Parametric study of biodiesel production from used soybean oil

For Peer ReviewParametric Study of Biodiesel Production from Used Soybean Oil

Journal European Journal of Lipid Science and Technology

Manuscript ID EJLT-2007-0229R2

Wiley - Manuscript type Research Paper

Date Submitted by the Author

14-Feb-2008

Complete List of Authors Allawzi Mamdouh Jordan University of Science and Technology Chemical Engineering kandah Munther Jordan University of Science and Technology Chemical Engineering

Keywords Biodiesel Transesterification Glycerin KOH

Wiley-VCH

European Journal of Lipid Science and Technology

For Peer Review

1

Parametric Study of Biodiesel Production from Used Soybean Oil

Mamdouh Allawzi Munther Issa KandahChemical Engineering Department

Jordan University of Science and TechnologyPO Box 3030 ndashIrbid ndash 22110 ndash Jordan

mallawzijustedujo

Abstract

Biodiesel an alternative diesel fuel derived from vegetable oil animal fat or

waste vegetable oil (WVO) is obtained by reacting the oil or fat with an alcohol

(transesterification) in the presence of a basic catalyst to produce the corresponding

monondashalkyl esters In this work the effect of catalyst KOH to WVO ratio ethanol

concentration and time of reaction on the biodiesel yield were investigated The

transesterification reaction was performed at a constant temperature (35oC) in order to

minimize the cost of heating and ethanol evaporation A 23 complete factorial design on

biodiesel yield (Y) was performed using low and high levels of operating variables KOH

concentration (9 to 14 gl) ethanol concentration (30 to 40 vol) and time (30 to 40

min) The complete factorial model that can be used to fit the data was determined The

model shows that interaction exists among the parameters and that the parameters or

factors do not operate independently on the response (biodiesel yield) Highest yield

was obtained in the first 30 minutes of the reaction time Results indicate that the highest

yield was 785 vol using KOH to WVO ratio of 12 gl and 30 vol ethanol The

ASTM tests indicate that biodiesel properties are within the biodiesel standard limits

Keywords Biodiesel Transesterification KOH Ethanol Glycerin

Correspondence author mallawzijustedujo

Page 1 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

2

1 Introduction

The use of plant oil as fuel in a compression ignition engine is as old as the engine

itself [1] Although vegetable oil alone proved too viscous for continuous use in the

diesel engine plant-derived fuel oil was developed during the early part of the 20th

century [2] Lately increased demand has driven research that examines the production

of biodiesel especially after petroleum prices skyrocketed and uncertainty over the

anticipated scarcity of energy resources in the future Biodiesel provides a market for

excess production of vegetable oil and animal fat while decreasing a countrys

dependence on petroleum Biodiesel is renewable and its impact on global warming is

less than that of fossil fuels due to its closed carbon cycle exhaust emissions from

biodiesel are lower than that of regular diesel fuel Biodiesel provides substantial

reductions in carbon monoxide unburned hydrocarbons and particulate emissions from

diesel engines [3] Biodiesel has excellent lubricating properties when added to regular

diesel fuel it can convert fuel with poor lubricating properties into an acceptable fuel

Biodiesel is defined as the mono-alkyl esters of fatty acids derived from vegetable

oil or animal fat [4] Biodiesel is the product obtained when vegetable oil or animal fat is

chemically reacted with an alcohol to produce a new compound that is known as a fatty

acid alkyl ester A catalyst such as sodium or potassium hydroxide is required and

glycerol is produced as a byproduct Numerous references can be found in literature

regarding biodiesel production and properties [5-10]

Kasteren and Nisworo [11] described the conceptual design of a production

process in which waste cooking oil is converted via supercritical transesterification with

methanol to methyl esters (biodiesel) Results showed that biodiesel by supercritical

Page 2 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

3

transesterification can be scaled up resulting in a high purity of methyl esters (998)

and almost pure glycerol (964) obtained as a by-product The traditional acid and the

new two-step catalyzed processes for synthesis of biodiesel expressed as fatty acid

methyl ester (FAME) were comparatively studied to achieve an economic and practical

method for utilization of waste cooking oil (WCO) from Chinese restaurants [12] This

new two-step process showed the advantages of non-acidic wastewater high efficiency

low equipment cost and easy recovery of catalyst compared with the limitations of acidic

effluent no reusable catalyst and high cost of equipment in the traditional acid process

The reaction kinetics of acid-catalyzed transesterification of waste frying oil in excess

methanol to form fatty acid methyl esters (FAME) for possible use as biodiesel was

studied by Zhenga et al [13] In the presence of the large excess of methanol free fatty

acids present in the waste oil were very rapidly converted to methyl esters in the first few

minutes under the above conditions Little or no monoglycerides were detected during the

course of the reaction and diglycerides present in the initial waste oil were rapidly

converted to FAME The economic feasibilities of four continuous processes to produce

biodiesel including both alkali- and acid-catalyzed processes using waste cooking oil

and the standard process using virgin vegetable oil as the raw material were assessed by

Zhang et al [14] Although the alkalicatalyzed process using virgin vegetable oil had the

lowest fixed capital cost the acid-catalyzed process using waste cooking oil was

economically feasible overall providing a lower total manufacturing cost an attractive

after-tax rate of return and a lower biodiesel break-even price

Liu et al [15] studied the transesterification of soybean oil to biodiesel using CaO

as a solid base catalyst The reaction mechanism was proposed and the separate effects of

Page 3 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

4

the molar ratio of methanol to oil reaction temperature mass ratio of catalyst to oil and

water content were investigated The experimental results showed that a 121 molar ratio

of methanol to oil an addition of 8 CaO catalyst 65oC reaction temperature and 203

water content in methanol gave the best results the biodiesel yield exceeded 95 at 3 h

Demirbas [16] developed a non-catalytic biodiesel production route with supercritical

methanol that allows a simple process and high yield because of the simultaneous

transesterification of triglycerides and methyl esterification of fatty acids In the catalytic

supercritical methanol transesterification method the yield of conversion rises to 60ndash90

after the first minute Imahara et al[17] developed a non-catalytic biodiesel production

technology from oil and fat employing supercritical methanol The effect of thermal

degradation on cold flow properties was studied As a result it was found that all fatty

acid methyl esters including poly-unsaturated ones were stable at 270 oC17 MPa but at

350oC43 MPa they were partly decomposed to reduce the yield with isomerization from

cis-type to trans-type These behaviors were also observed for actual biodiesel prepared

from linseed oil and safflower oil which is high in poly-unsaturated fatty acids Ni and

Meunier [18] determined active and durable solid catalysts for the esterification of

palmitic acid (PA C16H32O2) dissolved in commercial sunflower oil with methanol

Contrary to the case of experiments realized at high dilution in solvents or in pure FFA

medium in which methanol is fully soluble a lack of full miscibility occurred in the

present case Azcan and Danisman [19] carried out a microwave assisted

transesterification of cottonseed oil in the presence of methanol and potassium hydroxide

(KOH) Parametric studies were conducted to investigate the optimum conditions

(catalyst amount reaction time and temperature) As a result a 7 min reaction time 333

Page 4 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

5

K temperature and 15 catalystndashoil ratio were obtained as optimum reaction parameters

for microwave heating Rathore and Madras [20] investigated the synthesis of biodiesel

from edible oil such as palm oil and groundnut oil and from crude non-edible oils like

Pongamia pinnata and Jatropha curcas in supercritical methanol and ethanol without

using any catalyst from 200 oC to 400oC at 200 bar The variables affecting the

conversion during transesterification such as molar ratio of alcohol to oil temperature

and time were investigated in supercritical methanol and ethanol Biodiesel was also

synthesized enzymatically with Novozym-435 lipase in the presence of supercritical

carbon dioxide The effect of reaction variables such as temperature molar ratio enzyme

loading and kinetics of the reaction was investigated for enzymatic synthesis in

supercritical carbon dioxide

In the transesterification reaction the forward reaction is only favored when an

excess of alcohol is used However it was found that when a large quantity of alcohol is

present the glycerol phase favors the alcohol phase and will not settle out [4] The main

alkali catalysts used are sodium hydroxide sodium ethoxide (methoxide) potassium

hydroxide and potassium ethoxide (methoxide) However it was found that sodium

hydroxide is only moderately soluble in ethanol and was also found to promote an

undesirable gel and emulsion formation during reaction [21] Methanol is usually used in

the production of biodiesel as it is more cost effective consequently there has been

substantial research conducted on the methanol-sodium hydroxide combination

Recognizing that methanol is relatively toxic its use has been restricted since

environmental considerations are given greater weight

Page 5 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

6

One common topic that occurs throughout most literature is the effect that water

and free fatty acids (FFA) have on both esterification and transesterification reactions

The fatty acids are problematic because they form soap with the metal ion from the alkali

catalysts which hinder separation decreases yield and leads to emulsion formation

Water has a two-fold effect the first when combined with soap and forming an emulsion

while the second occurs when the oil undergoes hydrolysis with the water forming fatty

acids Water was a more critical variable than was free fatty acids but more importantly

when both are present they acted synergistically to substantially decrease the yield of

esters [6] Washing with water is performed in order to purify the biodiesel by removing

catalyst residual glycerol and unreacted or partially reacted oils If the reaction mixture

was mixed for 20 minutes with pure glycerol and then 15 by weight of water was

added with a further 20 minutes of mixing washing becomes a lot simpler and requires

less total washes [22] This method also tended to remove all of the mono and di-

glycerides from the product phase However it was found that vigorous stirring before

all of the contaminates had been removed with glycerol caused emulsions to form that

were difficult to break [23] The objective of this work is to study the effect of different

parameters such as ethanol concentration KOH to WVO ratio and reaction time on the

biodiesel yield using waste soybean oil and investigating their effect on the density and

viscosity of the produced biodiesel

Page 6 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

7

2 MATERIALS AND METHODS

21 Materials

The waste vegetable oil (WVO) used in this work was soybean oil obtained from

local fast food restaurants The characteristics of the WVO are shown in Table 1 The

WVO was filtered in order to remove food residue and solid precipitate in the oil An oil

sample was heated and allowed to settle in order remove any water present in the oil In

the transesterification free fatty acids and water always produced negative effects since

the presence of free fatty acids and water causes soap formation consumes more catalyst

and reduces catalyst effectiveness The transesterification reaction performed in this work

was based on using potassium hydroxide (85) as catalyst and pure ethanol (999)

22 Design of Experiment

In order to study the effect of operating parameters on the biodiesel yield which

is the output response in this work a set of design experiments was performed Two

KOH concentrations (X1) two ethanol concentrations (X2) and two reaction times (X3)

were selected in order to perform the two-level factorial design These parameters were

selected at lower and upper levels as shown in Table 2 Each parameter is ranked as -1

and 1 at lower and upper levels respectively

23 Methods

In this work the concentration of KOH to oil ranged from 7 gl to 14 gl and

ethanol ranged from 30 to 40 vol of oil used The reaction was performed at 35oC

Two hundred milliliters of WVO were placed in a 300 ml flask the required amount of

KOH was separately dissolved in ethanol then added to the oil samples The samples

were kept under continuous mixing using a magnetic stirrer for 05 1 2 and 3 hours at

Page 7 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

8

35oC At the end of each run the samples were placed in separate funnels and were

allowed to settle for 24 hours two layers of glycerin and biodiesel were formed The two

layers were separated and each layer was heated to 80oC in order to evaporate the excess

ethanol The biodiesel samples were washed several times in order to remove any

impurities such as soap ethanol or KOH with the volume of glycerin and biodiesel being

recorded The specific density was measured for the biodiesel using a calibrated

picnometer The viscosity was then measured using a Saybolt viscosimeter Biodiesel

samples were analyzed at the Jordan Petroleum Refinery laboratory in order to determine

their characteristics The glyceride and ester content were tested using NMR (Bruker400

MHZ solvent CDCl3) at the chemistry department of Jordan University of Science and

Technology The biodiesel and glycerin yields are defined in Eq 1 and 2

Biodiesel Yield =fedethanolandWVOofvolumeTotal

producedBiodieselofVolumex100 helliphelliphelliphelliphelliphellip(1)

Glycerin Yield =fedethanolandWVOofvolumeTotal

producedGlycerinofVolumex100 helliphelliphelliphelliphelliphellip(2)

3 Results and discussions

31 Factorial design

A 23 complete factorial design was performed with the value of the operating

parameters shown in Table 2 These 8 tests include all possible combinations of X1 X2

and X3 The values of the biodiesel yield (Y) shown in Table 3 are the average values for

each test The complete factorial model that can be used to fit the data in Table 3 is

shown in Eq 3

Y= β0 + β1 X1 + β2 X2+ β3 X3 + β12 X1 X2 + β13 X1 X3

Page 8 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

9

+ β23 X2 X3 + β123 X1 X2 X3 (3)

where the parameters βi are responsible for the influence of the operating parameters Xi

on the response Y while βij and βijk are responsible for the possible interactions among

the operating variables i j and k and the effect of this interaction on the response Values

of these parameters can be determined by the least square method

Β = (XT X)-1 XTY helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip (4)

The matrix X represents the matrix of the operating parameters at different runs

The elements of the columns of X associated with the interaction terms X1X2hellipX1X2X3

are the products of the corresponding elements in the columns associated with X1X2 and

X3 The matrix XT X is an 8 x 8 symmetrical-square matrix with diagonal elements each

equal to 8 and off-diagonal elements each equal to zero Therefore (XT X)-1 is also an

8 x 8 symmetrical matrix square matrix with diagonal elements each equal to 18 and

off-diagonal elements each equal to zero Using the pseudo-level factors (ie -1 and +1)

along with Eq 4 matrix X and Y the least square fitted model of Eq 3 is

Y= 0736 - 0021 X1 - 0009 X2 - 0021X3 + 0003X1 X2 ndash 0003 X1 X3

- 0006 X2 X3 - 0009 X1 X2 X3 (5)

The analysis of variance for the Y model is shown in Table 4 Since there were no

replicates in the experiments an estimate of the error was selected as the SS of the three

factor interaction X123 which is equal to 00006 The F values are determined and

shown in Table 4 Using α of 05 and by comparing the Fcalc from the ANOVA table vs

Ftable (F05 1 8 = 532) it can be noted that the X1 X2 and X3 are highly significant

Meanwhile the interactions X12 X13 X23 are not significant however the three

Page 9 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

10

factor interaction X123 is significant since it is a presentation of the validity of the

model It can be observed that interaction exists among the three parameters and that the

parameters or factors do not operate independently on the response

32 Estimation of the main effects

Statistically the main effect of Xi is estimated from the difference between the

average high and low factor level responses as shown in Eq 6

Main effect of Xi = )( runsfactorialofnumbertheHalf

levellowatresponselevelhighatresponsesum summinus helliphelliphelliphelliphellip(6)

From Table 3 the main effect of increasing the concentration of KOH from 9 gl

to 14 gl over all levels of ethanol concentration and time is to decrease the biodiesel

yield (-015) Increasing the reaction time from 30 minutes to 180 minutes produces

similar results (-015) Also increasing ethanol concentration from 30 to 40 vol has

the least effect in decreasing the biodiesel yield ( -005)

33 Effect of KOH oil ratio

Figure 1 shows the effect of catalyst KOH to WVO ratio on the biodiesel yield It

is evident that as the ratio of KOH to WVO increased from 7 gl to 14 gl the biodiesel

yield decreased from 785 to 67 for both 35 and 40 vol ethanol This drop in

biodiesel yield can be explained by the fact that a side reaction may take place such as

soap formation as shown in Eqs 7 and 6 This means that proper concentration of

ethanol and KOH are required to give the maximum biodiesel production Another

disadvantage is the partial saponification reaction which produces soap Soap lowers the

yield of esters and makes the separation of ester and glycerol difficult [24]

Using 30 vol ethanol and KOH concentration from 7 gl to 8 gl neither

biodiesel nor glycerin were formed at 12 gl the biodiesel yield was the highest (ie

Page 10 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

11

785 vol) This is equivalent to using 35 or 40 ethanol with 7 gl KOH to WVO

ratio

Soap may be formed by an accompanying reaction during or subsequent to the

transesterification process

O ORC-OC2H5 + H2O ----------gt RC-OH + C2H5OH (7)

Ethyl Ester Water Fatty Acid Ethanol

This is an equilibrium reaction and any base will neutralize the acid formed

removing it and forcing the reaction to the right Also the reaction product of the base

and acid is an undesired substance (a soap which is an emulsifying agent)

O ORC-OH + KOH ------gt RC-OK + H2O helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip(8)

Fatty Acid A Base SaltSoap Water

34 Effect of ethanol and reaction time

To study the effect of ethanol percentage on the biodiesel yield density viscosity

and glycerin yield the ratio of KOH to oil was set at 12 gl Figures 2 and 3 show the

effect of excess ethanol and reaction time on biodiesel density and viscosity It is clear

from the figures that the density of biodiesel decreased significantly as the reaction goes

to completion But the excess ethanol does not have much impact on the biodiesel

density where it decreases from 09 to around 086 for all ethanol percentages The

change in biodiesel viscosity can be observed in Figure 3 where the biodiesel viscosity

changes from 53 to 78 21 and 18 Cst for 30 35 and 40 vol ethanol

respectively

Page 11 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

12

Figure 4 illustrates the effect of ethanol percentage and reaction time on the

glycerin yield It is apparent that the glycerin yield increased with the increase in both

ethanol percentage and reaction time Maximum yield of glycerin was around 21 for

both 35 and 40 vol ethanol while 30 vol ethanol shows a lower glycerin yield

around 185 This was confirmed with biodiesel yield data shown in Figure 5 where

the maximum biodiesel production was achieved at 30 ethanol with a yield value of

785 vol From Figures 4 and 5 one can observe that as the ethanol percentage

increases the biodiesel yield decreases This is true because an excess ethanol will shift

the transesterification reaction backwards where the ester produced in the reaction also

reverses and reverts into fatty acids On the other hand below 30 vol ethanol there

was no conversion and the formation of the two layers (ie glycerin and biodiesel) were

not observed unless the ethanol percentage was 30 vol or higher

Regarding the reaction time samples collected at a time of 10 and 20 minutes

were kept for 24 hours and no layer formation was observed The highest biodiesel yield

was obtained at 30 minutes as the reaction proceeds a drop in biodiesel yield was

observed as seen in Figure 5 and an increase in glycerin yield as shown in Figure 4 The

reaction approaches steady state after 3 hours where the yield remains constant Adding

water stops the transesterification reaction by drawing much of the water-soluble ethanol

KOH and glycerol out of the biodiesel and into the glycerol layer This can help reduce

soap inhibit emulsion formation and make further washing easier

35 Biodiesel Characteristics

The Biodiesel layer was heated to 80oC in order to evaporate the excess ethanol

then it was washed three times with water in order to remove any water-soluble ethanol

Page 12 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

13

KOH and glycerol Analysis for ester content mon- di- and tricglycerde were performed

using NMR (Bruker400 MHZ solvent CDCl3) From 1H spectrum shown in figure 6

and using spectrum references (CDCl3) [25] it is clear that mono- and diglyceride is not

present While the spectrum shows no presence of triglyceride it might be present in

traces under the peak (40 ndash 43) and (51 ndash 54) another test was performed using the

spectrophotometer at a wavelength of 505 nm where the triglyceride was detected with a

concentration of 025 wt which is an indication of the completion of the

transesterfication reaction These results indicate that the ester content is above 99 wt

which meet the ASTM and EN standards

The produced biodiesel additional characteristics were tested at the Jordan

Petroleum Refinery laboratory Results are shown in Table 5 vs literatures values [26

27] From these results one observes that the biodiesel properties are within the literature

values as described below

bull Flash Point ndash Minimum temperatures are required for proper safety and handling

of fuels Note that the biodiesel component must meet flash point criteria prior to

blending for the purpose of assuring that the biodiesel component does not

contain ethanol Results indicate that the flash point (126oC) is 1oC above the

allowable value (125oC)

bull Kinematic Viscosity ndash Affects injector lubrication and fuel atomization

Biodiesel fuel blends generally have improved lubricity however their higher

viscosity levels tend to form larger droplets on injection which can cause poor

combustion and increased exhaust smoke In this work the kinematic viscosity

was around 78 mm2sec which is close to the biodiesel literature value

Page 13 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

14

bull Sulfur ndash Levels in fuel are regulated by various governmental agencies to assure

compatibility with emission standard requirements Biodiesel blends may not

exceed the applicable maximum sulfur levels as defined for petroleum diesel The

value of total sulfur for the biodiesel produced was 0005 wt which is much

lower than the literature value of 005 wt

bull Cetane Number ndash A measure of the fuelrsquos ignition and combustion quality

characteristics biodiesel blend stock typically has a higher minimum cetane level

than that of petroleum diesel Fuels with low cetane numbers will cause hard

starting rough operation noise and increased smoke opacity The value of cetane

number is 501 and is within the literature value (48-65)

bull Carbon Residue ndash Testing of residue is intended to provide some indication of

the extent of carbon residue that results from the combustion of a fuel The value

of the carbon residue for the biodiesel produced was 00594 wt which is lower

than the literature value of 015 wt

bull Acid Number ndash Is a measure of acids in the fuel These acids emanate from two

sources (i) acids utilized in the production of the biodiesel that are not completely

removed in the production process and (ii) degradation by oxidation For

biodiesel blends the acid number will change as a result of the normal oxidation

process over time The acid number for the biodiesel produced was 00078

which is lower than the literature value of 008

Page 14 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

15

Conclusions

Results obtained in this study lead to the following conclusions

1 The main effect of operating variables shows that ethanol concentration has the least

effect in decreasing biodiesel yield (-005) while KOH concentration and time has a

higher effect in decreasing biodiesel yield (-015)

2 In this work biodiesel produced from WVO of soybean oil meet the ASTM standard

properties

3 The highest yield of biodiesel produced was 785 vol using KOH to WVO ratio of

12 gl and 30 vol ethanol

4 An increase in KOH to WVO ratio and ethanol concentrations resulted in a reduction

of biodiesel yield due to side reactions

References

[1] G Knothe Historical perspectives on vegetable oil-based diesel fuels Ind Oils

2001 12 1103ndash1107

[2] G Knothe The history of vegetable oil-based diesel fuels The Biodiesel Handbook

(Knothe G Krahl J Van Gerpen J eds) Champaign IL AOCS Press 2005

[3] C Sharp S Howell J Jobe The Effect of Biodiesel Fuels on Transient Emissions

from Modern Diesel Engines Part II Unregulated Emissions and Chemical

Characterization Technical Paper 2000-01-1968 Warrendale PA SAE 2000

[4] U Schuchardt R Sercheli R M Vargas Transesterification of vegetable oils a

review J Braz Chem Soc 1998 9(3) 199ndash212

Page 15 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

16

[5] B Freedman E H Pryde T L Mounts Variables affecting the yields of fatty esters

from transesterified vegetable oils J Am Oil Chem Soc 1984

[6] R C Lago R R Szpiz F H Jablonka D A Pereira L Hartman Extraction and

transesterification of vegetable oils with ethanol Oleacuteagineux 1985 40 (3) 147ndash151

[7] A Froumlhlich B Rice G Vicente The conversion of waste tallow into biodiesel grade

methyl ester 1st World Conference and Exhibition on Biomass for Energy and

Industry Proceedings of the Conference held in Seville (Spain) 2000 695ndash697

[8] S Gryglewicz Rapeseed oil methyl esters preparation using heterogeneous catalysts

Bioresour Technol 1999 70 (3) 249ndash253

[9] G Vicente A Coteron M Martiacutenez J Aracil Application of the factorial design of

experiments and response surface methodology to optimize biodiesel production Ind

Crops Prod 1998 8 29ndash35

[10] Y Shimada Y Watanabe T Samukawa A Sugihara H Noda H Fukuda Y

Tominag Conversion of vegetable oil to biodiesel using immobilized Candida

antarctica lipase J Am Oil Chem Soc 1999

[11] JMN Kasteren AP Nisworo A process model to estimate the cost of industrial

scale biodiesel production from waste cooking oil by supercritical transesterification

Resour Conserv Recycling 2007 50 442ndash458

[12] YWang S Ou Liu F Xue S Tang Comparison of two different processes to

synthesize biodiesel by waste cooking oil J Mol Catal A Chem 2006 252 107ndash

112

[13] S Zhenga M Katesb MA Dube DD McLeana Acid-catalyzed production of

biodiesel from waste frying oil Biomass Bioenergy 2006 30 267ndash272

Page 16 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

17

[14] Y Zhang MA Dub DD McLean M Kates Biodiesel production from waste

cooking oil 2 Economic assessment and sensitivity analysis Bioresour Technol

2003 90 229ndash240

[15] X Liu H He Y Wang S Zhu X Piao Transesterification of soybean oil to

biodiesel using CaO as a solid base catalyst Fuel 2008 87 (2) 216-221

[16] A Demirbas Comparison of transesterification methods for production of

biodiesel from vegetable oils and fats Energy Convers Manage 2008 49(1) 125-

130

[17] H Imahara E Minami S Hari S Saka Thermal stability of biodiesel in

supercritical methanol Fuel 2008 87(1) 1-6

[18] J Ni F C Meunier Esterification of free fatty acids in sunflower oil over solid

acid catalysts using batch and fixed bed-reactors Appl Catal 2007 333(1) 122-130

[19] N Azcan A Danisman Alkali catalyzed transesterification of cottonseed oil by

microwave irradiation Fuel 2007 86(17-18) 2639-2644

[20] V Rathore G Madras Synthesis of biodiesel from edible and non-edible oils in

supercritical alcohols and enzymatic synthesis in supercritical carbon dioxide Fuel

2007 86(17-18) 2650-2659

[21] Y Ali M A Hanna Physical properties of tallow ester and diesel fuel blends

Bioresour Technol 1994a 47 131-4

[22] C L Peterson D L Reece J C Thompson S M Beck C Chase Ethyl ester of

rapeseed used as a biodiesel fuel - a case study Biomass Bioenergy 1995 10 331-

336

Page 17 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

18

[23] F Ma L Clements A Hanna Biodiesel fuel from animal fat Ancillary studies

on transesterification of beef tallow Ind Eng Chem Res 1998 37 3768-3771

[24] F Ma M A Hanna Biodiesel production a review Biosour Technol 1999

70(1) 1-15

[25] F Jin K Kawasaki H Kishida K Tohji Z ZhouT Moriya and H Enomoto

NMR spectroscopic study of methanolysis reaction of vegetable oil Fuel 2007 86

1201-1207

[26] National Biodiesel Board ldquoSpecification for Biodiesel (B100)rdquo December (2001)

[27] T K Shaine Biodiesel Handling and Use Guidelinesrdquo Report No NRELTP-

580-30004 National Renewable Energy Laboratory Golden (C) September (2001)

Page 18 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

19

Figure 1 Effect of KOH to WVO ratio on biodiesel yield

Figure 2 Effect of ethanol concentration and reaction time on biodiesel specific density

Figure 3 Effect of ethanol concentration and reaction time on biodiesel viscosity

Figure 4 Effect of ethanol concentration and reaction time on glycerin yield

Figure 5 Effect of ethanol concentration and reaction time on biodiesel yield

Figure 6 1 H NMR spectrum of Biodiesel

Page 19 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

Table 1 Waste vegetable oil properties used in this work

Properties ValueDensity (gl) 09Viscosity (Cst) 51Saponification value (mg KOHg oil) 2047Iodine Value (IV) 1397Free fatty acids () 057Moisture content () 004Refractive index (RI) 146Peroxide value 82Color 20 red

50 yellow

Page 20 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

Table 2 Values of operating parameters at both levels

Operating Parameters -1 1 X1 (KOH) 9 gl 14 glX2 (Ethanol) 30 Vol 40 Vol X3 (Time) 30 min 180 min

Page 21 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

For Peer Review

Table 3 Analysis of variance for the Y model

Source DF Sum square(SS)

Adjusted SS Adjusted Mean square (MS)

Fcalc

X1 1 00036 00036 00036 5100X2 1 00006 00006 00006 857X3 1 00036 00036 00036 5100X12 1 00001 00001 00001 140X13 1 00001 00001 00001 140X23 1 00003 00003 00003 428X123 1 00006 00006 00006 857Error 8 00006 00006 000007

Table 4 Experimental results of the factorial design for Biodiesel yield

Y(yield ) X1 X2 X3

079 -1 -1 -1 073 1 -1 -1 076 -1 1 -1 075 1 1 -1 075 -1 -1 1071 1 - 1 1073 -1 1 1069 1 1 1

aΣ+ 288 293 288aΣ- 303 298 303Difference effect -015 -005 -015aΣ+ corresponds to the run of the responses at high Xi while Σ- corresponds to the run of the responses at low Xi

Page 22 of 29

Wiley-VCH

European Journal of Lipid Science and Technology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960