Artículo en PDF
How to cite
Complete issue
More information about this article
Journal's homepage in redalyc.org
Sistema de Información Científica
Red de Revistas Científicas de América Latina y el Caribe, España y Portugal
OIL-WATER SEPARATION, USING A HYDROPHILIC
POLYSULFONE-
POLYVINYLPYRROLIDONE ULTRAFILTRATION MEMBRANE
Heriberto ESPINOZA-GÓMEZ y Shui Wai LIN
Centro de Graduados del Instituto Tecnológico de Tijuana, Boulevard Industrial S/N Mesa de Otay, C.P. 22500, Tijuana,
B.C. México. Correo electrónico: jhespinoza@uabc.mx
(Recibido junio 2003, aceptado febrero 2004)
Palabras clave: microemulsiones aceite en agua, polisulfona-polivinilpirrolidona, fuerza centrífuga
RESUMEN
Se investigó la ultrafiltración para la separación de microemulsiones aceite/agua. Este
procedimiento presenta el inconveniente de la reducción del flujo debido al ensuciamiento de
la membrana. La hidrofilicidad de las membranas evita la adhesión de aceite en la superficie,
mientras que la fuerza centrífuga evita la acumulación de aceite en las cercanías de la membrana,
manteniendo flujos elevados. Los mejores resultados se obtuvieron con la membrana HL9 (4:18
Polivinilpirrolidona:Polisulfona), al aplicar 1000 r.p.m., la cual tuvo una recuperación del 85% de
agua con una velocidad de flujo de 4553 L/m
2
.d y una reducción del flujo del 56%. El contenido
de aceite se determinó por espectroscopía de infrarrojo, con transformadas de Fourier y accesorio
de reflectancia total atenuada (HATR-FTIR).
Key words: oil-in-water microemulsions, polysulfone-polyvinylpyrrolidone, centrifugal force
ABSTRACT
Ultrafiltration for the separation of oil-in-water micro-emulsions was investigated. Membrane
fouling is the major problem in ultrafiltration (UF) system. The present work proposes the
rotary spinning ultrafiltration membrane system for the treatment of oil-in-water micro-emulsions.
The best results were obtained using the membrane 9 (4:18 polyvnylpyrrolidone:polisulfone)
with water recovering of 85% with 1000 rpm and a flux velocity of 4553 L/m
2
d, and 56% of flux
decline. The oil-in-water micro-emulsions were measured using a HATR-FTIR.
Rev. Int. Contam. Ambient. 20 (2) 77-82, 2004
INTRODUCTION
Oil-in-water emulsions are one of the main pollutants
emitted into water by industry and domestic sewage
(Daminger
et al.
1995). The discharge of crude oily
wastewater into the sea or rivers has been under in-
creasingly careful scrutiny in recent years. In addition to
oily wastes from the petrochemical, metallurgical and
processing industries, it should be remembered that the
production of crude oil is often accompanied, on aver-
age, by an equal volume of water. A production that sepa-
rates most of the oil from water is usually used to give
an initial separation of oil and water. The small quantity
of remaining oil in the water must be reduced to an ac-
ceptable limit before the water can be discharged into
sea or rivers or re-injected for water flooding (Zhen-
Liang
et al.
1999).
The particularly stable emulsions are generated during
several mechanical operations such as grinding, rolling,
alkaline degreasing and transportation (Koltuniewicz and
Field 1996). The standard method for treatment of emul-
sified oily wastes is chemical de-emulsification followed
by secondary clarification. The systems require the use
of a variety of chemicals including sulphuric acid, iron and
H. Espinoza-Gómez and S.W. Lin
78
alumina sulphates and proprietary chemicals such as poly-
mers, waste pickle acid, etc. (Nazzal and Wiesner 1996;
Koltuniewicz and Field 1996; Arnot
et al
. 2000).
Several new effective methods have been recently
developed to solve the problem of oily wastes. Biotech-
nology offers a new approach based on biodegradation
and biotransformation of fats and oily wastes
(Koltuniewicz
et al.
1995). One of the most effective
methods of oil emulsion separation from water is
microfiltration (MF) and ultrafiltration (UF) performed
using ceramic membranes (Koltuniewicz
et al.
1995).
The performance of these membranes is enhanced by
means of surface modifications to achieve maximum yield
and separation effectiveness.
The primary limitation of conventional cross-flow UF
system in the treatment of concentrated oily wastewa-
ters is the low flux observed at high oil concentrations.
Recirculating velocities are used in conventional UF sys-
tems to maintain a satisfactory flux by inducing hydrau-
lic turbulence, which scours accumulated solute molecules
from the membrane surface. However, as the oil is con-
centrated, it becomes difficult to maintain a high cross-
flow velocity due to an increase in waste viscosity. This
problem can be overcome by decoupling the hydraulic
cleaning action from feed recirculation-pressurization. In
the high-shear rotary ultrafiltration (HSRUF) system, disk
membranes are rotated at speeds up to 1750 revolutions
per minute (rpm) to generate the hydraulic turbulence
necessary to scour solute molecules from the membrane
surface. Pumping is required only to provide transmem-
brane pressure and a small amount of recirculation flow.
METHODOLOGY
Figure 1
shows the schematic view of our separa-
tion unit. The dispersion was transferred from the tank
by pump to the membrane rotary system, and the re-
jected material was recirculated into the feed tank, while
permeate was collected in a separated reservoir.
The emulsion was prepared as follows: Owing to the
fact that the composition of the actual waste machine
cutting fluid may vary from machine shop to machine
shop, for the purpose of evaluating the functionalities of
the innovative energy-saving design of the spinning mem-
brane system, a synthetic waste machine oil was used to
cut the fluid in this research. The new and undiluted
machine cutting oil SOLG-A, from elf lubricants of
México, was bought at a machine shop near the Re-
search Center. The stock synthetic waste machine cut-
ting fluid was prepared by diluting 1.0 g of the original
commercial machine cutting oil into 1000 mL of deion-
ized water.
Three membrane types were tested in the same man-
ner, as previously reported (Espinoza-Gómez and Lin,
2001). The permeability of each of the membranes for
pure water was tested after cleaning and before each
run. The flux for pure water was easily restored to the
original value through rinsing with distilled water.
The American Standard Test Method (ASTM) for
oil, grease and petroleum hydrocarbons in water described
in ASTM D 3921-85, Vol. 11.02. 1996 was adopted
throughout our entire work. The oil content in the test
solution was determined by Horizontal Attenuated Total
Reflection IR (HATRIR), using an IR spectrophotom-
eter (instrument model Perkin Elmer 1600 FTIR). A
quantitative calibration curve, for oil content in oil-water
emulsion, was constructed using standard solutions pre-
pared from the new and undiluted machine cutting oil.
Figure 2
is an IR spectrum of the cutting oil; the peak at
2913 cm
-1
was monitorated for the determination of oil
content (
Fig. 3
). The oil content in the permeate sample
solution was determined against this calibration curve
as
recommended by the ASTM.
Rejected
Ultrafiltration
membrane
Inlet
Feed tank
(60 L)
Permeate
Pump
Rotor
Optical
tachometer
r.p.m.
Outlet
Fig. 1.
Scheme of the ultrafiltration pilot plant
2913
3406
1456
1373
4000.0
3000
2000
1500
1000
650.0
cm-1
99.0
90
80
70
60
50
40
30
20
10
0.0
%T
3406
2913
1456
1373
Fig. 2.
IR spectrum of oil
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
79
RESULTS
As illustrated in
Figure 4
, the conventional spinning
membrane system suffers some energy waste. The pro-
cess of permeate water flow behind the membrane disk
to the central rotating hollow shaft, is hindered by the
outward pushing centrifugal force during the spinning of
the membrane filtration operation. Therefore the effec-
tive filtration pressure is cut short by the action of this
centrifugal force, this amount of energy is being wasted
during the filtration process.
The centrifugal force acting on this out-going perme-
ate water flowing inside this tube creates a suction ac-
tion to pull the permeate. This waste ends up in the prod-
uct-water channel in front of the spinning hollow shaft.
This in fact counteracts the centrifugal force, acting on
the permeate that is flowing toward the spinning hollow
shaft. The result of this is an increase in the effective
filtration pressure across the spinning membrane. Thus
for a given applied filtration pressure and at a given spin-
ning velocity of the membrane disk, our energy-saving
spinning membrane system (with tube), should have a
0
1
2
5
10
20
40
60
80
100
0.08
0.06
0.04
0.02
0
absorbance
ppm oil-in-water
R
2
= 0.9815
Fig. 3
. Calibration curve for the oil-in-water emulsion determination (
ν
=2930cm
-1
)
Ultrafiltration
membrane
Ultrafiltration
membrane
Outlet
Outlet
Inlet
Inlet
Permeate
Hollow shaft
Permeate
Hollow shaft
Fig. 4.
Energy-saving membrane spinning system, (with tube) (a); and conventional spinning membrane (without tube) (b)
(a)
(b)
H. Espinoza-Gómez and S.W. Lin
80
higher permeate flux per time unit than the conventional
spinning membrane system (without tube) (
Fig. 4
). In-
deed the experimental results of oil-water emulsion sepa-
ration demonstrate this energy-saving benefit.
Figure 5
shows the comparison on the membrane permeates ve-
locities for the spinning membrane systems equipped with
and without the energy-savings device (tube).
The permeate velocities for membranes 7, 8, and 9,
were determined to be 19.3, 18.5, and 15.5 10
3
L/m
2
.d
respectively for the energy-saving design. For the con-
ventional design, the corresponding permeates velocities
were found to be 13.5, 10.9, and 6.7 10
3
L/m
2
.d. These
correspond to 43.0, 69.7 and 131% increase in perme-
ates velocities for 7, 8, and 9 membranes respectively.
The other membranes elaborated in this project, did not
have good results in comparison with the three employed.
The flow speed and the oil rejection were not signifi-
cant.
The higher membrane permeates velocities achieved
by our energy-saving spinning membrane system does
not hinder the oil rejection by the membrane.
Figure 6
indicates that for the same membrane, the oil rejection
was found to be higher for the spinning membrane sys-
tem with our energy-saving design than for the conven-
tional design.
The great force created by the spinning membrane
disk gives the suspended solid or large soluble solute little
chance to settle on the membrane surface.
Figure 7
shows the dependence among the permeate velocity, the
series of membranes (7, 8, and 9), and on membrane
disk spinning velocity and filtration time. This spinning
membrane filtration operation keep it’s permeate veloc-
ity in a good level, throughout 6 hours of test run, while
the stationary membrane filtration process suffers de-
0
20
40
60
80
100
120
140
160
180
200
220
240
Time (min)
21000
18000
15000
9000
12000
6000
0
3000
Permeate
(L/m^2.d)
Time (min)
0
20
40
60
80
100
120
140
160
180
200
220
240
7 (with tube)
7 (no tube)
8 (with tube)
8 (no tube)
9 (with tube)
9 (no tube)
Fig. 5.
Membrane permeates flux velocity for the spinning membrane disk system with and without
the energy-saving device (1000 rpm oil-in-water in feed: 1000 mg/L)
0
40
80
120
160
200
240
Time (min)
100
90
80
70
60
50
40
30
20
10
0
Oil rejection (%)
Fig. 6
. Oil rejection (%) for the spinning membrane disc system
with and without the energy-saving device (1000 rpm oil-in-
water in feed: 1000 mg/L) (PVP: polyvinylpyrrolidone, PS:
polisulfone)
With energy-saving device
Without energy-saving device
7 (22% PS, 0% PVP)
8 (20% PS, 2% PVP)
9 (18% PS, 4% PVP)
7 (22% PS, 0% PVP)
8 (20% PS, 2% PVP)
9 (18% PS, 4% PVP)
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
81
crease in permeate velocity soon after the test run was
started and continued to decrease with time. From these
test runs, it can also be seen, that the negative surface
charges density at the membrane surface do not affect
the permeate velocity of the spinning membrane sys-
tem. This is may be due to the fact that the great shear
force created by the spinning membrane disk is so over-
whelmed that the repulsive force existing between the
fixed negative surface charge at the membrane and the
negatively charged solute like oil-water emulsion in the
feed, becomes insignificant.
However, for the stationary test runs (i.e., membrane
disk spinning velocity = 0 rpm), the membrane, having
higher negative surface charge density, indeed tends to
have higher oil rejection from the oil-water emulsion feed.
The oil rejection of the membrane series is in the order
of 9>8>7 (
Fig. 8
). This is in line with the reduction in
membrane surface charge density shown in
Figure 9.
0
20
40
60
80
100
120
140
160
180
200
220
240
Time (min)
0
40
60
80
100
120
140
160
180
200
220
240
20
Time (min)
7 (1000 rpm)
8 (1000 rpm)
9 (1000 rpm)
7 (0 rpm)
8 (0 rpm)
9 (0 rpm)
21000
18000
15000
12000
9000
6000
3000
0
Permeate (L/m^2.d)
Fig. 7
. Dependence of the permeate velocity on membrane series (7, 8 and 9), on membrane disc
spinning velocity and filtration time (oil-in-water in feed: 1000 mg/L of oil-in-water)
100 120 140 160 180 200 220 240
7 (1000 rpm)
8 (1000 rpm)
9 (1000 rpm)
7 (0 rpm)
8 (0 rpm)
9 (0 rpm)
100
90
80
70
60
50
40
30
20
10
0
Oil rejection (%)
0
20
40
60
80
100
120
140
160
180
200
220
240
Time (min)
7 (1000 rpm)
7 (0 rpm)
8 (1000 rpm)
8 (0 rpm)
9 (1000 rpm)
9 (0 rpm)
Fig. 8
. Dependence of the oil rejection (%) on membranes series (7, 8 and 9), membrane disc
spinning velocity and filtration time (oil-in-water in feed: 1000 mg/L)
H. Espinoza-Gómez and S.W. Lin
82
In spite of more than thirty years of research and
development works on membrane technology for indus-
trial, environmental and domestic applications, membrane
fouling is still a major obstacle facing us today. One of
the major factors contributing to the membrane fouling
process is the condition of the feed flow at the boundary
region above the membrane surface. Turbulent flow de-
creases the thickness of the concentration polarization
layer above the membrane surface, and also minimizes
the chance of suspended solid or large solute in the feed
to settle down at the membrane surface. In order to mini-
mize the membrane fouling, the spinning membrane sepa-
ration system should be a prime choice for many mem-
brane separation processes. However the conventional
spinning membrane disk separation system suffers one
short coming: the effective filtration pressure of the spin-
ning membrane system is cut short by the centrifugal
force acting upon the permeate flowing towards to the
spinning hollow shaft.
Our energy-saving design of the spinning membrane
system clearly demonstrated the energy-saving benefit,
Fig. 9
. Composition of polyvinylpyrrolidone (PVP) polisulfone
(PS) ultrafiltration membrane casting solution
the increase of permeate velocity can be as high as 132%
(using 9 membrane) over the conventional spinning mem-
brane system. This innovative design for the spinning
membrane system may have an impact on the waste-
water treatment industries in the future.
ACKNOWLEDGEMENT
The authors are grateful for the financial support pro-
vided by CONACyT (411074-5-28023-U) and for the
scholarship to one of the authors (HEG/92464).
REFERENCES
Arnot T.C., Field R.W., Koltuniewicz A.B. (2000) Cross-flow
and dead-end microfiltration of oily-water emulsions Part
II. Mechanisms and modelling of flux decline. J of Memb
Sci 169, 1-15.
Belkacem M., Matamoros H., Cabassud C., Aurelle Y. and
Cotteret, J. (1995) New results in metal working wastewa-
ter treatment using membrane technology. J. Memb. Sci.
106, 195-205.
Daiminger U., Nitsch W., Plucinski P. and Hoffman S. (1995)
Novel techniques for oil/water separation. J. Memb. Sci.
99, 197-203.
Espinoza-Gómez H. and Lin S.W. (2001). Development of acry-
lonitrile co-polymers ultrafiltration membranes. Polymer
Bulletin 47(3-4), 297-304.
Koltuniewicz A.B. and Field R.W. (1996) Process factors dur-
ing removal of oil-in-water emulsions with cross-flow
microfiltration. Desalination 105, 79-89.
Koltuniewicz A.B., Field R.W. and Arnot T.C., (1995) Cross-
flow and dead-end microfiltration of oily-water emulsions.
Part I. Experimental study and analysis of flux decline. J.
Memb. Sci. 102, 193-207.
Nazzal F.F. and Wiesner M.R. (1996) Microfiltration of oil-in-
water emulsions. Water Environ. Res, 68-7, 1187-1191.
Zhen-Liang Xu, Tai-Shung Chung, Kai-Chee Loh and Bee Chun
Lim (1999). Polymeric asymmetric membranes made from
polyetherimide/polybenzimidazole/poly(ethylene glycol)
(PEI/PBI/PEG) for oil-surfactant-water separation. J. Memb.
Sci. 158, 41-53.
logo_pie_uaemex.mx