Accepted Manuscript

Preparation and kinetics study of biodiesel production from waste cooking oil using
new functionalized ionic liquids as catalysts

Zahoor Ullah, M. Azmi Bustam, Zakaria Man, Amir Sada Khan, Nawshad
Muhammad, Ariyanti Sarwono

PII: S0960-1481(17)30712-7
DOI: 10.1016/j.renene.2017.07.085
Reference: RENE 9060

To appear in: Renewable Energy

Received Date: 5 December 2016

Revised Date: 18 July 2017
Accepted Date: 20 July 2017

Please cite this article as: Ullah Z, Bustam MA, Man Z, Khan AS, Muhammad N, Sarwono A,
Preparation and kinetics study of biodiesel production from waste cooking oil using new functionalized
ionic liquids as catalysts, Renewable Energy (2017), doi: 10.1016/j.renene.2017.07.085.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT
1 Manuscript for-Renewable Energy Journal
2
3 Preparation and kinetics study of biodiesel production from waste cooking oil using new
4 functionalized ionic liquids as catalysts
5
6 Zahoor Ullah*ab, M. Azmi Bustama, Zakaria Mana, Amir Sada Khana, Nawshad Muhammadc,

PT
7 and Ariyanti Sarwonoa

a
8 Centre of Research in Ionic Liquid, Department of Chemical Engineering Universiti

RI
9 Teknologi PETRONAS, Tronoh 31750, Malaysia.

SC
b
10 Department of Chemistry, The Balochistan University of IT, Engineering and Management
11 Sciences (BUITEMS) , Takatu Campus, Quetta-87100, Pakistan
c
12 Interdisciplinary Research Centre in Biomedical Materials, COMSATS Institute of

U
13 Information Technology, Lahore, Pakistan.
AN
14

15 *Corresponding author: Zahoor Ullah; zahoor7979@yahoo.com

M

16
D

17 Abstract
18 In this work, 1,4-sultone and benzimidazolium-based ionic liquids (ILs) with four different
TE

19 anions were synthesized, and their structures were confirmed by nuclear magnetic resonance
20 (NMR) and elemental analysis (CHNS). The acidity of the synthesized ILs was studied using
EP

21 Hammett acidity function and COSMO-RS. The waste cooking oil was used as a raw
22 material for biodiesel production and their different fatty acids were determined by gas
C

23 chromatography coupled with flame ionization detector (GC-FID). These four ILs, as
24 catalysts, were screened and comparatively IL 3-methyl-1-(4-sulfo-butyl)-benzimidazolium
AC

25 trifluoromethanesulfonate [BSMBIM][CF3SO3] was selected for further detailed optimization

26 study. This IL experimental efficiency results supported the Hammett acidity function and
27 COSMO-RS study. The catalyst performance was studied and optimised the different
28 parameters. The catalyst efficiency was studied in one and two-step reactions.
29 [BSMBIM][CF3SO3] as a catalyst showed the esterification of waste cooking oil up to
30 78.13% in a single step reaction. Potassium hydroxide was used in the second step to trans-
31 esterify the waste cooking oil up to 94.52%. The catalyst was reused for seven times with
32 high-yield production. The obtained biodiesel was characterised by GC, NMR, FTIR,
1
 ACCEPTED MANUSCRIPT
33 thermogravimetric (TGA) and their physicochemical properties were compared with the
34 already established standards. The kinetic study of this transesterification reaction was
35 evaluated and followed the first-order reaction mechanism.
36 Keywords: Benzimidazole; Waste cooking oil; Biodiesel; Gas chromatography; Kinetic
37 study.
38

PT
39 1. Introduction
40 The depletion of world petroleum reservoirs and the increasing environmental anxieties have

RI
41 motivated the search for alternative fuels [1]. Biodiesel is one of the well-known renewable
42 biofuel replacements for the mineral diesel. It is fitted well with the green energy standard

SC
43 because of having lower content of sulphur leading to decreased emissions of SOx [2, 3]. At
44 present, the biodiesel is commonly produced by the transesterification method from different
45 renewable biological sources such as algae oil, animal fats and vegetable oils with alcohol
46

U
(commonly methanol) in the presence of catalysts, which shifts the triglyceride's fatty acids
AN
47 into fatty acid methyl esters and glycerol as by-product as shown in Fig. 1 [4-7].
O
O
CH2-O-C-R1
M

CH3O-C-R1
O O CH2-OH
catalyst
CH-O-C-R2 + 3CH3OH
D

CH3O-C-R2 + CH2-OH (1)

O
O
CH2-OH
TE

CH2-O-C-R3
CH3O-C-R3

48 Triglycerides Methanol Methyl Ester Glycerol (byproduct)

EP

49
50 Fig.1. The transesterification scheme of triglycerides (vegetable oil), with methanol to
C

51 produce methyl esters.

AC

52
53 Diasakou et al, [8] suggested the overall transesterification reaction to be divided into three
54 consecutive reversible reactions with the formation of intermediates of di-glycerides (DGs)
55 and mono-glycerides (MGs). The stepwise reactions are:
56
TG + M eO H DG + ME
57 (2)
58 D G + M eO H M G +M E (3)
59 (4)
M G + M eO H GL + ME

2
 ACCEPTED MANUSCRIPT
60 Many kinds of vegetable oils such as palm, rapeseed, soybean, coconut, sunflower and canola
61 oil etc., are used as feedstocks for biodiesel production, but edible vegetable oils are not
62 favoured as raw materials [9]. Mostly, the edible oils are expensive for biodiesel production
63 process, as the raw material contributes up to 75% of the total production cost. So, the low-
64 cost non-edible oils and waste cooking oils as feedstock’s utilization are very important in
65 developing countries [10, 11]. Therefore, the use of waste cooking oil instead of virgin oil as

PT
66 a raw material for biodiesel production is a substantial way, not only to reduce the high cost
67 of biodiesel but to solve the waste-disposal problem as well. [12, 13]. Generally, biodiesel are

RI
68 produced by conventional alkali-homogeneous catalysts, [14, 15] but the waste cooking oil
69 contains a significant amount of free fatty acids (FFA) and water, which make them

SC
70 unfeasible for alkali-homogeneous catalyst's catalysed processes. Moreover, it leads to the
71 soap formation and results in separation problems [16, 17].
72 Recently, ionic liquids (ILs) have been developed as desirable substitutes for volatile,
73

U
flammable and toxic solvents, which are the main cause of environmental pollution. ILs are
AN
74 liquid organic salts, mostly composed of cations and anions. They have typical properties
75 such as negligible vapour pressure, high thermal stability, non-flammability, and good
M

76 solvation properties making them green solvents. In view of this feature, ILs are considered
77 as designer solvents for any specific application and has diverted the attention of researchers
D

78 due to their adjustable physicochemical properties [4, 18-23]. Liang et al, [24] cited some
79 ILs, which are efficient for biodiesel production from different feedstocks, and ILs with
TE

80 sulfonic acid functionalized group owned high activities for acid catalysed reactions. Han et
81 al, reported biodiesel preparation from waste cooking oil using novel Bronsted acidic ILs
EP

82 with an alkane sulfonic (-SO3H) group as catalysts with 93.5% yield [25]. In addition, such
83 types of ILs from the reaction mixture were easily separated by decantation and had good
C

84 reusability as well.
85 Due to the acidic nature of waste cooking oil, acidic ILs are required for their esterification to
AC

86 biodiesel/fatty acid alkyl esters. There is a limited amount of literature on acid ILs for
87 esterification of waste cooking oil, and most studies involve basic or neutral ILs used for
88 transesterification of triglycerides found in virgin cooking oil. However, according to the best
89 of author’s knowledge, there is no kinetics evaluation information in the literature using IL as
90 catalysts. The aim of this study is to synthesise dual-functionalized acidic ILs with different
91 anions as catalysts for the synthesis of biodiesel from waste palm cooking oil and to optimize
92 their various factors affecting the waste cooking oil conversion into biodiesel by

3
 ACCEPTED MANUSCRIPT
93 esterification/transesterification method. To determine their reaction mechanism, a detail
94 kinetic study is also described.
95
96 2. Experimental
97 All the chemicals were obtained from Sigma-Aldrich (Malaysia), Merck (Germany), and all
98 these chemicals were used as received.

PT
99
100 2.1. Ionic liquids synthetic procedure

RI
101 1,4-butanesultone was dissolved in toluene, in a round-bottom flask, followed by adding a
102 stoichiometric amount of 1-methyl-benzimidazole under continuous stirring at room

SC
103 temperature for 72 h. The resultant white solid precipitate (zwitterion) was filtered and
104 washed with diethyl ether and then dried under the vacuum at 100oC for 5h. The dried white
105 solid (zwitterion) was re-dissolved in distilled water, and stoichiometric amount of the
106

U
methanesulfonic acid, trifluoromethanesulfonic acid, trifluoroacetic acid and sulphuric acid
AN
107 was added dropwise at room temperature and was stirred. After 30 minutes, the reaction
108 temperature was increased to 60oC and stirred it for 10 h. Finally; deep yellow viscous ILs
M

109 obtained were washed with diethyl ether and dried in the vacuum.
110
D

111 2.2. Characterization

112 The structures of ILs were confirmed by NMR (Bruker, Advance 500 MHz spectrometer),
TE

113 and CHNS elemental analysis (EA-1110, CE Instruments).

114 2.3. Hammett acidity of the ILs

EP

115 Bronsted acidity of the synthesized ILs were calculated using the reported procedure [26] for
116 the determination of Hammett acidity functions, using UV-visible spectrometer (Shimadzu,
C

117 UV-1800). The Hammett function (Ho) was measured using equation (5):
AC

+
118 = (1) + ([ ]/[ ] ) (5)

119 2.4. Biodiesel production process

120 The collected waste cooking oil was purified and dried from moisture. The required amount
121 of waste cooking oil was taken in a three-necked flask equipped with a reflux condenser and
122 cool-water chiller. The waste cooking oil was heated at 100oC for 30 min to remove the
123 traces of water and then the mixture of IL and methanol was added. When the reaction was
124 completed according to the designed reaction parameters, the reaction mixture was

4
 ACCEPTED MANUSCRIPT
125 transferred into the separating funnel for phase separation. The biodiesel was collected from
126 the upper layer as the lower layer was contained of IL mixed with glycerol and methanol. The
127 IL was reused after removing the glycerol by adding ethyl acetate and centrifuge. The IL
128 esterification/transesterification was applied and obtained biodiesel was up to some
129 percentage; KOH was applied for the transesterification process to obtain a high yield. The
130 final product was subjected to GC analysis and then characterized with NMR, FT-IR and

PT
131 TGA.

132 3. Results and discussion

RI
133 In this study, six ILs were proposed to prepare by adding respective acids, i.e.
134 trifluoromethanesulfonic acid, methanesulfonic acid, trifluoroacetic acid, sulfuric acid, acetic

SC
135 acid and formic acid, into the zwitterions from 1,4-butane sultone and methyl-benzimidazole.
136 While conducting the reaction, it has been observed that four acids (comparatively strong)

U
137 could prepare the desired products. While the two weak acids, i.e. acetic acid and formic acid
138 were unable to convert the prepared zwitterions into the proposed ILs which might be due to
AN
139 low availability of H+ for weak sultone conjugate base that plays a key role in the synthesis.
140
M

141 3.1. NMR and CHNS analysis

142 The synthesized ILs NMR and CHNS results are listed below.
D

143 3-methyl-1-(4-sulfo-butyl) benzimidazolium methanesulfonate [BSMBIM][CH3SO3]

TE

1
144 H NMR δ (500 MHz, D2O): 1.66-1.73 (m, 2H), 1.99-2.05(m, 2H), 2.69 (s, 3H), 2.83-2.86 (t,
145 2H), 3.97 (s, 3H), 4.40-4.43 (t, 2H), 7.54-7.56 (d, 2H), 7.70-7.77 (d, 2H), 9.16 (s, 1H).
13
146 C NMR δ (500 MHz, D2O): 21.1, 27.2, 32.9, 38.4, 46.5, 50.0, 112.9, 126.7, 131.0, 131.9,
EP

147 141.1.

148 CHNS: C13H20N2O6S2, Found (%): C=37.01, H=5.73, N=7.59, S=17.68.

C

149 Calculated (%): C=42.73, H=5.79, N=7.67, S=17.55.

AC

150 3-methyl-1-(4-sulfo-butyl) benzimidazolium trifluoro methanesulfonate

151 [BSMBIM][CF3SO3]
1
152 H NMR δ (500 MHz, D2O): 1.61-1.67 (m, 2H), 1.92-2.98 (m, 2H), 2.77-2.80 (t, 2H), 3.90(s,
153 3H), 4.32-4.35 (t, 2H), 7.44-7.46 (d, 2H), 7.60-7.68 (d, 2H), 9.00 (s, 1H).
13
154 C NMR δ (500 MHz, D2O): 21.2, 27.2, 32.9, 46.5, 50.0, 112.9, 120.9, 126.7, 131.0, 132.0,
155 141.1.

156 CHNS: C13F3H17N2O6S2, Found (%): C=37.30, H=4.29, N=6.80, S=15.35. Calculated (%):

5
 ACCEPTED MANUSCRIPT
157 C=37.23, H=4.33, N=6.68, S=15.29.

158

159 3-methyl-1-(4-sulfo-butyl) benzimidazolium trifluoroacetate [BSMBIM][CF3CO2]

1
160 H NMR δ (500 MHz, D2O): 1.65-1.71 (m, 2H) 1.98-2.04 (m, 2H), 2.81-2.84 (t, 2H), 3.96 (s,
161 3H), 4.39-4.42 (t, 2H), 7.55-7.56 (d, 2H), 7.71-7.76 (d, 2H), 9.15 (s, 1H).

PT
13
162 C NMR δ (500 MHz, D2O): 21.1, 27.1, 32.8, 46.4, 49.9, 112.7, 116.8, 126.6, 130.8, 131.7,
163 141.0, 161.5.

RI
164 CHNS: C14F3H17N2O5S, Found (%): C=43.99, H=4.61, N=7.39, S=7.73. Calculated (%):

SC
165 C=43.86, H=4.73, N=7.31, S=8.36.

166 3-methyl-1-(4-sulfo-butyl) benzimidazolium hydrogensulfate [BSMBIM][HSO4]

U
1
167 H NMR δ (500 MHz, D2O): 1.65-1.71 (m, 2H), 1.97-2.03 (m, 2H), 2.81-2.84 (t, 2H), 3.95 (s,
AN
168 3H), 4.38-4.41 (t, 2H), 7.50-7.52 (d, 2H), 7.66-7.74 (d, 2H), 9.15 (s, 1H).

13
169 C NMR δ (500 MHz, D2O): 21.2, 27.2, 32.9, 46.5, 50.0, 112.9, 126.8 (126.8), 131.1, 132.1,
M

170 141.2.

171 CHNS: C12H18N2O7S2, Found (%): C=39.98, H=7.02, N= 7.11, S=16.20. Calculated (%):
D

172 C=40.72, H=6.05, N=7.31, S=16.72.

TE

173 3.2. Sigma profile study

174 The effect of structure variations of the anions can also be observed from their surface charge
EP

175 distributions and σ-profiles. In COSMO-RS, the interaction energy between two segments is
176 expressed by two terms: a misfit and hydrogen bonding interaction terms. The strength of
C

177 hydrogen bonding between compounds is higher when both compounds have a higher
178 fraction of donor (σ >-0.0084 e/Å2) and acceptor segments (σ <0.0084 e/Å2). As it can be
AC

179 seen from Fig. 2, the anions have major peaks in the hydrogen bond acceptor regions (σ > 1
180 e/nm2). Moreover, HSO4- anion shows a third minor peak in the hydrogen bond donor region
181 (σ < -1 e/nm2) due to the hydrogen attached to one of the four oxygen atoms. The peaks in the
182 non-polar regions (-1 and 1 e/nm2) are due to the CF3 group. As consequences, the presence
183 of noticeable peaks in the hydrogen bond acceptor region (σ > + 1) of the σ-profile of these
184 ILs indicates that all of them show hydrogen-bond acceptor capacity. Therefore, from this
185 analysis of σ-profile, it can be predicted that ILs with hydrogen-bond acceptor groups (which

6
 ACCEPTED MANUSCRIPT
186 means basic character) might be shifted to the strong hydrogen-bond acceptor region (to the
187 right) compared to the weak hydrogen-bond acceptor. Moreover, these anions are derived
188 from strong acids which show slight difference in strength for its conjugate-base. It is known
189 that stronger acid results in weak conjugate base and weak acid results vice versa [27].
190 Hence, CF3SO3- anion is derived from strong acid so its conjugate base peak appeared on the
191 far-left hand side of hydrogen-bond acceptor region in comparison to other conjugate based

PT
192 anions as shown in sigma profile of COSMO-RS predictions. Similar observation was also
193 found in the literature [28]. This shifting of CF3SO3- anion, identify it as a weakest conjugate

RI
194 base thereby confer more acidity to it as compared other studied anions. The predication also
195 shows correlation with experimentally measured Hammett acidity (Table 1), and moreover

SC
196 similar trend in catalytic conversion of waste cooking oil into biodiesel was observed as
197 given in the following section.

30

U
AN
Cation
25
CH3SO3
HSO4
M

20
CF3CO2
CF3SO3
15
p(σ)

10
TE

5
EP

-3 -2 -1 0 1 2 3
σ(e/nm2)
C

198
AC

199 Fig. 2: the studied ILs sigma-profile computed by COSMO-RS.

200
201 3.3. Hammett acidity
202 The acidity of ILs greatly affected the catalytic conversion during biodiesel production;
203 therefore, Hammett acidity of the ILs was investigated in methanol. The Hammett acidity
204 values for ILs are listed in Table 2 and shown graphically in Fig. 3. In these ILs the cation is
205 the same but having different anions. The Ho values show that the IL ([BSMBIM][CF3SO3])

7
 ACCEPTED MANUSCRIPT
206 having the strong acidity which was expected, lower the value of Ho stronger will be there
207 acidity.
208
209
210 Table 2: Hammett acidity functions calculation of the acidic benzimidazolium based ILs.
A I HI pka + log Ho
([I/[H+]s)

PT
4-Nitroaniline 1.042 100 0
[BSMBIM][CH3SO3] 0.537 53.7 46.3 1.054393295 1.054
[BSMBIM][HSO4] 0.538 53.8 46.2 1.0561403 1.056

RI
[BSMBIM][CF3CO2] 0.575 57.5 42.5 1.121278915 1.121
[BSMBIM][CF3SO3] 0.526 52.6 47.4 1.035207402 1.035
211 Note: A = Absorbance

SC
212 I = It is the molar concentration of unprotonated form of the indicator in the solvent.
213 HI = It is the molar concentration of protonated form of the indicator in the solvent.
214 Ho = Hammett acidity function.

U
AN
1.0
[BSMBIM][CF SO ]
M

3 3

[BSMBIM][CH SO ]
3 3

[BSMBIM][CF CO ]
0.8 3 2

[BSMBIM][HSO ]
4
D

Dye
Absorbance

0.6
TE

0.4
EP

0.2
C

0.0
300 350 400 450 500
AC

Wave length (nm)

215
216 Fig 3: Apsorption spectra of various synthesised acidic ionic liquids recorded in 4-
217 nitroaniline. The different lines represent, [BSMBIM][CH3SO3] (dark blue),
218 [BSMBIM][CF3SO3] (green), [BSMBIM][CF3CO2] (purple), [BSMBIM][HSO4] (dark red).
219
220 3.4. Characterization of waste cooking oil

8
 ACCEPTED MANUSCRIPT
221 The various physicochemical properties of waste cooking oil were determined by the earlier
222 established methods which are published in our previous paper [5] and the different fatty
223 acids were determined by GC-FID using AOAC (969.33) [29] procedure and compared with
224 the available GC standards. All the determined fatty acids are listed in Table 3.
225
226 Table 3 : Fatty acids (FFA) in waste palm cooking oil.

PT
Peaks no Retention time (min) Name of FFAs
1 2.1626 Caprylic Acid

RI
2 2.930 Capratic Acid
3 4.3296 Lauric Acid

SC
4 6.4342 Myristic Acid
5 9.1211 Palmitic Acid
6 9.5046 Palmitoleic Acid
7 11.7850
U Stearic Acid
AN
8 12.2178 Oleic Acid

9 12.8398 Linoleic Acid

M

10 13.9250 Linolenic Acid

D

11 14.0778 Arachidic Acid

12 15.968 Behenic Acid
TE

13 16.6668 Erucic Acid

14 17.4331 Lignoceric Acid

EP

227
C

228 3.5. Optimization of transesterification reaction parameters of 1,4-sultone

229 methylbenzimidazolium based ionic liquids
AC

230 The catalytic performance of 1,4-sultone methylbenzimidazolium based four different acidic
231 ILs were investigated for biodiesel production from WCO by the transesterification reaction.
232 The reaction was run in a three-necked round-bottom glass reactor assembled with a reflux
233 condenser, cooled water chiller and heating plate. The various factors such as the molar ratio
234 of alcohol to oil, reaction temperature, reaction time, agitation speed and catalyst
235 concentration, which affected the transesterification reaction were measured and are describe
236 in the following.

9
 ACCEPTED MANUSCRIPT
237 Initial screening of the four 1,4-sultone methylbenzimidazolium based ILs were used to
238 choose a better one for the further detailed optimization. Their results are summarized in
239 Table 4. In all these ILs, comparatively 3-methyl-1-(4-sulfo-butyl)-benzimidazolium
240 trifluoromethanesulfonate ([BSMBIM][CF3SO3]) showed a good performance. The ILs
241 acidity and structure had enormous effects on the WCO conversion into biodiesel. It is
242 already understood that the nature of the anion effect the catalytic activity of such SO3H-

PT
243 functionalized ILs, [30] which is mostly related to the Bronsted acidic strength of the anion.
244 Being a stronger organic acid; trifluoromethanesulfonic acid could ionize a proton quickly

RI
245 because of the fluorine atoms' negative induction effect between the CF3 and SO3H group.
246 The proton then combines with the triglyceride carbonyl group to form a carbocation

SC
247 intermediate, which reacts through the nucleophilic substitution reaction with methanol and
248 finally forms one molecule of diglyceride [31].

U
249 Additionally, the Hammett acidity of the synthesized IL was determined by UV technique as
250 shown in Fig. 3. The Fig. 3 shows that the acidity profile of the ILs compared with the 4-
AN
251 nitroaniline indicator and comparatively 3-methyl-1-(4-sulfo-butyl)-benzimidazolium
252 trifluoromethanesulfonate ([BSMBIM][CF3SO3]) IL revealed stronger acidity. This stronger
M

253 acidity of [BSMBIM][CF3SO3] is the evidence of their high efficiency. The values of
254 Hammett acidity calculated on base of UV spectra is given in Table 2. The trifluoroacetate
D

255 anion (CF3CO2-) is derived from the weak acid, and its biodiesel yield is lower compared to
TE

256 the other three anions. Therefore, IL 3-methyl-1-(4-sulfo-butyl)-benzimidazolium

257 trifluoromethanesulfonate ([BSMBIM][CF3SO3]) was selected for detailed optimization.
EP

258 Table 4: Experimental screening of 1,4-sultone methylbenzimidazolium based ILs.

Parameters [BSMBIM][CH3SO3] [BSMBIM][CF3SO3] [BSMBIM][CF3CO2] [BSMBIM][HSO4]

C

Time
AC

1h 35.89% 37.33% 29.02% 36.90%

2h 39.33% 46.29% 30.28% 41.30%
3h 45.12% 47.42% 32.83% 44.85%
MeOH:Oil
9:1 22.40% 25.5% 17.91% 26.30%
12:1 34.74% 34.10% 23.62% 32.21%
15:1 35.89% 37.43% 29.02% 36.90%

10
 ACCEPTED MANUSCRIPT
Temperature
60oC 32.29% 35.61% 24.20% 34.02%
80oC 35.89% 37.33% 29.02% 36.90%
o
100 C 45.81% 53.60% 36.38% 48.49%
259
260 For time Screening: temperature = 80oC, MeOH:Oil = 15:1, mixing speed = 500 rpm.
261 For methanol to oil ratio screening: temperature = 80oC, mixing speed = 500 rpm, time = 1h. For temperature

PT
262 screening: time = 1h, MeOH:Oil = 15:1, speed=500 rpm, and catalyst 5 wt.% in all reactions.
263
264 3.5.1. Effect of catalyst concentration

RI
265 The effect of catalyst i.e. [BSMBIM][CF3SO3] concentration on the conversion of WCO into
266 biodiesel was studied by varying its concentration in the range of 2 to 6 wt.%. To evaluate the

SC
267 performance of the catalyst, the amount of catalyst was varied while other parameters were
268 kept fixed such as methanol to oil ratio is 15:1, reaction time 3h, temperature 100oC and

U
269 agitation speed of 600 rpm. The results presented in terms of biodiesel yield against catalyst
270 concentration are shown in Fig. 4. It can be observed from the figure that the product
AN
271 formation increases with the increase in catalyst concentration. The high yield was obtained
272 at 4, 4.5 and 5 wt.%. Further increase in catalyst concentration did not enhance the
M

273 conversion yield. In fact, increasing the concentration of [BSMBIM][CF3SO3] more than 5
274 wt.% caused a decreased in biodiesel yield. Increasing the amount of IL catalyst increases the
D

275 viscosity of the reaction media (because at 100oC the calculated viscosity of this IL is 155.28
TE

276 mPa.s) and as a result decreases the contact between the two phases during the reaction [32].

80
EP

70

60
C Yield (%)

50
AC

40

30

20
2 3 4 5 6
Catalyst Concentration (wt %)
277

278 Fig. 4. Effect of IL [BSMBIM][CF3SO3] concentration on yield at time 3h, temperature

11
 ACCEPTED MANUSCRIPT
279 100oC, methanol:oil ratio 15:1 and agitation speed 600 rpm.
280
281 3.5.2. Effect of methanol and waste cooking oil molar ratio
282 In this work, the methanol:oil molar ratio was varied (6:1, 9:1, 12:1, 15:1, 18:1, 21:1 and
283 24:1) in order to determine the effect of the amount of methanol on the methyl ester
284 production using [BSMBIM][CF3SO3] IL as a catalyst. While the other parameters were kept

PT
285 constant and catalyst concentration of 4 wt.% was used. The results in terms of ester yield
286 with respect to methanol to oil molar ratio are presented in Fig. 5.

RI
287 Here, the lowest methanol:oil molar ratio i.e. 6:1 has shown the lowest product yield
288 (19.43%). This lowest yield could be related to the high viscosity of [BSMBIM][CF3SO3].

SC
289 When the methanol: oil ratio was increased, the product yield was also increased indicating
290 that large quantity of methanol is needed to reduce the viscosity of the reaction media and to
291 facilitate the contact between the two phases. In practice, the molar ratio of 6:1 is usually
292

U
used to force the equilibrium to favour the product formation since the reaction is a reversible
AN
293 one. Several researchers reported 6:1 as the optimal ratio for alkali-catalysed
294 transesterification [33, 34]. The result shows that the ester content was increased when the
M

295 molar ratio was increased from 6:1–15:1. However, in this work 12:1 molar ratio was found
296 to be the optimum one because after that there is no significant increase in the product yield.
D

80
TE

70

60
EP

50
Yield (%)

40
C

30
AC

20

10
6:1 9:1 12:1 15:1 18:1 21:1 24:1
MeOH:Oil
297

298 Fig. 5. Effect of methanol:oil ratio on yield at [BSMBIM][CF3SO3] concentration of 4 wt.%,

299 reaction time 3h, temperature 100oC and agitation speed of 600 rpm.
300

12
 ACCEPTED MANUSCRIPT
301 3.5.3. Effect of temperature
302 One of the crucial factors is the reaction temperature that highly affected the biodiesel yield.
303 For example, higher temperature reduces the oil's viscosity, increases the rate of reaction and
304 results in the shortening of the reaction time [35]. However, in most cases the conversion rate
305 increases with time [36]. The effect of reaction temperature was investigated by varying the
306 reaction temperature from 60 to 160°C, as shown in Fig. 6. The results show that by

PT
307 increasing the reaction temperature, the yield (%) of biodiesel production was increased and
308 the high yield was obtained at high-temperature, i.e. 120oC. However, when the temperature
was increased above 120oC, there is a slight increase in product yield (1 or 2%), indicating

RI
309
310 that 120oC is the optimum temperature for the conversion of waste cooking oil into ester,

SC
311 catalysed by [BSMBIM][CF3SO3].

80

U
AN
70

60
Yield (%)

50
D

40
TE

30
60 80 100 120 140 160
EP

o
Temperature ( C)
312

313 Fig. 6. Effect of reaction temperature on yield at [BSMBIM][CF3SO35] concentration of 4

C

314 wt.%, methanol: oil ratio of 12:1, 3h reaction time and agitation speed 600 rpm.
AC

315

316 3.5.4. Effect of agitation speeds

317 The reactants mixing is very important for transesterification reaction completion, and also, it
318 increases the product yield. Agitation speed through a mixing of the catalyst with reactants
319 and rate of reaction increases the collision between the particles and diffusion of one reactant
320 into another. When the stirrer speeds increases, it will shorten the reaction time and increases
321 the conversion. Beyond certain stirrer speed, there would not be a significant rise in the yield.

13
 ACCEPTED MANUSCRIPT
322 Thus, stirrer speed optimization is necessary for various feedstock’s based on the different
323 physical properties [37].
324 The stirring speed effect on the yield of biodiesel from the WCO transesterification reactions
325 was investigated while keeping the other parameters constant. The agitation speed effect was
326 studied from 200-800 rpm. The results are shown in Fig. 7 demonstrated that the yield of
327 biodiesel increases when increased the agitation speed, and a maximum 78.13% yield was

PT
328 obtained at the stirring speed of 700 rpm. From the results, it is also observed that when the
329 agitation speed was increased from 700 rpm, the final biodiesel yield was found almost same,

RI
330 hence, in this work stirring speed 700 rpm is considered as the optimum speed for the
331 reaction.

SC
90

80

70

U
AN
60
Yield (%)

50
M

40

30
D

20
TE

10
200 300 400 500 600 700 800
Agitation Speed
332
EP

333 Fig. 7. Effect of agitation speed on yield at [BSMBIM][CF3SO3] concentration of 4 wt.%,

334 methanol:oil ratio of 12:1 reaction time 3h and temperature 120oC.
C

335
AC

336 3.5.6. Effect of time

337 Reaction time plays a vital role in the transesterification of triglycerides. Therefore, a certain
338 period is required for the transesterification reaction of triglyceride to complete. Fig. 8, shows
339 the effect of time, investigated by varying the reaction time from 30 to 400 minutes, keeping
340 the other parameters constant. The experimental result shows that the biodiesel yield (%) was
341 increased with the increase of time till 240 minutes and the highest biodiesel yield was
342 obtained at 240 min. When the reaction time was increased more than 240 min, no effect on

14
 ACCEPTED MANUSCRIPT
343 biodiesel yield was observed. Hence, 240 min was concluded as the optimum reaction time
344 for the WCO conversion into biodiesel. This optimal time was effective up to 78.13% yield
345 after that there was no significant increase in the yield of biodiesel. Hence, two-step reaction
346 i.e. esterification and transesterification were applied to obtain a high yield. KOH was used as
347 a catalyst for the transesterification step and after treatment for 30 min the product yield was
348 reached to 85.73%, and after 1 h the yield reached to 94.52%, which shows that the second

PT
349 step was more effective. The detailed results of first and second step are shown in the Fig. 8.
350 At the beginning, the reaction was slower due to the mixing and dispersion of oil, methanol

RI
351 and catalyst phases. The conversion rate was increased with the time due to the minimization
352 of the mass transfer between alcohol–oil–catalyst and completed in a specific time period.

SC
353 However, an excess reaction time beyond the optimum value may lead to a reduction in the
354 biodiesel yield due to the backward transesterification reaction, which in turn lowers the ester
355 contents [38, 39].

U
AN
100

90
M

80

70
Yield (%)

60
TE

50

40

30
EP

20
0 50 100 150 200 250 300 350 400
Time (min)
C

356
AC

357 Fig. 7. The effect of reaction time on yield at [BSMBIM][CF3SO3] concentration of 4 wt. %,
358 methanol:oil ratio 12:1, temperature 120oC and agitation speed of 700 rpm.
359

360 3.5.7. Catalyst recovery and usability

361 Catalyst recovery is a main feature for determining the economic viability of a process. The
362 reusability tests were carried out on the [BSMBIM][CF3SO3] IL as a catalyst at optimised
363 reaction condition i.e. the catalyst amount of 4 wt.%, methanol–oil ratio 12:1, at 120oC, for

15
 ACCEPTED MANUSCRIPT
364 240 min at a mixing speed of 700 rpm. The catalyst was used repeatedly for eight times for
365 the conversion of WCO to biodiesel. The reusability experimental profile of
366 [BSMBIM][CF3SO3] IL is shown in Fig. 9. The results show that the catalyst possesses
367 substantial reusability, and could be used repeatedly for seven times with maximum yield.
368 This high percentage indicated that the activity of the catalyst was retained after recycling.
369 However, the biodiesel yield was decreased as the number of repeated use increased and at

PT
370 eighth time the yield decreased to 70.5%. This decrease might be because of traces of
371 glycerol which interact with anion of IL through possible hydrogen bonding which causes a

RI
372 decrease in its efficiency for catalytic transesterification.

SC
100

U
80
AN
Yield (%)

60
M

40

20
D

0
TE

1 2 3 4 5 6 7 8
Catalyst Recovery
373
EP

374 Fig. 9. Catalyst [BSMBIM][CF3SO3] reusability versus yield (%) of biodiesel.

375
C

376 3.6. Synthesized biodiesel characterization

377 The final biodiesel was collected and washed several times with deionized water. Then dried,
AC

378 determined their different physicochemical properties and compared with the ASTM and EN
379 standards. The biodiesel physicochemical properties are tabulated in Table 5 and their results
380 showed that the quality of final product achieved the high standard. Their qualitative analysis
381 was investigated by GC-FID and is listed in Table 6. These results show that the design
382 catalyst would be one of the good choices for the methyl ester production from WCO. The
383 qualitative FT-IR, TGA and 1HNMR results are provided in supporting materials (Figs S 3- S
384 7).

16
 ACCEPTED MANUSCRIPT
385 Table 5: Physicochemical properties of synthesized biodiesel.
Property Unit ASTM D-6751 EN14214 Synthesized
Biodiesel
Kinematic viscosity at 40 oC mm2/s 1.9-6.0 3.50-5.00 4.94
Density at 15 oC kg/m3 860-894 860-900 883
o
Flash point C >120 >120 161

PT
Moisture content % <0.05 <0.05 0.04
Acid value mgKOH/g ≤0.5 ≤0.5 0.45

RI
Cetane number ……. 48-60 >51 57.1
Calorific value J/g N.R N.R 40111

SC
386 N.R = Not resulted

387 Table 6: FAMEs composition of synthesized biodiesel.

U
Peaks no Retention time (min) Identified
compounds
AN
1 7.719 Methyl laurate
2 8.878 Methyl myristate
3 10.342 Methyl palmitate
M

4 10.533 Methyl palmitoleate

5 12.137 Methyl stearate
6 12.567 Methyl oleate
D

7 13.059 Methyl linoleate

8 14.745 Methyl lonolenate
TE

388

389 3.7. Fourier-transform infrared (FT-IR) spectroscopy analysis

EP

390 Fourier-transform infrared (FT-IR) spectroscopy was further used for the synthesized
391 biodiesel and WCO characterization. The WCO and biodiesel FT-IR spectrum is shown in
392 Fig. (10a), and Fig. (10b) individually. The peaks at 2920.30 cm-1 and 2851.91 cm-1 are
C

393 assigned relatively to the stretching vibrations of symmetric and asymmetric aliphatic C-H in
AC

394 the CH2 and the terminal CH3 groups. At 1743.10 cm-1 there is a strong absorption peak
395 which is attributed to the carbonyl groups (C=O) stretching vibration of triglycerides. The
396 peaks in the region of 1200–1400 cm-1 are mostly assigned to the CH2 and CH3 aliphatic
397 groups bending vibrations like symmetric H–C–H, bending at 1377.25 cm-1 and CH2
398 scissoring at 1463.79 cm-1. The peaks in the region at 1096-1116 cm-1 are assigned to the C–
399 O ester group stretching vibration and CH2 wagging. The rocking vibration overlapping peaks
400 of the (CH2)n appeared at 720.97 cm-1 [40, 41].

17
 ACCEPTED MANUSCRIPT
401 On the other hand, there is characteristic's absorption bands at 1743 cm-1 which are due to the
402 carbonyl group (C=O) stretching vibration, and the stretching vibration's peaks or bands at
403 1000-1500 cm-1 corresponding to that of symmetric and asymmetric C–O group [42, 43]. The
404 CH3, CH2 and CH groups stretching vibrations normally appear in 2950-2980 cm-1, 2850-
405 2950 cm-1 and 3000-3050 cm-1 regions, where the bending vibrations of these groups appear
406 in the 13-50-1475 cm-1, 1150-1350 cm-1 and 722.59 cm-1 regions respectively [43].

PT
407 The FTIR spectra of synthesized biodiesel were found very similar to waste cooking oil
408 because due to the chemical nature similarities of the methyl esters and triglycerides.

RI
409 However, it can be seen that there are very little differences between the FTIR spectrum of
410 WCO and its biodiesel. It is noted that the peaks of the WCO appeared at 1743.10 cm-1,
1377.26 cm-1, 1157.19 cm-1, 1031.96 cm-1, 871.18 cm-1 are shifted to 1743.95 cm-1, 1361.91

SC
411
412 cm-1, 1170.91 cm-1, 1016.31 cm-1, 880.38 cm-1 in the biodiesel respectively. So, the peak's
413 disappearance at 1463.79 cm-1, 1096.86 cm-1, and 965.93 cm-1 from the WCO spectrum and
414

U
in a biodiesel sample the appearance of the new peaks at 1436.14 cm-1, and 1196.25 cm-1
AN
415 clearly show the conversion of WCO into the biodiesel.
M

1656.69 1417.99 871.18

100 1377.25

3469.54 2026.78 1031.86

2680.16
90
3006.20
D

965.93
Transmittance (%)

80
1236.91
1463.79 720.97
TE

70 1457.13 1096.86

2851.91
60
1157.19 1116.87
2920.30
1743.10
50
EP

40
4000 3500 3000 2500 2000 1500 1000
-1
C

Wavenumber (cm )
416
417 Fig. 10a. FT-IR spectrum of WCO.
AC

18
 ACCEPTED MANUSCRIPT

90
880.38 851.95

80

70 1361.91
1016.71

Transmittance (%)
3005.67
60 722.59
1460.82 1118.63
50 1196.69
1170.91
40 1436.14

PT
30

2854.34 1743.95
20
2924.99

RI
10
4000 3500 3000 2500 2000 1500 1000 500

Wavenumber (cm-1)
418

SC
419 Fig. 10b. FT-IR spectrum of synthesized biodiesel.
420

U
421 3.8. Thermogravimetry (TGA) analysis
AN
422 Thermogravimetric analysis of various types of materials such as polymers and oils has been
423 widely used to evaluate their thermal stability. The samples thermal properties or behaviour
are measured as a function of several reaction parameters such as time, temperature and
M

424
425 heating rates. In the chemical engineering field, this type of technique is very popular as a
tool of thermal analysis [44].
D

426
427 Thermogravimetric analysis of the methyl ester was performed to check the water content
TE

428 and its volatility, as reported earlier that the volatility greatly affects the fuel's ignition quality
429 in the internal-combustion engine [45]. The TGA curves of both biodiesel and WCO are
EP

430 presented in Fig. 11. The thermal stability curve shows that biodiesel is less thermally stable
431 to WCO. The onset temperature of WCO is 385oC while for biodiesel is 191oC. The Fig. 11
432 shows that the biodiesel completely volatilized at 250oC. Therefore, the synthesized biodiesel
C

433 as an alternative fuel, ignition within the compression-ignition engine would not be a
AC

434 problem. Because Sivakumar et al, [45] reported that the biodiesel compared to diesel fuel are
435 less volatile. However, compression engine generates a very high temperature (800oC).
436 Moreover, the biodiesel obtained from WCO at the optimal process conditions represents
437 lower combustion temperature as compared to WCO, which indicates that the
438 transesterification reaction has made the biodiesel more ignitable.

19
 ACCEPTED MANUSCRIPT

100

80

WCO
Weight %
60

PT
40
Biodiesel

20

RI
0

SC
100 200 300 400 500
o
Temperature ( C)
439

U
440 Fig. 11. Thermogravimetric profile of biodiesel and WCO at a heating rate of 10oC/min.
AN
441
442
443 3.9. Proton nuclear magnetic resonance (1H NMR) analysis
M

444 Figs. (12a) and (12b) show the proton nuclear magnetic resonance spectra of WCO and their
445 biodiesel for additional characterization. The peak at 5.3 ppm is assigned to the alkene
D

446 containing hydrogen (CH2=CH). At 3.6 ppm the peak is attributed to the alkoxy group in the
TE

447 methyl ester (CO2CH3), and the peak appeared at 2.3 ppm is assigned to the ester α-
448 methylene group (CH2CO2CH3). These two distinct peaks confirm the presence of methyl
esters in the synthesized biodiesel. At 2.7 ppm the peak appeared and attributed to the two
EP

449
450 double bonds of the α-methylene group (C=CH–CH2–CH=C) [43, 46].
451 Similarly, the peak at 2.02 ppm was observed and attributed to one double bond of the α-
C

452 methylene group (=CH–CH2–), whereas the observed peak at 1.64 ppm is assigned to the
AC

453 ester β-methylene group (CH2–C–COOCH3). At 1.2 ppm a broad peak appeared, which
454 attributed to the backbone of methylenes (–CH2)n–). The corresponding peak of terminal
455 methyl protons (C–CH3) appeared at 0.90 ppm. However, the waste cooking oil exhibit two
456 additional peaks at 4.08 to 4.22 ppm and 5.20 ppm, corresponding to the glyceride methlene
457 protons at C1 and C3 (–CH2–COOR) and methine protons at C2 (–CH–COOR) respectively.
458 The disappearance of the glycerol protons from the biodiesel also confirms the conversion of
459 WCO into biodiesel [46, 47].

20
 ACCEPTED MANUSCRIPT

PT
Triglyceride glycerol
backbone protons peaks

RI
SC
6 4 2 0
ppm

460

Fig. 12a. Proton NMR spectrum of waste cooking oil.

U
461 AN
M
D

Methoxy proton peak

(biodiesel)
TE
EP

10 8 6 4 2 0
ppm
C

462
AC

463 Fig. 12b. Proton NMR spectrum of synthesized biodiesel.

464

465

466 3.10. Kinetic study

467 From the transesterification reaction, the samples were analysed by GC and the product along
468 with intermediates are plotted in Fig. 13. The Fig. 13 shows that the mixture composition of
469 the reaction is changed, in which the triglyceride (TG) concentration is being dropped and the

21
 ACCEPTED MANUSCRIPT
470 methyl ester (ME) concentration is increases. The similar observations were also made by
471 other researchers [48, 49]. As the reaction progressed, the concentration of triglyceride
472 decreases continuously and the concentrations of intermediates i.e. diglyceride (DG) and
473 monoglyceride (MG) first increased to some level and then decreased due to the formation of
474 methyl ester.

475 Eventually, the concentration of TG, DG, MG and ME stayed almost constant as the reaction

PT
476 reached to equilibrium. Since an excess of methanol was used, the ME final concentration
477 raised than that of TG. This is an expected behaviour in homogeneous reaction [50].

RI
SC
100

80
Product Percentage (wt.%)

U
AN
60 TG
DG
MG
40 GL
M

ME

20
D

0
TE

0 50 100 150 200 250

Time (min)
478
EP

479 Fig 13. The Products and intermediates concentration during the transesterification of WCO
480 at optimum conditions using IL [BSMBIM][CF3SO3] as a catalyst.
C

481 3.10.1. Chemistry of the reaction

AC

482 The reaction scheme for transesterification of vegetable oils is presented earlier in as Fig. 1
483 (equation 1). For the reaction requires a stoichiometric amount of 3 mol methanol and 1 mol
484 of triglyceride (TG) react to give 3 molecules of fatty acid methyl ester and 1 molecule of
485 glycerol. Usually, this reaction in turn, consists of three consecutive reversible reactions,
486 where 1 molecule of the fatty acid ester is liberated in each step along with monoglyceride
487 (MG), and diglyceride (DG) as the intermediates products.

22
 ACCEPTED MANUSCRIPT
488 We defined a mathematical model by ignoring the intermediate reaction step of diglycerides
489 and monoglyceride, so the three steps can be simplified into one step. The backward reaction
490 is neglected since excess methanol (>3:1) is used in order to shift the reaction equilibrium
491 towards the formation of methyl esters in the forward reaction. Thus, the transesterification
492 reaction kinetics was calculated only for the forward reactions.
493 The experimental results were correlated for IL based on the reaction time and four different

PT
494 temperatures. For each temperature, from kinetic analysis three rate constants were
495 established using this model, ignoring the intermediates steps. Therefore, the reaction mixture

RI
496 can be considered as a pseudo-homogeneous medium and the reaction kinetics was controlled
497 by the transesterification reaction. The overall reaction can be written as:

SC

498 TG + 3CH3OH 3RCOOCH3 + Glycerol (6)

U
AN
499 3.10.2. First-order kinetics
500 According to the first-order kinetics’ reaction, the forward reaction of the WCO
M

501 transesterification can be expressed as follows:

[ ]
502 = − = [ ]
= [ !][" # ]3 (7)
D

503 Researchers reported [51, 52] that when the molar ratio of methanol was higher than
TE

504 triglyceride, the forward reaction are believed to be the pseudo-first-order. After that, the rate
505 of reaction is dependent only on the TG concentration, and the methanol concentration is
EP

506 assumed to be constant and the reverse reaction is neglected, thus the rate law equation
507 becomes:
C

[ ]
508 = − = = [ !] (8)
AC

[ ]

509 Integrating the above equation (8) gives:

510
[ ]&
511 ln = . (9)
[ ]

512
513 [TG]o is the initial concentration of tri glycerides, [TG] is the glyceride's concentration after
514 ʽtʼ time, and kTG is the pseudo first order rate constant.

23
 ACCEPTED MANUSCRIPT
515 The fraction of the glyceride molecules that were converted into methyl ester (biodiesel) was
516 calculated from the equation 10 or 11.
517
[ ]& [ ] [ ]
518 ()* = = 1 − (10)
[ ]& [ ]&

519 or
[ ]
520 (()* − 1) = (11)

PT
[ ]&

521 Where XME represents the fractional conversion of the biodiesel esters.
For the calculation of reaction constant (k), the following equation 12 was used:

RI
522
523
ln(()* − 1) = −

SC
524 (12)
525
526 For the calculation of ln k the following equation 13 was used:

U
527
AN
528 + = 2.303 log (13)
529
Using equation 9, the glyceride’s concentrations were calculated for each reaction
M

530
531 temperature, and then a plot of the pseudo first order was plotted as shown in Fig. 14.
As shown in Fig. 14, the plots gave straight lines at 80, 100, 120 and 140oC, using IL as a
D

532
533 catalyst. The plots giving straight lines to fit the data and their corresponding correlation
TE

534 coefficient values evidenced that the first-order kinetics’ model to be the best model for this
535 transesterification study. Moreover, the second-order kinetic study is given in the
EP

536 supplementary data.

C
AC

24
 ACCEPTED MANUSCRIPT

3.0 (R2 = 0.996)

2.5
80 oC
100 oC
ln ([TG]o/[TG])

2.0 120 oC
140 oC

PT
1.5

1.0

RI
0.5

SC
0.0
0 50 100 150 200 250
Time (min)

U
537
Fig. 14. Pseudo first–order reaction plot for the transesterification reaction of WCO at 80,
AN
538
539 100, 120 and 140oC.
M

540

541
D

542
TE

543 3.10.3. Activation energy determination

EP

544 The Svante Arrhenius theory was used for the calculation of activation energy of
545 transesterification of WCO using IL. The equation 14 gives a quantitative basis for the
546 relationship between the activation energy, temperature and reaction rate constant.
C
AC

* /2
547 = 1 (14)

548 The logarithmic form of the equation (14) can be expressed as:

* 4
549 ln = ln 1 − 2
× (15)

550 Where k is the reaction rate constant, A is the Arrhenius constant or pre-exponential constant,
551 R is the universal gas constant (8.314 Jmol-1 K-1), T is the absolute temperature (K) and Ea is
552 the activation energy.

25
 ACCEPTED MANUSCRIPT
553 The biodiesel production from WCO, catalysed by IL, was studied at four different
554 temperatures, the ln k values were calculated and plotted against 1000/T to determine the
555 activation energy (Ea) as shown in Fig. 15.

-4.4 R2 = 0.983

PT
Intercept = 0.0684
-4.5 Slope = -2.3148

-4.6

RI
-4.7
ln k

SC
-4.8

-4.9

U
-5.0
AN
-5.1
1.90 1.95 2.00 2.05 2.10 2.15 2.20

1000/T (K-1)
M

556

557 Fig.15. Arrhenius plot of natural log of the rate constant (ln k) versus the inverse of
D

558 temperature (1000/T) in kelvin for the transesterification of WCO into biodiesel catalysed by
TE

559 IL.

560 The Arrhenius plot of ln k versus 1000/T (Fig.15) gives a straight line. The activation energy
EP

561 (Ea) and the frequency factor (A) were calculated from the slope and intercept of the linear
562 fitting of lnk vs 1000/T, respectively. The obtained value of the activation energy (Ea) is
19.24 kJ/mol, and that of frequency factor (A) is 1.07 min-1. In literature, different values for
C

563
564 activation energy are reported and this value of the conversion of WCO to biodiesel,
AC

565 catalysed by IL, is in good agreement with reported literature values such as 13.3, 20.7 and
566 16.7 kJ/mol [53-55]. Some researcher reported 36.03, 43.4, 58.5, 79 kJ/mol [56-58], 113, 148
567 kJ/mol [59]. Moreover, this calculated activation energy is reasonable and said it is concluded
568 that the ILs has the potential to produce biodiesel from WCO.

569 4. Conclusions

570 Four dual-functionalized Bronsted acids based ILs with different anions were prepared and
571 applied as catalysts for biodiesel synthesis from waste palm cooking oil. The catalyst’s better

26
 ACCEPTED MANUSCRIPT
572 efficiency was noted and proved by Hammett acidity and COSMO-RS study. The optimized
573 conditions for the biodiesel production from WCO were catalyst concentration 4 wt%,
574 methanol to oil ratio 12:1, temperature 120oC, agitation speed 700 rpm and time of 240 min.
575 The catalyst was recycled for eight times, and the reusability tests showed that the catalyst
576 had considerable utility for repeated use. The final product satisfied the ASTM and EN
577 standards. The kinetic study and the experimental results demonstrate that the reaction

PT
578 follows a pseudo first-order mechanism with the activation energy (Ea) of 19.24 kJ mol-1,
579 frequency factor (A) of 1.07 min-1. Moreover, the data did not fit well on the second-order

RI
580 mechanism. The calculated activation energy for such reaction in the selected waste cooking
581 oil for biodiesel production is reasonable.

SC
582
583 Acknowledgement
584 We especially acknowledge the Centre of Research in Ionic Liquid, all the research officers
585

U
and postgraduate students for helping in all aspects.
AN
586

587 References
M

588 [1] X. Liu, H. Ma, Y. Wu, C. Wang, M. Yang, P. Yan, U. Welz-Biermann, Esterification of
589 glycerol with acetic acid using double SO 3 H-functionalized ionic liquids as recoverable
590 catalysts, Green Chem 13(3) (2011) 697-701.
D

591 [2] X. Liang, Novel acidic ionic liquid polymer for biodiesel synthesis from waste oils, Appl
592 Catal A: Gen 455 (2013) 206-210.
TE

593 [3] D. Bradley, E. Levin, C. Rodriguez, P.G. Williard, A. Stanton, A.M. Socha, Equilibrium
594 studies of canola oil transesterification using a sodium glyceroxide catalyst prepared from a
595 biodiesel waste stream, Fuel Process Technol 146 (2016) 70-75.
EP

596 [4] S.H. Ha, M.N. Lan, S.H. Lee, S.M. Hwang, Y.-M. Koo, Lipase-catalyzed biodiesel
597 production from soybean oil in ionic liquids, Enzyme and Microb. Technol 41(4) (2007) 480-
598 483.
599 [5] Z. Ullah, M.A. Bustam, Z. Man, Biodiesel production from waste cooking oil by acidic
C

600 ionic liquid as a catalyst, Renew Energy 77(0) (2015) 521-526.

601 [6] Z. Zhang, J. Ji, Waste Pig Carcasses as a Renewable Resource for Production of Biofuels,
AC

602 ACS Sustain. Chem. Eng 3(2) (2014) 204-209.

603 [7] N. Muhammad, Y.A. Elsheikh, M.I.A. Mutalib, A.A. Bazmi, R.A. Khan, H. Khan, S.
604 Rafiq, Z. Man, I. khan, An overview of the role of ionic liquids in biodiesel reactions, Ind.
605 Eng. Chem. Res 21 (2015) 1-10.
606 [8] M. Diasakou, A. Louloudi, N. Papayannakos, Kinetics of the non-catalytic
607 transesterification of soybean oil, Fuel 77(12) (1998) 1297-1302.
608 [9] X. Liang, Synthesis of biodiesel from waste oil under mild conditions using novel acidic
609 ionic liquid immobilization on poly divinylbenzene, Energy 63 (2013) 103-108.
610 [10] S. Das, A.J. Thakur, D. Deka, Two-Stage Conversion of High Free Fatty Acid Jatropha
611 curcas Oil to Biodiesel Using Brønsted Acidic Ionic Liquid and KOH as Catalysts, Scientific
612 World J. 2014 (2014).

27
 ACCEPTED MANUSCRIPT
613 [11] S. Liu, Z. Wang, S. Yu, C. Xie, Transesterification of waste oil to biodiesel using
614 Brønsted acid ionic liquid as catalyst, Bull Chem Soc Ethiop 27(2) (2013) 289-294.
615 [12] X. Liang, Novel efficient procedure for biodiesel synthesis from waste oils using solid
616 acidic ionic liquid polymer as the catalyst, Ind. Eng. Chem. Res 52(21) (2013) 6894-6900.
617 [13] Y.-C. Lin, P.-M. Yang, S.-C. Chen, J.-F. Lin, Improving biodiesel yields from waste
618 cooking oil using ionic liquids as catalysts with a microwave heating system, Fuel Process
619 Technol 115 (2013) 57-62.
620 [14] A. De los Ríos, F.H. Fernández, D. Gómez, M. Rubio, G. Víllora, Biocatalytic
621 transesterification of sunflower and waste cooking oils in ionic liquid media, Process

PT
622 Biochem 46(7) (2011) 1475-1480.
623 [15] A.M. Socha, J.K. Sello, Efficient conversion of triacylglycerols and fatty acids to
624 biodiesel in a microwave reactor using metal triflate catalysts, Org. Biomol. Chem 8(20)

RI
625 (2010) 4753-4756.
626 [16] D.E. Lopez, K. Suwannakarn, D.A. Bruce, J.G. Goodwin, Esterification and
627 transesterification on tungstated zirconia: effect of calcination temperature, J. Catal 247(1)

SC
628 (2007) 43-50.
629 [17] Y. Xu, M. Hanna, Synthesis and characterization of hazelnut oil-based biodiesel, Ind
630 Crops Prod 29(2) (2009) 473-479.
631 [18] Z. Ullah, M.A. Bustam, Z. Man, N. Muhammad, A.S. Khan, Synthesis, characterization

U
632 and the effect of temperature on different physicochemical properties of protic ionic liquids,
633 RSC Adv 5(87) (2015) 71449-71461.
AN
634 [19] L. Zhang, M. Xian, Y. He, L. Li, J. Yang, S. Yu, X. Xu, A Brønsted acidic ionic liquid
635 as an efficient and environmentally benign catalyst for biodiesel synthesis from free fatty
636 acids and alcohols, Bioresour. Technol 100(19) (2009) 4368-4373.
M

637 [20] B.S. Caldas, C.S. Nunes, P.R. Souza, F.A. Rosa, J.V. Visentainer, S. Oscar de Olivera,
638 E.C. Muniz, Supercritical ethanolysis for biodiesel production from edible oil waste using
639 ionic liquid [HMim][HSO 4] as catalyst, Appl. Catal. B 181 (2016) 289-297.
D

640 [21] A.M. Socha, R. Parthasarathi, J. Shi, S. Pattathil, D. Whyte, M. Bergeron, A. George, K.
TE

641 Tran, V. Stavila, S. Venkatachalam, Efficient biomass pretreatment using ionic liquids
642 derived from lignin and hemicellulose, Proceedings of the National Academy of Sciences
643 111(35) (2014) E3587-E3595.
644 [22] T. Welton, Room-temperature ionic liquids. Solvents for synthesis and catalysis, Chem.
EP

645 Rev 99(8) (1999) 2071-2084.

646 [23] Z. Ullah, M.A. Bustam, N. Muhammad, Z. Man, A.S. Khan, Synthesis and
C

647 thermophysical properties of hydrogensulfate based acidic ionic liquids, J. Solution Chem
648 44(3-4) (2015) 875-889.
AC

649 [24] X. Liang, H. Xiao, C. Qi, Efficient procedure for biodiesel synthesis from waste oils
650 using novel solid acidic ionic liquid polymer as catalysts, Fuel Process Technol 110 (2013)
651 109-113.
652 [25] M. Han, W. Yi, Q. Wu, Y. Liu, Y. Hong, D. Wang, Preparation of biodiesel from waste
653 oils catalyzed by a Brønsted acidic ionic liquid, Bioresour. Technol 100(7) (2009) 2308-
654 2310.
655 [26] X. Tong, Y. Li, Efficient and Selective Dehydration of Fructose to 5‐
656 Hydroxymethylfurfural Catalyzed by Brønsted‐Acidic Ionic Liquids, ChemSusChem 3(3)
657 (2010) 350-355.
658 [27] N. Muhammad, G. Gonfa, A. Rahim, P. Ahmad, F. Iqbal, F. Sharif, A.S. Khan, F.U.
659 Khan, Z.U.L.H. Khan, F. Rehman, I.U. Rehman, Investigation of ionic liquids as a

28
 ACCEPTED MANUSCRIPT
660 pretreatment solvent for extraction of collagen biopolymer from waste fish scales using
661 COSMO-RS and experiment, J. Mol. Liq 232 (2017) 258-264.
662 [28] S.S. AlSaleem, W.M. Zahid, I.M. AlNashef, M.K. Hadj-Kali, Solubility of Halogenated
663 Hydrocarbons in Hydrophobic Ionic Liquids: Experimental Study and COSMO-RS
664 Prediction, J. Chem. Eng. Data 60(10) (2015) 2926-2936.
665 [29] R. Wirasnita, T. Hadibarata, Y.M. Novelina, A.R.M. Yusoff, Z. Yusop, A modified
666 methylation method to determine fatty acid content by gas chromatography, Bull. Korean
667 Chem. Soc 34(11) (2013) 3239.

PT
668 [30] F. Tao, H. Song, L. Chou, Hydrolysis of cellulose in SO 3 H-functionalized ionic
669 liquids, Bioresour. Technol 102(19) (2011) 9000-9006.
670 [31] K.-X. Li, L. Chen, Z.-C. Yan, H.-L. Wang, Application of pyridinium ionic liquid as a

RI
671 recyclable catalyst for acid-catalyzed transesterification of Jatropha oil, Catal lett 139(3-4)
672 (2010) 151-156.
673 [32] A. Kawashima, K. Matsubara, K. Honda, Acceleration of catalytic activity of calcium

SC
674 oxide for biodiesel production, Bioresour. Technol 100(2) (2009) 696-700.
675 [33] F. Ma, M.A. Hanna, Biodiesel production: a review, Bioresour. Technol 70(1) (1999) 1-
676 15.

U
677 [34] D. Leung, Y. Guo, Transesterification of neat and used frying oil: optimization for
678 biodiesel production, Fuel Process Technol 87(10) (2006) 883-890.
AN
679 [35] M. Mathiyazhagan, A. Ganapathi, Factors affecting biodiesel production, Res Plant Bio
680 1(2) (2011) 1-5.
681 [36] L. Meher, D. Vidya Sagar, S. Naik, Technical aspects of biodiesel production by
682 transesterification—a review, Renew. Sustainable Energy Rev 10(3) (2006) 248-268.
M

683 [37] A. Gnanaprakasam, V.M. Sivakumar, A. Surendhar, M. Thirumarimurugan, T.

684 Kannadasan, Recent Strategy of Biodiesel Production from Waste Cooking Oil and Process
685 Influencing Parameters: A Review, J. Energy 2013 (2013).
D

686 [38] L. Meher, D.V. Sagar, S. Naik, Technical aspects of biodiesel production by
687 transesterification—a review, Renew. Sustainable Energy Rev 10(3) (2006) 248-268.
TE

688 [39] D.Y. Leung, X. Wu, M. Leung, A review on biodiesel production using catalyzed
689 transesterification, Appl. energy 87(4) (2010) 1083-1095.
690 [40] A. Rohman, Y.B.C. Man, Fourier transform infrared (FTIR) spectroscopy for analysis of
691 extra virgin olive oil adulterated with palm oil, Food Res Int 43(3) (2010) 886-892.
EP

692 [41] Q. Zhang, C. Liu, Z. Sun, X. Hu, Q. Shen, J. Wu, Authentication of edible vegetable oils
693 adulterated with used frying oil by Fourier Transform Infrared Spectroscopy, Food chem
694 132(3) (2012) 1607-1613.
C

695 [42] I.P. Soares, T.F. Rezende, R.C. Silva, E.V.R. Castro, I.C. Fortes, Multivariate calibration
696 by variable selection for blends of raw soybean oil/biodiesel from different sources using
AC

697 fourier transform infrared spectroscopy (FTIR) spectra data, Energy Fuels 22(3) (2008) 2079-
698 2083.
699 [43] M. Tariq, S. Ali, F. Ahmad, M. Ahmad, M. Zafar, N. Khalid, M.A. Khan, Identification,
700 FT-IR, NMR (1 H and 13 C) and GC/MS studies of fatty acid methyl esters in biodiesel from
701 rocket seed oil, Fuel Process Technol 92(3) (2011) 336-341.
702 [44] S. Jain, M. Sharma, Application of thermogravimetric analysis for thermal stability of
703 Jatropha curcas biodiesel, Fuel 93 (2012) 252-257.
704 [45] P. Sivakumar, K. Anbarasu, S. Renganathan, Bio-diesel production by alkali catalyzed
705 transesterification of dairy waste scum, Fuel 90(1) (2011) 147-151.
706 [46] J.K. Satyarthi, D. Srinivas, P. Ratnasamy, Estimation of free fatty acid content in oils,
707 fats, and biodiesel by 1H NMR spectroscopy, Energy Fuels 23(4) (2009) 2273-2277.

29
 ACCEPTED MANUSCRIPT
708 [47] D.C. Deka, S. Basumatary, High quality biodiesel from yellow oleander (Thevetia
709 peruviana) seed oil, Biomass Bioenergy 35(5) (2011) 1797-1803.
710 [48] H. Noureddini, D. Zhu, Kinetics of transesterification of soybean oil, J. Am. Oil Chem.
711 Soc 74(11) (1997) 1457-1463.
712 [49] G. Vicente, M. Martínez, J. Aracil, A. Esteban, Kinetics of sunflower oil methanolysis,
713 Ind. Eng. Chem. Res. 44(15) (2005) 5447-5454.
714 [50] V. Mao, S.K. Konar, D.G. Boocock, The pseudo-single-phase, base-catalyzed
715 transmethylation of soybean oil, J. Am. Oil Chem. Soc 81(8) (2004) 803-808.

PT
716 [51] P. Sivakumar, S. Sindhanaiselvan, N.N. Gandhi, S.S. Devi, S. Renganathan,
717 Optimization and kinetic studies on biodiesel production from underutilized Ceiba Pentandra
718 oil, Fuel 103 (2013) 693-698.
719 [52] B. Freedman, R.O. Butterfield, E.H. Pryde, Transesterification kinetics of soybean oil 1,

RI
720 . Am. Oil Chem. Soc 63(10) (1986) 1375-1380.
721 [53] E. Sendzikiene, V. Makareviciene, P. Janulis, S. Kitrys, Kinetics of free fatty acids
722 esterification with methanol in the production of biodiesel fuel, European journal of lipid

SC
723 science and technology 106(12) (2004) 831-836.
724 [54] M. Chai, Q. Tu, M. Lu, Y.J. Yang, Esterification pretreatment of free fatty acid in
725 biodiesel production, from laboratory to industry, Fuel processing technology 125 (2014)

U
726 106-113.
727 [55] E.-S. Song, J.-w. Lim, H.-S. Lee, Y.-W. Lee, Transesterification of RBD palm oil using
AN
728 supercritical methanol, The Journal of Supercritical Fluids 44(3) (2008) 356-363.
729 [56] O.S. Stamenković, M.D. Kostić, N.M. Joković, V.B. Veljković, The kinetics of base-
730 catalyzed methanolysis of waste cooking oil.
731 [57] A. Birla, B. Singh, S. Upadhyay, Y. Sharma, Kinetics studies of synthesis of biodiesel
M

732 from waste frying oil using a heterogeneous catalyst derived from snail shell, Bioresource
733 technology 106 (2012) 95-100.
734 [58] J. Kansedo, K.T. Lee, Process optimization and kinetic study for biodiesel production
D

735 from non-edible sea mango (Cerbera odollam) oil using response surface methodology,
736 Chemical Engineering Journal 214 (2013) 157-164.
TE

737 [59] R. Ibbett, S. Gaddipati, D. Greetham, S. Hill, G. Tucker, The kinetics of inhibitor
738 production resulting from hydrothermal deconstruction of wheat straw studied using a
739 pressurised microwave reactor, Biotechnology for biofuels 7(1) (2014) 1-14.
EP

740
C
AC

30
 ACCEPTED MANUSCRIPT
Highlights

• Waste cooking oil as a promising source for biodiesel production.

• SO3H functionalised ILs are acting as dual functionality.
• Acidic ILs are required for acidic oils esterification process.
• This IL follows the first order mechanism.

PT
RI
U SC
AN
M
D
TE
C EP
AC