See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/283445595

Surfactant-free alternative fuel: Phase behavior and diffusion properties

Article  in  Journal of Colloid and Interface Science · November 2015

DOI: 10.1016/j.jcis.2015.10.040

CITATIONS READS

6 6,258

4 authors:

Ibrahim Kayali Jyothi Chembolli Kunhatta

Al-Quds University 4 PUBLICATIONS   46 CITATIONS   
40 PUBLICATIONS   359 CITATIONS   
SEE PROFILE
SEE PROFILE

Khawla Qamhieh Ulf Olsson

Al-Quds University Lund University
25 PUBLICATIONS   227 CITATIONS    294 PUBLICATIONS   9,713 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Small-Angle Scattering for the study of cortical bone nano-structure View project

Peptide self-assembly View project

All content following this page was uploaded by Ibrahim Kayali on 09 December 2017.

The user has requested enhancement of the downloaded file.

 Journal of Colloid and Interface Science 463 (2016) 173–179

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science

journal homepage: www.elsevier.com/locate/jcis

Surfactant-free alternative fuel: Phase behavior and diffusion properties

Ibrahim Kayali a,b,⇑, Chemboli K. Jyothi a, Khawla Qamhieh a,b, Ulf Olsson a
a
Physical Chemistry Division, Chemistry Department, Lund University, Lund, Sweden
b
College of Science & Technology, Al-Quds University, Jerusalem, Palestine

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o a b s t r a c t

Article history: Phase behavior of the three components, 1-propanol, water and oil is studied at 10, 25, and 40 °C.
Received 20 August 2015 Biodiesel, limonene and diesel are used as oil phases. NMR self-diffusion measurements are performed
Revised 13 October 2015 to investigate the microstructure of the one-phase regions. Tie lines in the two-phase regions are deter-
Accepted 15 October 2015
mined both by proton NMR analysis and compared with theoretical calculations. NMR self-diffusion
Available online 23 October 2015
results for the different components in these systems do not show any sign of confinement or obstruc-
tions, demonstrating these mixtures to be structureless solutions. A good agreement between the exper-
Keywords:
imental and calculated phase behavior is obtained. The determined tie lines in the two-phase regions
Surfactant-free microemulsion
Fuel microemulsion
show higher affinity of 1-propanol to water than to oil.
NMR self-diffusion coefficient Ó 2015 Elsevier Inc. All rights reserved.
Tie lines

1. Introduction monly known as biodiesel, is one of the candidates that actually

The concern over environmental pollution and the expected
depletion of fossil fuel have raised the attention toward finding
alternative renewable fuel. Several promising non-fossil oxy-
genated fuels with the potential of reducing the pollutant emis-
sions like soot, NOx, CO and CO2 have been investigated. The
methyl ester of fatty acids made from different vegetable oils, com-
⇑ Corresponding author at: B.O. Box 20287, Jerusalem, Palestine.
E-mail address: ikayali@science.alquds.edu (I. Kayali).
has been tested as an alternative and as an additive to ordinary die-
sel [1]. The fatty acid composition corresponds to that of its parent
oil. Several specifications like cold-flow, cetane number and oxida-
tive stability can vary according to this difference in composition
[2]. For example, the cetane number, which is related to the igni-
tion quality, increases with increasing chain length and saturation
of the parent fatty acids. Viscosity, which is the main reason why
oils can’t be used directly as fuel and have to be esterified, usually
increases with increasing number of carbons in the chain and
decreases with increasing unsaturation. Oxidative stability

http://dx.doi.org/10.1016/j.jcis.2015.10.040
0021-9797/Ó 2015 Elsevier Inc. All rights reserved.
174 I. Kayali et al. / Journal of Colloid and Interface Science 463 (2016) 173–179
depends on the number of double bonds present in the unsatu-
rated fatty acids. Oleic acid, which constitute 52–65% of rapeseed
oil, has relatively higher oxidative stability than linoleic acid,
which is a main constituent of soybean oil. Therefore the enrich-
ment of biodiesel with methyl oleate would improve its stability
[3]. Alcohol is another attractive non-fossil oxygenated renewable
resource [4]. The use of ethanol as a 10 % dry ethanol (E10) in gaso-
line is a common practice in order to reduce vehicle exhaust emis-
sions. Blending ethanol with diesel reduces the soot formation
substantially [5]. However adding ethanol to diesel can also lower
the cetane number, decrease the energy content and have limited
miscibility at lower temperatures. Adding a suitable emulsifier will
solve the problem of miscibility and will contribute positively
toward improving the cetane number [4].
Limonene oil extracted from citrus peel has also been investi-
gated as a renewable source of fuel. Combustion experiments with
30 % orange oil-diesel blend resulted in reducing the CO, the hydro-
carbon and smoke emissions while increasing NOx emissions [6].
The use of water in fuel emulsion and microemulsion in com-
bustion engines has also been investigated [7]. The presence of
water facilitated atomization of fuel through the microexplosions,
therefore improving the combustion efficiency. For water in diesel
microemulsion, significant reduction in soot, NOx, CO and CO2
emissions were observed compared to neat diesel [8]. On the
nanometer length scale, microemulsion has a structure with oil
and water domains separated by surfactant films at the interface
[9]. Microemulsion may consist of water droplets in oil (W/O), oil
droplets in water (O/W) or they can be bicontinuous, in which both
oil and water form domains that are continuous in all three dimen-
sions. Due to the high interfacial volume in microemulsion, high
amount of surfactants are needed in the formulation. In addition,
the surfactant to be used should burn readily without forming
smoke and should not contain nitrogen or sulfur [10].
Surfactant-free microemulsions have also been suggested.
Using conductivity measurements and visual examination, a micro
emulsified ternary solution composed of water, hexane and 2-
propanol was indicated [11]. A single-phase region composed of
water, oleic acid and n-propanol was also studied [12]. The
microstructure of the single-phase region investigated by electrical
conductivity, showed the existence of three different micro
regions, water in oleic acid, a bicontinuous and oleic acid in water
region. The results were further confirmed by freeze-fracture and
cryogenic transmission electron microscopy. The ternary system
water/ethanol/octanol has been studied extensively by various
scattering methods [13–15]. A microemulsion region, composed
of water rich and octanol rich domains, respectively, has been pro-
posed in the vicinity of the plait point. Concentration fluctuations,
observed in small angle scattering as an Ornstein–Zernike behav-
ior, increases upon approaching phase separation. Solutions near
a plait point are clearly non-ideal. Whether the proposed
‘‘surfactant-free microemulsions” are not simply enhanced con-
centration fluctuations in near-critical conditions has not yet been
clearly demonstrated. Mixtures of octanol, ethanol and water have
also been studied by molecular dynamics simulations [16].
Recently, work has been extended to other polar solvent [17].
NMR self-diffusion studies can provide a reliable method to
examine the internal structure of microemulsions [18]. In fact,
the first experimental evidence that microemulsions may be bicon-
tinuous came from NMR self-diffusion data [19]. The term bicon-
tinuous has been used to imply the case where large fractions of
both oil and water molecules form channels extending over macro-
scopic distances in all three dimensions. Multi-component self-
diffusion measurements can also provide direct insight into other
types, namely W/O and O/W droplet microemulsion. The solvent
forming the continuous phase will have a diffusion coefficients
near that of neat liquid while the other solvent confined to dis-
crete, disconnected droplets will have a much lower diffusion coef-
ficients due to confinement and obstruction of molecules.
In addition to the above three types of microemulsions, a fourth
possibility of structureless solutions may exist when weak amphi-
philes are being used [20]. Physical properties for these mixtures,
such as surface tension and diffusion coefficients, have been found
to be similar to those of conventional microemulsion [21]. Struc-
tureless microemulsion has also been reported for some non-
aqueous systems even in the presence of surfactant [22,23]. Using
self-diffusion coefficient measurements, the non-aqueous glycerol/
hexanol/SDS isotropic system did not show any obvious confine-
ment characteristics and the authors suggested a structureless
simple solution. The same conclusion was also given for the four-
component system with the added p-xylene.
Surfactant free mixtures of water, hexane and isopropanol have
been used as a medium for enzymatic reactions [24]. In the ternary
phase diagram, the one phase transparent region was suggested to
consist of 3 different sub regions: a normal ternary solution with
no detected microstructure, a microemulsion composed of aque-
ous microdroplets dispersed in hexane-rich continuous phase
and an intermediate region of H-bonded aggregates of water and
isopropanol dispersed in hexane-rich medium [25].
In this article, we have investigated the ternary phase diagrams
of 1-propanol, water and oils. NMR self-diffusion measurements
are conducted to elucidate the microstructure of the one-phase
regions. Proton NMR measurements, as well as theoretical calcula-
tions, are used to determine the tie lines in the two-phase regions
for these systems. NMR self-diffusion has been used previously to
characterize self-assembly in near-critical mixtures [26], where it
may be particularly useful as the self-diffusion often is not influ-
enced by the composition fluctuations.
2. Materials and methods
2.1. Materials
1-propanol (99.8%) and R(+)-Limonene, technical grade (90%)
were purchased from Sigma–Aldrich, Sweden. European ultra low
sulfur diesel was obtained from a local gasoline station in Lund
Sweden. The equivalent composition for diesel is C12H23. The fatty
methyl ester, biodiesel, from rapeseed oil, with methyl oleate as its
main constituent, was from Tecosol (Ochsenfurt, Germany). Milli-
pore water was used in all experiments.
2.2. Methods
2.2.1. Ternary phase diagrams
In order to determine the location and boundaries of the differ-
ent phases on the ternary phase diagrams, samples were prepared
by adding water drop wise to pre-weighed mixtures from the other
two components in glass test tubes with screw caps. Vigorous stir-
ring followed all the aqueous phase additions using a vortex mixer.
The time for equilibration between each addition was typically
1–2 min. Phase diagrams were investigated at 10, 25 and 40 °C.
The number of phases were detected by bare eyes and between
cross polarizers.
2.2.2. NMR self-diffusion experiments
Self-diffusion coefficients were measured by the stimulated
echo pulsed field gradient (PGSE-NMR) method using a Bruker
AVANCE II 200 spectrometer with Bruker field gradient probe oper-
ating at proton resonance frequency of 200 MHz. The temperature
was kept at 25 °C with an accuracy of ±0.1 °C. The spin-echo ampli-
tude, I, of a given NMR resonance is given by [27]:
I ¼ I0 expðkDÞ ð1Þ
 I. Kayali et al. / Journal of Colloid and Interface Science 463 (2016) 173–179 175

where I0 is the echo amplitude in the absence of field gradient

pulses, D is the self-diffusion coefficient of the molecule and
k = c2g2d2(D  d/3), where c is the gyromagnetic ratio, g is the
magnetic field gradient strength, d is the gradient duration and D
is the time between the gradient pulses. Experiments were carried
out at constant D, 50 and 100 ms and at appropriate g. The self-
diffusion coefficients of water were obtained from the spin-echo
signal at around 4.8 ppm.

2.2.3. Tie lines by 1H NMR spectroscopy

Samples from the two-phase regions were prepared such that
the ratio of water to oil was held constant at 1:1, while the amount
of 1-propanol was increased. The samples were then thoroughly
separated into the constituent phases. 1H NMR spectra were
obtained for the top and bottom layers of the systems using a BRU-
KER AVANCE III 500 MHz. 1H NMR spectra were obtained under the Fig. 1. Ternary phase diagram for biodiesel, 1-propanol and water at temperatures
following conditions: spectral width, 2000 Hz; repetition time, 10, 25 and 40 °C. 1U indicates the one-phase region and 2U represents the two-
60 s; pre-acquisition delay, 58 s; pulse width, 90°; data points, phase region.
7998 and number of scans, 8.

trum. This is crucial for demonstrating any microstructural

2.2.4. Tie lines by theoretical calculations
The regular solution model [28] was used to calculate the tie
lines of the three ternary phase diagrams, propanol/water/biodie-
sel, propanol/water/diesel, and propanol/water/ limonene systems.
The theory describes the behavior between three molecular com-
ponents, which are assumed to mix in a fully random pattern,
and the interactions are assumed to be restricted to the nearest
neighbors. The different phases at equilibrium can be determined
by minimizing the total free energy G for the system, which can
be given as
G ¼ RT½Ntot fu1 lnðu1 Þ þ u2 lnðu2 Þ þ u3 lnðu3 Þ þ v12 u1 u2
þ v13 u1 u3 þ v23 u2 u3 g
where ui represents the mole fraction of i component, R is the gas
constant, T is the temperature in Kelvin, and Ntot is the total size
of the lattice.
We optimized the values of the interaction parameters between
propanol–water (vPW), propanol–oil (vPO), and water–oil (vWO), to
fit the experimental phases. Throughout, the simulation package
POLYMER, Version 3.4.2, P. by Per Linse, at Lund University, in Swe-
den, was used.
3. Results and discussion
transition from W/O to bi continuous to O/W microemulsion. In
addition, very large variations in relative diffusion coefficients
should be detected for such a transition.
In Fig. 2, the NMR spectrum of pure biodiesel is shown. The
spectrum displays the following: in the aliphatic region, a methyl
peak around 0.9 ppm and a methylene peak in the range of
1.3–1.4 ppm. The methylene beta to the ester group is at 1.6 ppm
and the methylene alpha is at 2.3 ppm. The peaks at 2.1 ppm,
2.8 ppm and 5.4 ppm are due to the protons located at or near
the double bonds within the unsaturated methyl esters, 18:1,
18:2 and 18:3. A sharp peak at 3.7 ppm is related to the ester
methyl located next to the carbonyl carbon. It should be noted that
ð2Þ
the results agree very well with previously published data [29].
Fig. 3 shows 1H NMR spectrum for sample [7] on the biodiesel
phase diagram as an example. Diffusion coefficients for all compo-
nents are determined for the peaks indicated and the values are
presented in Table 1. For samples [1–4], the diffusion coefficients
for biodiesel ranges between 3.0 and 7.4 ⁄ 1010 m2/s and that
for water between 2.2 and 7.4 ⁄ 1010 m2/s, indicating the absence
of any obstruction or confinement characteristics in the solution.
For 1-propanol, the diffusion coefficients increased from 4.5 to
11.6 ⁄ 1010 m2/s at higher amount of 1-propanol. This may be

3.1. Phase behavior and self-diffusion properties

3.1.1. Biodiesel microemulsion

Fig. 1 presents the ternary phase diagram for biodiesel,
1-propanol and water at temperatures 10, 25 and 40 °C. The region
labeled 1U represents the single, clear and homogeneous phase
while the 2U corresponds to the unstable emulsion region that
tend to separate rapidly into two phases upon standing. Varying
temperature from 10 to 40 °C had little effect on the phase bound-
aries between the one-phase and two-phase regions as shown in
the figure. Increasing the temperature to 40 °C slightly increased
the one-phase region while at 10 °C, there was a slight decrease
compared to that at 25 °C.
In order to explore the microstructure in the one-phase region,
self-diffusion coefficient measurements were performed for sam-
ples [1–4] along the line extending from the biodiesel corner
toward the 1-propanol–water axis. In addition, samples [5–8] were
chosen at constant 1-propanol ratio for the biodiesel-rich and
water-rich domains. The great advantage of the self-diffusion
NMR technique, besides being noninvasive, is the ability to deter-
mine the diffusion for all three components from the same spec- Fig. 2. Proton NMR spectrum of neat biodiesel. See text for details.
176 I. Kayali et al. / Journal of Colloid and Interface Science 463 (2016) 173–179

Fig. 4. Ternary phase diagram for limonene, 1-propanol and water at temperatures
10, 25 and 40 °C.

Fig. 3. Proton NMR spectrum of biodiesel microemulsion. Diffusion coefficients for

the three components biodiesel (BD), 1-propanol (P) and water (W) are determined
for the indicated peaks.

due to the fact that 1-propanol has lower viscosity (2.1 mPa s) than
biodiesel (6.8 mPa s) at room temperature. The results for samples
[5–9] at constant amount of alcohol showed similar values for
1-propanol, meanwhile the diffusion of biodiesel ranges between
2.9 and 4.4 ⁄ 1010 m2/s and that for water between 10.4 and
5.5 ⁄ 1010 m2/s. Again these variations may be attributed to the
high viscosity of biodiesel compared to that for water. There were
no large variations in the relative diffusion coefficients (D/D°) for
the three components for samples from different locations of the
one-phase region, again indicating the absence of any obstruction
effect.

3.1.2. Limonene microemulsion

Limonene produced from a renewable source, is an inexpensive
monoterpenic hydrocarbon that has been extensively used in
industry, especially in the fragrance sector. Limonene microemul-
sions have also been investigated using mainly nonionic surfac-
tants [30–32]. The transition in microstructure from W/O to
bicontinuous to O/W was noticed from the sharp variations in
the diffusion coefficients of both water and oil upon dilution
[32]. Fig. 4 presents the ternary phase diagram for limonene,
1-propanol and water at temperatures 10, 25 and 40 °C. The
single-phase region is relatively smaller than the one obtained
for biodiesel. Again, varying temperatures between the range
10–40 °C had limited effect on the phase boundaries. Fig. 5 shows
the NMR spectrum of pure limonene. The spectrum displays the
following known features from limonene: a methyl group on the
cyclohexene ring at 1.6 ppm and a terminal methyl group at
Fig. 5. Proton NMR spectrum of neat limonene. See text for details.
around 1.7 ppm. The peaks at 1.5 ppm, 1.9 ppm and 2.1 ppm are
due to protons located on the cyclohexene ring. The peak at
4.7 ppm is due to protons located at the terminal double bond
while the peak at 5.4 ppm is related to the proton at the double
bond on the ring. Fig. 6 presents the NMR spectrum for sample 5,
as an example, on the limonene phase diagram with the composi-
tion: 60 wt.% 1-propanol; 20 wt.% water and 20 wt.% limonene.
Diffusion coefficients for all components are determined from the
peaks indicated and values are presented in Table 2. Samples
[1–3] are from the water-rich area, while samples [6–8] are from
the oil-rich area and samples [4,5] are in the region of equal

Table 1
Diffusion coefficients of the components in the biodiesel micro emulsions.

Micro emulsion compositions Water 1-Propanol Biodiesel

D ⁄ 1010 m2/s D/D° D ⁄ 1010 m2/s D/D° D ⁄ 1010 m2/s D/D°
(1) 20%P 2%W 78%BD 2.2 0.10 4.5 0.57 3.0 1.40
(2) 50%P 6%W 44%BD 3.0 0.13 6.1 0.78 4.0 1.90
(3) 72%P 8%W 20%BD 4.3 0.19 6.4 0.83 4.3 2.05
(4) 86%P 9%W 5%BD 7.4 0.32 11.6 1.49 7.4 3.50
(5) 60%P 40%W – 9.0 0.39 6.2 0.80 – –
(6) 60%P 34%W 6%BD 10.4 0.42 5.5 0.74 2.9 1.36
(7) 60%P 20%W 20% BD 6.7 0.29 5.6 0.72 3.1 1.50
(8) 60%P 6%W 34% BD 5.5 0.24 5.5 0.71 3.4 1.63
(9) 60%P – 40%BD – – 6.3 0.80 4.4 2.10
 I. Kayali et al. / Journal of Colloid and Interface Science 463 (2016) 173–179 177

Fig. 7. Ternary phase diagram for diesel, 1-propanol and water at 25 °C.

Fig. 6. Proton NMR spectrum of limonene microemulsion. Diffusion coefficients for

the three components, limonene (L), 1-propano (P) and water (W) are determined
for the indicated peaks.
amounts of water and oil. As seen in Table 2, the diffusion coeffi-
cients for water and limonene are in the same size range for most
of the samples regardless whether the sample is located in the
water-rich or oil-rich region. This clearly demonstrates the absence
of any aggregates or structure that might obstruct or hinder molec-
ular diffusion. The diffusion coefficients for 1-propanol varies in
the range 5.4 –18.6 ⁄ 1010 m2/s, having the higher values for
samples from the oil-rich region. This may be due to the fact that
limonene has lower viscosity (0.9 mP s at room temperature) than
1-propanol. The same conclusion, i.e. the absence of structure, is
reached by examining the D/D° values for the different compo-
nents presented in Table 2.
3.1.3. Diesel microemulsion
Surfactant stabilized water in diesel microemulsion has been
studied by NMR diffusometry [33]. The obtained diffusion coeffi-
cients for the different components suggest the presence of spher-
ical water domains especially at low water contents. The large
amount of the nonionic surfactant used in that study, helped to
stabilize the interface between water and diesel.
As stated in the introduction, blending diesel with alcohol has
been receiving attention as a way for reducing particulate matter
emissions. Surfactant is usually needed to prevent any phase sep-
aration due to the presence of water or at low temperatures.
Among a number of short-chain alcohol tested to formulate
microemulsion with water and hexane in the absence of surfactant,
only 1-propanol or 2-propanol could stabilize such systems [25]. In
another study, a system consisting of water and diesel: ethanol
(9:1) was turbid and needed at least 14 wt.% from the nonionic sur-
factant to become clear one-phase microemulsion [34].
In the present investigation, ternary phase diagram composed
of water, diesel and 1-propanol is constructed and presented in
Fig. 7. The one-phase region is much smaller than the ones
obtained for biodiesel or limonene. At least 75 wt.% of 1-propanol
is needed to stabilize the microemulsion at water: diesel ratio of
(1:1) compared to 50 wt.% when diesel is replaced by biodiesel.
Because of the smaller solution phase, no NMR diffusion measure-
ments were performed to explore the microstructure.
3.2. Tie lines in the two- phase regions
3.2.1. Tie lines by 1H NMR spectroscopy
Fig. 8 shows the proton NMR spectra for the sample with com-
position of 30 wt.% 1-propanol, 35 wt.% biodiesel and 35 wt.%
water. The sample separates into two phases, the top layer consist-
ing of biodiesel and 1-propanol and the bottom layer consisting of
water and 1-propanol. The top layer spectrum (A) displays the
peaks at 5.4 ppm, 2.8 ppm and 2.1 ppm due to the protons located
at or near the double bond within the unsaturated methyl esters.
The peak at 3.6 ppm is due to the methyl ester located next to
the carbonyl carbon and the peak around 0.9 ppm is from the ter-
minal alkyl methyl of the methyl ester. The methylene alpha to the
ester group is at 2.3 ppm and the beta group is at 1.6 ppm. The
remaining –CH2 group protons have similar resonance frequencies
and overlap in the range of 1.3–1.5 ppm. The bottom layer (B)
spectrum displays the peak at 0.9 ppm due to the methyl group
from 1-propanol, as also shown in the top layer A. The peak at
1.4–1.6 ppm is due to the protons from –CH2 group while the peak
at 3.5–3.6 ppm is due to protons adjacent to –OH group. The peak
in the range of 4.7–5.0 ppm is related to protons from both water
and the –OH group in the 1-propanol.
The signal area in the NMR spectrum is exactly proportional to
the relative number of protons. Thus the mole ratios for all compo-
nents can be easily calculated using the integration of each signal

Table 2
Diffusion coefficients of the components in the limonene micro emulsions.

Micro emulsion compositions Water 1-Propanol Limonene

D ⁄ 1010 m2/s D/D° D ⁄ 1010 m2/s D/D° D ⁄ 1010 m2/s D/D°
(1) 52%P 44%W 4%L 11.5 0.50 5.4 0.69 5.5 0.36
(2) 64%P 26%W 10%L 7.3 0.32 7.7 0.99 7.7 0.51
(3) 86%P 10%W 4%L 8.8 0.38 7.3 0.93 7.4 0.49
(4) 90%P 5%W 5% L 9.9 0.43 8.0 1.03 9.3 0.62
(5) 60%P 20%W 20%L 8.2 0.35 7.0 0.90 7.2 0.48
(6) 72%P 8%W 20%L 11.6 0.50 10.4 1.33 11.2 0.75
(7) 74.5%P 25.5%L 0 0 18.6 2.39 14.7 0.98
(8) 50%P 6%W 44%L 16.6 0.72 17.0 2.19 17.7 1.18
178 I. Kayali et al. / Journal of Colloid and Interface Science 463 (2016) 173–179
Fig. 8. Proton NMR spectra for sample from the two-phase region with composition
of 30 wt.% 1-propanol, 35 wt.% biodiesel and 35 wt.% water. A: top layer and B:
bottom layer.
Fig. 9. Proton NMR spectra for sample from the two-phase region with composition
of 30 wt.% 1-propanol, 35 wt.% limonene and 35 wt.% water. A: top layer and B:
bottom layer.
Fig. 10. Calculated phase boundaries and tie lines from the regular solution model
for the system: biodiesel, 1-propanol and water.
Fig. 11. Calculated phase boundaries and tie lines from the regular solution model
for the system: limonene, 1-propanol and water.
Fig. 12. Calculated phase boundaries and tie lines from the regular solution model
for the system: diesel, 1-propanol and water.
in the spectrum. The compositions of the top and bottom layers are
determined and tie lines are drawn accordingly. As shown in Fig. 1,
tie lines connect solutions rich in biodiesel with solutions that con-
tain considerable amount of both 1-propanol and water with only
small amount of biodiesel. The direction of the tie lines indicate a
higher affinity between 1-propanol and water than that between
1-propanol and biodiesel.
Fig. 9 presents the proton NMR spectra for the sample with the
composition of 30 wt.% 1-propanol, 35 wt.% limonene and 35 wt.%
water. The sample separates into two phases, the top layer com-
posed of limonene and 1-propanol and the bottom layer composed
of water and 1-propanol. The top layer spectrum (A) displays the
known features from limonene and that from 1-propanol. The bot-
tom layer spectrum (B) shows the known characteristic peaks of
water and 1-propanol as described earlier. Compositions of the
top and bottom layers are determined, yielding the tie lines shown
in Fig. 4. The lines slope up more than the ones with biodiesel. The
slopes of the lines indicate higher affinity of 1-propanol to water
than its affinity to limonene.
3.2.2. Tie lines by theoretical model
The approach of adding a third component to increase the
mutual solubility of two immiscible liquids has been known for
more than a 100 years [35,36]. When the third component is mis-
cible with both the other liquids a one phase solution can be
obtained when the concentration of the third component is suffi-
ciently high that it effectively acts as the solvent of the two immis-
cible liquids. The situation can be approximately described by
regular solution theory [37], where miscibility is dictated by the
short range interaction parameter v. Two liquids are miscible
when v < 2 and only partly miscible when v > 2.
 I. Kayali et al. / Journal of Colloid and Interface Science 463 (2016) 173–179
The phase diagrams and tie-lines where simulated using regular
solution theory, giving quantitative estimates of the interaction
parameters. When doing so, the water–oil v parameter as first
determined from the solubilities of water in oil, and oil in water,
respectively. Then the propanol–water and propanol–oil v param-
eters were determined from the amount of propanol needed to mix
the water and the oil and combined with the slope of the tie-line
connecting the coexisting phases. For the 1-propanol (P)-water
(W)-biodiesel (B) system, the optimized v values are: vPW = 0.5,
vPB = 0.8 and vWB = 4.3. For the system 1-propanol, water and limo-
nene (L): vPW = 0.5, vPL = 1 and vWL = 4.7. The v parameters for the
third system, 1-propanol, water and diesel (D) are: vPW = 0.5,
vPD = 1.5 and vWD = 7.9. The obtained v parameters describing
the strength of interaction among the components are effective
parameters adapted for the present systems. The calculated phase
diagrams and tie lines are presented in Figs. 10–12. The slopes of
the tie lines in Figs. 10 and 11 are in good agreement with the
experimentally determined ones. Tie lines for the system
1-propanol, water and diesel are only theoretically determined.
Fig. 12 shows the determined tie lines, sloping up even higher than
the two previous ones, connecting solutions that are rich in diesel
with solutions that contain large amount of both 1-propanol and
water with only small amount of diesel.
4. Conclusions
The phase behavior of the ternary 1-propanol, water and
biodiesel, limonene or diesel, as the oil phase, is examined. These
particular systems are selected because of their suitability as
alternative fuel. The microstructure of the one-phase regions is
investigated using NMR self-diffusion technique. The results show
rapid diffusion for all components, indicating the absence of any
confinement or obstruction. The typical transition from O/W to
W/O type via a bicontinuous microemulsion is not found. Rather,
we conclude that the one phase regions of these surfactant-free
systems are to be considered as simple solutions, without any
significant water rich and oil rich domains. Structured surfactant
free microemulsions have been suggested recently in similar sys-
tems [15,17]. However, as discussed above, one cannot exclude
that the observed scattering from those systems correspond to
simple composition fluctuations as the composition appear to be
near plait points.
The phase boundaries and tie-lines of the present systems were
simulated using regular solution theory. The obtained interaction
parameters can be useful in future comparisons between different
alternative fuel systems.
Acknowledgments
Financial support for the stay of I. K. and K. Q. as visiting
professors at Lund University was generously provided by the Arab
View publication stats
179
Fund Fellowship Program in Kuwait. The authors would like to
thank Goran Carlstrom for his valuable help with NMR analysis
and Per Linse for introducing us to the POLYMER software.
References
[1] A.A. DantasNeto, M.R. Fernandes, E.L. barrosto, T.N. Castro Dantas, M.C.P.A.
Moura, Braz. J. Petrol. Gas. 7 (2013) 141.
[2] G. Knothe, Energy Fuels 22 (2008) 1358.
[3] A.J. Kinney, T.E. Clemente, Fuel Process. Technol. 86 (2005) 1137.
[4] A.C. Hansen, Q. Zhang, P.W.L. Lyne, Bioresour. Technol. 96 (2005) 277.
[5] M.L. Botero, Y. Huang, D.L. Zhu, A. Molina, C.K. Law, Fuel 94 (2012) 342.
[6] K. Purushothaman, G. Nagarajan, Fuel 88 (2009) 1732.
[7] A. Lif, K. Holmberg, Adv. Colloids Interface Sci. 123–126 (2006) 231.
[8] I. Kayali, M. Karaein, K. Qamhieh, S. Wadaah, W. Ahmad, U. Olsson, J.
Dispersion Sci. Technol. 36 (2015) 10.
[9] I. Danielsson, B. Lindman, Colloids Surf. 3 (1981) 391.
[10] A.P. Andheria, S.S. Bhagwat, J. Colloid Interface Sci. 171 (1995) 211.
[11] G.D. Smith, C.E. Donelan, R.E. Barden, J. Colloid Interface Sci. 60 (1977) 488.
[12] J. Xu, A. Yin, J. Zhao, D. Li, W. How, J. Phys. Chem. B 117 (2013) 450.
[13] M.L. Klossek, D. Touraud, T. Zemb, W. Kunz, Chem. Phys. Chem. 13 (2012)
4116.
[14] M. Klossek, D. Touraud, W. Kunz, Phys. Chem. Chem. Phys. 15 (2013) 1097.
[15] O. Diat, M.L. Klossek, D. Touraud, B. Deme, I. Grillo, W. Kunz, T. Zemb, J. Appl.
Crystallogr. 46 (2013) 1665.
[16] S. Schottle, J. Marcus, O. Diat, D. Touraud, W. Kunz, T. Zemb, D. Horinek, Chem.
Sci. 5 (2014) 2949.
[17] V. Fischer, J. Marcus, D. Touraud, O. Diat, W. Kunz, J. Colloid Interface Sci. 453
(2015) 186.
[18] B. Lindman, K. Shinoda, U. Olsson, D. Anderson, G. Karlström, H. Wennerström,
Colloids Surf. 38 (1989) 205.
[19] B. Lindman, N. Kamenka, T.M. Kathopoulis, B. Brun, P.G. Nilsson, J. Phys. Chem.
84 (1980) 2485.
[20] J.F. Bodet, H.T. Davis, L.E. Scriven, W.G. Miller, Langmuir 4 (1988) 455.
[21] P.K. Kilpatrick, H.T. Davis, L.E. Scriven, W.G. Miller, J. Colloid Interface Sci. 118
(1987) 270.
[22] K.P. Das, A. Ceglie, B. Lindman, S.E. Friberg, J. Colloid Interface Sci. 166 (1987)
390.
[23] S.B. Rananavare, A.J.I. Ward, D.W. Osborne, S.E. Friberg, J. Phys. Chem. 92
(1988) 5181.
[24] Y.L. Khmelnitsky, A.V. Hoek, C. Veeger, A.J.W.G. Visser, J. Phys. Chem. 93 (1989)
872.
[25] G.D. Smith, R.E. Barden, in: K.L. Mittal, E.J. Fendler (Eds.), Solution Behavior of
Surfactants, vol. II, Plenum, New York, 1982, p. 1225.
[26] P.G. Nilsson, H. Wennerstrom, B. Lindman, J. Phys. Chem. 87 (1983) 1377.
[27] P. Stilbs, Prog. NMR Spectros. 19 (1987) 1.
[28] D.F. Evans, H. Wennerstrom, The Colloidal domain where Physics, Chemistry
and Biology Meet, second ed., Wiley-VCH, New York, 1999.
[29] M.T. Horst, S. Urbin, R. Burton, C. MacMillan, Lipid Technol. 21 (2009) 1.
[30] S.C. Sharma, G.G. Warr, Langmuir 28 (2012) 11707.
[31] I. Kayali, A. Khan, B. Lindman, J. Colloid Interface Sci. 297 (2006) 792.
[32] N. Garti, I.A. Yuli, A. Spernath, R.E. Hoffman, Phys. Chem. Chem. Phys. 6 (2004)
2968.
[33] A. Lif, M. Nyden, K. Holmberg, J. Dispersion Sci. Technol. 30 (2009) 881.
[34] I. Kayali, M. Karaeen, W. Ahmed, K. Qamhieh, U. Olsson, J. Dispersion Sci.
Technol. http://dx.doi.org/10.1080/01932691.2015.1071267.
[35] W.D. Bancroft, J. Phys. Chem. 1 (1896) 414.
[36] W.D. Bancroft, J. Phys. Chem. 1 (1896) 647.
[37] R.S. Berry, S.A. Rice, J. Ross, Physical Chemistry, second ed., Oxford University
Press, New York, 2000.