A.I. Ahmed / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–13

100

Ethanol : oil = 6:1 Temperature = 70 °C Time = 90 minutes Stirring rate = 600 rpm

90 89.2

80

80.45

70 FOEE,wt.%

87.12

68.34

76.52

Before purification After purification

60 50 40.22

40 30 20 10 0 0.25

0.75

1.25 KOH, wt.%

1.75

Fig. 4. Influence of KOH amount on FOEE yield.

105

KOH = 0.75% Temperature = 70 °C Time = 90 minutes Stirring rate = 600 rpm

95

FOEE, wt.%

85

89.2

90.62

93.22 Before purification

86.21

84.22

After purification

75 65 61.23 55 45 35 27.22

25 3

5

7 9 Ethanol to oil molar ratio

11

13

Fig. 5. Influence of ethanol to fat molar ratio on FOEE yield.

process. The acid value of FO used in the present study is 1.23 mg KOH/g which corresponds to 0.615% FFA content. As a result, a onestep process, namely the alkali-catalyzed transesterification was used in the production of FOEE via optimized protocols. 3.3.1. Influence of catalyst concentration One of the important variables that affect the transesterification reaction is the catalyst concentration. The presence of the catalyst increases the reaction rate of the reaction and thus increases conversion of oil to its corresponding ester. As a result, the catalyst concentration must be optimized. As shown in Fig. 4, different concentrations of KOH ranged from (0.25– 1.50%) in which 0.25% increments were tested, in order to select the optimal concentration of KOH. Ethanol to oil molar ratio, reaction temperature, time and stirring rate were fixed at 6:1, 70 °C, 90 min and 600 rpm, respectively. It was observed that FOEE yield was low with the low concentrations of KOH which is in agreement with findings reported elsewhere [39–42]. However, the yield increases with the increase of KOH concentration. The best yield of FOEE (89.20% ∼ 86.55 ester content) was obtained at 0.75% w/w. Suppalakpanya et al. [2] tested various amounts of KOH on ethanolysis of the esterified palm oil by microwave with dry washing by bleaching earth and found that 1.50 wt.% was the optimal concentration. Such variation may be

attributed to the varied chemical nature of the raw oils used for BD production as well as type of adsorbent used in the purification step. Nevertheless, yield of FOEE reduced with higher concentrations of KOH owing to the excess catalyst favoring the process of saponification. As a result, the yield decreases [40,41]. Thus, 0.75% w/w was established as the optimum concentration. 3.3.2. Influence of ethanol to oil molar ratio Alcohol to oil molar ratio is one of the most important variables affecting transesterification of oil into ester. The stoichiometric molar ratio of alcohol to triglyceride for transesterification is 3:1. However, higher molar ratios are required to drive the reaction to the right. Moreover, alcohol to oil molar ratio affects the production cost. Thus, it should be optimized. The influence of ethanol to oil molar ratio was investigated by testing several ethanol to oil molar ratio (3:1, 5:1, 6:1, 8:1, 9:1, 10:1 and 12:1) as shown in Fig. 5. Concentration of KOH, temperature, time and stirring rate were fixed at 0.75 wt. %., 70 °C, 90 min and 600 rpm, respectively. The yield of FOEE was very low (33.23%) at the stoichiometric ratio of ethanol. Moreover, it was very hard to separate the ethyl esters from the glycerin phase. As a result, molar ratios higher than 3:1 were used to drive the reaction to completion [43]. It was observed that the yield of FOEE increased with the increase of ethanol to oil molar ratio. The highest yield of FOEE

Please cite this article as: A.B. Fadhil, A.I. Ahmed, Ethanolysis of fish oil via optimized protocol and purification by dry washing of crude ethyl esters, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.06.010

ARTICLE IN PRESS

JID: JTICE

[m5G;July 23, 2015;21:45]

A.B. Fadhil, A.I. Ahmed / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–13

7

98 KOH = 0.75% Ethanol : oil = 9:1 Time = 90 minutes Stirring rate = 600 rpm

97

FOEE, wt.%

96

95.45

95

Before purification

94

After purification

93.33

93

93.22

92.21

92 91 90

89.23

89

88 40

50

60

70

80

90

Temperature, C Fig. 6. Influence of temperature on FOEE yield.

(93.22% ∼ 91.45% w/w ester content) was produced at 9:1 ethanol to oil molar ratio. Suppalakpanya et al. [2] investigated ethanolysis of the esterified palm oil by microwave with dry washing by bleaching earth using various ethanol to oil molar ratio and found that 8.5:1 ethanol to oil was the optimal molar ratio. Similar findings were also reported by Anastopoulos et al. [5] on ethanolysis of various vegetable oils. This difference in the optimum ethanol to oil molar ratio required to reach optimum conversion could be attributed to the variation of the chemical composition of the feedstock oils as well as type of the purification method used. Molar ratios beyond 9:1 decreased the yield of FOEE. The excess ethanol hinders gravity separation of the ethyl ester and glycerin phases. Remaining of glycerin in the solution drives the equilibrium back to the reactants side resulting in lower yield of ethyl ester. The glycerol remaining in the solution drives the equilibrium back to the left side of the reaction resulting in the lower yield of ethyl ester [3,6]. Consequently, 9:1 was established as the optimum molar ratio. 3.3.3. Influence of the reaction temperature Temperature accelerates the transesterification process through increasing collisions among the molecules of the reactants [39]. Moreover, it facilitates the reaction by weakening the chemical bonds and thus eases the formation of new product bonds. From an economical point of view, temperature affects the production cost of BD. As a result, it should be optimized. Ethanolysis of FO was conducted at various temperatures (40, 50, 60, 70 and 80 °C), while other parameters were kept fixed. As shown in Fig. 6, yield of FOEE increases with the increment of the reaction temperature owing to the rate constant increases with temperature. Maximum yield of FOEE (95.45% w/w ∼ 93.88% w/w ester content) was obtained at 60 °C which is far below the boiling point of ethanol. Thus, the cost of production will be less. Anastopoulos et al. [5] conducted ethanolysis of sunflower, rapeseed and used frying oils at various temperatures and found that 80 °C was the optimal temperature. This finding could be attributed to the variation of chemical composition of the feedstocks as well as the purification method applied. However, temperatures above 60 °C favor the side reaction viz. saponification by the base catalyst at the expense of transesterification. As a result, part of the esters will convert to soap. Thus, the ethyl esters yield decreases [39–43]. Therefore, 60 °C was chosen as the optimal temperature. 3.3.4. Influence of reaction time The reaction time affects the production cost of BD. Thus, it must be optimized. The influence of time on ethanolysis of FO was

performed at different times (30, 45, 60, 90 and 120 min) as depicted in Fig. 7. With the optimal conditions obtained, it was observed that yield of FOEE increases with the increase of time. Maximum yield of FOEE (98.04% ∼ 97.34% w/w ester content) was obtained at 60 min. With long reaction periods, FOEE yield decreased. This occurred as a result of the hydrolysis of the formed ester resulting in conversion of more fatty acids to soaps [39]. Ginting et al. [44] investigated alkaline in situ ethanolysis of Jatropha curcas and found that a duration of 120 min was the optimal. Similar findings were also reported by Encinar et al. [16] on ethanolysis of used frying oil. 3.3.5. Influence of the catalyst type The alkali catalysts are widely used in the transesterification process. Many researchers tested various alkali catalysts during transesterification reaction so as to find the optimum catalyst [11,15,16,41,43]. In the present study, different alkali catalysts such as NaOH, CH3 ONa and CH3 CH2 ONa were also tested in addition to KOH so as to find the optimal catalyst. For 0.75 KOH w/w of oil was the optimal concentration for ethanolysis FO during the present study, the same concentration of above alkali catalysts was tested so as to make the results comparable as depicted in Fig. 8. With the optimal conditions obtained, the potassium hydroxide catalyst (KOH) exhibited the highest ethyl ester yield among other alkaline catalysts. Fadhil et al. [35] tested various alkaline catalysts in a process of transesterification of melon seed oil, Silybum marianum L. seed oil and Silurus triostegus Heckel fish oil and found that KOH was the best [4,8,15]. Anwar et al. [43] reported similar findings on methanolysis of rapeseed oil. 3.4. Fuel properties of the produced fuels The fuel properties of FOEE produced was determined in accordance with the ASTM standards and listed in Table 3. Each property was measured in triplicate and result was recorded as the mean ± (SD). For comparison purpose, FOEE was also produced from fish oil using optimal conditions obtained during the present study but with purification by using water washing method. Density of FOEE is 0.8798 g/mL. It was also higher than that of conventional diesel fuel. Moreover, the density of FOEE was comparable to density of the ethyl ester produced from a mixture of chicken and swine fats (0.8700 g/mL) [21] and from palm oil (0.8753 g/mL) [2]. The density was also within the safe limits as prescribed by the ASTM standards. The viscosity of FOEE was found to be 2.93 mm2 /s which is much lower than that of the parent oil. Purification of the crude

Please cite this article as: A.B. Fadhil, A.I. Ahmed, Ethanolysis of fish oil via optimized protocol and purification by dry washing of crude ethyl esters, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.06.010

ARTICLE IN PRESS

JID: JTICE 8

[m5G;July 23, 2015;21:45]

A.B. Fadhil, A.I. Ahmed / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–13

100

KOH = 0.75 wt.% Ethanol : oil = 9:1 Temperature = 60 °C Stirring rate = 600 rpm

98 98.04

Before purification

96

FOEE, wt.%

After purification

95.45 94 92.11

92

90 88.33

88 87.45 86 30

50

70 Time, minutes

90

110

130

Fig. 7. Influence of time on FOEE yield.

100

Before purification

99.01 98.04

Catalyst wt.% = 0.75 Ethanol : oil = 9:1 Temperature = 60 °C Time = 60 minutes Stirring rate = 600 rpm

After purification

98

96 FOEE, wt.%

94 91.02

92 90

87.33

88

89.22 87.71

88.23 86.64

86 84 82 80 KOH

NaOH

CH3ONa

CH3CH2ONaO

Catalyst type Fig. 8. Influence of the alkali catalyst type on FOEE yield.

Table 3 Properties of FOEE compared to standard fatty acid methyl esters. Property

Method

ASTM limits

CEE

FOEE1

FOEE2

Density @ 15.6° C Kinematic viscosity @ 40 ° C Flash point ° C Acid value, mg KOH/g oil Iodine value, mg I2 /100 oil Cloud point ° C Pour point ° C Conradson Carbon Residue,wt. % Refractive index @ 20 ° C Cetane number Glycerol content (wt.%) Soap (ppm) Yield% Ester content, w/w%

ASTM D4052-91 ASTM D445 ASTM D93 ASTM D664 Hanus method ASTM D 2500 ASTM D 2500 ASTM D 4530 D1747–09 ASTM D-613 Pisarello et al., (2010) AOCS Cc 17-95 – Bindhu et al., (2012)

0.9000 max. 1.9–6.0 130 min. 0.80 max. 120 max. – – 0.05 max. – Min. 47 – – 96.0 –

0.9002 7.42 189 0.62 101.11 5 1 0.17 1.4580 – 0.3402 4.22 – –

0.8798 ± 0.0021 2.93 ± 0.10 160 ± 1 0.14 ± 0.01 103.20 ± 1.0 3 ± 0.50 –1 ± 0.50 0.025 ± 0.001 1.4540 ± 0.0003 51.25 ± 0.34 0.0803 ± 0.0012 0.11 ± 0.01 99.01 ± 0.87 97.43 ± 0.56

0.8992 6.22 184 0.52 102.11 7 2 0.10 1.4561 52.32 0.2211 2.0 85.10 81.22

1 2

purified by the AC washing. purified by water washing.

Please cite this article as: A.B. Fadhil, A.I. Ahmed, Ethanolysis of fish oil via optimized protocol and purification by dry washing of crude ethyl esters, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.06.010

JID: JTICE

ARTICLE IN PRESS

[m5G;July 23, 2015;21:45]

A.B. Fadhil, A.I. Ahmed / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–13

ethyl esters using the AC reduced the viscosity by around 60.51% compared to 16.17% for that purified by water washing method. Cunha Jr et al. [11] reported viscosity of the ethyl esters produced from a mixture of chicken and swine fats to be 4.82 mm2/ s, whereas viscosity of palm oil ethyl esters produced via microwave heating was found to be (5.51 mm2 /s) [42]. Besides, viscosity of FOEE purified by the AC was lower than that reported for the ethyl esters produced from the esterified palm oil with microwave heating and purified by bleaching earth (5.78) mm2 /s [2]. This difference may be attributed to the verified chemical composition of the feedstocks oils, the purification method used and type of the adsorbent used in the purification step. The viscosity of FOEE was also within the acceptable ranges specified by the ASTM standard D 6751 (1.9–6.0 mm2 /s). Fish oil ethyl ester had a flash point (160 °C) greater than that of conventional diesel fuel. The reduction in the flash point after purification by the AC was 15.34% compared to 2.64% for that purified by using washing by water. Suppalakpanya et al. [2] reported the flash point of the esterified palm oil with microwave heating and purified by bleaching earth to be 178 °C. The low flash point of FOEE compared to the ethyl esters purified by bleaching earth could be attributed to the surface area and surface nature of the adsorbent used in the purification. The AC has higher surface area and various functional groups on its surface. As a result, impurities such as mono, di or tri-glycerides which its presence in biodiesel increases its flash point can be further removed by the AC. The AV offered by FOEE is 0.14 mg KOH/g. The use of the AC in the purification of the CCE reduced the acid value by around 77.41% compared to 16.12% for that purified by wet washing method. The AV of ethyl esters produced from the esterified palm oil with microwave heating and purified by bleaching earth was found to be 0.73 mg KOH/g [2]. This difference in the AV may be attributed to reasons mentioned previously. Furthermore, the AV of FOEE is within the prescribed limits in the ASTM biodiesel standards. Conradson carbon residue (CCR) was used to determine the amount of carbon leftover after pyrolysis of oil. It is very important for BD because it indicates the presence of glycerides, free fatty acids, soaps, polymers and inorganic impurities [45]. The CCR observed for FOEE satisfied the limits of <0.05% max (0.025%). The reduction in CCR was 85.29% after purification of the CEE by the AC compared to 41.17% for wet washing method. The CCR of BD produced from the marine fish oil was reported to be 0.76% [30]. The difference in the CCR of FOEE compared to BD produced from the marine fish oil may be attributed to the type of purification method applied, namely dry washing method which helps in removing some impurities such as mono, di or tri-glycerides as their removal via wet washing method is impossible. The cetane number (CN) of FOEE is 52.25. Encinar et al. [16] reported cetane index of ethyl esters produced from waste cooking oil to be 49.3, while the CN of BD produced from okra seed oil was found to be 55.20 [43]. The CN of BD greatly relates to the content of the saturated fatty acids in an oil or fat. As a result, FOEE has higher CN than that reported for okra seed oil BD. The refractive index (RI) at some reference condition (i.e., 20 °C and 1 atm) is another useful characterization parameter to estimate the composition and quality of petroleum fractions. The RI is also used to estimate other physical properties such as molecular weight, the critical constants, or transport properties of hydrocarbon system and for comparison purposes and as a clue for purity [10]. It was found that FOEE possesses lower value of the RI compared to its parent oil. This finding indicates conversion of oil to its corresponding ester. The RI of FOEE was 1.4540 after purification by the AC. Furthermore, it was also lower than that reported for soybean ethyl ester (1.4641) and linseed ethyl esters (1.4606) [46]. This variation may be attributed to the varied fatty acid composition in the parent oil as well as the method used in the purification of BD. The completion of the transesterification process can be determined by determination of the glycerin content in BD [47]. The glycerin content of the CEE was 0.3402 wt. %. This value reduced by

9

around 76.39% after purification of the CEE by the AC, while it reduced by around 35.05% using wet washing method. Manique et al. [47] reported the glycerin content of BD produced from waste frying oil and purified by using 5.0% w/w rice husk ash and 1.0% w/w magnesol to be 0.46 and 0.45%, respectively. This variation attributes greatly to the nature of adsorbent used, its surface area and its porous structure. Manique et al. [47] stated that the surface of the rice husk ash contain large pores which was proved by the SEM. The AC surface contains various pores as well as it contains various functional groups which have an important role on adsorption and removing impurities. The glycerin content of FOEE was much lower than the limits prescribed by the ASTM and EN standards. The IV of the FOEE was 103.20 mg (I2 /100 g oil). Cunha Jr et al. [11] reported the IV of the ethyl esters produced from a mixture of chicken and swine fats to be (77.70 mg I2 /100 g fat). The difference between the two fuels could be ascribed to the varied content of the unsaturated fatty acids of the feedstock oils used for BD production. The IV of FOEE was satisfactory according to the ASTM standards (<120 max.). The flow properties of BD can be determined in terms of cloud point and pour point having values 6 and –1 °C, respectively. The cloud and pour points reported for the esterified palm oil with microwave heating and purified by bleaching earth to be 7and 4 °C, respectively [2]. The unsaturated fatty acid content of palm oil is 49.0% compared to 61.79% for FO [42]. As a result, the latter has lower cloud and pour points than the former. The cloud and pour points of FOEE are suitable for use in cold weather conditions prevailing in northern Iraq in winter. Pose of corrosion, damage of the injectors as well as clogging of filters could be attributed to the presence of soaps in BD. Thus, soap content should be as less as possible [48]. Soap content of the CEE reduced by around 98.10% after purification by the AC compared to 52.60% for that purified by water washing method. As seen from Table 3, FOEE purified by the dry washing method using the AC had better yield, ester content and properties compared to those observed for FOEE purified using the wet washing method. Regarding the yield, it is well known that water washing method involves addition of excessive amounts of water with gentle shaking to remove impurities such as unreacted alcohol, glycerol and the catalyst. As a result, part of the esters is lost in the form of emulsion and soaps and thus decreasing the yield. For dry washing method, the surface of the AC contains various oxygen groups which play an important role in removing impurities from the CEE. Consequently, purification using the said adsorbent is better. As shown in Table 3, the density, viscosity, flash point, cloud and pour points and CCR of FOEE purified by the dry method using the AC were much better than those of FOEE purified by wet washing method. This may be attributed to the presence of some impurities such as mono, di or tri-glycerides in FOEE purified by wet washing method. The presence of such impurities in biodiesel affected greatly those impurities. Moreover, such impurities are water insoluble and thus adsorption using the AC is more effective to remove those impurities than wet washing method. Results in Table 3 imply that properties of FOEE were in agreement with those prescribed by the ASTM standards. Therefore, FOEE can be used as fuel in the internal combustion engines without any modification in the engine. 3.5. Analysis of fish oil ethyl esters Fish oil ethyl ester was analyzed using different methods such as the thin layer chromatography (TLC) and the Fourier Transform Infra Red (FTIR) spectroscopy. The TLC was used as a fast and an easy technique to monitor ethanolysis of FO. Fig. 9 depicts the TLC photographs of the CEE and those purified by the dry and wet washing methods. It was noticed that the purification using the AC resulted in removal all impurities from the CEE such as glycerides as well as free fatty acids, while some impurities found in that were purified using wet washing method.

Please cite this article as: A.B. Fadhil, A.I. Ahmed, Ethanolysis of fish oil via optimized protocol and purification by dry washing of crude ethyl esters, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.06.010

ARTICLE IN PRESS

JID: JTICE 10

[m5G;July 23, 2015;21:45]

A.B. Fadhil, A.I. Ahmed / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–13

Ester

Wet washing by water

Dry washing by AC

Crude ethyl esters

Fig. 9. The TLC photographs of CEE and FOEE purified by the AC and water washing method.

This indicates that the dry washing method is more successful in purification of BD than wet washing method [8,15]. The most important functional groups present in the CEE and that purified by the AC were determined using the FTIR spectroscopy as shown in Fig. 10. The observed bands were as follows: the absorption band at 2854–3010 cm−1 attributes to asymmetric and symmetric CH3 stretching vibrations (–CO–O–CH3 ), 1466–1404 cm−1 is due to (–C–H) in an alkane bonding. The absorption band observed at 1244–1117 cm−1 attributes to stretching vibration of the (–C–O–) ester groups, whereas that observed at 1662 cm−1 is attributed to (C═C) stretching modes characteristic of olefins. The characteristic absorption band at 1739 cm−1 attributes to the ester carbonyl group (C═O). The absorption band at 3437 cm−1 is attributed to the stretching mode of vibration of O–H group. The observed absorption bands are similar to those observed in our previous investigations as well as those reported in the literature [4,8,15]. It is clear from Fig. 10 that functional groups present in the CEE were similar to those present in the purified ethyl ester but with a bit shift toward higher or lower frequency. The differences between the spectra of the CEE and that purified by the AC were clear so that the area under the absorption peak of C–H group, the carbonyl and the stretching O–H groups were smaller than those observed in the CEE. Moreover, the area under the absorption band which refers to stretching mode of vibration of O–H group in the CEE was greater than that in the purified ethyl ester. This confirms that the CEE purified by the suggested adsorbent contains as low as possible amounts of pollutants such as glycerol or soaps [4,8,15] . 3.6. Kinetics study The kinetics of the transesterification process was widely investigated by many researchers [6,49–52]. However, most of those researchers considered that the transesterification will proceed through a pseudo first order kinetics as a function of the triglyceride concentration [TG] and temperature of transesterification. Although the use of ethanol in excess is required to shift the reaction to product side (forward reaction), it would not be involved in the reaction as a limiting reactant [6,49]. The transesterification process involves reaction of 1 mole of the triglyceride with 3 moles of alcohol (ethanol) to yield 3 moles of ester and 1 mole of glycerol. Ignoring the intermediate step of the reaction;

the reaction can be expressed as follows:

RCOOR + 3CH3 OH → 3RCOOCH3 + Glycerol The rate equation of any reaction can be expressed as follows:

Rate = − [concentration]/dt

(1)

As we stated earlier, the triglyceride is the limiting reactant of the transesterification, as a result, the rate equation can be rewritten as follows:

Rate = − d[T G]/dt

(2)

As we are studying the kinetic of the reaction; the rate constant must be introduced. Thus, the equation can be written as follows [15]:

−d[T G]/dt = k.[T G]

(3)

Integrating Eq. (3) by considering that [TG0 ] is the initial concentration of TG at time, t = 0 and [TGt ] is the concentration of triglyceride at time, t = t. Thus, the equation can be rewritten as follows [6]:

−ln([T G0 ]/[T Gt ]) = k.t

(4)

Rearrangement of the above equation gives the following equation:

−ln[T G0 ] − ln[T Gt ] = k.t

(5)

Plotting of –ln [TG0 ] – ln [TGt ] versus time will yield the slope which represents the rate constant as depicted in Fig. 11. It is well known that the increment of the reaction temperature increases the rate of the reaction. As a result, the rate constant for the forward reaction increases with the rise of temperature. Thus, the obtained results (Table 4) agree with the assumption of irreversible reaction .Similar results were reported by other authors as well [6]. The Arrhenius equation was used to calculate the activation energy of ethanolysis of FO. The relation between the reaction rate constant versus temperature can be described by the equation:

k = Ae−Ea/RT

(6)

where, A is the frequency factor, Ea is the activation energy, R is the universal molar gas constant and T is the temperature (°K). As the activation energy is temperature dependent; the rate constant at any

Please cite this article as: A.B. Fadhil, A.I. Ahmed, Ethanolysis of fish oil via optimized protocol and purification by dry washing of crude ethyl esters, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.06.010

ARTICLE IN PRESS

JID: JTICE

[m5G;July 23, 2015;21:45]

A.B. Fadhil, A.I. Ahmed / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–13

11

Fig. 10. The FTIR spectra of the CEE and FOEE purified by the AC.

3.5 KOH = 0.75 wt. % Ethanol : oil = 9:1 Stirring rate = 600 rpm

3

ln[TG0]- ln [TGt]

2.5 2 50 C 1.5

60C 70 C

1 0.5 0 0

1000

2000 Time, sec

3000

4000

Fig. 11. ln [TG0 ] – ln [TGt ] versus time plot at different temperatures for ethanolysis of FO.

temperature can be expressed using the following Eq.:

ln k = ln A − Ea /RT

(7)

Fig. 12 shows the linear plot of ln k versus 1/T which gives the activation energy. The obtained results validate the first order reaction for ethanolysis of FO. The activation energy for the transesterification reaction of FO with ethanol was found to be 14.45 kJ/mole from the first order reaction.

Table 4 The rate constants of the transesterification of fish oil at various temperatures. Temperature (°K)

Rate constant, k (s– 1 )

R2 values

323 333 343

0.000414 0.000390 0.000299

0.9201 0.9978 0.9955

Please cite this article as: A.B. Fadhil, A.I. Ahmed, Ethanolysis of fish oil via optimized protocol and purification by dry washing of crude ethyl esters, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.06.010

ARTICLE IN PRESS

JID: JTICE 12

[m5G;July 23, 2015;21:45]

A.B. Fadhil, A.I. Ahmed / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–13

KOH = 0.75 wt.% Ethanol to oil = 9:1 Stirring rate = 600 rpm

-8.15 -8.1 -8.05

R² = 0.9019

ln k

-8 -7.95 -7.9 -7.85 -7.8 -7.75 0.0029 -7.7

0.00295

0.003

0.00305

0.0031

0.00315

1/T (X 1000) (K-1) Fig. 12. Arrhenius plot, ln k versus 1/T for ethanolysis of FO. Table 5 Properties of (FOEE + PD) blends. Sample

Density @ 16 ° C g/mL

KV @40 ° C mm2 /sec

FP° C

PP° C

AV mg KOH/g

RI @ 20° C

PD B100 B10 B20 B30 B40 B50

0.8300 0.8798 0.8422 0.8461 0.8540 0.8581 0.8601

2.14 2.93 2.33 2.45 2.68 2.89 3.98

76 160 78 83 87 93 98

–16 –1 –16 –14 –15 –14 –12

0.47 0.14 0.45 0.41 0.37 0.28 0.24

1.4800 1.4540 1.4711 1.4612 1.4600 1.4588 1.4553

3.7. Blending evaluation Biodiesel and petro diesel (PD) can be blended with each other at all ratios because they are miscible. Furthermore, BD has some characteristics which are not found in PD, and vice versa. Thus, their blending will enhance properties of each other. In the present study, various blends of BD and PD (B10 , B20 , B30 , B40 and B50 ) were prepared and assessed for several interesting properties such as the density, kinematic viscosity, pour point, flash point, acid value and refractive index as shown in Table 5. It can be seen that the density of PD increased markedly with the increment of BD content in the blends. However, the density remained within the limits prescribed by the ASTM standards. The increment of BD content in the blends increased viscosity of PD. However, the viscosity remained within the standard limits. The flash point of PD is 76 °C, while that of BD was 160 °C. As in the case of viscosity, the flash point of PD increased with increasing BD content in the blends. One of the disadvantages of BD is its high flow properties such as pour point. This property is connected greatly to the content of unsaturated fatty acids in the oil. Table 5 shows that the pour point of PD increases with the increment of BD content in the blends due to the high pour point of BD compared to PD. However, the pour point of PD remained low and suitable for use in some cold weathers such as those prevailing in north Iraq in winter. The AV and RI of PD are much higher than that of BD. Therefore, blending of PD with BD results in a reduction of the AV and RI of the former. High values of the density, viscosity and flash point of BD compared to PD is due to the high molecular mass of BD compared to the conventional diesel fuel [4,15]. Similar findings were also reported by Fadhil and Ali [15]. Moreover, the same findings were observed by several researchers as well [52]. 4. Conclusions In the present research work, fish oil was extracted from fish waste without any chemical treatment and used in the production of fatty acids ethyl ester (FOEE). The optimized KOH-catalyzed

transesterification with ethanol was used to produce FOEE. The dry washing method was used to purify the crude ethyl esters using the activated carbon which was produced from the de-oiled fish waste. The following results were concluded: 1. The best yield of FOEE (98.04% ∼ 97.11% ester content w/w) was produced at 0.75 wt. % KOH, 9:1 ethanol to oil molar ratio, 70 °C reaction temperature and 60 min of reaction. 2. It was found that the yield, ester content and properties of FOEE purified by the dry washing method were better than those observed for FOEE purified by the water washing method. 3. The fuel properties of the produced FOEE were evaluated and found comparable with diesel fuel. Moreover, the properties were complied with the limits prescribed in the ASTM D6751 standards. 4. Blending of the produced ethyl ester with petro diesel was also investigated. The results showed properties of petro diesel were influenced by the addition of biodiesel. However, the properties of the blends were within the limits prescribed in the ASTM D6751 standards. 5. Ethanolysis of fish oil was found to follow first order kinetics, and the activation energy was found to be 14.45 kJ/mol. Acknowledgment The authors would like to thank Mosul University, College of Science, Department of Chemistry for supporting this work. References [1] Demirbas A. Progress and recent trends in biodiesel fuels. Energy Convers Manage. 2009;50:14–34. [2] Suppalakpanya K, Ratanawilai SB, Tongurai C. Production of ethyl ester from esterified crude palm oil by microwave with dry washing by bleaching earth. Appl Energy 2010;87:2356–9. [3] Rashid U, Ibrahim M, Yasin S, Yunus R, Taufiq-Yap YH, Knothee G. Biodiesel from Citrus reticulata (mandarin orange) seed oil, a potential non-food feedstock. Ind Crops Prod. 2013;45:355–9. [4] Fadhil AB. Optimization of transesterification parameters of melon seed oil. Int J Green Energ. 2013;10:763–74.

Please cite this article as: A.B. Fadhil, A.I. Ahmed, Ethanolysis of fish oil via optimized protocol and purification by dry washing of crude ethyl esters, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.06.010

JID: JTICE

ARTICLE IN PRESS A.B. Fadhil, A.I. Ahmed / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–13

[5] Anastopoulos G, Zannikou Y, Stournas S, Kalligeros S. Transesterification of vegetable oils with ethanol and characterization of the key fuel properties of ethyl esters. Energies 2009;2:362–76. [6] Sivakumar P, Sindhanaiselvan S, Nagendra N, Sureshan G, Devi S, Renganathan S. Optimization and kinetic studies on biodiesel production from underutilized Ceiba Pentandra oil. Fuel 2013;103:693–8. [7] Ahmad M, Zafar M, Sultana S, Azam A, Khan MA. The optimization of biodiesel production from a novel source of wild non-edible oil yielding plant Silybum marianum. Int J Green Energ. 2014;11:589–94. [8] Fadhil AB, Ahmed KM, Dheyab MM. Silybum marianum L. seed oil: a novel feedstock for biodiesel production. Arab J Chem. 2012. http://dx.doi.org/ 10.1016/j.arabjc.2012.11.009. Article in Press. [9] Alptekin E, Canakci M. Optimization of pretreatment reaction for methyl ester production from chicken fat. Fuel 2010;89:4035–9. [10] Fadhil AB. Biodiesel production from beef tallow waste using alkali-catalyzed transesterification. Arab J Sci Eng. 2013;38:41–7. [11] Cunha Jr AC, Feddern V, De Pra MC, Higarashi MM, Abreu PG, Coldebella A. Synthesis and characterization of ethylic biodiesel from animal fat wastes. Fuel 2013;105:228–34. [12] Wyatt VT, Hess MA, Dunn RO, Foglia TA, Haas MJ, Marmer WN. Fuel properties and nitrogen oxide emission levels of biodiesel produced from animal fats. J Am Oil Chem Soc. 2005;82(8):585–91. [13] Ejikeme PM, Anyaogu ID, Egbuonu CAC, Eze VC. Pig-fat (Lard) derivatives as alternative diesel fuel in compression ignition engines. J Petroleum Technol Altern Fuels 2013;4(1):7–11. [14] Hong IK, Park JW, Lee SB. Optimization of fish-oil-based biodiesel synthesis. J Ind Eng Chem. 2013;19:764–8. [15] Fadhil AB, Ali LH. Alkaline-catalyzed transesterification of Silurus triostegus Heckel fish oil: optimization of transesterification parameters. Renew Energy 2013;60:481–8. [16] Encinar JM, Gonzalez JF, Rodriguez-Reinares A. Ethanolysis of used frying oil: biodiesel preparation and characterization. Fuel Process Technol. 2007;88:513– 22. [17] Berrios M, Martin MA, Chica AF, Martin A. Purification of biodiesel from used cooking oils. Appl Energy 2011;88:3625–31. [18] Berrios M, Skelton RL. Comparison of purification methods for biodiesel. Chem Eng J. 2008;144:459–65. [19] Faccini CS, Cunha ME, Moraes MSA, Krause LC, Manique MC, Rodrigues MRA. Dry washing in biodiesel purification: a comparative study of adsorbents. J Braz Chem Soc 2010;22:558–63. [20] Fadhil AB, Dheyab MM, Abdul-Qader AY. Purification of biodiesel using activated carbons produced from spent tea waste. J Assoc Arab Univ Basic Appl Res. 2012;11:45–9. [21] AC Jr, Feddern V, Marina CDP, Martha M, Higarashi PG, Coldebella A. Synthesis and characterization of ethylic biodiesel from animal fat wastes. Fuel 2013;105:228–34. [22] Ilvania C, Véras FAL, Silvaa AD, Ferrão-Gonzales VHM. One-step enzymatic production of fatty acid ethyl ester from high-acidity waste feedstocks in solvent-free media. Bioresour Technol. 2011;102:9653–8. [23] Armenta RE, Vinaturo M, Burja MA, Kralovec JA, Barrow CJ. Transesterification of fish oil to produce fatty acid ethyl esters using ultrasonic energy. J Am Oil Chem Soc 2007;84:1045–52. [24] Predojevic ZJ. The production of biodiesel from waste frying oils: a comparison of different purification steps. Fuel 2008;87:3522–8. [25] Bindhu CH, Reddy JRC, Rao BVSK, Ravinder T, Chakrabarti PP, Karuna MSI, et al. Properties and evaluation of biodiesel from Sterculia foetida seed oil. J Am Oil Chem Soc 2012;89:891–6. [26] Pisarello ML, Dall-Csta BO, Veizaaga NS, Querini CA. Volumetric method for free and total glycerin determination in biodiesel. Ind Eng Chem Res. 2010;49:8935– 41. [27] Wua Qf, Zhang FS. A clean process for activator recovery during activated carbon production from waste biomass. Fuel 2012;94:426–32.

[m5G;July 23, 2015;21:45] 13

[28] Cerato AB, Lutenegger AJ. Recommended method for determining surface area of fine-grained soils. Geotech Test J ASTM 2002;25(3):315–21. [29] Fadhil AB, Deyab MM. Conversion of some fruit stones and shells into activated carbons. Arab J Sci Eng. 2008;33(2A):175–84. [30] Lin CY, Li RJ. Fuel properties of biodiesel produced from the crude fish oil from the soapstock of marine fish. Fuel Process Technol. 2009;90(1):130–6. [31] Dias JM, Maria CM, Alvim-Ferraz, Almeida MF. Production of biodiesel from acid waste lard. Bioresour Technol. 2009;100:6355–61. [32] Morshed M, Ferdous K, Khan MR, Mazumder MSI, Islam MA, Uddin MdT. Rubber seed oil as a potential source for biodiesel production in Bangladesh. Fuel 2011;90:2981–6. [33] Encinar JM, Sanchez N, Martinez G, Garcia L. Study of biodiesel production from animal fats with high free fatty acid content. Bioresour Technol. 2011;102:10907– 14. [34] El-Mashad HM, Zhang R, Roberto J, Avena B. A two-step process for biodiesel production from salmon oil. Biosyst Eng. 2008;99:220–7. [35] Costa JF, Almeida MF, Alvim-Ferraz MCM, Dias JM. Biodiesel production using oil from fish canning industry wastes. Energy Convers Manage. 2013;74:17–23. [36] García-Moreno PJ, Khanum M, Guadix A, Guadix EM. Optimization of biodiesel production from waste fish oil. Renew Energy 2014;68:618–24. [37] Bestani B, Benderdouche N, Benstaali B, Belhakem M, Addou A. Methylene blue and iodine adsorption onto an activated desert plant. Bioresour Technol. 2008;99:8441–4. [38] Sinha S, Agarwal AK, Garg S. Biodiesel development from rice bran oil: transesterification process optimization and fuel characterization. Energy Convers Manage. 2008;49:1248–57. [39] Kafuku G, Mbarawa M. Biodiesel production from Croton megalocarpus oil and its process optimization. Fuel 2010;89:2556–60. [40] Fei C, Milford A, Hanna DZ, Hu L, Quan Z, Bao-An S, et al. Production of biodiesel from non-edible herbaceous vegetable oil: Xanthium sibiricum Patr. Bioresour Technol. 2013;140:435–8. [41] Rashid U, Anwar F. Production of biodiesel through optimized alkaline-catalyzed transesterification of rapeseed oil. Fuel 2008;87:265–73. [42] Suppalakpanya K, Ratanawilai SB, Tongurai C. Production of ethyl ester from crude palm oil by two-step reaction with a microwave system. Fuel 2010;89:2140–4. [43] Anwar F, Rashid U. Production of biodiesel through optimized alkaline-catalyzed transesterification of rapeseed oil. Fuel 2008;87:265–73. [44] Ginting MSA, Azizan MT, Yusup S. Alkaline in situ ethanolysis of Jatropha curcas. Fuel 2012;93:82–5. [45] Shambhu, Bhattacharya VB, Nayak1 TK, Das LKS. Studies on characterization of raw Jatropha oil and its biodiesels with relevance of diesel. Int J Emerg Technol Adv Eng. 2013;3(4):48–54. [46] Guzatto R, Defferrari D, Reiznautt QB, Cadore IR, Samios D. Transesterification double step process modification for ethyl ester biodiesel production from vegetable and waste oils. Fuel 2012;92:197–203. [47] Manique MC, Faccini CS, Onorevoli B, Benvenutti EV, Caram EB. Rice husk ash as an adsorbent for purifying biodiesel from waste frying oil. Fuel 2012;92:56–61. [48] Atadashi IM, Aroua MK, Abdul Aziz AR, Sulaiman NMN. Refining technologies for the purification of crude biodiesel. Appl Energy 2011;88:4239–51. [49] Birla A, Singh B, Upadhyay SN, Sharma YC. Kinetics studies of synthesis of biodiesel from waste frying oil using a heterogeneous catalyst derived from snail shell. Bioresour Technol. 2012;106:95–100. [50] Berchmans HJ, Morishita K, Takarada T. Kinetic study of hydroxide-catalyzed methanolysis of Jatropha curcas–waste food oil mixture for biodiesel production. Fuel 2013;104:46–52. [51] Zhang AL, Sheng B, Xin Z, Liu Q, Sun S. Kinetics of transesterification of palm oil and dimethyl carbonate for biodiesel production at the catalysis of heterogeneous base catalyst. Bioresour Technol. 2010;101:8144–50. [52] Silitonga AS, Ong HC, Mahlia TMI, Masjuki HH, Chong WT. Characterization and production of Ceiba pentandra biodiesel and its blends. Fuel 2013;108:855–8.

Please cite this article as: A.B. Fadhil, A.I. Ahmed, Ethanolysis of fish oil via optimized protocol and purification by dry washing of crude ethyl esters, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.06.010