Anion and cation effects of ionic liquids and ammonium salts evaluated as dehydrating agents for super-heavy crude oil: Experimental and theoretical points of view

Article abstract of DOI:10.1016/j.molliq.2014.03.044

One series of ionic liquids (IL), [TOA]⁺ [Y] (where [Y] ^- = [Cl]^-, [HSO₄]^- or [H ₂PO₄]^-), and new ammonium salts, [OCD] ⁺ [Y]^- (where [Y] = [HSO₄]^-, [MeSO₃]^- or [MePhSO₃]^-), were evaluated as demulsifying agents for super-heavy crude oil, at API = 6.39° (MO1) and API = 7.13°(MO2), with 8.46 and 8.94 wt.% of asphaltenes, respectively, at dosages of 1000 and 1500 ppm, using a conventional heating bottle test. At 1000 ppm, the ranking of water removal efficiency for the [TOA]⁺ [Y]^- series (in MO1 and MO2, in order to reach 96.7% and 95.4%, respectively) was [Cl]^- > [HSO₄] ^- > [H₂PO₄]^-. For the [OCD] ⁺ [Y]^- series, the rankings for efficiency at 360 min were [HSO₄]^- > [MeSO₃]^-, reaching approximately 27% in MO2, whereas [MePhSO₃]^- reached 1% in MO2. At a dosage of 1500 ppm, the [TOA]⁺ [Y]^- series achieved 100% water removal, showing the same order at 1000 ppm. For [OCD] ⁺ [Y]^-, the order at 24 h was [HSO₄] ^- > [MeSO₃]^- > [MePhSO₃] ^- for MO1 and MO2 at maximums of 85.5% and 92.3%, respectively. [OCD]⁺ [HSO₄]^- was an efficient demulsifier in tests of long duration. Quantum parameters were obtained at the semiempirical RM1 level, in order to establish a correlation with the experimental results. The molecular volume (V_M), molecular refractivity (MR), dipolar moment (μ) and mean polarizability (α) of the cation and anion fragments were important factors driving the dewatering process. V_M, MR and α were shown to have a direct relationship with water removal efficiency. Furthermore, absolute softness and the electrophilicity index of the cation fragments were calculated at the density functional theory (DFT) level. An increase in the softness and electrophilicity of the cations correlated with lesser effectiveness as demulsifiers. Finally, the partition coefficient of the cations showed that if the viscosity of the crude oil is low, it is possible to use ILs with a low partition coefficient (log P), but if the viscosity is high, the log P of IL should be near the [TOA]⁺ [Cl]^- value.

Full text of DOI:10.1016/j.molliq.2014.03.044

Journal of Molecular Liquids 196 (2014) 249–257
Contents lists available at ScienceDirect
Journal of Molecular Liquids
journal homepage: www.elsevier.com/locate/molliq
Anion and cation effects of ionic liquids and ammonium salts evaluated
as dehydrating agents for super-heavy crude oil: Experimental and
theoretical points of view
a
a
César A. Flores a, , Eugenio A. Flores , Edgar Hernández , Laura V. Castro a,b, Antonieta García c,
⁎
Fernando Alvarez a, Flavio S. Vázquez a
Instituto Mexicano del Petróleo, Programa de Ingeniería Molecular, Eje Central Lázaro Cárdenas No. 152, San Bartolo Atepehuacan, C.P. 07730 Distrito Federal, Mexico, Mexico
Instituto Politécnico Nacional, Escuela Superior de Ingeniería Química e Industrias Extractivas, Unidad Profesional Adolfo López Mateos, C.P. 07738 Distrito Federal, Mexico, Mexico
Instituto Politécnico Nacional, Centro de Investigación e Innovación Tecnológica Cerrada de Cecati S/N, C.P. 02250, Distrito Federal, Mexico, Mexico
a
b
c
a r t i c l e i n f o
a b s t r a c t
One series of ionic liquids (IL), [TOA]+ [Y] (where [Y]− = [Cl]−, [HSO4]− or [H2PO4]−), and new ammonium salts,
[OCD]+ [Y]− (where [Y] = [HSO4]−, [MeSO3]− or [MePhSO3]−), were evaluated as demulsifying agents for
super-heavy crude oil, at API = 6.39° (MO1) and API = 7.13° (MO2), with 8.46 and 8.94 wt.% of asphaltenes,
respectively, at dosages of 1000 and 1500 ppm, using a conventional heating bottle test. At 1000 ppm, the ranking
of water removal efficiency for the [TOA]+ [Y]− series (in MO1 and MO2, in order to reach 96.7% and 95.4%,
respectively) was [Cl]− N [HSO4]− N [H2PO4]−. For the [OCD]+ [Y]− series, the rankings for efficiency at
360 min were [HSO4]− N [MeSO3]−, reaching approximately 27% in MO2, whereas [MePhSO3]− reached 1% in
MO2.
Article history:
Received 12 December 2013
Received in revised form 17 February 2014
Accepted 25 March 2014
Available online 13 April 2014
Keywords:
ILs
Ammonium salts
Water removal
Crude oil
At a dosage of 1500 ppm, the [TOA]+ [Y]− series achieved 100% water removal, showing the same order at
1000 ppm. For [OCD]+ [Y]−, the order at 24 h was [HSO4]− N [MeSO3]− N [MePhSO3]− for MO1 and MO2 at
maximums of 85.5% and 92.3%, respectively. [OCD]+ [HSO4]− was an efficient demulsifier in tests of long dura-
tion. Quantum parameters were obtained at the semiempirical RM1 level, in order to establish a correlation
with the experimental results. The molecular volume (VM), molecular refractivity (MR), dipolar moment (μ)
and mean polarizability (α) of the cation and anion fragments were important factors driving the dewatering
process. VM, MR and α were shown to have a direct relationship with water removal efficiency. Furthermore,
absolute softness and the electrophilicity index of the cation fragments were calculated at the density functional
theory (DFT) level. An increase in the softness and electrophilicity of the cations correlated with lesser effective-
ness as demulsifiers. Finally, the partition coefficient of the cations showed that if the viscosity of the crude oil is
low, it is possible to use ILs with a low partition coefficient (log P), but if the viscosity is high, the log P of IL should
be near the [TOA]+ [Cl]− value.
RM1 level
DFT level
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
done in order to reach two targets: diminishing, as far as possible,
corrosion damage in refinery installations, and preparing the oil for
During its extraction, crude oil is produced as an emulsion with
water. There are different types of such emulsions: water/oil (W/O),
oil/water (O/W), and even ternary emulsions (W/O/W and O/W/O).
These emulsions are stabilized by natural surfactants that are present
in crude oil, as well as by resins, clays of colloidal size, asphaltenes,
carboxylic derivatives, and transition metal salts.
exportation [1].
To break up crude oil emulsions, it is common to add mixtures of
chemical products known as demulsifiers. Different families of these
products have been used, as well as alkylphenol aldehyde resin [2],
polyoxypropylenated–polyoxyethylenated amines [3], alkoxylated
polyhydric alcohols [4], polyalkylene oxide-modified polydimethylsi-
loxane chains [5], aromatic sulfonic acids [6], alkoxylated polyethy-
leneimines [7], orthoester-based polymers [8], hyperbranched
polymers (polycarbonates, polyesters, polyethers, polyurethanes,
polyureas, polyamides, polyetheramines, polyesteramides) [9], phos-
phoric ester composition [10], and magnetic demulsifiers [11]; all of
these are formulations which consist mainly of three components:
breaking, coalescing, and clarifying agents. The aim of such agents is
From the economic and operational points of view, it is imperative
that these emulsions be broken down into their constituents. This is
⁎
Corresponding author at: Programa de Ingeniería Molecular, Instituto Mexicano del
Petróleo, CP 07730, Mexico.
E-mail address: caflores@imp.mx (C.A. Flores).
http://dx.doi.org/10.1016/j.molliq.2014.03.044
0167-7322/© 2014 Elsevier B.V. All rights reserved.

250
C.A. Flores et al. / Journal of Molecular Liquids 196 (2014) 249–257
-
to break up the emulsion in the shortest possible amount of time, as
well by using microwave heating technology [12].
+
Y
NH3
Ionic liquids (ILs) are a new class of chemical products; they are
considered “green solvents,” due to their low-pressure vapor, and
recent reviews give an overview of their potential utilization. A large
number of these molecules have been synthesized and are available
on the market, while some are made only for laboratory use [13,14].
ILs have also been applied in the oil industry — for example, as
asphaltene dispersants [15]; as potential solvents for extracting
asphaltenes from coal residue [16]; as agents in the oxidative desulfur-
ization of fuel oil [17], in the desulfurization of natural gasoline [18], and
in naphthenic acid removal from crude oil by forming ILs [19]; and as
agents for upgrading heavy oil through the separation of asphaltenes
[20], for removing mercaptan from hydrocarbon streams [21], and for
inhibiting clathrate hydrate [22]. Recently, the effect of 1-dodecyl-3-
methylimidazolium chloride [C12mim]+ [Cl]− was measured for its
dynamic interfacial tension with heavy crude oil [23]. ILs have also
been used in oil dehydration. In this vein, two research groups, in the
same year, recently described the application of ILs as demulsifying
agents for crude oil, using microwave (MW) irradiation [24,25], which
has been employed to enhance the demulsification rate and to provide
high efficiency in the breaking up of W/O emulsions under conventional
heating procedures [26–30].
On the other hand, reactivity descriptors such as absolute electro-
negativity, χ; absolute hardness, η; and absolute softness, S, have been
widely used to illuminate the nature of molecules in terms of their
stability, and consequently, their reactivity [31,32]. Moreover, electro-
philicity, as a descriptor of reactivity, provides a quantitative classifica-
tion of the global electrophilic nature of a molecule on a relative scale;
it is effectively the power of a system to “soak up” electrons [33].
The aim of this work was to explore the behavior of emulsions of
heavy and super-heavy crude oil under the influence of one group of
ILs ([TOA]+ [Y]−) and ammonium salts ([OCD]+ [Y]−) in a conventional
bottle test. The first group, [TOA]+ [Y]− (see Fig. 1), was synthesized
previously [34]; the second group, [OCD]+ [Y]− (see Fig. 2), is new —
its anion sizes were varied. Theoretical calculations at the semiempirical
and density functional theory (DFT) levels were carried out in order to
establish the correlation between the quantum parameters and the
experimental behavior of the groups, and therefore, to explore the effect
of anion size on the efficiency of these substances as demulsifiers.
Y- = HSO4, CH3SO3,
SO3
Fig. 2. Ammonium salt [OCD]+ [Y]− series.
2.2.2. Nuclear magnetic resonance (NMR) spectrometry
1H and 13C NMR spectra were collected at 300 K, using a JEOL_Eclipse
spectrometer at 300 MHz and 75.4 MHz for 1H and 13C nuclei, respec-
tively, using CDCl3 as a solvent. The chemical shifts were referenced to
tetramethylsilane (TMS) (1H, δ = 0.0 ppm, and 13C, δ = 0.0 ppm).
2.3. General procedure for the synthesis of ILs and ammonium salts
The trioctylmethylammonium [Y]− ([TOA]+ [Y]−) series was
synthesized according to the procedure described in the literature
[33]. The new [OCD]+ [Y]− series, derivatives of octadecylammonium,
was synthesized according to the route of synthesis presented in Fig. 3.
2.3.1. [OCD]+ [Y]− general synthesis procedure
Toluene (250 mL), with equal mol amounts of octadecylamine and a
strong acid (sulfuric, p-toluene-sulfonic, or methanesulfonic) were
placed in a flask equipped with a magnetic stirrer, thermometer and
reflux condenser. The reaction mixture was refluxed for 24 h; after
that time, the solvent was removed under reduced pressure. Finally,
the solid product was crushed in hexane and vacuum-filtered.
2.4. Spectral characterization of synthesized ammonium salts
2.4.1. Octadecylammonium bisulfate ([OCD]+ [HSO4]−)
IR (KBr, cm−1): 3050, 2923, 2849, 2020, 1628, 1470, 1196, 1052; 1H
NMR (CDCl3) δ 2.916 (m, 2H, J = 7.5 Hz, H1), 1.652 (m, 2H, J = 7.2 Hz,
H2), 1.291 (m, 30H, H3–H17), 0.905 (t, 3H, J = 6.6 Hz, H18); 13C NMR
(CDCl3) δ 40.96, 33.21, 30.92, 30.61, 30.38, 28.73, 27.59, 23.88, 14.63.
2. Experimental procedure
2.4.2. Octadecylammonium mesylate ([OCD]+ [MeSO3]−)
IR (KBr, cm−1): 3040, 2920, 2490, 2000, 1620, 1520, 1460,
1150, 1040; 1H NMR (CDCl3) δ 3.204 (m, 3H, J = 7.2 Hz, HT), 2.916
(m, 2H, J = 7.5 Hz, H1), 1.644 (m, 2H, J = 7.2 Hz, H2), 1.310 (m, 30 H,
J = 6.3 Hz, H3–H17), 0.900 (t, 3H, J = 6.3 Hz, H18); 13C NMR (CDCl3) δ
73.86, 40.94, 33.21, 30.93, 30.39, 28.69, 27.61, 23.88, 23.44, 14.61.
2.1. Materials and methods
The following reagents were purchased from Aldrich: sulfuric
acid (98%), methanesulfonic acid (99.5%), p-toluenesulfonic acid
monohydrate (99.5%), and octadecylamine (97%). These were used
with no further purification.
2.4.3. Octadecylammonium tosylate ([OCD]+ [MePhSO3]−)
IR (KBr, cm−1): 3060, 3050, 2920, 2360, 1620, 1520, 1470, 1120,
1040, 1010, 717, 690; 1H NMR (CDCl3) δ 7.709 (d, 2H, J = 8.1 Hz, HX),
7.24 (dd, 2H, J = 8.4, 0.6 Hz, HW), 2.892 (t, 2H, J = 6.9 Hz, H1), 2.374
(s, 3H, HU), 1.628 (m, 2H, J = 7.2 Hz, H2), 1.290 (t, 30H, J = 6.6 Hz,
H3–H17), 0.902 (t, 3H, J = 6.6 Hz, H18); 13C NMR (CDCl3) δ 143.63,
141.86, 129.99, 127.09, 40.92, 33.23, 30.95, 30.69, 30.39, 28.72, 27.60,
23.89, 14.61, 21.49.
2.2. Apparatus
2.2.1. Fourier transform infrared (FTIR) spectroscopy
FTIR spectra were recorded on a Nicolet FT-IR 5DX FT spectrometer.
+
N
-
Y
2.5. Demulsifier evaluation
n
CH3
The Mexican crude oil samples (MO1 and MO2) were characterized
as described previously [35]. Tables 1 and 2 list the characteristic
physico-chemical properties of the super-heavy Mexican crude oils.
Evaluation by bottle testing was performed roughly as follows: to
graduated bottles (150 mL volume), crude oil was added in quantities
n=3, Y- = Cl, HSO4, H2PO4
Fig. 1. Structures of the [TOA]+ [Y]− series.

C.A. Flores et al. / Journal of Molecular Liquids 196 (2014) 249–257
251
H2SO4
SO4H
-
H3N
+
12
10
4
6
14
16
2
+
CH3SO3H
(Z)
SO3CH3
-
H3N
+
H2N
18
(W)
(U)
(V)
SO3H
(Y)
SO3
H3N
+
(X)
-
Fig. 3. Synthesis of the [OCD]+ [Y]− series of ILs.
of 100 mL. The bottles were placed in a temperature-controlled water
bath at 80 °C for a period of 30 min, after which they were removed.
Subsequently, one aliquot of a dissolution of ILs in xylenes, or ammoni-
um salts in methanol, was added; in the case of untreated oil, an aliquot
of xylene was added; all the bottles were shaken for 2 min at a rate of 2
shakes per second. After mixing, the bottles were placed in the
controlled temperature bath, and the water removal from petroleum
was monitored for 360 min at 1000 ppm and for 24 h at 1500 ppm.
At a dosage of 1000 ppm, [TOA]+ [Cl]− broke up the emulsion of
MO1 to an extent of 0.9% at 5 min (Fig. 4a), and the MO2 crude oil to
an extent of 23% (Fig. 4b). At 10 min, [TOA]+ [Cl]− was able to reach
77.4% and 92.3%, for MO1 and MO2, respectively, whereas [TOA]+
[HSO4]− and [TOA]+ [H2PO4]− in MO1 were able to reach 35.5% and
69.2%, respectively, and with MO2 were able to reach 32.3% and 61.5%,
respectively. The superior effectiveness of [Cl]− as an anion, as opposed
to [HSO4]− and [H2PO4]−, is significant. The maximum WRE values
obtained with MO1 for [TOA]+ [Cl]−, [TOA]+ [HSO4]− and [TOA]+
[H2PO4]− were 96.8%, 96.8% and 93.5%, respectively, whereas with
MO2 the three ILs reached 95.4%. Regarding the ammonium salts,
for MO1 crude oil, the WRE values for [HSO4]−, [MeSO3]−, and
[MePhSO3]− were 3.2%, 1.6% and 0.8%, respectively, and for the MO2
crude oil 27.7%, 22.1% and 0%, respectively.
2.6. Computational procedure
The ILs and ammonium salt structures were fully optimized at the
RM1 level, using Hyperchem 8.0 software [36], using a RMS =
1E−06. The quantitative structure-activity relationship (QSAR) param-
eters of the ILs and ammonium salts were obtained using Hyperchem.
Frontier orbital energies were obtained at the B3LYP/6-31G + (d,p)
level, using Gaussian 03 software [37].
At 1500 ppm and at 5 min, the WRE values of [TOA]+ [Cl]−, [TOA]+
[HSO4]− and [TOA]+ [H2PO4]− with the MO1 crude oil were 80.6%,
77.4% and 35.5%, respectively. [TOA]+ [Cl]− reached 100% at 10 min,
whereas [TOA]+ [HSO4]− and [TOA]+ [H2PO4]− reached maximums of
96.7% and 93.5%, respectively, at 180 min. With the MO2 crude oil, the
WRE values for [TOA]+ [Cl]−, [TOA]+ [HSO4]− and [TOA]+ [H2PO4]−
were 93.9%, 92.3%, and 90.8%, respectively. In this case, a WRE of 100%
for [Cl]−, [HSO4]− and [H2PO4]− was reached at around 15, 25, and
180 min, respectively. In regard to ammonium salts, the emulsion
with MO1 broke up at 20 min for [OCD]+ [HSO4]− and [OCD]+
[MeSO3]−, at 4.5% and 4.8%, respectively. [OCD]+ [MePhSO3]− started
to break up at 40 min, with an efficiency of 3.2%. At 360 min, the
WRE values for [OCD]+ [HSO4]−, [OCD]+ [MeSO3]− and [OCD]+
[MePhSO3]− were 35.7%, 29.0% and 22.1%, respectively, and at 24 h,
85.5%, 77.3%, and 39.3%. The emulsion of MO2 crude oil broke up at
10 min, with [OCD]+ [HSO4]− and [OCD]+ [MeSO3]− yielding WRE
values of 46.2% and 23.1%, respectively. [OCD]+ [MePhSO3]− broke up
the emulsion at 45 min. At 360 min, the WRE values for [OCD]+
[HSO4]−, [OCD]+ [MeSO3]− and [OCD]+ [MePhSO3]− were 76.9%,
73.8% and 43.1%, respectively; and at 24 h, 92.3%, 76.9%, and 53.8%.
3. Results and discussion
New ammonium salts, [OCD]+ [Y]− (where [Y] = [HSO4]−, [MeSO3]−
or [MePhSO3]−), were synthesized according to the established experi-
mental procedures, and the compounds were characterized by FTIR and
NMR spectroscopy. Subsequently, the ammonium salts ([OCD]+ [Y]−
)
and the ILs ([TOA]+ [Y]−) were evaluated as demulsifying agents for
super-heavy crude oil. The anion size was modified in both series to
evaluate the effect on water removal efficiency (WRE). Figs. 4 and 5
show the WRE patterns of the [TOA]+ [Y]− and [OCD]+ [Y]− series at a
dosage of 1000 ppm for MO1 and MO2 crude oils; Fig. 5 shows the
WRE patterns for both crude oils at a dosage of 1500 ppm.
Table 1
Physical data for Mexican super-heavy crude oils.
Parameter
MO1
MO2
Table 2
API density
6.39
7.13
Chemical properties of Mexican super-heavy crude oils ((wt./wt.%).
Salt (lb/1000 bls of oil)
Water by distillation (vol.%)
Kinematic viscosity (mm2 s−1
Pour point (°C)
21,655
56
19,513
60
Type
Saturates
Aromatics
Resins
Asphaltenes
)
35,456
−6
16,073
−6
MO1
MO2
16.09
16.81
21.97
22.51
53.42
51.70
8.46
8.94
Asphaltenes from n-C7
15.71
13.78

252
C.A. Flores et al. / Journal of Molecular Liquids 196 (2014) 249–257
Fig. 4. Water removal efficiency of the [TOA]+ [Y]− series at doses of 1000 ppm: (a) MO1 crude oil; (b) MO2 crude oil.
In contrast to these findings, Guzmán-Lucero et al. [24] carried out an
heavy crude oil after 9 h of heating. With super-heavy crude oil,
microwave-heated at the same dosage, [TOA]+ [Cl]− removed 98% of
the volume of water after 30 min of irradiation. However, the super-
heavy crude oil used by Guzmán-Lucero had only 1245 lb/1000 bbl of
salt, notably lower than the amount of salt present in the MO1
(21,655 lb/1000 bls.) or MO2 (19,513 lb/1000 bls.) crude oils [23]. It is
well known that an increase in salt content causes a reduction of water
removal in the microwave heating test [29]. Guzmán-Lucero mentioned
that the efficiency of the ILs depended only on the alkyl chain size.
In addition, Lemos et al. [25] evaluated the compounds 1-octyl-3-
methylimidazolium tetrafluoroborate ([Omim]+ [BF4]−) and 1-octyl-
evaluation of ten ionic liquids, cation-containing alkylimidazolium
([MIM-C14]+ and [MIM-C18]+), alkylpyridinium ([Py-C14]+ and
[Py-C18]+), and ammonium – in the form of triethylhexylammonium
([TEA-C6]+), triethyldodecylammonium ([TEA-C12]+), triethyloc-
tadecylammonium ([TEA-C18]+), tripentylmethylammonium ([TPA]+),
trihexylammonium ([THA]+) and trioctylmethylammonium ([TOA]+
–
using [TOA]+ [Cl]−) as demulsifier at a dosage of 1000 ppm, and using
the conventional heating test on medium (API = 29.59°), so-called
heavy (API = 21.27°), and super-heavy crude oils (API = 9.88°).
[TOA]+ [Cl]− was the only substance with good performance in super-
Fig. 5. Water removal efficiency of the [OCD]+ [Y]− series at doses of 1500 ppm: (a) MO1 crude oil; (b) MO2 crude oil.

C.A. Flores et al. / Journal of Molecular Liquids 196 (2014) 249–257
253
3-methylimidazolium hexafluorophosphate ([Omim]+ [PF6]−) as
demulsifiers of crude oil, using both microwave heating and conven-
tional heating tests. They used a crude oil of API = 23.3° with a 1.9%
weight of asphaltenes, making a stable emulsion with a range of 30–
50% (v/v) of water. The conventional heating test yielded WRE values
of 24.7% and 42.8% for [Omim]+ [BF4]− and [Omim]+ [PF6]− at dosages
of 5100 and 6200 ppm, respectively. However, using microwave heating
at the same dosage, the efficiency values reached 87.4% and 95.6%,
respectively.
In order to understand their effectiveness for breaking up water/oil
emulsions, the [TOA]+ [Y]− and [OCD]+ [Y]− series, as well as the ILs
described by Guzmán-Lucero and Lemos, were fully optimized at the
semi-empirical RM1 level to obtain the molecular volume (VM), molec-
ular refractivity (MR), and polarizability (α). MR and α parameters are
related to molecular volume and steric effects. As the best demulsifying
agent for MO1 and MO2 crude oils, [TOA]+ [Cl]− is considered a refer-
ence compound. Thus, Figs. 6 and 7 show a graph of the VM and MR of
[TOA]+ [Cl]− versus the gap of [TOA]+ [Cl]−, compared to other ILs
Fig. 7. (a) Graph of MR of [X]+ [Y]− versus the gap of MR of [X]+ [Y]− − [TOA]+ [Cl]−
(b) graph MR of [X]+ versus the gap of MR of [X]+ − [TOA]+; (c) graph MR of [Y]− versus
the gap of MR of [Y]− − [Cl]−
;
.
and ammonium salts; this gap value enables us to assess the behavior
of ILs and ammonium salts in relation to the reference compound. In
Figs. 6 and 7, each IL and ammonium salt is labeled as it appears in
Table 3. It is noteworthy, as seen in Fig. 6a, that the gap value is negative,
so that the IL or ammonium salt is not efficient as a demulsifier of super-
heavy crude oil; conversely, a positive gap value within the 2.3 limit
between compounds 6 and 1 is typically considered acceptable. As for
cations, [TOA]+ shows higher VM and MR values than the other cations
(Fig. 6b). [OCD]+ shows a value near [THA]+, being the lowest negative
gap value, and it shows medium performance as a demulsifier, depend-
ing on which of the anions is employed. In this case, in the comparisons
of [TOA]+ [HSO4]− with [OCD]+ [HSO4]− and [TOA]+ [Cl]− with [THA]+
[Cl]−, the effect of cation size on dewatering efficiency is significant,
showing that the cation must have a specific VM or MR value to be con-
sidered an efficient demulsifier. [MIM-C18]+ and [Py-C18]+ show
lower negative gap values than those of [MIM-C14]+ and [Py-C14]+
;
therefore, ILs containing [MIM-C18]+ or [Py-C18]+ should perform
better as demulsifiers. This assertion is contrary to the observations of
Guzmán-Lucero; however, because they did not give a water efficiency
value for the conventional heating bottle test (only face-scale was
given), it is not possible to arrive at a valid comparison [24]. Indeed,
other factors can influence WRE; these are discussed below.
Fig. 6. (a) Graph of VM of [X]+ [Y]− versus the gap of VM of [X]+ [Y]− − [TOA]+ [Cl]−
(b) graph VM of [X]+ versus the gap of VM of [X]+ − [TOA]+; (c) graph VM of [Y]− versus
the gap of VM of [Y]− − [Cl]−
;
.

254
C.A. Flores et al. / Journal of Molecular Liquids 196 (2014) 249–257
Table 3
Polarizability, molecular volume and molecular refractivity values for several ILs.
LI
VM (Å3)
α (a.u.)
MR (Å3)
LI
Cation
Anion
LI
Cation
Anion
LI
Cation
Anion
1
2
3
4
5
6
7
8
[TOA]+ [Cl]−
1495.8
1571.6
1613.5
1291.1
1315.8
1497.1
1172.9
1012.0
789.1
1111.3
1431.1
1176.1
1358.7
1142.8
1360.8
850.9
1454.8
1454.8
1454.8
1126.6
1126.6
1126.6
1134.3
973.3
134.6
235.3
252.3
235.3
266.4
482.6
134.6
134.6
158.8
158.8
158.8
158.8
158.8
158.8
158.8
194.3
251.5
212.1
224.6
228.0
167.9
172.4
223.2
165.4
142.1
112.6
155.9
201.7
171.2
204.4
178.1
203.3
114.9
119.3
204.4
204.4
204.4
146.9
146.9
146.9
157.8
134.4
102.8
149.3
195.7
156.5
187.9
158.6
189.7
109.3
109.3
0
17.2
19.7
17.2
22.1
72.0
0
0
0
0
0
0
0
0
0
7.4
10.6
125.8
126.9
133.5
94.9
98.0
110.3
98.2
84.4
68.8
96.5
124.1
97.8
119.6
119.6
119.6
86.4
86.4
86.4
92.0
78.2
60.1
87.7
115.3
89.0
107.4
89.9
108.3
63.0
63.0
6.1
12.9
13.6
12.9
16.3
41.1
6.1
6.1
8.7
8.7
8.7
8.7
8.7
8.7
8.7
6.0
9.0
[TOA]+ [HSO4]−
[TOA]+ [H2PO4]−
[OCD]+ [HSO4]−
[OCD]+ [MeSO3]−
[OCD]+ [MePhO3]−
[THA]+ [Cl]−
[TPA]+ [Cl]−
9
[TEA-C6]+ [Br]−
[TEA-C12]+ [Br]−
[TEA-C18]+ [Br]−
[MIM-C14]+ [Br]−
[MIM-C18]+ [Br]−
[Py-C14]+ [Br]−
[Py-C18]+ [Br]−
[Omim]+ [BF4]−
[Omim]+ [PF6]−
724.7
10
11
12
13
14
15
16
17
1046.4
1387.8
1099.9
1310.2
1070.4
1285.8
769.0
116.2
98.7
117.1
66.2
938.4
769.0
67.8
In regard to the anions in Table 3, it is worth noting that there is an
inverse relationship between the VM and MR values of the anion frag-
ments and the water removal efficiency of ILs and ammonium salts.
[BF4]− shows a lower positive gap value than [HSO4]− (Fig. 6c), which
implies that using this anion with an [OCD]+ fragment to obtain the am-
monium salt [OCD]+ [BF4]− would be more efficient than using [OCD]+
[HSO4]−. Fig. 7a and b show the same pattern of behavior as Fig. 6a and
b, except for [Py-C18]+ [Br]− versus [MIM-C18]+ [Br]−, which shows a
less negative gap; this is the same behavior observed for cationic frag-
ments (Fig. 7b). Accordingly, a smaller anion volume presents less steric
hindrance to the natural surfactants that stabilize the emulsion in crude
oil (asphaltenes and paraffins), making small anions more effective for
breaking up the emulsion. In this sense, the steric effect plays an impor-
tant role in the water removal process, although this behavior is oppo-
site to that observed for cation size, as mentioned above. In ILs
containing [BF4]− and [PF6]−, this behavior is not observed; thus,
other factors should be considered. In general, these results help us
decide whether to select or reject the use of ILs or cationic fragments
for dehydrating heavy or super-heavy crude oil without microwave
heating. In sum, ILs or ammonium salts, when the cation size is
increased and the anion size decreased, can perform more efficiently
as dehydrating agents for heavy and super-heavy crude oils.
Table 3 lists the mean polarizability values for ILs, ammonium salts,
cations and anions, and Fig. 8 shows a graph of mean polarizability ver-
sus the gap value for cations. If the cation size increases, as in [TPA]+
,
[THA]+, and [TOA]+, the α values also increase, yielding a high WRE
value. [OCD]+ gave a mean polarizability 57.5 a.u. smaller than
[TOA]+, reflecting the low efficiency of both as demulsifiers. The mean
polarizability of [Cl]− and [Br]− anions was zero, whereas for [BF4]−
,
[PF6]−, [HSO4]−, [H2PO4]−, [MeSO3]−, and [MePhSO3]− anions, it was
7.4, 10.6, 17.2, 19.7, 22.1, and 72.0, respectively. It is significant that an
increase in the polarizability of the anions, mainly in the [TOA]+ [Y]−
and [OCD]+ [Y]− series, provokes a decrease of WRE values. In addition,
quantum parameters of ILs and ammonium salts were obtained to
establish a correlation with the experimental results. The dipolar mo-
ments (μ) of the ILs and ammonium salts do not show any correlation
with the experimental results, whereas the μ of the cations and anions
do show a correlation (Table 4). [OCD]+ shows the highest μ value; not-
withstanding, the [OCD]+ [Y]− series is capable of removing emulsified
water from crude oil of API b 10°, in comparison with the ILs with [TEA-
C6]+, [TEA-C12]+, [TEA-C18]+, [MIM-C14]+, [MIM-C18]+, [Py-C14]+
,
and [Py-C18]+, which are inefficient as demulsifiers for super-heavy
crude oil. μ[Omim]+ is 8.7 Debye higher than μ[TOA]+. ILs with an
[Omim]+ fragment are efficient for removing water (N90%) only at IL
dosages higher than 3000 ppm and only with the use of microwave
heating for the test; furthermore, the crude oil evaluated has a low vis-
cosity and a low amount of asphaltenes, favoring the water removal
process. In the case of the anions, for [Cl]−, [HSO4]− and [H2PO4]−, the
μ
value increased to 0, 2.3, and 2.7 Debye, respectively, as
the demulsification capability decreased (Table 4). [Br]−, [BF4]− and
Table 4
Dipolar moment (μ) of the cation and anion fragments of different ILs.
Cation
μ (Debye)
Anion
μ (Debye)
[TOA]+
4.8
50.8
3.4
2.6
6.8
20.1
35.4
32.4
44.9
29.7
40.9
13.5
[Cl]−
0
0
2.3
2.7
[OCD]+
[Br]−
[THA]+
[HSO4]−
[H2PO4]−
[CH3SO3]−
[MePhSO3]−
[BF4]−
[TPA]+
[TEA-C6]+
[TEA-C12]+
[TEA-C18]+
[MIM-C14]+
[MIM-C18]+
[Py-C14]+
[Py-C18]+
[Omim]+
3.5
10.7
1.6E−04
2.7E−04
[PF6]−
Fig. 8. Graph of α[X]+ versus the gap of α[X]+ − α[TOA]+
.

C.A. Flores et al. / Journal of Molecular Liquids 196 (2014) 249–257
255
Fig. 9. (a) Graph of absolute softness for cations; (b) Graph of electrophilicity for cations.
[PF6]− anions have the same μ value as [Cl]−, but their molecular
volume is different, and as previously mentioned, this parameter has
an important influence on the effectiveness of ILs.
In regard to reactivity parameters, the absolute softness (S) and the
electrophilicity index (ω) of the ILs and ammonium salts did not show
any relationship with the WRE. However, S and ω for the cations
showed a relationship with the experimental results. Fig. 9 shows the
correlation between the S and ω of [TOA]+ and the gap value. Softer cat-
ions than [TOA]+, specifically [OCD]+, are less efficient at removing
water from super-heavy crude oil. [TEA-C12]+, [MIM-C14]+ and
[Py-C14]+ are regular demulsifiers for so-called heavy crude oil, being
harder cations than those with a larger alkyl-chain, which are ineffec-
tive for this type of crude oil. Thus, the WRE values with [MIM-C14]+
should be higher than with [MIM-C18]+; unfortunately, because of
the lack of values, it is not possible to corroborate this. In addition, if
the hardness increases with [THA]+, [TEA-C6]+, and [TPA]+, the WRE
decreases. Even when VM or MR are important factors to be considered
in respect to the ILs employed, the absolute softness parameters seem to
be important parameters to take into account for ILs and ammonium
salts.
Fig. 10. (a) Graph of ω versus EA for cation fragments; (b) amplification of the 7.70–7.88 region. Both graphs are in eV.

256
C.A. Flores et al. / Journal of Molecular Liquids 196 (2014) 249–257
On the other hand, the electrophilicity index, ω, measures the capa-
4. Conclusion
bility of an agent to accept electrons. In this sense, cation-containing
imidazolium and pyridinium structures present high electronic delocal-
ization via resonance effects, and therefore, they show high ω values.
Parr et al. have noted that ω should be related to electron affinity (EA),
because both parameters measure the capability of an agent to accept
electrons; in particular, EA reflects a capability to accept only one elec-
tron from the environment [33]. Considering this correlation between
both parameters, Fig. 10a shows a graph of ω versus EA for cationic frag-
ments where R2 = 0.9928. Fig. 10b shows an amplification of the 7.70–
7.88 region; it is noteworthy that [TOA]+ showed the lowest EA values
for cations with alkyl-fragments. A correlation was also made for ILs;
however, in this case R2 = 0.6645. In sum, an increase in the
capability for accepting electrons (high EA and ω values) reduces the
effectiveness of the ILs and ammonium salts as demulsifying agents,
mainly due to the cation-containing imidazolium and pyridinium
structures.
The IL series [TOA]+ [Y]− and the ammonium salt series [OCD]+ [Y]−
were evaluated as demulsifying agents for super-heavy crude oil, and the
size of the anion was varied. ILs with [TOA]+ were able to reach a WRE
greater than 95%, whereas in the case of the ammonium salts, [OCD]+
[HSO4]− and [OCD]+ [MeSO3]− were more efficient as demulsifiers
than [OCD]+ [MePhSO3]−, which removed 53.8% of water after 24 h.
Quantum parameters at the semiempirical RM1 and DFT levels, such
as VM, MR, S, and ω, enabled us to establish a correlation with the exper-
imental results, and the data reported in the literature evaluating ILs as
demulsifier agents were included. Therefore, it is important to consider
the following: For an IL or an ammonium salt to serve as an effective
demulsifier, it must have: a) a high molecular volume of IL and cations,
b) a high polarizability value for the cations, and c) low molecular re-
fractivity and molecular volume values for the anions.
Recently, Hezave et al. [38] carried out a study of the surfactant effect
of 1-dodecyl-3-methylimidazolium chloride on water/oil interfacial
tension; the partition coefficient (log P) was calculated for the cations,
and it was observed that they were more lipophilic than the cationic
heads of conventional surfactants such as dodecyl(trimethyl)azanium
bromide (DTAB). Considering this study, log P was obtained for the
ILs, ammonium salts, cations and anions (Table 5). There was no lineal
relationship of log P versus WRE; however, if the ILs or ammonium
salts have a log P value lower or higher than 8.69, corresponding to
[TOA]+ [Cl]−, the WRE is low. In the case of the anions for the [TOA]+
[Y]− series, if the log P[Y]− value is near zero, the WRE is higher; thus,
when the anion fragment is more hydrophilic, the ILs are more efficient
demulsifiers. In the [OCD]+ [Y]− series, this statement obtained for the
[HSO4]− and [MePhO3]− anions; however, the [MeSO3]− anion showed
a lower log P than [HSO4]−, but it was a little less efficient than [HSO4]−;
hence, other factors are implicated in this case. In the comparison of
[TOA]+ [HSO4]− with [OCD]+ [HSO4]−, it is noteworthy that with a
more hydrophobic alkyl cation, the IL is more capable of diffusing into
the oil phase, but being less compatible with the hydrophilic phase, it
has a low removal efficiency. Other factors involved, such as π–π inter-
actions of aromatic rings, could also affect the WRE. Hydrophobic
anions, such as [PF6]− and [BF4]−, show intermediate log P values,
whereas their respective ILs show low log P values and are capable of
removing a certain amount of water in a conventional bottle test from
crude oil of API = 23.3°. Hence, if the crude oil is lighter, it is possible
to use ILs with a low log P, but if the viscosity of the crude oil is high,
the log P of the IL should be near the [TOA]+ [Cl]− value.
Acknowledgments
The current study was supported by the Instituto Mexicano del
Petróleo, Molecular Engineering Program project number D.01480.
The authors also would like to thank Henry Jankiewicz for the editing
work that he did for this paper. The authors would also like to thank
M. García Murillo for her assistance.
References
[1] M.F. Ali, M.H. Alqam, Fuel 79 (2000) 1309–1316.
[2] D. Leinweber, E. Wasmun, Resins comprised of alkyl phenols and of glyoxylic acid
derivatives, and their use as demulsifiers, US Patent 2004/0102586A1.
[3] N.N. Zaki, N.E.S. Maysour, A.A.A. Abdel-Azim, Petrol. Sci. Technol. 18 (2000) 1009–1025.
[4] X.R. Xu, J.Y. Yang, B.L. Zhang, J.S. Gao, Petrol. Sci. Technol. 25 (2007) 1375–1390.
[5] D. Daniel-David, I. Pezron, D. Clausse, C. Dalmazzone, C. Noik, L. Komunjer, Phys.
Chem. Chem. Phys. 6 (2004) 1570–1574.
[6] R. Varadaraj, C.H. Brons, Aromatic sulfonic acid demulsifier for crude oils, US Patent
2003/0092779A1.
[7] G. Elfers, W. Sager, H.H. Vogel, K. Oppenlaender, Petroleum emulsion breakers, US
Patent 5445765.
[8] P.E. Hellberg, I. Uneback, Enviromentally-friendly oil/water demulsifiers, WO Patent
2007/115980 A1.
[9] B. Bruchmann, K.H. Buchner, M. Guzmann, G. Brodt, S. Frenzel, Hyperbranched poly-
mers for use as demulsifiers for cracking crude oil emulsions, US Patent 2008/
0153931 A1.
[10] F.T. Lang, Phosphoric ester demulsifier composition, US Patent 2006/0036057A1.
[11] J. Peng, Q. Liu, Z. Xu, J. Masliyah, Energy Fuels 26 (2012) 2705–2710.
[12] A.N.I. Anisa, A.H. Nour, A.H. Nour, J. Appl. Sci. 11 (2011) 2898–2906.
[13] T. Welton, Chem. Rev. 99 (1999) 2071–2084.
[14] J.P. Hallet, T. Welton, Chem. Rev. 111 (2011) 3508–3576.
[15] M. Boukherissa, F. Mutelet, A. Modarressi, A. Dicko, D. Dafri, M. Rogalski, Energy
Fuels 23 (2009) 2557–2564.
[16] Y. Nie, L. Bai, Y. Li, H.F. Dong, X.P. Zhang, S.J. Zhang, Ind. Eng. Chem. Res. 50 (2011)
10278–10282.
[17] D.S. Zhao, Y.A. Wang, E.H. Duan, Molecules 14 (2009) 4351–4357.
[18] N.V. Likhanova, D. Guzman-Lucero, E.A. Flores, P. Garcia, M.A. Dominguez-Aguilar, J.
Palomeque, R. Martinez-Palou, Mol. Divers. 14 (2010) 777–787.
[19] L.J.J. Shi, B.X.X. Shen, G.Q.Q. Wang, Energy Fuels 22 (2008) 4177–4181.
[20] M. Siskin, M.A. Francisco, R.M. Billimoria, Upgrading of heavy hydrocarbons by the
separation of asphaltenes using ionic liquids, WO Patent 2008/124042A1.
[21] D.J. O'Rear, L.C. Boudreau, M.S. Driver, C.L. Munson, Removal of mercaptans from
hydrocarbon streams using ionic liquids, WO Patent 2002/034863.
[22] H. Adidharma, C.W. Xiao, Dual function gas hydrate inhibitors, WO Patent 2009/
114674.
[23] A.Z. Hezave, S. Dorostkar, S. Ayatollahi, M. Nabipour, B. Hemmateenejad, J. Mol. Liq.
187 (2013) 83–89.
[24] D. Guzmán-Lucero, P. Flores, T. Rojo, R. Martínez-Palou, Energy Fuels 24 (2010)
3610–3615.
[25] R.C.B. Lemos, E.B. da Silva, A. dos Santos, R.C.L. Guimarães, B.M.S. Ferreira, R.A.
Guarnieri, C. Dariva, E. Franceschi, A.F. Santos, M. Fortuny, Energy Fuels 24 (2010)
4439–4444.
[26] L. Xia, S. Lu, G. Cao, J. Colloid Interface Sci. 271 (2004) 504–506.
[27] A.H. Nour, R.M. Yunnus, J. Appl. Sci. 6 (2006) 1698–1702.
[28] A.H. Nour, R.M. Yunnus, J. Appl. Sci. 6 (2006) 1868–1872.
[29] A.H. Nour, R.M. Yunnus, J. Appl. Sci. 6 (2006) 2307–2311.
[30] M. Fortuny, C.B. Oliviera, R.L. Melo, M. Nele, R.C. Coutinho, A.F. Santos, Energy Fuels
21 (2007) 1358–1364.
Table 5
Partition coefficients of the ILs.
Compound
Cation
Anion
[TOA]+ [Cl]−
8.69
9.55
8.89
6.32
5.13
3.64
6.02
8.40
6.78
8.37
7.75
9.33
11.95
10.47
13.15
4.60
5.77
8.06
8.06
8.06
5.86
4.49
2.71
5.09
7.47
5.85
7.43
6.81
8.40
5.70
5.70
5.70
2.26
2.26
0.63
1.41
1.88
0.63
0.63
0.94
0.94
0.94
0.94
0.94
0.94
0.94
1.41
1.17
3.32
1.37
2.06
[TOA]+ [HSO4]−
[TOA]+ [H2PO4]−
[THA]+ [Cl]−
[TPA]+ [Cl]−
[TEA-C6]+ [Br]−
[TEA-C12]+ [Br]−
[TEA-C18]+ [Br]−
[MIM-C14]+ [Br]−
[MIM-C18]+ [Br]−
[Py-C14]+ [Br]−
[Py-C18]+ [Br]−
[OCD]+ [HSO4]−
[OCD]+ [MeSO3]−
[OCD]+ [MePhO3]−
[Omim]+ [BF4]−
[Omim]+ [PF6]−
[31] R.G. Parr, R.A. Donnelly, M. Levy, W.E. Palke, J. Chem. Phys. 68 (1978) 3801–3807.
[32] R.G. Parr, R.G. Pearson, J. Am. Chem. Soc. 105 (1983) 7512–7516.
[33] R.G. Parr, L.V. Szentpály, S. Liu, J. Am. Chem. Soc. 121 (1999) 1922–1924.

C.A. Flores et al. / Journal of Molecular Liquids 196 (2014) 249–257
257
[34] J.P. Mikkola, P. Virtanen, R. Sojholm, Green Chem. 8 (2006) 250–255.
[35] L.V. Castro, F. Vazquez, Energy Fuels 23 (2009) 1603–1609.
[36] Hyperchem 6.0, 1993. Hypercube Inc. Florida.
[37] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, J.A.
Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar, J.
Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson, H.
Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P. Hratchian, J.
B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin,
R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P.
Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O.
Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachair, J.B. Foresman, J.V. Ortiz, Q. Cui,
A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara,
M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A.
Pople, Gaussian 03, Revision B.03, Gaussian Inc., Pittsburgh, 2003.
[38] A.Z. Hezave, S. Dorostkar, S. Ayatollahi, M. Nabipour, B. Hemmateenejad, Colloid
Surf. A 421 (2013) 63–71.