A reverse membrane emulsification process based on a hierarchically porous monolith for high efficiency water-oil separation

Article abstract of DOI:10.1039/c2ta00007e

A hierarchically porous monolith with macro- and meso-pores was synthesized via a sol-gel and phase separation process. Due to the surface modification by organic silanes, the wettability of the silica material was effectively controlled. A series of hydrophobic porous silica monoliths (HPSM) were obtained. Using a "reverse membrane emulsification" process, the HPSM not only cleared oil away from water, but also broke the micro-emulsion efficiently, even when emulsion stabilizer was in the system. As the filtration layer, 1.0 g of HPSM could treat 944 mL of oil containing water or 667 mL of surfactant-stabilized micro-emulsion. HPSM materials could remove at maximum 96.5% of the oil in water and 100% surfactant in the micro-emulsion. In addition, the material could be reused through a simple treatment. The excellent separating effect was kept even after 8 times of regeneration. Special selectivity, easy operation and excellent recyclability make the material have great potential for practical application.

Full text of DOI:10.1039/c2ta00007e

Journal of
Materials Chemistry A
View Article Online
View Journal | View Issue
PAPER
A reverse membrane emulsification process based on a
hierarchically porous monolith for high efficiency
water–oil separation†
Cite this: J. Mater. Chem. A, 2013, 1,
1701
*
Yuchao Wang, Shengyang Tao and Yonglin An
A hierarchically porous monolith with macro- and meso-pores was synthesized via a sol–gel and phase
separation process. Due to the surface modification by organic silanes, the wettability of the silica
material was effectively controlled. A series of hydrophobic porous silica monoliths (HPSM) were
obtained. Using a “reverse membrane emulsification” process, the HPSM not only cleared oil away from
water, but also broke the micro-emulsion efficiently, even when emulsion stabilizer was in the system.
As the filtration layer, 1.0 g of HPSM could treat 944 mL of oil containing water or 667 mL of
surfactant-stabilized micro-emulsion. HPSM materials could remove at maximum 96.5% of the oil in
water and 100% surfactant in the micro-emulsion. In addition, the material could be reused through a
simple treatment. The excellent separating effect was kept even after 8 times of regeneration. Special
selectivity, easy operation and excellent recyclability make the material have great potential for practical
application.
Received 21st August 2012
Accepted 20th November 2012
DOI: 10.1039/c2ta00007e
www.rsc.org/MaterialsA
necessary to fabricate a kind of material which has a strong
ability to selectively absorb the micro oil drops, catch the
Introduction
Surface wettability is an important property for solids, espe-
cially for porous materials.1–5 It can make materials themselves
become hydrophobic or hydrophilic.6–8 This characteristic can
be used to develop porous adsorbents to selectively uptake
different polar or nonpolar liquids, such as water, ethanol,
benzene and gasoline. Among these adsorbents, the oil-
absorption materials with nonpolar surfaces are generally
considered meaningful and valuable.9–12 For a long time, due to
the exploitation, transportation and application of petroleum,
oil spills have occurred worldwide, like the events in the Gulf of
Mexico.13–15 The leaking oil causes serious environmental
pollution and terrible ecological disasters, especially to the
living beings in the water.16 High performance oil-adsorption
materials are needed urgently.17–20 Meanwhile, a small amount
of water in oil also needs to be separated, especially in the crude
oil exploration and transmission. Surfactants are used for this
purpose, however, water forms various emulsions with oil and
surfactants, creating even further challenges. The droplets in
these emulsions are much more stable and are difficult to
separate. Chemicals are commonly added to the oil–water
mixture for demulsication, but they easily cause further
pollution and are difficult to clear.21–23 For this reason, it is
amphiphilic stabilizer (surfactant) and break the emulsion.
Since porous membranes are widely used to produce emul-
sions, it is possible that the “reverse membrane emulsication”
process could turn the emulsion back to the separated water
and oil phase. To realize this function, rst, there should be
plenty of space inside the material to ensure the success of uid
ow. Additionally, the materials should be hydrophobic and
oleophilic to produce high selectivity of oils against water.
Finally, the surface of the material should have strong interac-
tions with the oil molecules or surfactants in order to break the
emulsion.
Hierarchically porous silica is a new kind of porous material,
which has gained great attention from chemists worldwide.24–28
Due to the interconnected pore structure, the hierarchically
porous materials have been used in catalysis,29–31 absorp-
tion,32–35 separation,36–38 energy storage39–41 and sensing.42–45 It
was reported that the porous silica monolith could be used as a
high performance column packing material in HPLC.46 This
indicates that the pore structure and surface groups may have a
great effect to improve the separation property of the mono-
lithic materials.
Herein, we report a novel kind of hydrophobic porous silica
monolith (HPSM) with controlled surface wettability and hier-
archical microstructure for capturing tiny oil droplets and
breaking emulsions via a “reverse membrane emulsication”
method. This material shows several advantages: rst, the
monolithic shape allows the silicate bulk to be used as a
ltering material without further modication. Second, the
Department of Chemistry, Dalian University of Technology, Dalian, Liaoning, P. R.
China. E-mail: taosy@dlut.edu.cn; Fax: +86-411-84986035; Tel: +86-411-84986035
† Electronic supplementary information (ESI) available: SEM images of HPSM,
TEM images, FT-IR spectrum, TGA of HPSM and the images of the contact
angles with Span 80 of the samples. See DOI: 10.1039/c2ta00007e
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 1701–1708 | 1701

View Article Online
Journal of Materials Chemistry A
Paper
hydrophobic surface, covered by long-chain alkanes, could help
to trap tiny oil droplet because of its high affinity to nonpolar
organics. Third, the stability of porous silica allows the material
to be applied in various liquid phases. Finally, facile regenera-
tion means that the materials have a long service life and can be
economically utilized.
Preparation of the emulsion
The simple emulsion was prepared by dispersing 0.50 g of
mercantile soybean oil into 1000 mL of distilled water homo-
geneously. The oil content in the emulsion was 0.5 g Lꢀ1
.
The micro-emulsion with surfactant as the stabilizer was
prepared as follows: distilled water and the solution of n-decane
with Span-80 (sorbitan monooleate) at a concentration of 0.1 wt
% were mixed together at a volumetric ratio of about 1 : 9. The
mixed solution was converted to homogeneous emulsion by a
high-speed shear emulsier at a shear rate of about 5000 rpm
for 20 min.
Experimental
Materials
Tetramethoxysilane (TMOS, 99.0%) was purchased from the
Chemical Factory of Wuhan University (Wuhan, China). Poly-
(ethyleneoxide)-block-poly-(propyleneoxide)-block-poly-(ethylene-
oxide) (P123) was obtained from Aldrich Chemical Co. Inc.
Nitric acid (AR) was purchased from Beijing Chemical
Reagents Factory (Beijing, China). Active carbon (AR), 25 wt%
ammonia (AR), n-decane (AR), Span-80 (AR), toluene (AR) and
petroleum ether (AR) were purchased from Fuyu Fine Chem-
ical of Tianjin Co. Ltd (Tianjin, China). N-Octyltrimethoxy-
silane (NOEO, 96%) was obtained from Dalian Onichem
Specialities Co. Ltd. N-Octadecyltrimethoxysilane (95%),
methyltrichlorosilane (99%), phenyltriethoxysilane (98%) and
Testing the effect of adsorption and separation
A block of modied silica monolith xed on a funnel was used
for the test by vacuum ltration of the emulsion. The ratio of
demulsication could be obtained by measuring the volumes of
the claried water phase and oil phase. The amounts of residual
oil and surfactant were tested by UV (ultraviolet) spectrometer.
Testing the density of the as-prepared materials
A block of the as-prepared material with regular shape was
weighed 3 times to get the mass (m) and the volume (v) was
measured by a ruler. The density of the materials (r) could be
calculated by the following formula.
(tridecauoro-1,1,2,2-tetrahydrooctyl)-triethoxysilane
(95%)
were brought from Meryer Chemical Co. Ltd (Shenzhen,
China).
AR stands for analytical reagent, whose purity was above
99.7%. All the reagents mentioned above were used without
further purication.
m
r ¼
v
Preparation of the porous silica monolith (PSM)
Characterization
PSM was prepared according to the previously reported
method.30 First, P123 (poly-(ethyleneoxide)-block-poly-(propyle-
neoxide)-block-poly-(ethyleneoxide), 4.0 g) was homogeneously
dissolved in an aqueous solution of nitric acid (12.0 mL of 1.0
mol dmꢀ3). Subsequently TMOS (tetramethoxysilane, 5.15 g)
was added and violently stirred for about 15 min for hydrolysis
The scanning electron microscopy (SEM) images were taken
with a JEOL JSM-6700F eld emission scanning electron
microscope (FESEM, 20 kV). The transmission electron
microscopy (TEM) images were taken with FEI Tecnai G2 Spirit
(120 kV). Mercury porosimeter (PORESIZER-9320, Micro-
meritics Co., USA) was used to measure the size of macro-
pores. The nitrogen adsorption and desorption isotherms were
measured at 77 K using an 3H-2000PS1/2 static volume
method (Beishide Instrument-ST Co. Ltd). Samples needed to
be degassed under vacuum at 373 K for 4 h before measure-
ment. Surface areas were calculated by the Brunauer–Emmett–
Teller (BET) method and the pore volume and pore size
distributions were calculated using the Barrett–Joyner–
Halenda (BJH) model. FT-IR spectra (4000–500 cmꢀ1) in KBr
were collected on a Nicolet Avatar 360 FT-IR spectrometer. The
functional materials were mixed with KBr and pressed to a
thin disc for FT-IR detection. Thermal gravimetric analysis was
carried out using TGA/SDTA851e coming from Mettler Co. Ltd,
ꢁ
at 0 C. 15 min later, the semitransparent solution was trans-
ferred into PE tubes. The PE (poly ethylene) tubes were sealed
ꢁ
and kept at 40 C for gelation and then aged for about 5 times
the gelation time. The resultant wet gels were immersed into the
1.0 mol dmꢀ3 aqueous solution of ammonia at 40 ꢁC for solvent
exchange for about 4 days. The wet silica gels were carefully
ꢁ
evaporation-dried at 40 C, followed by calcination from room
temperature to 650 ꢁC with a heating rate of 1 ꢁC minꢀ1 and
holding at 650 ꢁC for 5 h in air in order to remove organic
substances.
Surface modication
HPSMs were prepared due to the modication with silane Switzerland. The samples were heated progressively from 25 to
coupling agents (including methyltrichlorosilane, n-octyl- 800 C at a heating rate of 10 C minꢀ1 in air. UV spectra were
trimethoxysilane, n-octadecyltrimethoxysilane, phenyl- obtained on a Helios Alpha UV spectrophotometer manufac-
ꢁ
ꢁ
triethoxysilane and (tridecauoro-1, 1, 2, 2-tetrahydrooctyl)- tured by Unicam Limited (Britain). A SL200B (Solon Tech. Inc.
triethoxysilane). PSM (1.0 g) was allowed to react with an Ltd, Shanghai) contact-angle goniometer was used for the
appropriate amount of reagent (1.6 mmol) in toluene (50 mL) at static contact-angle measurements. All the drops in the
reux for 12 h. The as-prepared sample was repeatedly washed images were 5 mL. The soware offered by the instrument
ꢁ
with petroleum ether and dried at 100 C.
manufacturer calculates the contact angles based on the
1702 | J. Mater. Chem. A, 2013, 1, 1701–1708
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
method of the tangent. The data of Differential Scanning
Calorimetry (DSC) was obtained from TA Q20ꢀ1DSC (TA
Instruments, USA) at a heating rate of 5 ꢁC min and the
temperature range was from room temperature to 80 ꢁC in a
puried nitrogen atmosphere.
All the measurements were performed at room temperature
with droplets of deionized water.
Results and discussion
Fabrication and characterization of HPSM
The hierarchically porous silica monolith (PSM in Fig. 1a) was
prepared through a phase separation method at the process of
sol–gel transition. The shape of the monolith depended on the
container employed during the gelation of the silica sol. Plastic
tubes were used to form PSM with uniform shape and the same
diameter. A white silica cylinder was formed aer gelation and
it was easy to cut into a certain length and use for emulsion
ltration. Aer calcination in air, the inner surface of the PSM
could be fully exposed with the absence of structure directing
agent. It was convenient to be modied by different organic
silanes to obtain hydrophobic material (abbreviated as HPSM).
The density of the HPSM was just 0.35 g cmꢀ3. And due to the
hydrophobic surface formed by some alkyl groups, the mono-
lith could ow on the water (as Fig. 1b shown). The synthesis
procedure was simple, the silica sol precursor could be scaled
up to 10 L in one pot and more than 1.2 kg of HPSM could be
obtained at once, indicating the possibility of industrial
production. The SEM images (Fig. 1c) showed that the PSM had
a porous structure with interconnected porous channels. The
pore diameter was nearly 3 mm. Aer modication by n-octyl
groups, the porous structure of HPSM-8 was kept as Fig. 1d
shows. But the pore diameter (around 2.5 mm) became smaller
than for the unmodied ones. HPSMs modied with other
silanes (see Fig. S1, ESI†) showed similar results. The results
were in good accordance with the experimental data by the
mercury intrusion method. The shrinking of the silica skeleton
during the modication process might cause the decrease of the
macropore diameter. The large macropore size would be a great
benet to the transfer of uid in the porous monolith and the
rough surface formed by the macroporous structure is advan-
tageous for high hydrophobicity.47,48
Fig. 1 (a) Optical image of the PSM. (b) The HPSM-8 in contact with water. (c)
and (d) SEM images of PSM and HPSM-8, respectively. (e) N2 adsorption–
desorption isotherms.
was necessary for the breaking of emulsion. The H1-type
hysteresis loops in the adsorption–desorption isotherms
revealed that there were uniform mesopores with narrow pore
size distribution.50,51 A well-dened step happened at approxi-
mately P/P0 ¼ 0.85–0.96, indicating the lling of the mesopores
via capillary condensation. As shown in Table 1, the diameter of
the mesopores in unmodied material was 17.01 nm. When n-
octyl groups were anchored on the surface, the mesopore size
decreased to 13.88 nm. The decline of pore diameter was nearly
two times the length of the n-octyl siloxane group (about
1.45 nm), which indicated that there might be a single layer of
alkyl groups on the surface. The mesopore volume was 1.64 cm3
gꢀ1 for HPSM-8, smaller than PSM (1.95 cm3 gꢀ1), but was larger
than many other reported hybrid mesoporous silica mate-
rials.52–56 The larger pore volume was useful for adsorbing and
storing oils and surfactant molecules. The slight shrinking of
the mesopores did not affect the adsorption ability of the
materials since the mesopores were still large enough for the
diffusion of organic molecules.
The mesopores inside the silica skeleton were characterized
via TEM (Fig. S2†) and N2 adsorption–desorption (Fig. 1e). The
large mesopores could be observed obviously. Just as Fig. S2b†
shows, the mesoporous structure had been kept during modi-
cation. As shown in Fig. 1e, the N2 adsorption–desorption
The surface properties of HPSM
isotherms of the PSM and HPSMs all belong to the type-IV According to the previous study,57–61 the wettability of materials
isotherm according to IUPAC classication.49 It illuminated that depends on the surface energy and microstructure. The surface
the mesopores in the hierarchically porous system was not of a material with low energy and rough structure could usually
destroyed. As shown in Table 1, the surface area of PSM reached get ideal hydrophobicity. According to SEM images, the porous
377 m2 gꢀ1 and could still keep as 318 m2 gꢀ1 aer n-octyl structure of HPSM obviously provided a rough surface. The
modication. The decrease of surface area depended on the surface energy is directly bound up with the kinds of surface
molecular size of silane modiers. For example, the surface area groups. The surface groups of HPSM were identied by the
of HPSM-18 was 289 m2 gꢀ1, which was smaller than that of Fourier-transform infrared (FT-IR) spectrum (see Fig. S3, ESI†).
HPSM-1 (349 m2 gꢀ1) and HPSM-8. The large area could The bands at 1082, 2980 and 3480 cmꢀ1 were assigned to the
enhance the contact between the surface and emulsion, which stretching vibrations of Si–O–Si, –CH3 and –OH, respectively.62
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 1701–1708 | 1703

View Article Online
Journal of Materials Chemistry A
Paper
Table 1 Structural properties of the PSM and modified materials
Dosage of
Macropore
size
(nm)
SBET
,
Vp,
Dp,
modier
specic surface area
pore volume
pore diameter
(nm)
Sample
Modier
(mol gꢀ1
)
(m2 gꢀ1
)
(cm3 gꢀ1
)
PSM
3.01
2.51
2.48
2.52
2.47
2.50
377
349
318
289
265
227
1.95
1.79
1.64
1.58
1.77
1.45
17.01
14.00
13.88
15.85
16.10
13.75
HPSM-1
HPSM-8
HPSM-18
HPSM-B
HPSM-F
Methyltrichlorosilane
n-Octyltriethoxysilane
n-Octadecyltrimethoxysilane
Phenyltriethoxysilane
(Tridecauoro-1,1,2,2-
1.59 ꢂ 10ꢀ3
1.59 ꢂ 10ꢀ3
1.59 ꢂ 10ꢀ3
1.59 ꢂ 10ꢀ3
1.59 ꢂ 10ꢀ3
tetrahydrooctyl)-triethoxysilane
Fig. 3 The images of the emulsion before (a) and after (b) filtration by HPSM-8
and the diameter distribution of droplets in emulsion (c). A conjecture of the
adsorption process of oil droplets by HPSM-8 (d).
Due to the modications the HPSMs were expected to have
ideal selectivity to oil and surfactants, the contact angles were
tested to examine the surface wettability of the materials. In
Fig. 2a, the water contact angle of the PSM was less than 4ꢁ,
meaning that it was superhydrophilic. The abundant –OH
groups on the surface and the porous microstructure was
Fig. 2 The images of the water contact angles (a–f), the oil (soybean oil) contact believed to lead to this phenomenon. Aer modication, HPSM-
8 became hydrophobic with the contact angle of 144ꢁ (as Fig. 2c
shown). The situations were similar for HPSM-1, HPSM-18,
HPSM-B and HPSM-F (Fig. 2b and d–f). Their surfaces were all
hydrophobic with water contact angles between 102ꢁand 150ꢁ.
Due to the reduction of the surface energy, long-chain alkyl or
angles (g–l) and the contact-angle data (m). The materials include PSM (a and g),
HPSM-1 (b and h), HPSM-8 (c and i), HPSM-18 (d and j), HPSM-B (e and k) and
HPSM-F (f and l).
When octadecyl was covered on the silica surface, the vibration uoro-alkyl groups made the surface have better hydropho-
strength of –CH3 was distinctly enhanced. As shown in Fig. S4 bicity. In another aspect, according to the oil contact angle (in
(see ESI†), thermogravimetric analysis (TGA) was utilized to Fig. 2g–l), all the HPSMs showed oleophilic properties (the oil
explore the coverage degree of alkyl groups on the modied contact angles were nearly 0ꢁ), except the HPSM-F. The surface
surfaces. The graed density of alkyls was about 0.8–1.1 per of HPSM-F showed an amphiphobicity, which is usually found
nm2, which was calculated through the mass loss. This result in uoro-alkyl modied materials. This property meant the
provided further evidence to prove the formation of the single HPSM-F had a weak interaction with both water and oil. The
alkyl layer on the pore surface. The properly graed density, results of the contact angles with the surfactant (Span-80) could
which was not too low or too high, made the HPSM have both be obtained from Fig. S5 (see ESI†). They were similar to the oil
low energy surface and large pore volume.
contact angles.
1704 | J. Mater. Chem. A, 2013, 1, 1701–1708
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
Fig. 6 (a) Image of the filtrated solution which were partly, A, and completely, B,
demulsified. (b) The diameter distribution of droplets in micro-emulsion and (c)
the mechanism of the demulsification by HPSM-18.
Fig. 4 The oil removal ability of HPSM-8 for 0.5 wt% (soybean oil) emulsion (a),
1.1 cm was sealed in the epoxy resin plaster. The upper and
lower ends of the column were exposed to air and connected
with plastic tube to let uid pass through. The water–oil
emulsion was ltrated through the column under vacuum.
Comparing with PSM, the HPSM-8 showed an outstanding
separation capacity. Aer ltration, the tiny oil droplets nearly
disappeared in the ltrate solution, as shown in Fig. 3a and b.
500 mL emulsion (containing 0.5% oil in weight) could be
treated by 0.53 g of HPSM-8. 96.5% of oil in emulsion was
removed. According to the dynamic light scattering results, the
average diameter of the oil droplet was 6.7 nm (Fig. 3c), much
smaller than the macropore diameter of the HPSMs. When the
oil droplets owed through the macropore with water, they
could easily reach the inner pore surface and be adsorbed into
the mesopores (Fig. 3d).
the regeneration capacity (b) and the comparison with some other materials (c).
Removal of oil droplets from emulsion by HPSM
The separation properties of HPSM modied by different
groups were explored by using them as ltration materials. A
block of HPSM column with a length of 0.53 cm and diameter of
The hydrophobic surfaces and the van der Waals forces
between surface and oil molecules were thought as the main
reason. Meanwhile, the large pore volume provided space to
store the adsorbed oil droplets. For other hydrophobic mate-
rials, HPSM, HPSM-1 and HPSM-B showed a poor selectivity for
water and oil. A great amount of water was also adsorbed. The
emulsion was still very cloudy aer ltration. This might be
caused by the poor hydrophobicity and weak interactions
between the surface and oil. The carbon chain on the surface
was too short to produce enough interactions. Although HPSM-
F showed a large contact angle, its surface was both hydro-
phobic and oleophobic. Also, it couldn't produce a selective
adsorption. Compared with the HPSM-8, the HPSM-18 exhibi-
ted similar selective adsorptivity to the tiny oil droplets in the
emulsion, but its adsorption capacity was smaller than that of
HPSM-8. This might be due to its smallest pore volume among
all the hydrophobic HPSMs, which was caused by the modi-
cation of long octadecyl groups. The adsorption capacity of
HPSM-8 was also better than many popular adsorbents. SBA-15
and commercial silica gels were modied with n-octyl groups,
Fig. 5 The demulsification and surfactant (Span-80) removal abilities of HPSM-
18 (a), the regeneration capacity (b) and the comparison with some other
materials (c).
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 1701–1708 | 1705

View Article Online
Journal of Materials Chemistry A
Paper
Table 2 The demulsification efficiency of some state-of-the-art materials
Demulsication
efficiency
Material
Emulsion type
Emulsier
Method
Reference
HPSM-18
W/O, 10% (v/v)
O/W, 1.5 wt%
W/O, 40 wt%
O/W, 1350 ppm
W/O, 5 wt%
Span-80, 0.05%
99.95%
76%
96.78%
99.5%
90%
80%
73%
95%
98–99%
Filtration
Rotating disk ltration
Shake
Filtration
Gravity settling
Filtration
Filtration
Filtration
Filtration
Polyvinylidene diuoride
Polyether modied by nano-SiO2
Polysaccharide Nanobrous
Ethyl cellulose polymers
Modied resin
Fibrous lter
ZnO nanorod array-coated mesh lms
Polymer membrane
63
64
65
66
67
68
69
70
O/W, 1 g Lꢀ1
O/W, 0.2 wt%
O/W, 30% (v/v)
W/O, 1 wt%
SDBSa
SDBSa, 5 ꢂ 10ꢀ5 wt%
D2EHPAb, 5 wt%
a
SDBS meaning sodium dodecyl benzene sulfonate. b D2EHPA meaning di-2-ethylhexyl phosphoric acid.
emulsier) was ltrated with the HPSM-18, the white emulsion
turned transparent immediately. The aqueous phase and oil
phase clearly separated. According to the light scattering
results, the ratio of demulsication reached 99.95%. The
removal of emulsier from the emulsion was considered the
main reason for the demulsication. The content of Span-80
was small in the initial ltrate and changed with the increase
of the volume of lter liquor (as Fig. 5a shown). The ltration
material adsorbed most of the surfactants. Lacking enough
emulsier, the emulsion became unstable and phase separa-
tion happened. When the demulsication capacity of HPSM-18
reached saturation, the residual rate of emulsier was close to
100%. It meant that the HPSM-18 could not capture more
emulsier from the emulsion. Fig. 6a showed the ltrated
emulsion. When demulsication was thorough, there was a
distinct phase interface between water and oil and the solution
was transparent. When the solution was halfway demulsied,
the ltrated solution was cloudy and the interface was not
clear.
The HPSM-18 could also be totally recyclable. Aer being
washed by petroleum ether, the emulsier-capturing ability of
the HPSM-18 was recovered and the demulsication ability
didn't show an obvious decrease aer 8 cycles (Fig. 5b).
Compared with the HPSM-18, other HPSMs showed a lower
demulsication capacity and fewer emulsiers were captured by
them (Fig. 5c). It was surprising that the HPSM-8 didn't show a
satised performance on non-surfactant emulsion removal. The
material modied by octadecyl groups displayed a large
adsorptive capacity, which was quite different from HPSMs
modied by other groups. This might be due to the interactions
between the octadecyl group on the silica surface and the alkyl
tail in Span-80, which was an octadecanoic acid ester. The
similar length of long carbon chains could produce a strong van
der Waals force and hydrophobic interaction. In addition, active
carbon could do well in demulsication, but couldn't removal
the emulsier (Span-80) efficiently. Comparing with many other
kinds of state-of-the-art materials, HPSMs also had outstanding
demulsication efficiency (Table 2). Different from the
membranes, HPSMs could adsorb the emulsier to avoid that
the emulsion formed again.
Fig. 7 Differential scanning calorimetry (DSC) of PSM and HPSMs composited
with Span-80 from 25 ꢁC to 80 ꢁC.
but they exhibited a very low oil removal ability. As the general
adsorbent widely used, active carbon didn't show a better oil
removal ability than HPSM-8. The hierarchical structure and
larger pore volume should greatly improve the oil-capturing
capacity of HPSMs. The adsorption property of HPSM-8 could
be totally recovered by organic solvent washing. Aer recycling 6
times the ratio of oil removal still reached nearly 100% (Fig. 4).
Demulsication of emulsions stabilized by surfactants
In addition to the simple water–oil emulsion, there are many
complex emulsion systems which are stabilized by surfactants.
Since lots of non-ionic surfactants are used in the petroleum
industry, water is easily induced into oil during the exploitation
and transmission procedure. The residual water is an important
parameter to the quality of petroleum. The emulsion droplets
are usually smaller than those in a simple water–oil mixture
system and it is difficult to directly separate water from oil by
physical ltration. Some demulsiers can break these emul-
sions, but they bring secondary pollution at the same time. It is
interesting that demulsication can be perfectly realized as the
emulsion passes through our HPSMs.
Since the size of the emulsion droplets was small (35–45 nm,
shown in Fig. 6b), the supposed mechanism is shown in
The HPSM-18 exhibited the best demulsication ability.
Aer the emulsion stabilized by Span-80 (a commonly used
1706 | J. Mater. Chem. A, 2013, 1, 1701–1708
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
Fig. 6c. Firstly, the emulsier-stabilized droplet was captured
by the surface of HPSM-18. Then the micelle broke under the
impact of ow. The surfactants remained on the silica surface
and the droplet was released. Finally, the unprotected small
droplets got together to form another continuous phase. This
was just like a “reverse membrane emulsication” process. In
membrane emulsication, small emulsion droplets were
produced when liquid passed through the porous substrate.
Pores in the membrane were used to help to form the emul-
sion. But for HPSM, when the surface properties were changed,
droplets were forced to combine together in the porous struc-
ture and the emulsion was broken. So “reverse membrane
emulsication” was employed to describe this phenomenon.
The DSC results further support our conjecture. The same
amount of emulsier was adsorbed on the surface of different
HPSMs, whose masses were also the same. The adsorptive layer
of the emulsier could produce an interaction with the alkyl
groups on the silica surface and the intensity of this interaction
related to the thermal properties of the materials. As shown in
Fig. 7, endothermic peaks appeared aer the emulsier was
adsorbed. For PSM, the peak was sharp and narrow. The
enthalpy change was small, just 0.5 J gꢀ1. When alkyl groups
were graed on the surface, the endothermic peaks became
broad, and the enthalpy change greatly increased. The enthalpy
change of HPSM-18 was 3.6 J gꢀ1, and larger than that of other
HPSMs, indicating that HPSM-18 had a wide and strong
Acknowledgements
We are genuinely grateful to the National Natural Science
Foundation of China (20903018 and 51273030) and the
Fundamental Research Funds for the Central Universities
(DUT11LK23) for support of this research.
Notes and references
1 R. N. Wenzel, Ind. Eng. Chem., 1936, 28, 988.
2 A. B. D. Cassie and S. Baxter, Trans. Faraday Soc., 1944, 40,
546.
3 R. Blossey, Nat. Mater., 2003, 2, 301.
4 I. P. Parkin and R. G. Palgrave, J. Mater. Chem., 2005, 15,
1689.
5 T. L. Sun, L. Feng, X. F. Gao and L. Jiang, Acc. Chem. Res.,
2005, 38, 644.
6 H. Y. Erbil, A. L. Demirel, Y. Avci and O. Mert, Science, 2003,
299, 1377.
7 L. Feng, S. H. Li, Y. S. Li, H. J. Li, L. J. Zhang, J. Zhai,
Y. L. Song, B. Q. Liu, L. Jiang and D. B. Zhu, Adv. Mater.,
2002, 14, 1857.
8 R. Wang, K. Hashimoto, A. Fujishima, M. Chikuni,
E. Kojima, A. Kitamura, M. Shimohigoshi and
T. Watanabe, Nature, 1997, 388, 431.
9 C. H. Su, Appl. Surf. Sci., 2010, 256, 2122.
interaction with Span-80, which was benecial for the demul- 10 J. Li, L. Shi, Y. Chen, Y. Zhang, Z. Guo, B.-L. Su and W. Liu, J.
sication process.
Mater. Chem., 2012, 22, 9774.
11 A. Marmur, Langmuir, 2004, 20, 3517.
12 S. Y. Tao, Y. C. Wang and Y. L. An, J. Mater. Chem., 2011, 21,
11901.
Conclusions
In summary, we have demonstrated a fast and efficient 13 Q. Wei, R. Mather, A. Fotheringham and R. Yang, Mar.
demulsication method with hydrophobic porous silica Pollut. Bull., 2003, 46, 780.
monolith (HPSM), based on the combination of excellent 14 D. Ceylan, S. Dogu, B. Karacik, S. D. Yakan, O. S. Okay and
surface performance (high selective adsorption for oil and non- O. Okay, Environ. Sci. Technol., 2009, 43, 3846.
ion surfactant) and unique structure properties of hierarchical 15 M. O. Adebajo, R. L. Frost, J. T. Kloprogge, O. Carmody and
pores (ow passing through and storage of oil). Oil droplets in S. Kokot, J. Porous Mater., 2003, 10, 159.
a simple emulsion system (without emulsier) were easily 16 S. E. Allan, B. W. Smith and K. A. Anderson, Environ. Sci.
removed via ltration. The ratio of the oils reached nearly Technol., 2012, 46, 2033.
100% and the materials could be recycled several times. More 17 J. Yuan, X. Liu, O. Akbulut, J. Hu, S. L. Suib, J. Kong and
importantly, using “reverse membrane emulsication” F. Stellacci, Nat. Nanotechnol., 2008, 3, 332.
process, the emulsion stabilized by surfactant could be 18 X. Gui, J. Wei, K. Wang, A. Cao, H. Zhu, Y. Jia, Q. Shu and
completely broken. The long-chain hydrocarbon modied D. Wu, Adv. Mater., 2010, 22, 617.
HPSM-18 exhibited the best demulsication ability via 19 X. Gui, Z. Zeng, A. Cao, Z. Lin, H. Zeng, R. Xiang, T. Wu,
adsorbing the emulsier from the emulsion. The demulsi- Y. Zhu and Z. Tang, J. Mater. Chem., 2012, 22, 18300.
cation ratio reached 99.95% and the materials also could be 20 H. W. Liang, Q. F. Guan, L. F. Chen, Z. Zhu, W. J. Zhang and
reused. The surface interactions between the modier and S. H. Yu, Angew. Chem., Int. Ed., 2012, 51, 5101.
a
emulsiers was the primary driving force to remove these 21 R. C. Little, Environ. Sci. Technol., 1981, 15, 1184.
stabilizers from the emulsion and realize demulsication. 22 M. A. Krawczyk, D. T. Wasan and C. Shetty, Ind. Eng. Chem.
Because of the facile preparation and high performance of the
modied HPSMs in the treatment of emulsion, it could be 23 H. N. Abdurahman, R. M. Yunus and Z. Jemaat, J. Appl. Sci.,
integrated into ltration systems as a new component to 2007, 7, 196.
separate the water–oil mixed system. In addition, the devel- 24 Y. Shin, J. Liu, L. Wang, Z. Nie, W. Samuels, G. Fryxell and
opment of a monolith with different pore structure, control- G. Exarhos, Angew. Chem., Int. Ed., 2000, 39, 2702.
Res., 1991, 30, 367.
lable surface properties and which shows considerable 25 M. Hartmann, Angew. Chem., Int. Ed., 2004, 43, 5880.
adsorptivity is benecial for the design of efficient separation 26 D. Kuang, T. Brezesinski and B. Smarsly, J. Am. Chem. Soc.,
materials.
2004, 126, 10534.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 1701–1708 | 1707

View Article Online
Journal of Materials Chemistry A
Paper
˚
´
27 J. Smatt, S. Schunk and M. Linden, Chem. Mater., 2003, 15, 50 D. Zhao, J. Feng, Q. Huo, N. Melosh, G. H. Fredrickson,
2354. B. F. Chmelka and G. D. Stucky, Science, 1998, 279, 548.
28 S. Hartmann, D. Brandhuber and N. Husing, Acc. Chem. Res., 51 N. Nakamura, R. Takahashi, S. Sato, T. Sodesawa and
2007, 40, 885. S. Yoshida, Phys. Chem. Chem. Phys., 2000, 2, 4983.
29 A. Corma, U. Dıaz, T. Garcıa, G. Sastre and A. Velty, J. Am. 52 C. B. Gao and S. A. Che, Adv. Funct. Mater., 2010, 20, 2750.
¨
´
´
Chem. Soc., 2010, 132, 15011.
30 T. Amatani, K. Nakanishi, K. Hirao and T. Kodaira, Chem.
Mater., 2005, 17, 2114.
53 M. C. Burleigh, M. A. Markowitz, E. M. Wong, J. S. Lin and
B. P. Gaber, Chem. Mater., 2001, 13, 4411.
54 W. Wang, W. Zhou and A. Sayari, Chem. Mater., 2003, 15,
4886.
31 J. Ge, T. Huynh, Y. Hu and Y. Yin, Nano Lett., 2008, 8, 931.
32 Q. Chen, F. Fan, D. Long, X. Liu, X. Liang, W. Qiao and 55 B. J. Melde, B. T. Holland, C. F. Blanford and A. Stein, Chem.
L. Ling, Ind. Eng. Chem. Res., 2010, 49, 11408. Mater., 1999, 11, 3302.
33 T. Ma, X. Zhang and Z. Yuan, J. Phys. Chem. C, 2009, 113, 56 N. Bion, P. Ferreira, A. Valente, I. S. Goncalves and J. Rocha,
12854. J. Mater. Chem., 2003, 13, 1910.
34 W. Cai, J. Yu and M. Jaroniec, J. Mater. Chem., 2010, 20, 4587. 57 D. Y. Ryu, K. Shin, E. Drockenmuller, C. Hawker and
35 G. Drisko, V. Luca, E. Sizgek, N. Scales and R. Caruso,
Langmuir, 2009, 25, 5286.
36 M. Snyder and M. Tsapatsis, Angew. Chem., Int. Ed., 2007, 46,
7560.
T. P. Russell, Science, 2005, 308, 236.
58 J. Lahann, S. Mitragotri, T. Tran, H. Kaido, J. Sundaram,
I. S. Choi, S. Hoffer, G. A. Somorjai and R. Langer, Science,
2003, 299, 371.
37 S. Dai, Chem.–Eur. J., 2001, 7, 763.
38 F. Akhtar, Q. Liu, N. Hedin and L. Bergstrom, Energy Environ.
59 A. Nakajima, K. Hashimoto and T. Watanabe, Langmuir,
2000, 16, 7044.
¨
Sci., 2012, 5, 7664.
60 K. Acatay, E. Simsek, C. Ow-Yang and Y. Z. Menceloglu,
Angew. Chem., Int. Ed., 2004, 43, 5210.
61 L. Zhai, F. Cebeci, R. E. Cohen and M. F. Rubner, Nano Lett.,
2004, 4, 1349.
39 Y. S. Hu, P. Adelhelm, B. M. Smarsly, S. Hore, M. Antonietti
and J. Maier, Adv. Funct. Mater., 2007, 17, 1873.
40 J. Biener, M. Stadermann, M. Suss, M. A. Worsley,
M. M. Biener, K. A. Rose and T. F. Baumann, Energy 62 N. J. Shirtcliffe, G. McHale, M. I. Newton, C. C. Perry and
Environ. Sci., 2011, 4, 656. P. Roach, Chem. Commun., 2005, 3135.
41 J. Chen, Y. Tan, C. Li, Y. Cheah, D. Luan, S. Madhavi, 63 L. Li, L. Ding, Z. Tu, Y. Wan, D. Claussea and J. Lanoiselle, J.
´
F. Boey, L. Archer and X. Lou, J. Am. Chem. Soc., 2010,
132, 6124.
Membr. Sci., 2009, 342, 70.
64 F. H. Wang, L. B. Shen, H. Zhu and K. F. Han, Pet. Sci.
Technol., 2011, 29, 2521.
´
42 O. Sel, S. Sallard, T. Brezesinski, J. Rathousky, D. R. Dunphy,
A. Collord and B. M. Smarsly, Adv. Funct. Mater., 2007, 17, 65 H. Ma, C. Burger, B. S. Hsiao and B. Chu, Biomacromolecules,
3241.
2011, 12, 970.
43 L. C. Jia and W. P. Cai, Adv. Funct. Mater., 2010, 20, 3765.
44 S. Tao, J. Yin and G. Li, J. Mater. Chem., 2008, 18, 4872.
66 X. Feng, S. Wang, J. Hou, L. Wang, C. Cepuch, J. Masliyah
and Z. Xu, Ind. Eng. Chem. Res., 2011, 50, 6347.
45 B. Fang, J. Kim, M. Kim, M. Kim and J. Yu, Phys. Chem. Chem. 67 Y. B. Zhou, L. Chen, X. M. Hu and J. Lu, Ind. Eng. Chem. Res.,
Phys., 2009, 11, 1380. 2009, 48, 1660.
46 K. Nakanishi and N. Tanaka, Acc. Chem. Res., 2007, 40, 863. 68 S. Bansal, V. Arnim, T. Stegmaier and H. Planck, J. Hazard.
47 N. A. Patankar, Langmuir, 2004, 20, 7097.
Mater., 2011, 19, 45.
48 M. Cathilde and D. Quere, So Matter, 2005, 1, 55.
49 M. Vila, J. L. Hueso, M. Manzano, I. Izquierdo-Barba,
69 D. Tian, X. Zhang, Y. Tian, Y. Wu, X. Wang, J. Zhai and
L. Jiang, J. Mater. Chem., 2012, 22, 19652.
´
´
A. Andres, J. Sanchez-Marcos, C. Prietoc and M. Vallet- 70 N. M. Kocherginsky, C. L. Tan and W. F. Lu, J. Membr. Sci.,
´
Regı, J. Mater. Chem., 2009, 19, 7745.
2003, 220, 117.
1708 | J. Mater. Chem. A, 2013, 1, 1701–1708
This journal is ª The Royal Society of Chemistry 2013