Efficient removal of phenolic compounds from model oil using benzyl Imidazolium-based ionic liquids

Article abstract of DOI:10.1016/j.molliq.2017.05.111

Three benzyl imidazolium-based room temperature ionic liquids (RTILs) with various substituents namely allyl, benzyl, and vinyl were synthesized and used as solvents in liquid-liquid extraction for the removal of phenolic compounds from hexane as the model oil. The RTILs were characterized using ¹H NMR, ¹³C NMR, FT-IR, and CHN elemental analyses. Their density and viscosity were also measured. Five main parameters were evaluated through the removal process; the effect of IL substituent, the phase volume ratio of IL and model oil, phase contact time, and temperature. IL containing allylic substituent showed outstanding performance with approximately 95% efficiency under selected optimized conditions. To ensure that the RTIL can be used as a solvent for the removal of phenol, various other types of model oil apart from hexane such as petroleum ether, heptane, and cyclohexane were also used. The RTIL exhibited good recyclability and negligible loss of mass even after six cycles. Further mechanistic interactions between RTIL and phenol were studied by ¹H NMR and FT-IR.

Full text of DOI:10.1016/j.molliq.2017.05.111

Accepted Manuscript
Efficient removal of phenolic compounds from model oil using
benzyl Imidazolium-based ionic liquids
Nadiah Sidek, Ninie S.A. Manan, Sharifah Mohamad
PII:
DOI:
Reference:
S0167-7322(16)33368-2
doi: 10.1016/j.molliq.2017.05.111
MOLLIQ 7403
To appear in:
Journal of Molecular Liquids
Received date:
Revised date:
Accepted date:
28 October 2016
12 May 2017
14 May 2017
Please cite this article as: Nadiah Sidek, Ninie S.A. Manan, Sharifah Mohamad , Efficient
removal of phenolic compounds from model oil using benzyl Imidazolium-based ionic
liquids, Journal of Molecular Liquids (2017), doi: 10.1016/j.molliq.2017.05.111
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
Efficient Removal of Phenolic Compounds from Model Oil Using Benzyl Imidazolium-
Based Ionic Liquids
Nadiah Sidek a,b, Ninie S. A. Manan a,b,*, Sharifah Mohamad a,b
a Department of Chemistry, Faculty of Science, University Malaya, 50603 Kuala Lumpur.
b University Malaya Centre for Ionic Liquid (UMCiL), University Malaya, 50603 Kuala
Lumpur.
*Corresponding Author; Email: niniemanan@um.edu.my; Tel: +60379677022 Ext 3873
Abstract
Three benzyl imidazolium-based room temperature ionic liquids (RTILs) with various
substituents namely allyl, benzyl, and vinyl were synthesized and used as solvents in liquid-
liquid extraction for the removal of phenolic compounds from hexane as the model oil. The
1
13
RTILs were characterized using H NMR, C NMR, FT-IR, and CHN elemental analyses.
Their density and viscosity were also measured. Five main parameters were evaluated
through the removal process; the effect of IL substituent, the phase volume ratio of IL and
model oil, phase contact time, and temperature. IL containing allylic substituent showed
outstanding performance with approximately 95% efficiency under selected optimized
conditions. To ensure that the RTIL can be used as a solvent for the removal of phenol,
various other types of model oil apart from hexane such as petroleum ether, heptane, and
cyclohexane were also used. The RTIL exhibited good recyclability and negligible loss of
mass even after six cycles. Further mechanistic interactions between RTIL and phenol were
studied by 1H NMR and FT-IR.
Keywords: Room temperature ionic liquid, Benzyl imidazolium, Phenol, Liquid-liquid
extraction, Model oil
1. Introduction
Phenol is a major industrial chemical usually produced from a multi-step process with
benzene as the feedstock [1]. Other major sources of phenolic compounds are coal
liquefaction oil, coal tar, petroleum, and biomass via pyrolysis [2-4]. Phenol is separated
from coal via pyrolysis of low-rank coals, carbonization, hydropyrolysis, and liquefaction. It
is used industrially for the production of phenolic resins, bisphenol A, caprolactam, and other
phenolic compounds. In addition, they have been widely employed in many industrial
processes as synthesis intermediates or as raw materials in the manufacturing of herbicides,
pesticides, insecticides, pharmaceuticals, and dyes [5].
Coal tar has a relatively high concentration of phenolic compounds. Thus, it is desirable to
separate phenolic compounds from the oil mixture before further refining for economic

ACCEPTED MANUSCRIPT
reasons. The traditional method currently employed to separate phenols from oil mixtures is
through the use of strongly alkaline and acidic chemicals. However, this method produces
excessive amounts of phenol-containing wastewater [6, 7]. Therefore, an environmentally
friendly and efficient method becomes necessary to exhaustively extract phenols from oil.
In recent times, a new type of green solvent has emerged as an alternative reaction media to
traditional organic solvents. Ionic liquids (ILs) can be broadly defined as ionic species that
melt at or below 100 C. Their structure typically consist of organic cations and either
inorganic or organic anions [8]. Room temperature ionic liquids (RTILs) are a subset of ILs
and they exist as liquids at room temperature [9, 10]. ILs have caught considerable attention
in the analytical chemistry field due to their unique solvent properties including negligible
vapour pressures, high thermal stabilities, variable viscosities, adjustable miscibility,
reusability, and non-flammability [11-13].
The choice of both cations and anions allows us to tweak the properties of the resulting ILs to
obtain the desired solvent properties [14, 15]. In order to enhance the properties of ILs as
extractants, their substituent alkyl group can be customized. In fact, increasing the chain
length of the substituent groups on the cations modifies the viscosity, hydrophobicity, and
melting point [15].
Recently, there has been a growing interest in manipulating ionic liquids and other types of
salts for the removal of phenolic compounds from oil. In 2013, Yucui Hou and co-workers
studied several imidazolium-based ILs with different cationic alkyl chain length to separate
phenol from hexane model oil. The study proved that imidazolium IL with shorter alkyl chain
length in particularly [Bmim]Cl extracted phenol with a higher efficiency compared to its
longer alkyl counterparts. The anions effect was also studied and the extraction efficiency
-
-
using Cl- surpassed Br-, BF4 , and PF6 due to the higher electronegativity of Cl- [16]. Tiantian
Jiao et. al designed imidazole and its homolog compounds as new extraction agents for the
separation of phenolic compounds from coal tar. They managed to prove that imidazole could
extract phenols via the formation of hydrogen bonds [7].
Thus, in order to increase their extraction capability toward phenolic compounds, it is in our
interest to synthesize a series of benzyl imidazolium-based ILs with allyl, benzyl and vinyl
substituents. By employing substituents with double bonds and aromatic rings, phenols can
be extracted by enhancing the π-π interaction. Due to the aromatic characteristic of phenolic
compounds, the designation of new ionic liquids with aromatic and double bond moieties
could open up a new frontier in the separation studies of phenolic compounds. In this study,
the allylic benzyl imidazolium IL was selected for the removal of phenol and the effects of
experimental conditions such as phase volume ratio, phase contact time, temperature and type
of model oil used were studied. The regeneration of IL and the reaction mechanism were also
investigated.
2. Experimental
2.1 Reagents and Materials

ACCEPTED MANUSCRIPT
1-benzylimidazole (99%), allyl chloride (98%) and 1-vinylimidazole (≥99%) were purchased
from Aldrich whereas benzyl chloride was commercially available from Merck. Acetonitrile
(AcN) supplied by Merck were used as solvents in the synthesis part. For the model oil, n-
hexane for liquid chromatography was purchased from Merck whereas n-heptane and
petroleum ether were supplied by Friedemann and Schmidt while cyclohexane was obtained
from Fisher Scientific. Crystallized phenol (Ph) was supplied by Scharlau, 2-nitrophenol (2-
NP) (98%), 4-nitrophenol (4-NP) (≥99%), 2-chlorophenol (2-CP) (≥99%), 2,4,6-
trichlorophenol (2,4,6-TCP) (98%), 2,4-dinitrophenol (2,4-DNP) of analytical standard and
2,4-dichlorophenol (2,4-DCP) were supplied by Sigma-Aldrich. 2,4,6-trinitrophenol (2,4,6-
TNP) was purchased from R&M Chemicals while 3-chlorophenol (3-CP), 4-chlorophenol (4-
CP), toluene and xylene were commercially available from Merck. Standard fluoranthene and
fluorene were purchased from Sigma-Aldrich. Ethyl acetate from R&M Chemicals was used
in the synthesis part. Ultrapure water was used throughout this experiment. All chemicals
were used as received without further purification.
2.2 Instrumentation
The Fourier Transform Infrared (FTIR) spectra were recorded by Perkin-Elmer Spectrum 400
FTIR spectrometer at room temperature equipped with a diamond attenuated total reflectance
attachment. All spectra were recorded with characteristic peaks in the range of 450–4000
cm−1. The Proton Nuclear Magnetic Resonance (1H NMR) and 13C NMR spectra were
recorded by Lambda JEOL 400 MHz FT-NMR spectrometer. All the samples were dissolved
in deuterated dimethyl sulfoxide (DMSO-d6) with tetramethylsilane (TMS) as an internal
reference for the calibration of proton chemical shift.
Elemental analyses (CHN) were performed using Pelkin-Elmer 2400 Series II Elemental
Analyser. Electronic spectra were recorded by Shimadzu UV-Vis-NIR 3600 spectrometer in
the range of 200-800 nm using a 1 cm quartz cell.
The densities of all products were obtained from Automatic Density Meter Rudolph Research
Analytical DDM 2910 at 25 C. The viscosities were measured using Anton Paar Physica
MCR 301 rheometer at 25 C.
A HPLC system consisting of a Shimadzu (Tokyo, Japan) LC-20AT pump, SPD-M20A
diode array detector, SIL-20A HT auto sampler and CTO-10AS VP column oven was used
for the separation and quantification of the analytes. The separation was conducted using a C-
18 reverse phase column (250 mm x 4.6 mm; particle size 5 μm) hypersil gold, Thermo
science USA. A mobile phase of methanol:water (80:20, v/v) was used at a flow rate of 1.0
mL min-1. The injection volume was 10 μL. The detection for phenol was found at a retention
time of 3.56 min and a wavelength of 272 nm; for 2,4-DCP it is at 4.37 min and 286 nm; for
fluorene, 6.91 min and 262 nm ; for fluoranthene, 8.34 min and 286 nm; and for 2,4-DNP, the
peak is found at 9.37 min and 267 nm.

ACCEPTED MANUSCRIPT
2.3 Synthesis of ILs
All ILs were synthesized according to the literature with some modifications [17]. The
synthesis involved a 1:2 molar ratio for the imidazole precursor to alkyl halide. Excess
amount of alkyl halide was used so that imidazole can be completely reacted.
2.3.1 1-Allyl-3-benzylimidazolium chloride
Allyl chloride (0.10 mol, 7.652 g) was added dropwise over a period of 30 minutes to a
round-bottomed flask containing 1-benzylimidazole (0.05 mol, 7.910 g) in 20 mL AcN under
continuous stirring. The mixture was then stirred for about 1 hour and subsequently refluxed
at 70 °C for 24 hours till a light brown liquid forms. Excess reactants and solvent were
removed by rotary evaporation. The crude product was then diluted in an appropriate volume
of water and washed three times with 15 mL ethyl acetate in a separating funnel. Finally,
excess water was removed by rotary evaporation and dried under vacuum at 60 °C overnight
to yield a light golden brown liquid (11.418 g, 97.29%). Density (ρ, g cm-3) of the liquid is
1.1578 and the viscosity (η, Pa.s) is 3.2593.
1H NMR (DMSO-d6, δ, ppm): 4.869 (2H, d, NCH2); 5.259-5.353 (2H, dd, C=CH); 5.476
(2H, s, NCH2); 6.035 (1H, m, CH); 7.344-7.459 (5H, m, CH aromatic); 7.780 (1H, s,
NC=CH); 7.897 (1H, s, NC=CH); 9.550 (1H, s, N=CH).
FT-IR (υ, cm-1): 3383.02 (Quaternary N), 3127.65 (sp2 C-H vinyl), 3054.50 (sp2 C-H aromatic),
2977.69 (sp2 C-H imidazolium ring), 2851.38 (sp3 C-H), 1644.60 (C=N), 1605.74 (C=C vinyl),
1558.55, 1497.69 (C=C aromatic), 1451.94 (CH2 methylene), 1153.90 (C-N).
13C NMR (DMSO-d6, δ, ppm): 136.870 (C4), 135.526 (C6), 132.284 (C5), 129.500 (C8),
129.252 (C10), 128.928 (C9), 123.388 (C11), 123.188 (C2), 120.881 (C1), 52.346 (C7),
51.479 (C3).
Anal. calcd. (%) for C13H15N2Cl: C, 66.520; H, 6.440; N, 11.930. Found: C, 65.337; H,
6.867; N, 11.698.
2.3.2 1,3-Dibenzylimidazolium chloride
The same method was used to synthesize 1,3-dibenzylimidazolium chloride [DBZIM]Cl with
1-benzylimidazole (0.05 mol, 7.910 g) and benzyl chloride (0.10 mol, 12.658 g). A pale
yellow viscous liquid (13.976 g, 98.15%) is obtained. Density (ρ, g cm-3): 1.1871. Viscosity
(η, Pa.s): 1395.3.
1H NMR (DMSO-d6, δ, ppm): 5.432 (4H, s, NCH2); 7.321-7.420 (10H, m, CH aromatic);
7.847 (2H, s, NC=CH); 9.638 (1H, s, N=CH).

ACCEPTED MANUSCRIPT
13C NMR (DMSO-d6, δ, ppm): 136.927 (C6), 135.440 (C7), 129.529 (C4), 129.290 (C3),
128.947 (C2), 123.388 (C1), 52.461 (C5).
FT-IR (υ, cm-1): 3382.37 (Quaternary N), 3054.10 (sp2 C-H aromatic), 2981.98 (sp2 C-H
imidazolium ring), 2847.55 (sp3 C-H), 1626.25 (C=N), 1557.37, 1497.33 (C=C aromatic),
1454.98 (CH2 methylene), 1148.99 (C-N).
Anal. calcd. (%) for C17H17N2Cl: C, 71.70; H, 6.020; N, 9.840. Found: C, 69.933; H, 6.406;
N, 9.534.
2.3.1 1-Benzyl-3-vinylimidazolium chloride
1-vinylimidazole (0.05 mol, 4.706 g) and benzyl chloride (0.10 mol, 12.658 g) were used to
synthesize 1-benzyl-3-vinylimidazolium chloride [BZVIM]Cl. A dark brown viscous liquid is
obtained (10.833 g, 98.17%). Density (ρ, g cm-3): 1.1948. Viscosity (η, Pa.s): 39.898.
1H NMR (DMSO-d6, δ, ppm): 5.384-5.362 (1H, dd, C=CH); 5.471 (2H, s, NCH2); 5.965-
6.004 (1H, dd, C=CH); 7.314 (1H, q, NCH) 7.345-7.469 (10H, m, CH aromatic); 7.974 (1H,
s, NC=CH); 8.262 (1H, s, NC=CH); 9.921 (1H, s, N=CH).
13C NMR (DMSO-d6, δ, ppm): 136.165 (C3), 135.068 (C5), 129.510 (C4), 129.433 (C7),
129.357 (C8), 129.090 (C9), 123.789 (C10), 120.061 (C2), 109.315 (C1), 52.632 (C6).
FT-IR (υ, cm-1): 3382.68 (Quaternary N), 3119.73 (sp2 C-H vinyl), 3055.57 (sp2 C-H aromatic),
2998.95 (sp2 C-H imidazolium ring), 2861.01 (sp3 C-H), 1650.44 (C=N), 1566.79 (C=C vinyl),
1548.26, 1497.71 (C=C aromatic), 1454.44 (CH2 methylene), 1160.21 (C-N).
Anal. calcd. (%) for C12H13N2Cl: C, 64.782; H, 5.448; N, 12.63. Found: C, 65.310; H, 5.940;
N, 12.690.
All synthesized ILs were stored in a desiccator due to their hygroscopic property. The name
and structure for all ILs are tabulated in Table 1. The three ILs are hydrophilic and thus they
are used as solvents for the removal of phenol from model oil in this work.
Table 1
Name and structure of benzyl imidazolium-based ILs.
RTIL
Structure
Abbreviation

ACCEPTED MANUSCRIPT
1-Allyl-3-
[ABZIM]Cl
benzylimidazolium
chloride
1,3-
[DBZIM]Cl
Dibenzylimidazolium
chloride
1-Benzyl-3-
vinylimidazolium
chloride
[BZVIM]Cl
2.4 Solubility Test
A preliminary study was carried out in order to investigate the solubility of IL in the model
oil. A specific amount of IL was charged in a capped glass vial and then pure hexane was
added to the IL. Fig. 1 shows the two immiscible phases of [ABZIM]Cl and the model oil.
The IL forms the bottom layer while the model oil forms the top layer. The vial was then
placed in a shaker at room temperature. The vial was shaken for 30 min and then allowed to
1
settle. The upper hexane layer was decanted and analysed by UV-Vis spectroscopy and H
NMR spectroscopy. For UV-Vis analysis, pure IL was dissolved in water to a concentration
of 0.10 g/L and its spectrum was then recorded. The spectrum of final hexane phase was also
recorded. For 1H NMR analysis, pure hexane and final hexane phase were dissolved in CDCl3
while the ionic liquid phase after extraction was dissolved in DMSO-D6. Their NMR spectra
were recorded and analysed accordingly.

ACCEPTED MANUSCRIPT
Fig. 1. Golden brown [ABZIM]Cl at the bottom while colourless model oil on the top.
For UV-Vis analysis, the absorption of the final hexane phase was compared with the
absorption of pure ionic liquid. The maximum absorption of the ionic liquid appeared at 256
nm. No ionic liquid absorption was found in the final hexane phase after extraction. Thus, it
can be concluded that the ionic liquid does not contaminate the model oil (refer to
Supplementary Information, Fig. S1).
Moreover, the 1H NMR spectrum of the final hexane phase showed no evidence of any ionic
liquid contamination. The NMR spectrum of the ionic liquid was also recorded and no
hexane chemical shift was observed. Thus, it can also be concluded that the ionic liquid is not
soluble in the model oil (refer to Supplementary Information, Fig. S2).
2.4.1 Liquid-liquid extraction for quantitative solubility analysis
A quantitative analysis was also carried out in order to determine the solubility of
[ABZIM]Cl in hexane. The procedure involved the mixing of 0.l mL [ABZIM]Cl with 4 mL
hexane under the optimized conditions. A calibration curve was set up by preparing various
concentrations concentration (50ppm, 200ppm, 400ppm, 600 and 800ppm) of standard
solution of [ABZIM]Cl in water, Fig. S3. A standard solution of [ABZIM]Cl with a
concentration of 500 ppm was also prepared as a reference.
After the LLE procedure, the hexane layer was decanted out and the [ABZIM]Cl layer was
diluted with water until its concentration reaches 500 ppm. This sample was then analysed by
UV-Vis spectrophotometer where the maximum absorption of [ABZIM]Cl appeared at
wavelength, λ = 256 nm.
This experiment was done in triplicate and it was found that the solubility of [ABZIM]Cl in
hexane is (0.00215±0.00007) wt%. Almost a similar result was observed when aromatic
compound was used as model oil where the solubility of [ABZIM]Cl in toluene is
(0.0032±0.0004) wt%. The results prove that the solubilities of the IL in hexane and toluene
are very low which indicates that the IL has good properties and it is important for its
application.
2.5 Preparation of model oil and synthetic oil

ACCEPTED MANUSCRIPT
Each phenolic compound was dissolved in hexane to obtain a standard stock solution with a
concentration of 100 ppm. All stock solutions were stored in the refrigerator. Working
solutions were prepared daily by diluting the standard stock solutions with hexane.
In order to study the extractive performance of the ionic liquid in a real sample, a synthetic
oil was prepared simulating those of real oil [18]. In this study, 2,4-DCP, 2,4-DNP and
phenol were chosen as representatives for phenolic compounds while two types of
polyaromatic hydrocarbons (PAHs), fluorene and fluoranthene were chosen to represent the
aromatic compounds. 50 ppm of 2,4-DCP, 2 ppm of 2,4-DNP, 5 ppm of phenol, 0.5 ppm of
fluorene and 1 ppm of fluoranthene were completely dissolved in hexane and then diluted to
25 mL in a volumetric flask.
2.6 Separation process in model oil and synthetic oil
First, a specific amount of IL was charged in a capped glass vial and a certain amount of
model oil with a concentration of 5 ppm was added to the IL. The vial was then placed in a
shaker at room temperature. After shaking for a known time and allowing it to settle, the
upper hexane layer was decanted and the phenol content was analysed by UV-Vis
spectroscopy. Concentrations of the phenol in hexane before and after the extraction were
determined by UV-Vis spectrophotometer in which the maximum absorption of phenol in
hexane appeared at wavelength, λ = 271 nm. In order to determine the concentration of
phenol, various concentrations of standard solution of phenol were prepared and measured to
obtain a calibration curve. Removal efficiency could be calculated by using Eq. 1 as shown
below:
퐶 −퐶
표
% 푅푒푚표푣푎푙 =
푓 × 100%
(Eq. 1)
퐶
표
where Co and Cf represent the phenol concentration in the original model oil and the upper
layer after reaction respectively.
The separation process of phenolic compounds in synthetic oil was performed under
optimized conditions. A specific amount of IL and synthetic oil was charged in a capped
glass vial. After shaking for a set amount of time and allowing it to settle, the upper synthetic
oil layer was decanted and filtered using a PTFE syringe filter (13 mm, 0.22 μm pore size).
The filtrate was transferred into a vial and subsequently injected into the HPLC system for
analysis. The % removal of phenolic compounds was determined by the difference in
chromatograms peak area of the synthetic oil before and after extraction.
3. Results and discussion
3.1 Effect of structure of RTILs
The influences of various cation substituents (benzyl, vinyl and allyl groups) on the
performance of the RTIL in the removal of phenol were investigated.

ACCEPTED MANUSCRIPT
Fig. 2 shows the phenol removal efficiency for the three RTILs. [ABZIM]Cl managed to
remove 78.33% phenol from the model oil while [DBZIM]Cl and [BZVIM]Cl were able to
remove 73.17% and 63.83% phenol respectively. The substituent groups of the RTIL from
the best to the worst performing follows the order of allyl > benzyl > vinyl. There are two
explanations for how the substituents affect the phenol removal performance.
Fig. 2. Effect of RTILs structure on the phenol removal efficiency.
The first explanation is related to the ability of the RTIL to form hydrogen bonds with
phenol. The main mechanism involves the chloride ion (Cl-) of the IL which acts as the
electron donor to the hydroxyl group on the phenol. The Cl- ion has a higher probability to
come in contact with the phenol, therefore causing a higher degree of hydrogen bonding to
occur in the system. Thus, freely moving Cl- anions are favourable in order to have a high
removal efficiency of phenol. It is important to note that the distance of Cl- from the
imidazolium ring in IL is governed by the interionic force between the anion itself and the
hydrogen on the imidazolium ring attached to the C(2) carbon [19, 20]. The numberings of
the ring atoms for the imidazolium structures are depicted in Fig. 3. The strength of the C(2)-
1
H bond can be seen from the H NMR chemical shift value (δ, ppm) of the ILs. The higher
the C(2)-H chemical shift, the stronger its interaction with the anion thus lowering the
1
probability of the anion to form H-bond with phenol [21, 22]. By comparing the H NMR
spectra of all synthesized ILs, it can be seen that the C(2)-H of [ABZIM]Cl had the least
chemical shift (δ, ppm). Thus, it can be concluded that the Cl- of [ABZIM]Cl can move rather
freely to form stronger hydrogen bonds with phenol compared to the other two ILs. The
chemical shift of C(2)-H for the three RTILs are tabulated in Table 2.

ACCEPTED MANUSCRIPT
Fig. 3. The numberings of the ring atoms of imidazolium ring
Table 2
Chemical shift of carbon C(2) proton
RTILs
Chemical shift of carbon C(2) proton δ, ppm
[ABZIM]Cl
[DBZIM]Cl
[BZVIM]Cl
9.550
9.638
9.921
The second explanation of how various substituents on ILs affect its performance can be
attributed to the alkyl chain length on the imidazolium ring. According to literature, the
longer the N-alkyl side chains, the weaker the intramolecular hydrogen bonding between the
C(2)-H and the anions [23]. As a consequence, the ability of longer alkyl chain ILs to remove
phenol increases. This phenomenon can be observed in [BZVIM]Cl whose vinyl side chain is
shorter than the allyl group on [ABZIM]Cl, resulting in lower interaction to form hydrogen
bonds with phenol due to the close distance between the cation and Cl- [3, 7]. The ability of
[ABZIM]Cl to remove phenol from the model oil surpassed that of [DBZIM]Cl due to the
larger volume occupied by the benzyl group compared to that of the allyl substituent. The
large volume of the substituent also increased the molecular volume of [DBZIM]Cl, thus
affecting the distance between the charge centres of the anion and the cation. As a result, the
steric hindrance caused by bigger benzyl group hindered the ability of the chloride ion to
form H-bond with phenol. Therefore, it should be noted that there is an optimum distance
between the chloride ion and cation to form hydrogen bond with phenol [3, 7, 24].
3.2 Effect of phase volume ratio
The phase volume ratio is an important factor for the extraction process. In order to determine
the effect of phase volume ratio of the model oil and IL on the phenol removal efficiency, an
experiment was conducted using [ABZIM]Cl as the representative at 25 ⁰C for 30 min.
Hexane-to-[ABZIM]Cl volume ratios of 20 (4 mL/0.2 mL) to 500 (4 mL/0.008 mL) were
used (Fig. 4). It can be seen that the % removal dropped significantly with increasing phase
volume ratios. The decrease in removal efficiency could be due to the much greater volume
of oil phase compared to IL phase used in this extraction. The performance of [ABZIM]Cl in
removing phenol achieved 99.46% when a phase volume ratio of 20 is used. It is also

ACCEPTED MANUSCRIPT
observed that the removal efficiency remained sufficiently high at 95.24% even when a
higher volume ratio of 40 (4 mL/0.1 mL) is used. Therefore, taking into account the
economic efficiency, a volume ratio of 4:0.1 of model oil to IL was selected as the optimal
value and used in succeeding steps in order to decrease the usage of IL in this extraction.
Fig. 4. Effect of phase volume ratio of [ABZIM]Cl to model oil.
3.3 Effect of phase contact time
The phase contact time is another important parameter in the extraction process because the
mass transfer of analyte between two immiscible phases requires some time to reach
equilibrium [25]. The effect of contact time was investigated comprehensively from 10 min
to 60 min at a volume ratio of 40 and reaction temperature at 25 ⁰C. From Fig. 5, the removal
efficiency increased sharply from 10 min to 30 min. However, the % removal reached a
plateau after 30 min. Further increase of contact time from 40 min to 60 min showed a slight
decrease in phenol removal efficiency which may be due to the evaporation of the hexane
after long extraction times. The removal of phenol reached equilibrium after 30 min and thus,
a duration of 30 min was employed as the optimum time for phenol extraction.

ACCEPTED MANUSCRIPT
Fig. 5. The effect of time on the phenol removal efficiency.
3.4 Effect of temperature
The effect of temperature on the phenol removal efficiency was studied thoroughly at
temperatures 25 ⁰C to 60 ⁰C at a volume ratio of 40 and contact time of 30 min. Fig. 6 shows
the temperature dependence of the phenol removal efficiencies. It can be seen that increasing
the reaction temperature decreased the phenol removal efficiencies. The % removal decreased
gradually from 25 ⁰C to 30 ⁰C and then exhibited a more significant decrease at elevated
temperature. Therefore, from the results obtained, the extraction procedure is best performed
at room temperature.

ACCEPTED MANUSCRIPT
Fig. 6. Effect of temperature on phenol removal efficiency.
3.5 Types of model oil
In order to study the effectiveness of [ABZIM]Cl in removing phenols from oil, n-heptane,
cyclohexane and petroleum ether were selected as the model oils as they represent non-
aqueous environments. Fig. 7 shows that [ABZIM]Cl managed to remove approximately 85
% phenol from all model oils used. Thus, it can be concluded that [ABZIM]Cl is a suitable
solvent to use for liquid-liquid extraction in non-aqueous media.
Fig. 7. Effect of types of model oil on phenol removal efficiency

ACCEPTED MANUSCRIPT
3.6 Selectivity study
Nine phenolic compounds with different nitro and chlorine substituents as well as two other
aromatic compounds; namely xylene and toluene were used as analytes in investigating the
influence of various substituents on the removal efficiency of phenols from model oil. The %
removal efficiency increased in the order of Ph > 4-CP > 3-CP > 2CP > 2,4-DCP > 2,4,6-
TCP for chlorinated phenols while for nitrophenols, it increased in the order of 4-NP > 2,4-
DNP > 2,4,6-TNP under the same experimental conditions. Higher removal efficiencies were
observed for phenols containing lesser amounts of the hydrophobic substituents chlorine and
nitro. This is because hydrophilic interaction is favourable for the reaction between IL and
phenols. As an example, 2,4,6-TCP has the lowest % removal from the oil and this is
attributed to the higher number of chlorine substituents [25-28].
There are usually aromatic compounds other than phenols found in oils. Thus, it is important
to study the selectivity of IL towards those compounds. In our study, there was no extraction
of toluene and xylene from the model oil by the IL. This is because there are no available
sites where H-bond can form between IL and aromatic compounds which is vital for the
extraction. The reaction mechanism is further discussed in the next section. Thus, it can be
concluded that the synthesized IL is only selective towards phenolic compounds and not
towards other aromatic compounds that may be present in the oil.
3.7 Reaction mechanism
The RTILs enable the separation of phenol from the model oil via dissolution of phenol from
the oil phase into the ILs phase. According to literature, there are two significant interactions
that govern the ability of RTILs to extract phenol. The first is the formation of hydrogen
bonds in which the chloride ion acts as the electron donor to the hydroxyl group in the phenol
structure. The high electronegativity of chloride ion favours hydrogen bonding and thus,
plays a crucial role in the extraction process. The second driving force for the removal of
phenol by imidazolium-based IL is the π-π conjugation between the imidazolium ring and the
benzene ring of phenol [29]. Moreover, additional π-π interaction is contributed by the
presence of benzyl and allyl substituents.
The formation of hydrogen bond between phenol and [ABZIM]Cl were analysed from the
1
FT-IR and H NMR spectra. FT-IR is an important tool in determining the presence of
hydrogen bonding [3, 7, 24]. In order to confirm the presence of H-bonding, 1:1 mole ratio
of [ABZIM]Cl and phenol were prepared and analysed. The OH vibrational frequency of
pure phenol could be observed at wavenumber of 3217.16 cm-1 shown in Fig. 8. However,
after extraction using [ABZIM]Cl, this peak shifted to a higher frequency of 3370.84 cm-1.
This change in the absorption peak is mainly due to the migration of the electron cloud in
oxygen atom on the phenol to a higher wavenumber [30].

ACCEPTED MANUSCRIPT
Fig. 8. FT-IR spectra of (a) [ABZIM]Cl-Phenol, (b) [ABZIM]Cl and (c) Phenol
1
In addition, a H NMR analysis was also employed in order to confirm the formation of
hydrogen bond between [ABZIM]Cl and phenol. From Fig. 9, it can be deduced that a
hydrogen bond was formed due to the shifting of OH chemical shift on phenol from 9.2920
ppm to 9.4760 ppm after extraction with [ABZIM]Cl. Moreover, the sharp peak of OH on the
pure phenol spectrum also broadened.
Interestingly, the role of aromatic moieties and unsaturated substituents in strengthening π-π
1
interaction of ILs with phenyl ring of phenol can be proven from their H NMR spectrum.
This is because cationic imidazolium ring modified with unsaturated substituents such as
benzyl and allyl groups could attract aromatic compounds more efficiently compared to their
alkyl counterparts [31]. Table 3 shows the chemical shift changes (Δδ, ppm) for [ABZIM]Cl
after extraction of phenol. The allylic protons of [ABZIM]Cl-Phenol mixture experienced a
downfield shift suggesting that there was a strong interaction between the IL and phenol.
Moreover, greater chemical shift changes are observed on aromatic protons of [ABZIM]Cl
confirming the ability of aromatic moieties in strengthening the π-π interaction and on
imidazolium protons of [ABZIM[Cl namely H10 and H11 proving that imidazolium ring
could also form π-π stacking with phenol. Fig. 10 illustrates the π-π interaction of aromatic
benzene ring of phenol with aromatic and allylic moieties of [ABZIM]Cl as well as H-
bonding of Cl- with hydroxyl group of phenol.

ACCEPTED MANUSCRIPT
Table 3
Proton
δ, ppm
Δδ, ppm
[ABZIM]Cl
5.2805
5.3400
6.0430
4.8685
9.5500
5.4760
7.3725
7.4100
7.4945
7.8970
7.7800
-
Phenol
-
-
-
-
-
-
-
-
-
-
-
[ABZIM]Cl-Phenol
5.2955
5.3630
6.0590
4.8715
9.4760
5.4730
7.3935
7.4230
7.4500
7.7760
7.8850
9.4760
6.7340
6.7730
7.1420
H1
H2
H3
H4
H5
H6
H7
H8
H9
H10
H11
Ha
Hb
Hc
Hd
+ 0.0150
+ 0.0230
+ 0.0160
+ 0.0030
- 0.0740
- 0.0030
+ 0.0210
+ 0.0130
- 0.0445
- 0.1210
+ 0.1050
+ 0.1840
+ 0.0320
+ 0.0535
+ 0.0300
9.2920
6.7020
6.7195
7.1120
-
-
-
(a)
(b)
Fig. 9. 1H NMR spectra of (a) pure phenol and (b) [ABZIM]Cl-Phenol.

ACCEPTED MANUSCRIPT
Fig. 10. The π-π interaction of aromatic benzene ring on phenol with aromatic and allylic moiety of
[ABZIM]Cl as well as H-bonding of Cl- with hydroxyl group on phenol
3.8 Recyclability study
To test the recyclability of [ABZIM]Cl, it is recovered after the first extraction and then
reused for subsequent cycles. Ethyl acetate was used as an efficient back-extraction agent to
recycle ILs due to the significant solubility difference between ILs and phenol in ethyl
acetate. Upon addition of ethyl acetate to IL layer, the phenol reached equilibrium in ethyl
acetate after shaking for 30 min. The ethyl acetate layer was then decanted and the process
was repeated three times to ensure the recycled IL is free from impurities. The same
processes were repeated for each cycle. The phenol removal efficiencies were finally
calculated to evaluate the extraction ability of the recycled IL. Fig. 11 shows that the recycled
IL can be reused several times and the phenol removal performance remains constant after 5
cycles. However, the efficiencies dropped gradually after 6th cycles but the efficiency is still
maintained at approximately 90%.

ACCEPTED MANUSCRIPT
Fig. 11. Phenol removal efficiency versus regeneration cycle of RTIL
To confirm the structure of the recycled RTIL, 1H NMR and FT-IR analyses were carried out.
In Fig. 12, the characteristic absorption peaks of the recycled [ABZIM]Cl were identical to
those of pure [ABZIM]Cl. The identity of the recycled [ABZIM]Cl was also confirmed by
the 1H NMR spectrum shown in Fig. 13.
Fig. 12. FT-IR spectra of pure [ABZIM]Cl and recycled [ABZIM]Cl.

ACCEPTED MANUSCRIPT
(a)
(b)
Fig. 13. 1H NMR spectra of (a) pure [ABZIM]Cl and (b) recycled [ABZIM]Cl.
3.9 Extraction of phenols in synthetic oil
The calculated HPLC chromatogram peak area of the synthetic oil before and after the
extraction showed that [ABZIM]Cl was able to remove about 87.72% of phenol, 64.77% of
2,4-DNP and 63.47% of 2,4-DCP. This result proved that the performance of [ABZIM]Cl is
excellent even in the presence of different types of phenolic compounds and polyaromatic
hydrocarbons (PAHs). The % removal of fluorene and fluoranthene was found to be
negligible. Thus, we concluded that benzyl imidazolium-based ionic liquid is only selective
towards phenolic compounds and not selective to polyaromatic hydrocarbons.
4. Conclusion
It was found that benzyl imidazolium-based IL is a potential alternative extraction medium
for separating phenols from model oil. The presence of benzyl and double bond groups
significantly improve the ability of ILs to remove phenol via hydrogen bonding and enhanced
1
π-π interaction of aromatic moieties. As revealed by the H NMR of [ABZIM]Cl-Phenol
mixture, the benzene ring on phenol interacts with both allylic and benzylic substituents of
[ABZIM]Cl as well as the imidazolium ring. Thus, it can be concluded that [ABZIM]Cl form

ACCEPTED MANUSCRIPT
strong bonds with phenol under optimized conditions with good recyclability up to 6 cycles
without significant loss in mass. This study also revealed that the long double bond group on
the benzyl imidazolium IL allow strong interaction with the phenol due to the freely moving
chloride anion forming hydrogen bonds with phenol compared to the shorter double bond due
to the tightly-attached chloride anion to the C(2)-H on imidazolium ring.
Acknowledgement
This work was supported by UMRG Programme RP020B-16SUS, Fundamental Research
Grant Scheme (FRGS): FP045-2014A. The author would like to thank Ministry of Higher
Education Malaysia for its financial support through MyBrainSc.
References
[1] H.H. Schobert, C. Song, Chemicals and materials from coal in the 21st century, Fuel, 81
(2002) 15-32.
[2] T. Jiao, M. Gong, X. Zhuang, C. Li, S. Zhang, A new separation method for phenolic
compounds from low-temperature coal tar with urea by complex formation, Journal of
Industrial and Engineering Chemistry, 29 (2015) 344-348.
[3] W. Guo, Y. Hou, W. Wu, S. Ren, S. Tian, K.N. Marsh, Separation of phenol from model
oils with quaternary ammonium salts via forming deep eutectic solvents, Green Chem., 15
(2013) 226-229.
[4] S. Yaman, Pyrolysis of biomass to produce fuels and chemical feedstocks, Energy
Conversion and Management, 45 (2004) 651-671.
[5] M. Pera-Titus, V. Garc´ıa-Molina, M.A. Baños, J. Giménez, S. Esplugas, Degradation of
chlorophenols by means of advanced oxidation processes: a general review, Applied
Catalysis B: Environmental 47 (2004) 219-256.
[6] K. Pang, Y. Hou, W. Wu, W. Guo, W. Peng, K.N. Marsh, Efficient separation of phenols
from oils via forming deep eutectic solvents, Green Chemistry, 14 (2012) 2398.
[7] T. Jiao, C. Li, X. Zhuang, S. Cao, H. Chen, S. Zhang, The new liquid–liquid extraction
method for separation of phenolic compounds from coal tar, Chemical Engineering Journal,
266 (2015) 148-155.
[8] M.M. Mahlambi, T.J. Malefetse, B.B. Mamba, R.W. Krause, β-Cyclodextrin-ionic liquid
polyurethanes for the removal of organic pollutants and heavy metals from water: synthesis
and characterization, Journal of Polymer Research, 17 (2009) 589-600.
[9] P. Sun, D.W. Armstrong, Ionic liquids in analytical chemistry, Analytica chimica acta,
661 (2010) 1-16.
[10] R. Hagiwara, Y. Ito, Room temperature ionic liquids of alkylimidazolium cations and
fuoroanions, Journal of Fluorine Chemistry 105, DOI (2000) 221-227.
[11] G.-T. Wei, Z. Yang, C.-J. Chen, Room temperature ionic liquid as a novel medium for
liquid/liquid extraction of metal ions, Analytica chimica acta, 488 (2003) 183-192.
[12] J.-M. Lee, Extraction of noble metal ions from aqueous solution by ionic liquids, Fluid
Phase Equilibria, 319 (2012) 30-36.
[13] T. Welton, Room-Temperature Ionic Liquids. Solvents for Synthesis and Catalysis,
Chem. Rev. , 99 (1999) 2071−2083.
[14] J.G. Huddleston, A.E. Visser, W.M. Reichert, H.D. Willauer, G.A. Broker, R.D. Rogers,
Characterization and comparison of hydrophilic and hydrophobic room temperature ionic
liquids incorporating the imidazolium cation, Green Chemistry, 3 (2001) 156-164.

ACCEPTED MANUSCRIPT
[15] A.E. Visser, R.P. Swatloski, W.M. Reichert, S.T. Griffin, R.D. Rogers, Traditional
Extractants in Nontraditional Solvents: Groups 1 and 2 Extraction by Crown Ethers in Room-
Temperature Ionic Liquids, Industrial & Engineering Chemistry Research, 39 (2000).
[16] Y. Hou, Y. Ren, W. Peng, S. Ren, W. Wu, Separation of Phenols from Oil Using
Imidazolium-Based Ionic Liquids, Industrial & Engineering Chemistry Research, 52 (2013)
18071-18075.
[17] S.A. Sakal, Y.-z. Lu, X.-c. Jiang, C. Shen, C.-x. Li, A Promising Ionic Liquid
[BMIM][FeCl4] for the Extractive Separation of Aromatic and Aliphatic Hydrocarbons,
Journal of Chemical & Engineering Data, 59 (2014) 533-539.
[18] K. Erdmann, T. Mohan, J.G. Verkade, HPLC and 31P NMR Analysis of Phenols in Coal
Liquefaction Oils, Energy & Fuels 10 (1996) 378-385.
[19] R.C. Remsing, J.L. Wildin, A.L. Rapp, G. Moyna, Hydrogen Bonds in Ionic Liquids
Revisited: 35/37Cl NMR Studies of Deuterium Isotope Effects in 1-n-Butyl-3-
Methylimidazolium Chloride, J. Phys. Chem. B, 111 (2007).
[20] P.A. Hunt, C.R. Ashworth, R.P. Matthews, Hydrogen bonding in ionic liquids, Chemical
Society reviews, 44 (2015) 1257-1288.
[21] A. Wulf, K. Fumino, D. Michalik, R. Ludwig, IR and NMR properties of ionic liquids:
do they tell us the same thing?, Chemphyschem : a European journal of chemical physics and
physical chemistry, 8 (2007) 2265-2269.
[22] P.A.Z. Suarez, S. Einloft, J.E.L. Dullius, R.F. de Souza, J. Dupont, Synthesis and
physical–chemical properties of ionic liquids based on 1-n-butyl-3-methyhlimidazolium
cation, J. Chim. Phys., 95 (1998) 1626-1639.
[23] K. Dong, S. Zhang, D. Wang, X. Yao, Hydrogen Bonds in Imidazolium Ionic Liquids, J.
Phys. Chem. A, 110 (2006).
[24] T. Jiao, X. Zhuang, H. He, C. Li, H. Chen, S. Zhang, Separation of Phenolic Compounds
from Coal Tar via Liquid–Liquid Extraction Using Amide Compounds, Industrial &
Engineering Chemistry Research, 54 (2015) 2573-2579.
[25] N. Deng, M. Li, L. Zhao, C. Lu, S.L. de Rooy, I.M. Warner, Highly efficient extraction
of phenolic compounds by use of magnetic room temperature ionic liquids for environmental
remediation, Journal of hazardous materials, 192 (2011) 1350-1357.
[26] J. Fan, Y. Fan, Y. Pei, K. Wu, J. Wang, M. Fan, Solvent extraction of selected
endocrine-disrupting phenols using ionic liquids, Separation and Purification Technology, 61
(2008) 324-331.
[27] V.M. Egorov, S.V. Smirnova, I.V. Pletnev, Highly efficient extraction of phenols and
aromatic amines into novel ionic liquids incorporating quaternary ammonium cation,
Separation and Purification Technology, 63 (2008) 710-715.
[28] M. Ahmaruzzaman, Adsorption of phenolic compounds on low-cost adsorbents: A
review, Advances in colloid and interface science, 143 (2008) 48-67.
[29] Y. Fan, Y. Li, X. Dong, G. Hu, S. Hua, J. Miao, D. Zhou, Extraction of Phenols from
Water with Functionalized Ionic Liquids, Industrial & Engineering Chemistry Research, 53
(2014) 20024-20031.
[30] T. Jiao, X. Zhuang, H. He, L. Zhao, C. Li, H. Chen, S. Zhang, An ionic liquid extraction
process for the separation of indole from wash oil, Green Chem., 17 (2015) 3783-3790.
[31] F. Zhang, Y. Li, L. Zhang, Z. Zhou, W. Sun, Z. Ren, Benzyl- and Vinyl-Functionalized
Imidazoium Ionic Liquids for Selective Separating Aromatic Hydrocarbons from Alkanes,
Ind. Eng. Chem. Res., 55 (2016) 747−756.

ACCEPTED MANUSCRIPT
Highlights
 The effects of the main parameters in liquid-liquid extraction were
investigated.
 Allyl substituted RTIL showed 95% removal efficiency.
 The mechanistic interaction of phenol removal by RTILs was proposed.
 π-π interaction and hydrogen bonding were found to be the ultimate removal
mechanism.
 The recyclability of RTIL was also discussed.