Colloids and Surfaces A: Physicochem. Eng.

Aspects 415 (2012) 112–118

Contents lists available at SciVerse ScienceDirect

Colloids and Surfaces A: Physicochemical and

Engineering Aspects
journal homepage: www.elsevier.com/locate/colsurfa

Evaluation of nanoemulsions in the cleaning of polymeric resins

Josane A. Costa ∗ , Elizabete F. Lucas, Yure G.C. Queirós, Claudia R.E. Mansur
Federal University of Rio de Janeiro, Institute of Macromolecules – Centro de Tecnologia, bloco J – Ilha do Fundão, Rio de Janeiro, 21945970, Brazil

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

 Polymeric resins are adsorbent mate-

rials used in the treatment of oily
water.
 But they need regeneration when
they reach saturation.
 The nanoemulsions present charac-
teristics to be utilized as regenera-
ting fluids.
 We produced stable nanoemulsions
based on Solbrax by the high-energy
method.
 These were used in the regeneration
of a polymeric resin microporous.

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

Article history: The development of nanotechnology has opened the way for applications in a wide range of fields.
Received 11 June 2012 Nanoemulsions have attracted great interest because the properties of emulsions are enhanced as the
Received in revised form interfacial film approaches equilibrium when formed by droplets of nanometric sizes. In the present
18 September 2012
study, Solbrax/water/polyoxyethylene nanoemulsions were produced by high-energy emulsification in
Accepted 1 October 2012
a high-pressure homogenizer. The droplets of the stable nanoemulsions had diameter size in the range of
Available online 9 October 2012
7–30 nm. These nanoemulsions were used in cleaning of a microporous polymer resin, which can be used
to treat oily water. This resin was contaminated with different crude oils. In all cases, the nanoemulsions
Keywords:
Amphiphilic resin
presented cleaning efficiency above of 90%.
Desorption © 2012 Elsevier B.V. All rights reserved.
High-pressure homogenizer
Nanoemulsions
Nanotechnology
Oily water
Polymer resins
Polyoxyethylene
Surfactants
Solbrax
Wastewater

1. Introduction must be lower than 29 ppm, has prompted the search for new
materials with ultrafiltration and absorbent properties, such as
The growing demand for systems to treat industrial wastewater membranes and polymer resins [1,2].
before discharge so that it complies with environmental legislation, These new materials have been quite successful, but they need
according to which the total oil and grease (TOG) concentration regeneration when they reach saturation, because disposing of
them after each saturation cycle makes the process costly and envi-
ronmentally unfeasible [3–6].
∗ Corresponding author. Tel.: +55 21 22701317. An example of systems that use this type of material is the sys-
E-mail address: josaneacosta@ima.ufrj.br (J.A. Costa). tem for treatment of oily water developed by the Macromolecules

0927-7757/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.colsurfa.2012.10.011
 J.A. Costa et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 112–118 113

and Colloids Laboratory of the Institute of Macromolecules, part of Table 1

Characteristics of Solbrax Eco 175/235a .
Rio de Janeiro Federal University (LMCP/IMA-UFRJ), composed of
columns filled with polymer resins. It is capable of reducing the Properties Unit Result
TOG of oily wastewater to levels below the legal threshold. These Benzene content V% 0.01
resins retain the oil in their pores, which have an average diameter Saybolt color 30
of around 50 nm, until they become saturated. However, for this Density (20/4 ◦ C) Kg/m3 0.8075
◦
system to become workable in practice, it is necessary to develop a Final boiling point C 216.6
◦
Initial boiling point C 186.1
method of regeneration of the saturated resins so that they can be ◦
Flash point C 64.5
reused, thus increasing their useful life and the overall efficiency of Aromatics content wt% 0.1
the process [1]. Sulfur content ppm 0
The regeneration of these new materials, which are generally a
Data from the manufacturer.
amphiphilic and micro and/or macroporous, requires a regener-
ation fluid with specific characteristics, such as: affinity for the
adsorbate with sufficient diffusion energy to overcome the adsorp- solvents and surfactant solutions, were utilized to assess the effi-
tion energy, thus promoting desorption; suitably small droplet size ciency of this cleaning.
to penetrate inside micropores; and sufficient mobility to carry the
adsorbate outside the adsorbent’s pores [1,7]. 2. Experimental part
Nanoemulsions are liquid dispersions in which droplets of one
liquid with nanometric diameters are dispersed in the continuous 2.1. Materials
phase of another liquid, so that they present the classic behavior
of colloids [8,9]. The small size, high kinetic stability and differ- The nonionic surfactants used, donated by Oxiteno – Brazil, are
entiated rheological behavior of their droplets, enabling them to of the ethoxylated lauryl type (Unitol® L line), which have different
easily penetrate macropores, are promising characteristics that numbers of ethylene oxide (EO) units in their chains.
allow nanoemulsions to be utilized as regenerating fluids for these The oil phase of the O/W emulsions consisted of the solvent
new ultrafiltration and adsorbent materials [10–12]. Solbrax Eco 175/225, produced by Petrobras and donated by that
The final characteristics of nanoemulsions depend on their com- company’s distribution subsidiary BR Distribuidora. Its character-
position, temperature, pressure and procedures used to prepare istics are described in Table 1.
them [13]. The energy necessary for their formation can be sup- Both the Unitol and Solbrax were used as received.
plied by mechanical devices, known as high-energy emulsification, Solbrax Eco 175/225 is mainly composed of aliphatic and naph-
or by using the chemical potential of their components, known as thenic hydrocarbons, with a distillation range between 175 and
the low-energy method [14,15]. 225 ◦ C. This range is similar to that of mineral spirit, but the con-
The stability of nanoemulsions comes from the small size of the centration of aromatic compounds is limited to a maximum of
particles, which are only weakly affected by gravity, allowing them 0.1 wt%. The low aromatics content is obtained by catalytic hydro-
to remain in constant Brownian motion. Therefore, effects such as genation, which allows converting the aromatics into naphthenic
flocculation, sedimentation or creaming are avoided. compounds, reducing the sulfur content, saturating the olefins and
The main phenomenon that destabilizes nanoemulsions is eliminating polar impurities. This technique makes the product
called Ostwald ripening, in which smaller droplets are attracted highly stable and drastically reduces its odor [18].
to larger ones and merge with them. This effect, caused by the dif- The resin used in this study was a poly(methyl methacrylate-
ference in solubility and chemical potentials of differently sized divinylbenzene) (MMA-DVB) copolymer with a specific surface
particles, can be minimized by obtaining nanoemulsions with nar- area, A, of 70 m2 /g, pore volume, V, of 0.56 cm3 /g, apparent density,
row size distribution of the dispersed droplets [8,10]. Dap , of 0.4 g/cm3 and mean pore diameter, Ø, of 463 Å. This resin is
The use of surfactants to prepare nanoemulsions makes them kept swollen in a solution of 10% ethanol and distilled and deionized
more stable, by steric impediment and reducing the interfacial ten- water. It was characterized in a previous study [1]. To contaminate
sion of the system, thus lessening the energy required for their the resin, samples of three different crude oils were used, charac-
formation [16]. terized in Table 2. We evaluated the affinity of three crude oil kind
Among the devices used to obtain emulsions are high-pressure with oil phase (Solbrax) of nanoemulsion, mainly on the content of
homogenizers (HPH), which can be divided into two main cate- saturated and more polar fractions, such as asphaltenes and resins.
gories according to the principle by which they function: devices
with a ring valve and devices with an interaction chamber. In this
2.2. Characterization of the surfactants and Solbrax
second type, the flow being processed is released by a pneumatic
pump that can pressurize compressed air (80–100 psi) to an even
The surfactants were analyzed by hydrogen nuclear magnetic
higher pressure (10,000 psi) and thus increase its flow rate. When
resonance (1 H-NMR), using a Varian 300 spectrometer operating
the fluid and air streams enter the interaction chamber, they collide,
at 300 MHz and 30 ± 0.1 ◦ C. This allowed identifying the number or
creating high shear forces that result in nanometer-scale particles
EO units in the surfactants’ chains and establishing their solubility
[17].
in water.
In this work, we produced stable nanoemulsions by the high-
energy method, using a high-pressure homogenizer. To stabilize
them we used nonionic ethoxylated surfactants, and as the oil phase Table 2
a commercial solvent (Solbrax), extracted from petroleum naphtha, Main properties of the crude oils useda .

with a low level of aromatics and sulfur. The oil phase was carefully Oil Density Saturates Aromatics Resins Asphaltenes
chosen to obtain nanoemulsions based on a solvent considered eco- (API) (wt%) (wt%) (wt%) (wt%)
logically correct and with a solubility parameter near those of the A 29.9 58.4 26.2 14.61 0.79
types of oils targeted for removal from the adsorbent material. The B 21.2 40.7 34.1 22.9 2.4
stable nanoemulsions produced were applied to clean adsorbent C 28.2 28.0 40.0 27.0 5.2
polymer resins previously contaminated with three types of crude a
Properties previously determined in the laboratories of the Petrobras Research
oil. For comparison purposes, other systems, such as petroleum Center (CENPES).
114 J.A. Costa et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 112–118

The Solbrax was also analyzed by 13 C-NMR, at a frequency of 2.5. Impregnation of the resin with three crude oils of different
75 MHz. The sample was placed without any preparation or dis- origins
solution in a 10-mm OD (outside diameter) test tube. The NMR
analysis was performed by placing a capillary in this tube with a Thirty aliquots weighing 4 g each of the polymer resin were
suitable deuterated solvent, chosen so as to produce different sig- weighed in tubes appropriate for centrifuging. The tubes were sep-
nals than those likely to be produced by the solvent. The tests were arated into three groups and 0.2 g of each of the three oils with
run within the standard conditions established by the device’s man- characteristics shown in Table 2 was added in each group. The tubes
ufacturer, with the temperature of the probe while obtaining the containing the mixture were processed in a Boeco model C-28A
spectra being 30 ◦ C. centrifuge at a 300 cycles/min for 15 min. The entire procedure was
The oil/water and oil/aqueous surfactant solutions’ interfacial carried out multiple times until the color was homogenous, thus
tensions (in mN m−1 ) were determined using a Krüss K-10 digi- assuring maximum adsorption of the oil in the resin.
tal tensiometer, which works on the Du Noüy ring method, at a
temperature of 25 ◦ C. The entire analytical method is described in 2.6. Cleaning of the resin with different systems
previous works [19,20].
The hydrophilic–lipophilic balance (HLB) of the oil phase (Sol- To assess the performance of the nanoemulsions in extracting
brax) was determined by the direct method, by which a surfactant’s the oil adsorbed in the resin, the response curves of these three oils
behavior is visually compared against that of others with known were obtained in the solvents Solbrax and chloroform, with the
HLB values, as described in an earlier work [21]. The HLB value aid of a Cary 50 ultraviolet-visible light spectrophotometer, using
of surfactant mixtures is then calculated by the average of the HLB cuvettes with an optical path of 2 mm and wavelength of 750 nm
numbers of the pure surfactants, weighted by their volumetric frac- (which presented the best correlation coefficient).
tions in the mixture. To extract the oil impregnated in the polymer resin was used
a volume of 50 mL of each of the extraction systems selected in
the work – chloroform; pure Solbrax; Unitol L70 solutions at con-
2.3. Methods of obtaining the Solbrax/water dispersions centrations of 10 and 12 wt% of micellar solutions and the stable
nanoemulsions obtained containing 10 and 12 wt% of Unitol L70.
The Solbrax/water dispersions, in the absence of a surfactant, All the extraction systems were added to the tubes containing
were prepared in an Avestin EmulsiFlex C5 HPH. the impregnated resins and placed in a shaker bath at a temperature
In the high-pressure homogenizer the dispersion is forced to of 30 ◦ C with agitation of 150 cycles/min for 1 h. The amount of oil
pass through a very narrow channel and is then released into the extracted was determined differently for the solvents (chloroform
interaction chamber where it abruptly expands, with the sudden and Solbrax) and for the systems containing surfactants (micellar
pressure drop breaking up the dispersed droplets. Due to the lack solutions and nanoemulsions).
of homogeneity of the flow, it is generally necessary to pass the In the case of the extraction systems composed of pure solvent,
fluid through the device several times until obtaining the desired aliquots were taken from the centrifuge tubes after the extraction
particle size range [22,23]. process and placed directly in the ultraviolet-visible spectropho-
The passage time of 30-mL aliquots through the device, at tometer for analysis under the same conditions employed for the
pressures of 5000, 10,000 and 15,000 psi, was timed with a solutions that served as a base for plotting the response curve.
chronometer to establish each operating cycle. To determine the amount of crude oil extracted with the micel-
The dispersions were prepared containing 5 wt% of Solbrax in lar solutions and the nanoemulsions, after the extraction in the
distilled deionized water and submitted to high-pressure homog- shaker bath the clean resin samples were removed from the extrac-
enization for periods from one to four cycles for each of the three tion solution by filtration and once again placed in the tubes and
pressures chosen. centrifuged. A volume of 50 mL of pure Solbrax was added to each
The droplet size and size distribution of the dispersed phases of the tubes. These were then placed in the shaker bath and the
of all the emulsions prepared, as well as their stability over time, extraction process was repeated. Aliquots were taken from the cen-
were evaluated in a Zetasizer Nano Series particle size analyzer trifuge tubes after the extraction process and placed directly in the
(Malvern) [21,24]. ultraviolet-visible spectrophotometer for analysis under the same
conditions employed to obtain the response curve of oils in the
solvent Solbrax.
2.4. Production of Solbrax/water/Unitol® L nanoemulsions

3. Results and discussion

Since without the addition of a surfactant, the Solbrax/water
dispersions produced in the HPH device showed phase sep-
3.1. Characterization of the surfactant and Solbrax samples
aration shortly after preparation, we chose two nonionic
ethoxylated surfactants of the Unitol L line to produce stable
The average number of EO units in the surfactants’ chains was
Solbrax/water nanoemulsions with reduced particle sizes. We pre-
calculated by 1 H-NMR and these results confirm those reported by
pared seven of these formulations, varying the proportions of the
supplier, as shown in Table 3. The table also shows the HLB num-
components.
bers given by the supplier. Each surfactant has a hydrophilic and a
Solutions of 8, 10 and 12% the two surfactants in distilled deion-
ized water were prepared and allowed to rest for 24 h for complete
solubilization. After addition of the oil phase (varying the concen- Table 3
tration of 5–8 wt%), the formulations were then processed in the Surfactants of the Unitol L line.
HPH. These surfactants and oil phase concentration values were Namea Description HLBa
attributed according to a previous study [21].
Unitol L20 Lauryl ether with 2 EO units 6.4
The droplet size and size distribution of the dispersed phases Unitol L60 Lauryl ether with 6 EO units 11.5
of all the emulsions prepared, as well as their stability over time, Unitol L70 Lauryl ether with 7 EO units 12.3
were evaluated in a Zetasizer Nano Series particle size analyzer Unitol L100 Lauryl ether with 10 EO units 13.9
(Malvern) [21,24]. a
Data from the manufacturer.
 J.A. Costa et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 112–118 115

lipophilic element. The balance of these two elements in the sur- phase to the nanometric range. Besides this device, we used the
factant gives the HLB. High values indicate a surfactant exhibiting surfactants Unitol L60 and L70 to reduce the interfacial tension and
mainly hydrophilic or polar properties, whereas low values repre- stabilize the nanoemulsions produced.
sent lipophilic or non-polar characteristics [25]. The results obtained show that the interfacial tension of the
Evaluation of the solubility of nonionic surfactants in aque- Solbrax/water system is 29.5 ± 0.1 mN m−1 . When the surfactants
ous solution is important to identify their capacity to stabilize were added to the dispersions at the concentrations used to prepare
emulsions, since this solubility depends on the temperature and the nanoemulsions (from 8 to 12 wt%), the interfacial tension of all
is directly related to the HLB value. these systems declined to levels below 1.0 mN m−1 . The exact num-
The cloud point or temperature of aqueous solutions of non- bers were not registered because they were outside the sensitivity
ionic surfactants is observed by increasing the temperature, which range of the device.
causes gradual dehydration of the ethylene oxide molecules, in turn
causing the surfactant to separate out of the solution [26,27]. 3.3. Evaluation of the Solbrax/water dispersions
The cloud points of the Unitol L100 and Unitol L20 surfactants
were determined in an earlier work [18]. Unitol L20 is insoluble The initial aim of this study was to determine the best conditions
throughout the entire range of concentrations and temperatures for preparing the emulsions, that is, the best pressures and num-
analyzed here (1–10% (w/v) and 20–80 ◦ C), while Unitol L100 has a ber of cycles that would be able to produce systems with smaller
cloud point of around 70 ◦ C at concentrations above 4% (w/v). The droplets of the dispersed phase and narrower size distribution. The
results obtained for the aqueous solutions containing pure Unitol device (an Avestin EmulsiFlex C5 high-pressure homogenizer) was
L60 and Unitol L70 (at concentrations between 1 and 12% (w/v)) chosen because it had proved to be efficient for other systems in
showed, as expected, that an increasing degree of ethoxylation earlier studies [21,24].
tends to increase their cloud point [26,27]. The cloud point mea- At this stage of the study, all the tests were performed only with
sured for Unitol L60 was ∼35 ◦ C and that for Unitol L70 was ∼50 ◦ C, the Solbrax/water phase (95 wt% and 5 wt%, respectively), with-
for all the concentrations analyzed. out the addition of surfactants. For this reason, all the dispersions
To try to identify the approximate chemical composition of produced had low stability (∼1 h).
the Solbrax, we subjected it to carbon nuclear magnetic reso- The Solbrax/water dispersions were processed at pressures of
nance (13 C-NMR), which produced various overlapping signals in 5000, 10,000 and 15,000 psi, the minimum, medium and maximum
the same chemical shift range (between 11 and 54 ppm). This can chosen for this study obeying maximum and minimum limit of the
be explained by the fact that Solbrax is a mixture of very similar device. At each pressure the dispersions were prepared for one to
structures. In this range, chemical shifts are found of 13 C atoms of four processing cycles. The time of each cycle was determined by
hydrocarbons with linear or branched chains [28]. Chemical shifts the procedure described in Section 2.3.
of the CH2 groups in monocyclic alkanes (from C3 H6 to C10 H20 ) The dispersions obtained in the HPH were evaluated in the
could also be observed from 22 to 28 ppm. particle size analyzer. The results showed that for pressures of
The presence of other compounds, such as alkenes or het- 10,000 and 15,000 psi in 2 and 3 cycles of operations the average
eroatoms, was not observed. Despite this, they might be present in diameter of the particles was around 200 nm, without significant
very small quantities, below the sensitivity level of the equipment, variations in the three sequential analyses, indicating that the sam-
as observed for the concentration of aromatics, which is 0.1 wt%, as ples produced were more stable than those obtained under the
shown in Table 1. other conditions tested, since the droplets produced were smaller
Each type of oil used in the emulsion formulation will require and had a narrower size distribution.
an emulsifier of a particular HLB value in order to ensure a sta-
ble product. This value can be estimated experimentally [25]. For 3.4. Preparation and evaluation of the stability of the
example, the required HLB for paraffin oil for simple O/W emul- Solbrax/water/surfactant emulsions
sions is proposed by some authors to lie between 10 [29] and 12
[30]. Since the Solbrax/water dispersions prepared in the HPH were
The HLB value of the Solbrax was determined by the visual unstable, it was necessary to add the surfactants Unitol L60 and
method, by evaluating the stability of oil/water emulsions obtained Unitol L70. This addition reduced the interfacial tension of these
in the presence of Unitol L20/Unitol L100 surfactant mixtures systems to very low levels (<1 mN m−1 ). This result combined with
[21,24]. A homogeneous appearance after 24 h was observed only their HLB values indicates these surfactants are suitable to prepare
for the surfactant system with an HLB of 9.5, so this was the value Solbrax/water nanoemulsions.
attributed to the Solbrax sample utilized in this work. As before, all the droplet size measurements were performed in
Previous studies by our group [21,24] showed there is a signifi- triplicate.
cant variation in the HLB number of oils, with a small contribution
of polar substances. Orange terpene, which is essentially apolar, 3.4.1. Evaluation of the Solbrax/water/Unitol L60 nanoemulsions
presented HLB values ≈6.4. However, orange essential oil, contain- Table 4 presents the composition of the formulations used in
ing approximately 2% polar substances, presented an HLB value of this study.
8.7. This indicates that the Solbrax used in this study, with an HLB Formulation 1, consisting of Solbrax (5 wt%)/water
value of 9.5, contains traces of even more polar compounds. (87 wt%)/Unitol L60 (8 wt%), was processed in the HPH at pressures
When surfactants are used with HLB numbers near that of the of 10,000 and 15,000 psi and two and three cycles.
oil used in the emulsion, it is possible to obtain smaller droplets The distribution curves showed that Unitol L60 is highly efficient
[31]. Therefore, we used only Unitol L60 and Unitol L70 to prepare in reducing the interfacial tension of the Solbrax/water system,
the nanoemulsions. even at the lowest surfactant concentration tested in this study
(8 wt%). Solbrax/water nanoemulsions with droplet sizes ranging
3.2. Determination of the Solbrax/water interfacial tension from 8 to 30 nm were produced at all pressures and operating cycles
in the high-pressure homogenizer so it from the lower pressure for
In this study we used an emulsifying device to overcome the two and three cycles already was obtained nanoemulsion.
internal and external pressure differences of the droplets gener- However, the stability of these nanoemulsions varied with tem-
ated, to enable reducing the size of the droplets in the dispersed perature. When the nanoemulsions produced were maintained at a
116 J.A. Costa et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 112–118

Table 4
Stability of the nanoemulsions produced by high-pressure homogenization with the respective compositions and particle diameter range.

Formulationa Unitol L60 (g)/(wt%) Unitol 70 (g)/(wt%) Oil (g)/(wt%) Water (g)/(wt%) Diameter (nm) Stabilityb

1 2.4/8 – 1.5/5 26.1/87 8–30 Stable <35◦ c

2 – 2.4/8 1.5/5 26.1/87 – Unstable
3 – 3.0/10 1.5/5 25.5/85 5–30 Stablec
4 – 3.6/12 1.5/5 24.9/83 4–20 Stable
5 – 3.6/12 1.8/6 24.6/82 4–20 Stable
6 – 3.6/12 2.1/7 24.3/81 4–30 Stablec
7 – 3.6/12 2.4/8 24.0/80 – Unstable
a
The composition is shown by the mass utilized (g) and by the final mass percentage (wt%) obtained for each component in the formulation.
b
The stable emulsions are those that did not present phase separation for 108 days after preparation (maximum time the stabilities were observed).
c
The systems only were stable for 108 days when prepared by HPH at a pressure of 15,000 psi and 3 processing cycles.

controlled temperature of 24 ◦ C, they remained stable, without any 3.5. Evaluation of the resin impregnation method
significant variation in the size distribution curves. On the other
hand, when the nanoemulsions were left at rest but submitted to The resin was impregnated directly with the different crude oil
higher temperatures, reaching about 35 ◦ C (the cloud point of Unitol samples in a centrifuge, without packing of this resin in a chromato-
L60), the emulsion totally lost stability. graphic column. This form of impregnation allowed better control
of the quantity of oil adsorbed in the polymer resin, for subsequent
quantification and comparison with the other systems.
3.4.2. Evaluation of the Solbrax/water/Unitol L70 nanoemulsions The impregnation of the oil in the resin was assessed visually and
To try to find a formulation to produce stable nanoemulsions by micrographs. These observations showed that the impregna-
less susceptible to temperature variations, we replaced Unitol L60 tion was successful, since the mass of oil used was well distributed
with Unitol L70 (with a cloud point around 50 ◦ C). The efficiency throughout the resin mass, including in the pores (Fig. 2).
of Unitol L70 in stabilizing the Solbrax/water emulsions was first
evaluated by repeating the same tests performed with Unitol L60, in
this case utilizing formulations 2–7 presented in Table 4. This table 3.6. Evaluation of the cleaning capacity of the resin with different
also shows the particle size range obtained in the size distribution extraction systems
curves.
The nanoemulsions produced with the lowest surfactant con- In all the samples contaminated with oil we took care not to
centration (8 wt% – formulations 1 and 2) remained stable for only reach the maximum saturation of the resin, based on the results of
24 h, even when processed at the highest two operating pressures a previous work [1]. The maximum oil concentration in the extrac-
in the HPH. tion solution should be 4000 ppm, since 0.2 g of oil was used to
So, we decided to test another formulation, containing the impregnate the resin and the total volume of the extraction solvent
same Solbrax concentration (5 wt%), but increasing the Unitol L70 was 50 mL. Table 5 shows the adsorption intensities, the maximum
concentration to 10 wt% (formulation 3 in Table 4). This new for- concentrations extracted with the solvents Solbrax and chloroform
mulation was only processed at pressures of 10,000 and 15,000 psi and their respective efficiencies for each type of oil.
and two and three cycles. Chloroform, because it is more polar, was better at extract-
Under all these conditions nanoemulsions were produced with ing crude oil containing a greater quantity of more polar fractions
droplet sizes ranging mainly between 5 and 20 nm, with only (crude oil C). Solbrax, because it is basically composed of a mixture
a small volume of larger droplets (bigger than 200 nm), in all of saturated linear hydrocarbons, was more efficient at extracting
cases except one. The exception was the dispersions processed at crude oil sample A. However, since this solvent contains traces of
15,000 psi and three cycles, where none of these larger droplets polar compounds, it also showed excellent efficiency in the oil sam-
were observed. For this reason, the most stable nanoemulsion was ples containing more polar fractions (crude oil C). Between the two
produced under these conditions (15,000 psi and three operating
cycles). This nanoemulsion remained stable for at least 108 days,
when the last stability analysis was performed. This showed only 35
a slight enlargement of the size range, to 5–30 nm, still within the 1 day
nanoemulsion range. 30
7 days
The surfactant concentration was increased to 12 wt% (formu-
lation 4 of Table 4), with the aim of finding other stable systems, 25 21 days
produced at lower pressures, and also with higher oil phase con-
35 days
20
centrations.
The droplet sizes again ranged mainly from 5 to 20 nm, but in 64 days
Volume (%)

15
this case instead of some larger droplets, there were some smaller 108 days
than this range. Just as for formulation 3, these systems remained
10
stable for 108 days, but in this case the last test showed no enlarge-
ment of the size range (Fig. 1). 5
These results indicate that an increased surfactant concen-
tration not only serves to stabilize the droplets formed, it also 0
reduces the system’s interfacial tension enough to produce stable 1 10 100 1000
nanoemulsions at pressures below 15,000 psi (10,000 psi). Size (d.nm)
The results of increasing the oil phase (Solbrax) concentration
in formulations 5–7 showed that it is only possible to obtain stable Fig. 1. Particle size distribution of the dispersed phase of the Solbrax/water
nanoemulsions produced high-pressure homogenization at a pressure of 15,000 psi
nanoemulsions, under the conditions tested in this study, up to a and stabilized with Unitol L70, at concentration of 12 wt%, at different preparation
concentration of 7 wt%. times.
 J.A. Costa et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 112–118 117

Fig. 2. Micrograph of the polymer resin: (a) virgin; (b), (c) and (d) impregnated with crude oils A, B and C respectively.

solvents, Table 5 clearly shows that Solbrax was more efficient than 100
E% = Ce × (3)
chloroform. 4000
In the case of extraction with the nanoemulsions and micel-
lar solutions, the residual oil that remained adsorbed in the resin Crc – Corrected residual oil concentration (ppm);
and was not extracted with the cleaning solutions was deter- Cr – Concentration of residual oil shown in Table 6 (ppm);
mined through a second extraction with Solbrax and the aid of the E – % cleaning of the resin using pure Solbrax (Table 5);
response curves of the crude oil samples in this solvent. E% – Extraction efficiency (%);
Table 6 shows the residual concentrations (Cr) and extraction Ce – Extraction solution concentration (ppm).
efficiency (E%) of the micellar solutions and nanoemulsions, which
were found by means of Eqs. (1)–(3). Since the Solbrax did not It can be seen in Table 6 that the extraction efficiencies cal-
have 100% extraction efficiency, we made corrections in the Cr by culated were better for the nanoemulsions than for the micellar
employing Eq. (1) and considering the extraction efficiency (E%) of solutions. This could be confirmed by observing the centrifuge
pure Solbrax shown in Table 5. The quantities of oil extracted by the tubes that contained the nanoemulsion and micellar solutions after
micellar solutions and nanoemulsions (Ce) were obtained by the the extraction. For all three types of oil, the contaminant extracted
difference between the maximum concentration (4000 ppm) capa- with the nanoemulsions was darker in color than those extracted
ble of being obtained if the extraction efficiency had been 100% and with the micellar solutions, indicating a greater oil concentration in
the value of the corrected residual concentration (Crc) by means of the nanoemulsions (Fig. 3), or a greater content of heavier fractions,
Eq. (2). Eq. (3) allows calculating the extraction efficiency of the such as resins and asphaltenes. In some cases, the adsorption inten-
micellar solutions and nanoemulsions considering the maximum sity values obtained when inserted in the corresponding response
efficiency (100%) for a sample at a concentration of Ce 4000 ppm. curves were equal to or slightly lower than the linear coefficients
of the equations, indicating a concentration of nearly zero. In these
100 cases we attributed a maximum value to the extraction efficiency,
Crc = Cr × (1)
ESolbrax because this occurred more frequently with crude oil A, which has
greater affinity with the oil phase (Solbrax) of the nanoemulsion
Ce = 4000 − Crc (2) (Table 5). Nevertheless, the possibility also must be considered of

Table 5
Adsorption intensity, concentration and extraction efficiency of the solvents.

Extraction solutions Resin with crude oil A Resin with crude oil B Resin with crude oil C
a a
Abs (%) Conc. (ppm) E (%) Abs (%) Conc. (ppm) E (%) Abs (%) Conc. (ppm) Ea (%)

Chloroform 0.0932 2000 50 0.1094 2350 59 0.0648 3162 79

Solbrax 0.0832 3880 97 0.1487 3756 94 0.0677 3820 96
a
E – extraction efficiency, calculated by the ratio between the concentration of oil extracted and the total oil (4000 ppm).
118 J.A. Costa et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 415 (2012) 112–118

Table 6
Adsorption intensity, residual concentration and percentage contents of the oil extracted with the micellar solutions and nanoemulsions.

Extraction solutionsc Resin with crude oil A Resin with crude oil B Resin with crude oil C

Abs (%) Crb (ppm) Ea (%) Abs (%) Crb (ppm) Ea (%) Abs (%) Crb (ppm) Ea (%)

Micellar 1 0.044 –d 100 0.070 1133 69.97 0.046 714 81.43

Micellar 2 0.048 400 89.7 0.106 2333 38.17 0.044 429 88.85
Nano 1 0.040 –d 100 0.057 700 81.45 0.040 –d 100
Nano 2 0.042 –d 100 0.039 100 97.35 0.045 571 85.72
Nano 3 0.040 –d 100 0.040 133 96.47 0.040 –d 100
a
E – Extraction efficiency, calculated from Eqs. (1)–(3).
b
Cr – Concentration of residual oil.
c
Micellar 1 – Aqueous solution of the surfactant Unitol L70 at 10 wt%; Micellar 2 – Aqueous solution of the surfactant Unitol L70 at 12 wt%; Nano 1 – Nanoemulsion formed
of Solbrax/aqueous solution of Unitol L70, wit 5% oil phase and 10% surfactant; Nano 2 – Nanoemulsion formed of Solbrax/aqueous solution of Unitol L70, with 5% oil phase
and 12% surfactant; and Nano 3 – Nanoemulsion formed of Solbrax/aqueous solution of Unitol L70, with 7% oil phase and 12% surfactant.
d
Concentration not calculated because the adsorption intensity of the extraction solution was equal to or slightly lower than the linear coefficient of the corresponding
calibration curve.

All the stable nanoemulsions produced in the HPH have great

potential for application in cleaning polymer resin impregnated
with crude oil, since the yields obtained were above 90%.

Acknowledgments

We thank Oxiteno do Brasil for donating the polyoxide sam-

ples and the Petrobras Research Center (CAPES), National Scientific
and Technological Research Council (CNPq) and Rio de Janeiro State
Research Foundation (FAPERJ) for funding.
Fig. 3. Photos of the extraction solutions after their use to clean the resin impreg-
nated with crude oils A, B and C, respectively, from left to right: (a) micellar solution References
2 and (b) nanoemulsion prepared with formulation 3.
[1] Y.G.C. Queiróz, M.D. Clarisse, R.S. Oliveira, B.D. Reis, E.F. Lucas, Ciência Tecnol.
16 (2006) 224.
errors in the analyses, attributed to the loss of resin sample material [2] Conselho Nacional do Meio Ambiente – CONAMA, Res. 393 (2007).
[3] N. Lesage, Chem. Eng. Process. 47 (2008) 303.
during the filtration process, the very small resin sample quantity [4] S. Rahman, Hydrocarb. Process 85 (2006) 93.
and petroleum mass and the very low absorbance values. [5] J.-F. Li, Z.-L. Xu, H. Yang, L.-Y. YU, M. Liu, Appl. Surf. Sci. 255 (2009) 4725.
Although the cleaning efficiency of the pure Solbrax was greater [6] S.M. Liu, Z.J. Zhang, M.H. Jiang, J. Harbin Inst. Technol. 6 (2007) 837.
[7] P. Cornel, H. Sontheimer, Chem. Eng. Sci. 41 (1996) 1791.
than that of the micellar solutions and one of the nanoemulsions, [8] C. Solans, P. Izquierdo, J. Nolla, N. Azemar, M.J. Garcia-Celma, Curr. Opin. Colloid
the differences were not significant when compared to the advan- Interface Sci. 10 (2005) 102.
tages of using the nanoemulsions. [9] O. Sonneville-Aubrun, J.T. Simonnet, F. L’Alloret, Adv. Colloid Interface Sci. 108
(2004) 145.
The idea behind the proposal to use a nanoemulsion in the [10] T. Tadros, P. Izquierdo, J. Esquena, C. Solans, Adv. Colloid Interface Sci. 303
regeneration of resins utilized to treat oily wastewater was to (2004) 108.
considerably reduce the quantity of pure solvent required in the [11] C. Qian, D.J. McClements, Food Hydrocolloids 25 (2011) 1000.
[12] S.K. Parida, S. Dash, S. Patel, B.K. Mishra, Adv. Colloid Interface Sci. 121 (2006)
regeneration of these types of systems. The use of pure solvents
77.
is not possible on an industrial scale because of large volume used [13] M.R. Noor El-Din, Colloids Surf. A 390 (2011) 189.
would need subsequent recovery, involving appreciable energy use. [14] K. Ouzineb, C. Lorda, N. Lesauze, C. Graillata, P.A. Tanguy, T. McKenna, Chem.
Eng. Sci. 6 (2006) 2994.
In this study it could be observed that the volume of nanoemul-
[15] P. Fernandez, V. Andréb, J. Riegera, A. Kühnle, Colloids Surf. A 251 (2004) 53.
sion used was the same as for the solvents, but the nanoemulsion [16] W.K. Kegel, J.T.G. Overbeek, H.N.W. Lekkerkerker, Thermodynamics of
contains at most 7% solvent by total volume and over 80% water. microemulsions 1, in: P. Kumar, K.L. Mittal (Eds.), Handbook of Microemulsion
The amount of energy utilized to produce the nanoemulsions Science and Technology, Marcel Dekker, New York, 1999, p. 13.
[17] J. Floury, J. Legrand, A. Desrumaux, Chem. Eng. Sci. 59 (2004) 1285.
might well be much lower than that utilized to recover high vol- [18] M. Fortuny, A.L.D. Ramos, C. Dariva, S.M.S. Egues, A.F. Santos, Quím. Nova 31
umes of pure solvents. (2008) 1553.
[19] C.M.F. Oliveira, E.F. Lucas, Polym. Bull. 24 (1990) 363.
[20] C.R.E. Mansur, S.P. Barboza, G. González, E.F. Lucas, J. Colloid Interface Sci. 271
4. Conclusions (2004) 232.
[21] L.R. Kourniatis, L. Spinelli, C.R. Piombine, C.R.E. Mansur, Colloid J. 72 (2010) 387.
At concentrations of 5 wt% of the oil phase and 10 wt% of the [22] P. Marie, J.M. Perrier-Cornet, P. Gervais, J. Food Eng. 53 (2002) 43.
[23] S.M. Jafari, Y. He, B. Bhandari, J. Food Eng. 82 (2007) 478.
surfactant Unitol L70, stable nanoemulsions were only produced [24] L. Spinelli, C.R.E. Mansur, G. González, E.F. Lucas, Colloid J. 72 (2010) 56.
in the HPH under processing conditions of 15,000 psi of pressure [25] T. Schmidts, D. Dobler, C. Nissing, F. Runkel, J. Colloid Interface Sci. 338 (2009)
and three cycles. When the concentration of this surfactant was 184.
[26] V.K. Srinastava, G. Kini, D. Rout, J. Colloid Interface Sci. 304 (2006) 214.
increased to 12 wt%, it was possible to obtain stable nanoemul- [27] T. Inoue, O. Haruyoshi, D. Murata, J. Colloid Interface Sci. 258 (2003) 374.
sions at pressures of 10,000 and 15,000 psi and with two and three [28] R.M. Silverstein, G.C. Bassler, T.C. Morril, Identificação Espectrométrica de Com-
processing cycles, at oil phase concentrations of 5 and 6 wt%. postos Orgânicos, Guanabara Koogan, Rio de Janeiro, 1994.
[29] P. Piccerelle, E. Cauture, G. Kalantzis, J.P. Reynier, J. Joachim, Int. J. Pharm. 163
The maximum Solbrax concentration that was able to produce
(1998) 73.
stable nanoemulsions containing 12 wt% Unitol L70 was 7 wt%, but [30] L.O. Orafidiya, F.A. Oladimeji, Int. J. Pharm. 237 (2002) 241.
only at a pressure of 15,000 psi and with three cycles. [31] W. Liu, D. Sun, C. Li, Q. Liu, J. Xu, J. Colloid Interface Sci. 303 (2006) 557.