Environ. Sci. Technol. 2001, 35, 2388-2394
Hybrid Organic/Inorganic Reverse
Osmosis (RO) Membrane for
Bactericidal Anti-Fouling. 1.
Preparation and Characterization of
TiO2 Nanoparticle Self-Assembled
Aromatic Polyamide
Thin-Film-Composite (TFC)
Membrane
SEUNG-YEOP KWAK* AND SUNG HO KIM
Hyperstructured Organic Materials Research Center (HOMRC)
and School of Materials Science and Engineering,
Seoul National University, San 56-1, Shinlim-dong,
Kwanak-ku, Seoul 151-744, Korea
SOON SIK KIM
Saehan Industries Incorporated, #14, Nongseo-Ri,
Kiheung-Eub, Yongin-City,Kyunggi-Do 449-900, Korea
Hybrid organic/inorganic reverse osmosis (RO) membranes
composed of aromatic polyamide thin films underneath
titanium dioxide (TiO2) nanosized particles have been
fabricated by a self-assembly process, aiming at breakthrough
of biofouling problems. First, positively charged particles
of the colloidal TiO2 were synthesized by a sol-gel process,
and the diameter of the resulting particles in acidic
aqueous solution was estimated to be ≈2 nm by analyzing
the UV-visible absorption characteristics with a quantum
mechanical model developed by Brus. Transmission
electron microscopy (TEM) further confirmed the formation
of the quantum-sized TiO2 particles (∼10 nm or less).
The TiO2 particles appeared to exist in the crystallographic
form of anatase as observed with the X-ray diffraction
(XRD) pattern in comparison with those of commercial 100%
rutile and commercial 70:30% anatase-to-rutile mixture.
The hybrid thin-film-composite (TFC) aromatic polyamide
membranes were prepared by self-assembly of the
TiO2 nanoparticles on the polymer chains with COOH
groups along the surface. They showed improved RO
performance in which the water flux even increased, though
slightly. Field-emission scanning electron microscopy
(FESEM) exhibited the TiO2 nanoparticles well adsorbed
onto the surface. X-ray photoelectron spectroscopy (XPS)
demonstrated quantitatively that a considerable amount
of the adsorbed particles were tightly self-assembled at the
expense of the initial loss of those that were loosely
bound, and became stabilized even after exposure to the
various washing and harsh RO operating conditions. The
antibacterial fouling potential of the TiO2 hybrid membrane
was examined and verified by measuring the viable numbers
and determining the survival ratios of the Escherichia
coli (E. coli) as a model bacterium, both with and without
UV light illumination. The photocatalytic bactericidal
efficiency was remarkably higher for the TiO2 hybrid
2388
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 11, 2001
membrane under UV illumination, compared to that of the
same membrane in darkness, as well as those for the
neat membranes under either light condition.
Introduction
One of the goals of research and industry in the reverse
osmosis (RO) membrane fields has been to enhance, or at
least maintain, water flux without sacrificing salt rejection
over a long period, in order to increase efficiency and reduce
the cost of operation. Nevertheless, the main difficulty in
accomplishing this goal is fouling (1, 2), where a serious flux
decline occurs as the actual operation time elapses. A detailed
assessment of the costs of biofouling was made for the RO
plant at Water Factory 21 in Orange County, CA (3). According
to the report, the membranes, owing to the additional
hydraulic resistance of the biofouling layer, operate at about
150% of their initial operating pressure over 80% of their life.
A regular amount of chlorine is added continuously to the
feedwater and membranes are cleaned periodically. The
bottom line is $727,816 spent each year to control membrane
biofouling. That represents about 30% of the total operating
costs for the facility. The principal types of fouling are
crystalline fouling (mineral scaling, or deposit of minerals
due to an excess of the solution product), organic fouling
(deposition of dissolved humic acids, oil, grease, etc.), particle
and colloid fouling (deposition of clay, silt, particulate humic
substances, debris, and silica), and microbial fouling (biofouling, adhesion and accumulation of microorganisms, and
forming biofilms) (3). Various approaches to reducing fouling
have been used, which generally involve pretreatment of the
feed solution, modification of the membrane surface properties (such as hydrophobic or hydrophilic and electronegative or electropositive), optimization of module arrangement and process conditions, and periodic cleaning
(4). However, these methods vary widely in applicability and
efficiency, thereby requiring a breakthrough to solve fouling
problems.
The most common RO membrane used for water treatment is the thin-film-composite (TFC) type composed of
aromatic polyamide. Particularly for such aromatic polyamide
TFC membranes fouling from the formation of biofilm on
the surface caused by microorganisms has been regarded as
of the uppermost importance (5). Microorganisms such as
bacteria and viruses in water adhere to membrane surfaces
and grow at the expense of nutrients accumulated from the
water phase. The attached microorganisms excrete extracellular polymeric substances (EPS) and, thus, form biofilms
(6). It has been reported that biofilm formation was related
to the depletion of residual disinfectant concentration, and
that no biofilm was formed from disinfectant-treated water,
such as chlorinated water containing a residual of 0.04-0.05
mg/L free chlorine (7, 8). However, it is noted that chlorination, although effective for the destruction of microorganisms,
generates harmful byproducts such as trihalomethanes and
other carcinogens. Motivated by these results, the present
paper is aimed at developing and characterizing a hybrid
membrane possessing inorganic nanoparticles capable of
killing the microbes without forming unwanted byproducts
as a means of precluding the formation of biofilms and, hence,
reducing fouling.
Titanium dioxide (TiO2) has been the focus of numerous
investigations in recent years, particularly because of its
* Corresponding author phone: +82-2 880-6082; fax: +82-2 8766086; e-mail: [email protected].
10.1021/es0017099 CCC: $20.00
 2001 American Chemical Society
Published on Web 05/04/2001
FIGURE 1. Schematic of the thin-film-composite (TFC) reverse osmosis (RO) membrane and the chemical structure of the aromatic
polyamide thin-film layer.
photocatalytic effects that decompose organic chemicals and
kill bacteria (9). TiO2 photocatalysis is known to generate
various active oxygen species, such as hydroxyl radical,
hydrogen peroxide, etc., by reductive reactions or oxidative
reactions under light (10, 11). These active oxygen species
further destroy the outer membrane of the bacterium cells
and decompose the endotoxin from them. Nanoscale technology manipulates things on the nanoscale (generally
regarded as 1-100 nm) which makes it possible to arrive at
fundamentally new types of devices with much improved
properties and/or novel functionality (12). Nanosized TiO2
particles, from the viewpoint of their photocatalytic capability
to break down bacteria and organisms, will also be very useful
because of their high surface area per unit volume and high
abrasive resistance when coated on the target materials (9).
Several different strategies to integrate TiO2 with target
materials have been reported, which include self-assembly
monolayer adsorption on functionalized surfaces, sol-gel
synthesis, vacuum vaporization, sputtering, metal organic
chemical vapor deposition (MOCVD), chemical vapor deposition (CVD), and the Langmuir-Blodgett (LB) method.
Among them, the method of self-assembly of TiO2 on surfaces
(for example, single-crystal silicon, quartz, and glass substrates), employing polymers with -CO2H or -SO3H functional groups, can be performed to fabricate multilayer
ultrathin films, overcoming the limitations (such as high
temperature, solvent involvement, costly fabrication, and
complex process control) inherently associated with other
methods (13-15). It is worth noting that the thin-film active
layer of aromatic polyamide TFC RO membranes is composed
of the cross-linked form of three amide linkages and the
linear form with pendant free carboxylic acid as shown in
Figure 1. The fraction of the linear carboxylic acid form has
been estimated to be 30% to 50%, depending on the
investigators (16-18). Thus, it is probable to self-assemble
the TiO2 nanoparticles on the aromatic polyamide TFC
membrane surface, thereafter expecting the appearance of
a novel organic/inorganic hybrid TFC membrane for the
photocatalytic bactericidal anti-fouling.
In this study, quantum-sized TiO2 particles are prepared
from the controlled hydrolysis of titanium tetraisopropoxide
(19) and then characterized. The particle size and crystal
structure of the resulting TiO2 nanoparticles are characterized
by UV-visible absorption spectroscopy, transmission elec-
tron microscopy (TEM), and X-ray diffraction (XRD). Introduction of the TiO2 nanoparticles within the aromatic
polyamide TFC membrane is performed by way of selfassembly of the nanoparticles through ionic interaction and
H-bonding force with -COOH functional groups of the
aromatic polyamide. The morphological structures of the
resulting TiO2 self-assembled TFC membrane are investigated
by field-emission scanning electron microscopy (FE-SEM),
as well as by X-ray photoelectron spectroscopy (XPS), which
analyzes the atomic concentrations of titanium. The RO
performance test is then carried out to see whether any
variation of water flux and salt rejection occurs in the presence
of the TiO2 nanoparticles on top of the TFC membrane
surface. Thereafter, XPS analysis of atomic concentrations
of Ti is also performed with the RO-tested membrane to
evaluate the binding durability of the self-assembled TiO2
nanoparticles, even after the harsh, actual operation conditions. Finally, to verify the photocatalytic bactericidal capability of this TiO2 nanoparticle self-assembled TFC membrane, the membrane surface is covered by a model
suspension of Escherichia coli (E. coli) bacterium cells grown
aerobically in the nutrient broth, which is in turn illuminated
by UV radiation. Then, the bactericidal effect is observed
and confirmed by counting the viable number of E. coli cells.
Experimental Section
Synthesis and Characterization of the Nanosized TiO2
Particles. Nanosized TiO2 colloids were prepared from the
controlled hydrolysis of titanium tetraisopropoxide, Ti(OCH(CH3)2)4, by following procedures in the literature (19). A
1.25-mL sample of Ti(OCH(CH3)2)4 (Aldrich, 97%) dissolved
in 25 mL of absolute ethanol (J. T. Baker) by injection was
added drop by drop under vigorous stirring to 250 mL of
distilled water (4 °C) adjusted to pH 1.5 with nitric acid. This
mixture was stirred overnight until it was clear and the
transparent colloidal suspension (1.34 g/L) resulted. TiO2
surfaces dissociatively absorb water to form surface hydroxyl
groups, which are believed to be the active sites for the
adsorption of reactants. The surface properties of TiO2
particles are described by acid-base equilibria involving
surface hydroxyl groups. As the isoelectric point of titanium
dioxide corresponded to pH ) 4.5-6.8 (20), the resulting
colloids would take the shape of stable cationic TiO2 complex
at pH 1.5.
VOL. 35, NO. 11, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2389
The UV-visible spectrum of the transparent TiO2 colloidal
suspension was recorded with a Hewlett-Packard HP8452
diode array spectrophotometer to analyze the optical absorption characteristics of TiO2 and, thus, to determine the
particle sizes theoretically. The particle sizes were also
determined by a JEOL JEM-200CX transmission electron
microscope (TEM) at 120 kV. For the TEM observation, the
TiO2 colloidal suspension was dropped on a carbon-coated
grid and then dried at room temperature. The crystal structure
of TiO2 was characterized by X-ray diffraction (XRD). XRD
analysis was performed on TiO2 powder samples with a MAC
Science X-ray diffractometer (MXP18X-MF22-SRA), operating
in the theta-theta geometry using 18kW Cu KR (λ ) 1.5418
Å) radiation. For comparison purposes, other commercial
TiO2 particles such as Sigma-Aldrich rutile TiO2 and DegussaHüls P25 TiO2 were also analyzed by XRD.
Preparation of TiO2 Nanoparticle Self-Assembled ThinFilm-Composite (TFC) Membranes and Measurement of
Transport Characteristics. The thin-film-composite (TFC)
membranes were prepared via interfacial polymerization of
m-phenylenediamine (MPD) in the aqueous phase (2 wt %)
and trimesoyl chloride (TMC) in the organic phase (0.1 wt
%) on the polysulfone supports reinforced by the nonwoven
fabric as schematically depicted in Figure 1. The polysulfone
layer acts as the support to give membranes the mechanical
strength to resist RO pressure. The thin-film layer governs
the actual separation of the solute and the passage of the
solvent. The resulting aromatic polyamide TFC membrane
was rinsed in a sodium carbonate solution (0.2 wt %) and
then washed with distilled water. The final membrane, with
an area of ca. 50.0 cm2, was dipped in the transparent TiO2
colloidal solution for 1 h to deposit TiO2 nanoparticles on
the membrane surface, then rinsed extensively with water.
Reverse osmosis (RO) performance tests were conducted
at 225 psi using 2000 ppm NaCl solution at 25 °C with the
apparatus of a continuous-flow type. The water flux was
determined by direct measurement of the permeate flow:
Flux (gfd) )
permeate (gallon)
membrane area (ft2)‚time (day)
(1)
The salt rejection was measured by the salt concentration
in the permeate obtained through measurements of the
electrical conductance of the permeate and the feed using
a conductance meter (Orion model 162):
(
Rejection (%) ) 1 -
permeate conductance
× 100 (2)
feed conductance
)
Characterization of Morphological and Chemical Structures of Membrane Surface. The surface topologies of the
TiO2 nanoparticle-introduced aromatic-polyamide-TFC membrane were investigated with a Philips XL30 FEG field
emission scanning electron microscope (FESEM). The surface
morphology of the neat aromatic-polyamide-TFC membrane
was also examined and was compared with that of the TiO2
self-assembled version. For the FESEM observation, the
membrane samples were cut into appropriate sizes and the
surfaces were coated with platinum or gold by a sputtercoating machine.
X-ray photoelectron spectroscopy (XPS) experiments were
carried out with a Kratos AXIS HS spectrometer using a Mg
KR X-ray source (1253.6 eV). The X-ray gun was operated at
10 kV and 1 mA, and the charge neutralization system was
used to obtain high-resolution spectra for the insulating
materials, such as polymers, by reducing the surface charge.
The spectrum was obtained at the photoelectron takeoff
angles (defined as the angle between the detected photoelectron beam and the membrane surfaces) of 30° and 90°
to give sampling depths of ca. 23 Å and ca. 45 Å, respectively.
2390
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 11, 2001
FIGURE 2. UV-visible absorption spectrum of the dilute nanosized
TiO2 colloidal suspension.
The elemental composition analysis was performed on
carbon, nitrogen, oxygen, and titanium, which constituted
the hybrid membrane as well as the neat membrane. The
sensitivity factors of individual elements were taken with the
values from the standard vision library provided by the
manufacturer, which were based on a combination of
photoelectric cross-section, transmission function, and
inelastic mean free path.
Evaluation of Photocatalytic Bactericidal Effect of TiO2
Self-Assembled TFC Membrane. Escherichia coli (E. coli)
bacterium cells (DH5R strain) were grown aerobically in 10
mL of nutrient broth (Luria-Bertani medium) at 37 °C for
12-16 h. The Luria-Bertani (LB) medium was prepared with
1 wt % Bacto-tryptone, 0.5 wt % yeast, and 1 wt % NaCl. The
grown cells were centrifuged at 10000 rpm for 1 min and
diluted to an appropriate concentration with sterilized water.
The E. coli cell dilution (150 µL, total 1.0 × 104 cells) was
pipetted onto either a TiO2 hybrid TFC membrane or a neat
membrane, which were placed in an incubator at a constant
temperature of 37 °C. Some of the individual membranes
were illuminated with an 8-W black light (VWR UVLS-28)
and some were not. The light intensity at the peak of 365 nm
was 500 µW cm-2 at 3 in., which was determined by the
procedure provided by the manufacturer. After illumination
up to the intended exposure time, the cells were pipetted out
and collected in 1.0 wt % aqueous sodium chloride solution.
The collected solutions were spread onto a LB agar plate and
incubated for 12-16 h to determine the number of viable
cells in terms of colony-forming units (CFU) as a function
of time. The initial cell number was determined to be 9420
in 150 µL of cell dilution suspension spread onto a LB plate
without illumination.
Results and Discussion
Particle Size and Crystal Structure of Synthesized TiO2.
Figure 2 shows the UV-visible spectrum of the TiO2 colloidal
solution obtained via sol-gel synthesis of titanium tetraisoproxide. The onset of absorption (λos) and the corresponding band gap energy (Eg) of the bulk TiO2 have been
determined to be λos ) 385 nm and Eg ) 3.2 eV for anatase
(21), and λos ) 415 nm and Eg ) 3.0 eV for rutile (22),
respectively. The band gap of the TiO2 colloidal solution is
measured to be 3.44 eV (361 nm) according to the spectral
analysis in the figure, which is in agreement with other
research. This corresponds to the 0.24 eV blue shift from the
bulk-phase band gap of anatase (3.2 eV), indicating that the
ultra-small TiO2 particles are formed. Particle size can be
further estimated according to a theoretical prediction
proposed by Brus (23):
∆Eg )
h2
1.8e2
2
R
8R µ
(3)
where ∆Eg is the band gap shift, R is the radius of the particle,
µ is the reduced mass of the exciton (µ ) 1.63 me; me is the
electron rest mass) for TiO2, and is the dielectric constant
of the semiconductor ( ) 184) for TiO2 (21). The band gap
shift of 0.24 eV for the TiO2 colloids corresponds to a particle
diameter of ca. 2.0 nm, which implies so-called quantum(Q) sized TiO2 particles. The particle size is also investigated
by transmission electron microscopy (TEM) as shown in
Figure 3, where black spots are the synthesized TiO2 particles
and all of them measure less than 10 nm. The discrepancy
between the particle sizes determined by TEM and UVvisible spectroscopy might ascribe to the agglomeration of
the Q-sized TiO2 particles during the drying procedure to
prepare TEM samples.
X-ray diffraction (XRD) analysis is employed to characterize the crystal structure of the TiO2 nanoparticles. It is
known that TiO2 particles are in two different crystal forms,
i.e., anatase and rutile (24). In the anatase (Tio[O2]c) structure,
the oxygens form a cubic closest packing, and the titanium
atoms lie in octahedral voids. In the rutile (Tio[O2]h) form,
the oxygens are arranged approximately in a hexagonal closest
packing, and the titanium atoms occupy a row pattern. Many
researchers claim that the anatase appears to be the most
photoactive and stable nanoparticles for widespread practical
applications (9, 25, 26), whereas the rutile is photocatallytically inactive (27-29) or much less active (30-33), although
it shows strong photoactivity selectively toward some cases
(9, 34). Figure 4 compares the X-ray diffraction patterns of
three types of TiO2 particles as designated in the figure. The
100% rutile TiO2, Figure 4(b), shows the characteristic peaks
located at 2θ of 27.45°. As for the diffraction pattern of Degussa
P25 TiO2, Figure 4(c), which is a nonporous 7:3 anatase-torutile mixture and is one of the most often used photocatalysts
(9), the 2θ of eminent peaks are 25.24° for anatase and 27.46°
for rutile. Comparing diffraction patterns of (a) with (b) and
(c), it is confirmed that our TiO2 nanoparticles are composed
entirely of anatase, which promises the highest photoreactivity and the best efficiency for destroying the microorganisms.
Reverse Osmosis Performance and Surface Characterization of TiO2 Self-Assembled and Neat Membranes. As
described earlier, the fabrication of TiO2 self-assembled thinfilm-composite (TFC) membranes was carried out by dipping
the neat aromatic polyamide membrane into the solution of
colloidal TiO2 particles followed by washing with water.
According to the recent investigation of the adsorption
behavior of carboxylic acid on TiO2 by virtue of diffuse
reflectance infrared Fourier transform (DRIFT), the process
of self-assembly between carboxylic acid and TiO2 was
explained by two different adsorption schemes (35). One
scheme was that TiO2 was bound with two oxygen atoms of
FIGURE 3. TEM micrograph of the TiO2 nanoparticles.
VOL. 35, NO. 11, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2391
FIGURE 4. XRD images of the synthesized TiO2 (a), commercial
rutile TiO2 (Sigma-Aldrich) (b), and commercial P25 TiO2 (DegussaHu1 ls) particles (c).
TABLE 1. Transport Characteristics of Aromatic TFC
Membranes
RO performance a
sample
neat TFC membrane
TiO2 hybrid TFC membrane
water flux (gfd) salt rejection (%)
13.2
14.4
96.5
96.6
a All the results were obtained with 2000 ppm NaCl in deionized
water and at the operating pressure of 225 psi and temperature of 25
°C.
a carboxylate group via a bidentate coordination to Ti4+
cations. The other scheme was to form a H bond between
a carbonyl group and the surface hydroxyl group of TiO2.
Although the majority of COOH groups are not dissolved
into the form of free ion at pH 1.5, however, the distribution
of electron density in a polar bond may be symbolized by
partial charges: δ+ (partial positive) and δ- (partial negative).
Two oxygen atoms of COOH groups are partial negative and
have ionic character. These facts suppose that ionic interaction between the positive surface of the metal oxide particles
and COOH groups with ionic character causes the adsorption,
and that TiO2 particles are strongly bonded to the membrane
surface by a bidentate coordination and a H bond.
The basic requirement of such a hybrid TFC membrane
is to preserve the reverse osmosis (RO) performance as much
as possible before the integration of TiO2 nanoparticles. Table
1 contains the RO performance data of water flux and salt
rejection for TiO2 self-assembled and neat aromatic polyamide TFC membranes; all the results were the arithmetic
means of four replications. The TiO2 hybrid TFC membrane
shows a slight increase in water flux as compared to that of
the neat membrane. The observed flux increase upon
integration of TiO2 nanoparticles can be explained by two
facts. During self-assembly of TiO2 nanoparticles, the TFC
membrane is exposed in the very-low-pH nitric acid, in which
the acid has been found to cause partial hydrolysis on the
membrane surface and increase hydrophilicity, and, hence,
water flux (36). The other explanation may involve the water
uptake characteristics of TiO2 particles (37), which are
considered to be a further contribution to the increase of
water flux. Shown in Figure 5 are the surface topologies of
the TiO2 hybrid and the neat TFC membranes investigated
by field-emission scanning electron microscopy (FESEM).
The neat aromatic polyamide membrane has the typical
surface morphology of a characteristic ridge-and-valley
structure, Figure 5(a), which has been observed by many
investigators as the analogous membrane under the trade
2392
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 11, 2001
FIGURE 5. FESEM micrographs of the neat (a) and the TiO2 hybrid
(b) aromatic polyamide TFC membranes.
name of FT-30 (38). Figure 5(b) displays the surface image
of the TiO2 self-assembled hybrid TFC membrane, where
TiO2 nanoparticles appear to exist as nodular shapes of ca.
10 nm or less on the surfaces of the ridges and valleys. To
confirm the self-assembly TiO2 nanoparticles within the
hybrid membrane and further estimate their abrasive
resistance to the washing steps involved in membrane
preparation, as well as the actual operating conditions, X-ray
photoelectron spectroscopic (XPS) analyses are carried out
for the neat TFC membrane and the TiO2 hybrid TFC
membrane treated under various conditions. The constituent
elements of the thin-film layer of the neat TFC membranes
are hydrogen, carbon, nitrogen, and oxygen, and additionally
titanium for the hybrid TFC membrane. Thus, XPS analyses
are performed on the elements of carbon, nitrogen, oxygen,
and titanium, but not on hydrogen because its photoelectron
cross-section is too small to be characterized by XPS. The
core-electron binding energies of the constituent elements
are typically 284.6 eV for C 1s, 397.9 eV for N 1s, 531.6 eV for
O 1s, 453.8 eV for Ti 2p3/2, and 460.0 eV for Ti 2p1/2. Figure
6 (a-d) shows the resulting XPS spectra, in which all the
photoelectron peaks appear at positions similar to the above
values and the presence of Ti peaks (d). The hybrid membrane
provides evidence of TiO2 self-assembling. On the basis of
the observed photoelectron peaks and corresponding sensitivity factors, the relative atomic concentrations of the
FIGURE 6. XPS spectra of carbon (a), nitrogen (b), oxygen (c), and titanium (d) for the neat and the TiO2 hybrid aromatic polyamide TFC
membranes.
individual elements can be calculated:
Ci )
AiSi
(4)
m
∑A /S
j
TABLE 2. Elemental Compositions of the TiO2 Hybrid TFC
Membranes under Various Washing Conditions and RO
Operational Hour
relative atomic concentration (%)
j
j
where Ai is the photoelectron peak area of the element i, Si
is the sensitivity factor for the element i, and m is the number
of elements in the sample. In Table 2, the elemental
compositions determined by an angle-resolved XPS analysis
are summarized for the hybrid membranes with various
washing conditions and RO operational hours. As seen in
the table, there is an initial drop in the relative atomic
concentration of Ti after washing the hybrid membrane which
has been just formed from dipping into the TiO2 colloidal
solution. An additional loss of TiO2 nanoparticles is observed
upon further RO operation with run time of 30 min.
Recognizing that the RO process in this study is operated in
the cross-flow mode where the feed solution is pumped across
the hybrid membrane parallel to its surface, it is thought
that the loosely bound TiO2 particles cannot overcome the
shear-flow force and are, thus, wiped out during the 30 min
RO operation. However, the TiO2 loss does not continue to
progress as the RO operational hours are increased, and the
amount of TiO2 levels off even after 7 days of RO operation.
This result indicates that a considerably substantial amount
of TiO2 nanoparticles remains tightly bound on the surface
of the membrane under actual RO running conditions, which
samplea
takeoff angle (°)
C
O
N
Ti
1
90
30
90
30
90
30
90
30
64.3
61.8
61.4
60.0
62.9
60.9
63.7
59.5
27.1
27.4
30.2
29.4
27.7
26.8
29.3
30.7
6.3
8.3
6.4
8.3
8.1
10.8
5.9
8.3
2.3
2.4
1.9
2.2
1.2
1.5
1.1
1.5
2
3
4
a Analyses were performed for the TiO self-assembled TFC RO
2
membranes (1) just after preparation, (2) after washing with flowing
water, (3) after RO operation of 30 min, (4) after RO operation for another
7 days.
is expected to act as photocatalyst and reduce the biofouling
as a result of destroying the bacteria on top of the membrane
surface.
Photocatalytic Bactericidal Effect of TiO2 Self-Assembled
Membrane. Figure 7 shows the plots of the survival ratios of
E. coli bacteria in both the TiO2 hybrid and the neat TFC
membranes with and without UV light illumination as a
function of time; the hybrid membrane was that of the 7-dayoperated RO. In the experimental setup, the natural diminution of cell population with time was unavoidable, probably
VOL. 35, NO. 11, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2393
FIGURE 7. Photocatalytic bactericidal effects of the TiO2 hybrid
and neat aromatic polyamide TFC membranes in the dark and with
UV light illumination.
due to an insufficient supply of the nutrients over a prolonged
time interval, and the survival ratio of E. coli cells for the neat
TFC membrane in the dark without illumination is inevitably
decreased by ca. 40%. The TiO2 hybrid TFC membrane in the
same dark condition is shown to affect and decrease slightly
the survival ratio, implying that the TiO2 itself, even in the
dark, may have minute photocatalysis on E. coli. For the
neat TFC membrane under UV light illumination, the survival
ratio of E. coli cells is reduced to ca. 40% within 3 h and ca.
37% within 4 h. The UV light causes more sterilization of the
TiO2 hybrid membrane; the 10% of E. coli cells survived only
after 3 h, reaching complete sterilization within 4 h in the
presence of TiO2. The TiO2 self-assembled hybrid membrane
under UV illumination possesses a remarkably higher photocatalytic bactericidal efficiency than that without illumination and the neat TFC membranes. The photocatalytic
bactericidal capability demonstrated by the TiO2 selfassembled hybrid TFC membrane offers a strong potential
for possible use as a new type of antifouling RO membrane.
The bactericidal efficiency of the different species of bacteria
and the actual antifouling performance of the hybrid
membrane, as well as an idea concerning the light source
inside the membrane module, will be revealed in a publication soon.
Acknowledgments
The authors are grateful to the Korea Science and Engineering
Foundation (KOSEF) for their support of this study through
the Hyperstructured Organic Materials Research Center
(HOMRC). They also express their appreciation to Saehan
Industries Inc. for their permission to publish this work.
Literature Cited
(1) Potts, D. E.; Ahlert, R. C.; Wang, S. S. Desalination 1981, 36, 235.
(2) Mulder, M. Basic Principles of Membrane Technology; Kluwer
Academic Publishers: Dordrecht, The Netherlands, 1996.
(3) Flemming, H. C. Exp. Thermal Fluid Sci. 1997, 14, 382.
(4) Fane, A. G.; Fell, C. J. D. Desalination 1987, 62, 117.
(5) Saad, M. A. Desalination 1992, 88, 85.
2394
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 11, 2001
(6) Flemming, H. C.; Schaule, G.; McDonogh, R.; Ridgeway, H. F.
Biofouling and Biocorrosion in Industrial Water Systems; Lewis
Publishers: Chelsea, MI, 1994.
(7) Lund, V.; Ormerod, K. Wat. Res. 1995, 29, 1013.
(8) Maggy, Momba N. B.; Cloete, T. E.; Venter, S. N.; Kfir, R. Wat.
Sci. Technol. 1998, 38, 283.
(9) Mills, A.; Hunte, S. L. J. Photochem. Photobiol. A: Chem. 1997,
108, 1.
(10) Kikuchi, Y.; Sunada, K.; Iyoda, T.; Hashimoto, K.; Fujishima, A.
J. Photochem. Photobiol. A: Chem. 1997, 106, 51.
(11) Sunada, K.; Kikuchi, Y.; Hashimoto, K.; Fujishima, A. Environ.
Sci. Technol. 1998, 32, 726.
(12) Reed, M. A. Chem. Eng. News 1993, March 22, 20.
(13) Liu, Y.; Wang, A.; Claus, R. J. Phys. Chem. B 1997, 101, 1385.
(14) Rizza, R.; Fitzmaurice, D. Chem. Mater. 1997, 9, 2969.
(15) Kovtyukhova, N.; Ollivier, P. J.; Chizhik, S.; Dubravin, A.;
Buzaneva, E.; Gorchinskiy, A.; Marchenko, A.; Smirnova, N. Thin
Solid Films 1999, 337, 166.
(16) Cadotte, J. E. Materials Science of Synthetic Membranes; Lloyd,
D. R., Ed.; American Chemical Society: Washington, DC, 1985.
(17) Koo, J. Y.; Petersen, R. J.; Cadotte, J. E. Polym. Prepr. (ACS, Div.
Polym. Chem.) 1986, 27, 391.
(18) Kwak, S.-Y.; Jung, S. G.; Kim, S. H. Macromolecules, submitted
for publication.
(19) Choi, W.; Termin, A.; Hoffmann, M. R. J. Phys. Chem. 1994, 98,
13669.
(20) Matijevic, E.; Budnik, M.; Meites, L. J. Colloid Interface Sci. 1977,
61, 302.
(21) Kormann, C.; Bahnemann, D. W.; Hoffmann, M. R. J. Phys. Chem.
1988, 92, 5196.
(22) Kandori, K.; Kon-no, K.; Kitahara, A. J. Colloid Interface Sci.
1988, 122, 78.
(23) Brus, L. E. J. Phys. Chem. 1986, 90, 2555.
(24) Lima-de-Faria, J. Structural Mineralogy; Kluwer Academic
Publishers: Dordrecht, The Netherlands, 1994.
(25) Rao, M. V.; Rajeshwar, K.; Vernerker, V. R.; Dubow, J. J. Phys.
Chem. 1980, 84, 1987.
(26) Nishimoto, S.; Ohtani, B.; Kajiwara, H.; Kagiya, T. J. Chem. Soc.,
Faraday Trans. 1 1985, 81, 61.
(27) Mills, A.; Davies, R. H.; Worsley, D. Chem. Soc. Rev. 1993, 22,
417.
(28) Mills, A.; Morris, S.; Davies, R. J. Photochem. Photobiol. A: Chem.
1993, 70, 183.
(29) Martin, S. T.; Morrison, C. L.; Hoffmann, M. R. J. Phys. Chem.
1994, 98, 13695.
(30) Karakitsou, K. E.; Verykios, X. E. J. Phys. Chem. 1993, 97, 1184.
(31) Noda, H.; Oikawa, K.; Kamada, H. Bull. Chem. Soc. Jpn. 1993,
66, 455.
(32) Ohtani. B.; Nishomoto, S. J. Phys. Chem. 1993, 97, 920.
(33) Mihaylov, B. V.; Hendrix, J. L.; Nelson, J. H. J. Photochem.
Photobiol. A: Chem. 1993, 72, 173.
(34) Hoffmann, M. R.; Martin, S. T.; Choi, W.; Bahnemann, D. W.
Chem. Rev. 1995, 95, 69.
(35) Lee, S. J.; Han, S. W.; Yoon, M.; Kim, K Vib. Spectrosc. 2000, 24,
265.
(36) Kulkarni, A.; Mukherjee, D.; Gill, W. N. J. Membr. Sci. 1996, 114,
39.
(37) Khare, S. Presented at International Conference on Polymer
Characteristics (POLYCHAR-8); University of North Texas,
Denton, TX; Jan. 11-14, 2000.
(38) Petersen, R. J.; Cadotte, J. E. Handbook of Industrial Membrane
Technology; Porter, M. E., Ed.; Noyes Publications: Park Ridge,
NJ, 1990.
Received for review September 28, 2000. Revised manuscript
received February 12, 2001. Accepted March 1, 2001.
ES0017099