Water Research 161 (2019) 27e34

Contents lists available at ScienceDirect

Water Research
journal homepage: www.elsevier.com/locate/watres

Differential ATR FTIR spectroscopy of membrane fouling:

Contributions of the substrate/fouling films and correlations with
transmembrane pressure
Hongguang Guo a, b, c, *, Xinyu Tang b, Gilbert Ganschow b, Gregory V. Korshin b
a
College of Architecture and Environment, Sichuan University, Chengdu, 610065, China
b
Department of Civil & Environmental Engineering, University of Washington, Box 352700, Seattle, WA, United States
c
Key Laboratory of Deep Earth Science and Engineering (Sichuan University), Ministry of Education, Chengdu, 610065, China

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

Article history: This study examined the formation of fouling films deposited on the surface of a polyethersulfone (PES)
Received 9 February 2019 membrane during the filtration of alginate solutions with various ionic strengths. Membrane fouling was
Received in revised form characterized by changes of the transmembrane pressure (TMP) and ex situ measured attenuated total
13 May 2019
reflectance (ATR) Fourier-transform IR (FTIR) spectra at varying stages of filtration runs. The ATR spectra
Accepted 25 May 2019
Available online 27 May 2019
that comprise the vibration bands characteristic of the PES substrate and the deposited film were pro-
cessed taking into the gradual weakening of the PES substrate-specific bands, whose intensity was
shown to depend on the wavenumber of IR radiation and the thickness of the deposited layer. Strongly
Keywords:
Alginate
linear correlations between ratios of first derivatives intensity and wavenumbers of the PES reference
Attenuated total reflectance (ATR) lines were established. Calculations of the PES bands’ attenuation coefficients allowed determining the
Differential spectroscopy apparent thickness and ATR FTIR vibrations of the fouling films per se. Strong correlations between TMP
Membrane fouling development and ATR-determined apparent thickness of the fouling layers were observed. The intensity
Fourier-transform infrared spectroscopy of ATR absorbance at 3200 cm
1 was linearly correlated with TMP development for small TMP values
(FTIR) before the point of rapidly developing failure of the hydraulic permeability of the system was reached.
Fouling film © 2019 Elsevier Ltd. All rights reserved.

1. Introduction accretion and consolidation with ensuing changes of membrane

Membrane-based technologies have demonstrated their un-
matched potential in surface water and wastewater treatment,
reverse osmosis and water reuse applications, yet membrane
fouling remains one of the greatest challenges for these technolo-
gies (Herzberg et al., 2009; Shannon et al., 2008). Extensive studies
have evaluated membrane fouling via the monitoring of flux
decline (constant pressure mode) or transmembrane pressure in-
crease (constant flux mode) combined with determinations of
filtered water quality (Thibault et al., 2017).
Given the complexity of membrane fouling processes and their
dependence on water chemistry, membrane material, trans-
membrane pressure and other operational conditions, determina-
tion of the nature of membrane foulants, mechanisms of their
* Corresponding author. College of Architecture and Environment, Sichuan Uni-
versity, Chengdu, 610065, China.
E-mail address: hgguo@scu.edu.cn (H. Guo).
permeability requires in situ methods that can maintain the
integrity of the solution/deposited layer/membrane system. A
number of in situ methods to quantify membrane fouling processes,
for instance spectrophotometry (Sim et al., 2018), Raman spec-
troscopy (Cui et al., 2011; Kogler et al., 2016), ultrasonic methods,
Nuclear Magnetic Resonance (NMR) micro-imaging, electro-
chemical impedance spectroscopy, optical coherence tomography
and others have been employed for that purpose (Dreszer et al.,
2014; Fortunato et al., 2017; Gao et al., 2014; Jing et al., 2016; Li
et al. 2003, 2005; Yao et al., 1995).
However, their use of in situ methods to study fouling has been
somewhat limited due to a number of experimental challenges
inherent to membrane systems. While the use of such methods and
their robustness are increasing, an array of ex situ techniques such
as scanning and transmission electron microscopy (SEM and TEM),
X-ray photoelectron spectroscopy (XPS), energy-dispersive spec-
troscopy (EDS), laser scanning microscopy (Meng et al., 2010),
synchrotron Fourier transform infrared imaging (Xie et al., 2017),
terahertz time-domain spectroscopy (Zhang et al., 2018),

https://doi.org/10.1016/j.watres.2019.05.086
0043-1354/© 2019 Elsevier Ltd. All rights reserved.
28 H. Guo et al. / Water Research 161 (2019) 27e34
excitation-emission matrix (EEM) and Raman spectroscopy of
membrane surfaces (Ao et al., 2016; Cui et al., 2015; Henderson
et al., 2011; Pawlowski et al., 2016; Yamamura et al., 2014; Yu
et al., 2019) and others have also been used to characterize
fouling films and foulants deposited on membrane surfaces (Le-
Clech et al., 2007; Marselina et al., 2009; Mendret et al., 2007;
Mores and Davis, 2001).
Attenuated Total Reflectance Fourier Transform Infrared (ATR
FTIR) spectroscopy is another ex situ method that has been widely
used to characterize membrane materials (Freger et al., 2002),
determine the thickness of polymer coating films (Bass and Freger,
2015), characterize their swelling, identify the chemical state of
water in them (Dias et al., 1998; Drazevic et al., 2014) and examine
fouling films formed on membrane surfaces (Gamage and Chellam,
2014; Sioutopoulos and Karabelas, 2012; Wang et al., 2015). Given
that ATR FTIR spectra comprise the IR bands characteristic of both
the membrane material and deposited films, the ATR spectra need
to be interpreted so as to separate the contributions of the substrate
itself and the deposited films whose thickness, structure and
chemistry tend to vary as function of the membrane material,
filtration conditions, water chemistry and other factors (Bass and
Freger, 2015; Belfer et al., 2000; Gamage and Chellam, 2014;
Huang et al., 2017; Wang et al., 2015; Xie et al., 2017).
ATR-based determination of the apparent thickness of thin
polymeric films formed on inert support has been carried out for
several membrane materials (i.e. polyamide and polyethersulfone,
PES) (Rao et al., 2003). ATR FTIR data have been used to compare
surface properties of pristine and fouled ultrafiltration PES mem-
branes that had varying pore sizes (Pihlajamaki et al., 1998). These
and other relevant studies have demonstrated that natural organic
matter (NOM), wastewater effluent organic matter (EfOM) and
extracellular polymeric substances (EPS) cause the deterioration of
the membrane performance (Howe and Clark, 2002; Kim and
Dempsey, 2013; Lee et al., 2004; Sioutopoulos and Karabelas,
2012). Given the site-specificity and heterogeneity of these com-
pound classes, individual model species (e.g., alginate and xanthan
polysaccharides, bovine serum albumin) have been employed to
study membrane fouling mechanisms. Alginate is known to
strongly influence the behavior of microfiltration, ultrafiltration,
nanofiltration and reverse osmosis membranes (Listiarini et al.,
2009; Mi and Elimelech, 2010; Sioutopoulos and Karabelas, 2016).
Effects of alginate in membrane fouling depend on its polymeric
properties (e.g., molecular weight of its macromolecules, their
charges, contributions of mannuronate and guluronate monomeric
blocks), its concentration, those of hardness cations, other solutes,
pH, and solution ionic strength whose increases tend to suppress
alginate's fouling action of alginate (Lee and Elimelech, 2006; Lu
et al., 2013; Zhang et al. 2017, 2018).
Despite the extent and depth of these studies, microscopic
mechanisms of membrane fouling, including those induced in the
specific case of the model foulant alginate remain to be established
in more detail (Kane et al., 2009). While ATR FTIR spectroscopy is an
ex situ technique with limitations applicable to any ex situ method,
its potential for ascertaining the nature of fouling films has been
somewhat limited by the need to separate the contributions of the
foulants from the potentially complex and intense vibrations of the
substrate itself (Lujan-Facundo et al., 2015). Besides, most of the
reported studies paid limited attention on the quantitative rela-
tionship between membrane fouling performance and IR vibra-
tions, as well as the thickness of the fouling layer. In the present
study, the ATR spectra that comprise the vibration bands charac-
teristic of the PES substrate and the deposited film were processed,
which addressed this limitation. A rigorous approach was devel-
oped that compares the ATR FTIR spectra of pristine and fouled PES
membranes. It subsequently allows calculating apparent
thicknesses of the fouling films, their effects on the IR vibrations of
the substrate, and examining relationships between the ATR-
derived parameters of the deposited films (e.g., their spectro-
scopic signature and apparent thickness) and the development of
the transmembrane pressure.
2. Materials and methods
Sodium alginate salt (MW 20 kDa, Lot# MKBG5630V) was pur-
chased from sigma-Aldrich (St. Louis, USA). Sodium Chloride, hy-
drochloric acid, sodium hydroxide and other reagents were
analytical grade and obtained without further purification. All so-
lutions were prepared using Milli-Q water (18.2 MU cm
1, Millipore
Corp., MA, USA). Commercial flat-sheet polyethersulfone (PES)
micro-filtration membrane was obtained from NADIR Corp. The
membrane (MP005, Microdyn-Nadir, USA) had a nominal pore size
of 0.05 mm. Before being installed into a filter holder cartridge
(surface area of 9.62 cm2), a membrane disk-shaped sample with a
47 mm diameter was cut from a flat PES membrane sheet, soaked in
deionized water overnight and carefully rinsed. The area of mem-
brane in contact with solution was 5 cm2.
Membrane tests used solutions containing 10 mg/L alginate
added as its sodium salt. Solution pH was adjusted to 7 using HCl or
NaOH. Solution pH was monitored using Orion™ Versa Star pH
detector (Thermo Fisher Scientific, USA). Ionic strength was
controlled by adding requisite amounts of sodium chloride stock
solutions. TMP measurements were carried out for 0.001, 0.01 and
0.1 M ionic strengths. The membrane unit was operated under a
constant flux mode (100 L.m
2hr
1, LMH). The flow was controlled
using a constant flow peristaltic pump, which was calibrated via
adjusting the rate every hour to avoid the fouling-response effect.
The inlet of the filter cartridge was connected to a pressure trans-
ducer (Omega Engineering, CT, USA) by a Y-junction and subse-
quently connected to a data logger linked to a desktop computer
which was used for data acquisition and processing. Under the filter
cartridge, a three-way pinch valve was used for outlet sampling and
permeate reservoir. In a typical filtration run, a PES membrane
placed in a standard polycarbonate holder was initially exposed to
deionized water that was fed to the unit to stabilize the membrane
and get control TMP data. After reaching a TMP value that was
stable for 30 min, the feed solution was switched to an alginate
solution. Filtration experiments were conducted using runtimes
that ranged from 2 min to above 8 h without stirring during fil-
trations. In the end of each filtration run with a predetermined
duration, the membrane was withdrawn from the cartridge, air-
dried overnight by employing the similar method in the previous
studies until the weight loss of membrane is stable, and used for
further analyses (Doumeche et al., 2007; Howe et al., 2002; Lujan-
Facundo et al., 2015). Comparison of TMP results for varying
filtration runtimes showed a good reproducibility of the measure-
ments as demonstrated by the data for a 0.001 M ionic strength
shown Fig. S1 in the Supplementary Information, SI section. Mea-
surements of dissolved organic carbon (DOC) concentrations using
a total organic carbon analyzer (Shimadzu TOC-L, Japan) in the
filtrate showed that typically less than 10% of the DOC of the feed
solution was retained on the membrane surface for the 0.1 M ionic
strength solution (data not shown).
Weight loss measurements showed that the weight of mem-
brane was stable after 30e60 min of drying. ATR FTIR spectra for
the membrane samples were acquired with an FTIR Vertex 70 in-
strument (Bruker, USA) equipped with a diamond ATR element. In a
typical test, the fouling surface of the selected membrane sample
was put against the diamond. The ATR spectra were acquired in a
400 to 4000 cm
1 wavenumber range with a 1 cm
1 wavenumber
increment. Three to five ATR spectra were acquired for randomly
 H. Guo et al. / Water Research 161 (2019) 27e34 29

selected spots on the surface of each membrane sample. The data

were averaged and numerically processed as described in the sec-
tions that follow.

3. Results and discussion

TMP measurements showed (Fig. 1) that the filtration of alginate

solution that had a 0.001 M ionic strength resulted in an initially
gradual increase of TMP values, following which, the TMP increased
abruptly thus indicating the consolidation of the fouling film and a
rapid decrease of its hydraulic permeability. Data for a 0.01 M ionic
strength show consistently lower TMP values compared with those
for 0.001 M ionic strength. TMP values observed in experiments
with 0.01 M alginate solution increased almost linearly with time
but no rapidly developing fouling was seen. TMP measurements for
alginate solution with a 0.1 M ionic strength showed little TMP
changes.
These observations are in agreement with the data of prior
research that has demonstrated the occurrence of lower membrane
fouling rates at increasing ionic strengths (Kim and Dempsey, 2013;
Lee et al., 2017; Lee and Elimelech, 2006; Listiarini et al., 2009; Mi
and Elimelech, 2010; Sioutopoulos and Karabelas, 2016; van den
Brink et al., 2009; Zhang et al., 2017). Our results are also consis-
tent with the notion that increases of the ionic strength affect
conformations of alginate macromolecules inducing their coiling,
compressing the electric double layer around these macromole-
cules and ultimately decreasing their propensity to form fouling
layers on membrane surfaces (Ang et al., 2005; Lee and Elimelech,
2006).
ATR FTIR spectra of the membranes exposed to alginate solu-
tions were compared with that of the pristine washed PES mem-
brane whose ATR spectrum is shown in Fig. S2 in the SI section. The
data shown in Fig. S2 are close to those reported in prior research
(Pihlajamaki et al., 1998). They indicate the presence of multiple
lines associated with the vibrations of the sulfonic groups in the
range of wavenumbers 400 to 1800 cm
1, as well as relatively weak
lines associated with anomeric carbon in the range of wave-
numbers 2400 to 3800 cm
1. The FTIR spectrum in the latter range
of wavelength also had weak bands associated with the presence of Fig. 2. ATR FTIR spectra of PES membrane for varying filtration run times. Ionic
residual water in the membrane. strength 0.001 M, pH 7, alginate concentration 10 mg/L.
ATR spectra of the PES membrane samples generated at varying
runtimes and ionic strengths of alginate solutions are exemplified
by those for a 0.001 M ionic strength (Fig. 2). Similar sets of ATR in Fig. S3 and Fig. S4, respectively. Examination of these spectra
FTIR spectra obtained for 0.01 and 0.1 M ionic strengths are shown indicates that they contain both the lines associated with the mo-
lecular vibrations of PES polymer and those associated with the
deposited alginate film. This is especially evident for 0.001 M and
0.01 M ionic strengths while the changes of the FTIR spectra for a
0.1 M ionic strength are subtle. This is consistent with the slight
changes of the transmembrane pressure (TMP) in these
experiments.
Further interpretation of the ATR data makes it necessary to
separate the contributions of the vibrations of the membrane
substrate and those of the deposited film. Given that in the ATR
configuration, the intensity of the vibrations of the PES substrate
becomes progressively weaker as the thickness of the film depos-
ited on the membrane surface increases, the overall FTIR spectrum
will have contributions of the substrate and the deposited film as
described by the following equation:

1
AðnÞ ¼ APES ðnÞ þ Aff ðnÞ (1)
aff ðvÞ
In Equation (1), AðnÞ, APES ðnÞ and Aff ðnÞ are the experimental ATR
Fig. 1. Development of the transmembrane pressure in filtration experiments carried spectrum of the membrane, those of the PES substrate without a
out at varying ionic strengths. Concentration of alginate 10 mg/L, pH 7. fouling film and of the fouling film per se, respectively. aff ðvÞ is a
30 H. Guo et al. / Water Research 161 (2019) 27e34

dimensionless coefficient that quantifies effects of the fouling film varying runtimes are shown in Fig. 3.
in the intensity of the ATR spectrum of the substrate. As per this Fig. 3 demonstrates that the intensity of the derivatives of the
definition, when aff ðvÞ equals to one, the fouling film is either ab- PES lines in the ATR spectra of the fouled membranes decreases
sent or it is too thin to affect the vibration lines of the substrate. with the filtration runtime but the shape of these derivatives re-
Conversely, as the fouling occurs, aff ðvÞ values increase due to the mains the same. It also shows that the decrease of the intensity of
growth of the fouling film whose presence results in a physical the derivatives becomes more prominent with the increase of the
separation between the ATR element and the substrate. For suffi- observation wavenumber. This is in accord with the assumption
ciently thick fouling layers, the value of aff ðvÞ becomes close to that the attenuation coefficient increases with the wavenumber.
infinity, and the ATR spectra will have only the contribution of the app
Thus Equation (4) shown below was used to determine the dff
fouling film. values based on the linear fitting dependence, concerning the ratio
Coefficients aff ðvÞ are expected to depend on the thickness of of the first derivatives values in the initial PES IR lines to that in the
the fouling layer and the wavenumber-dependent penetration presence of fouling layer.
depth (denoted henceforth as dp ðvÞÞ of the evanescent electro-
magnetic field generated by the ATR element. In principle, dp ðvÞ " !, ! #
values can be determined by the following equation (Boccaccio app 1 dAPES
v dAffv
dff ¼
1 (4)
et al., 2002; Mohr et al., 1991; Pihlajamaki et al., 1998): v dv dv
dAPES dAff
In the above equation, dvv and dvv are the values of the de-
1
dp ðvÞ ¼
 2 0:5 (2) rivatives of the reference PES lines in the ATR FTIR spectra of the
2pvn1 sin2 ðfÞ
nn21 PES membrane per se and that with a deposited layer, respectively,
! !
dAPES dAffv
and dapp
ff
is the best fit estimate of the slope between v
dv
= dv
In that equation, n1 and n2 are the refractive indexes of the
sample and the ATR element, respectively, f is the angle of inci- ratios and the wavenumbers of the IR radiation.
dence of light in the ATR element, and n is the wavenumber. Cal- Correlations between the wavenumber of IR radiation and ratios
culations that use equation (2) show that for the refraction indexes of the intensities of the first derivatives of the ATR spectra (Fig. 3)
of diamond and PES (Hart et al., 2002), the penetration depth of the are presented in Fig. 4. The values of derivatives used in these fig-
ATR evanescent field into the PES substrate changes from ca. 10 to ures and employed in subsequent calculations were the averages of
! !
1 mm when the wavenumber increases from 400 to 4000 cm
1.
dAPES dAffv
For more complex systems that include both the membrane the v
dv
= dv
ratios at wavenumbers located on the left and
substrate (e.g., PES layer) and a fouling film, Equation (2) can be
right sides from the zero value of the derivatives (corresponding to
processed to derive another expression that allows estimating and
the maximum of any particular PES reference line) shown in Fig. 3.
separating the contributions of the membrane per se and a fouling
Fig. 4 clearly demonstrates the existence of strongly linear cor-
film of its surface (The apparent thickness of such film estimated ! !
based on the data of ATR measurements is denoted henceforth as dAPES dAffv
app relations between v
dv
= dv
ratios and wavenumbers of the
dff ; the term “apparent” is introduced here to reflect the fact that
the ATR-based determinations of the thickness of surface layers PES reference lines. The slope of these correlations increases at
formed on membrane surfaces ultimately need to be compared longer filtration runtimes which is consistent with the expectation
with those of direct measurements of the thickness of the fouling app
that the apparent thickness of the fouling layer (dff ) increases in
films). Further derivation of the equations relevant to this reasoning
these conditions.
is given in the SI section. Suffice it to mention here that the overall app
Once dff values are established as described above, the differ-
ATR spectrum of a two-film system can be shown to be described
ential ATR FTIR spectra of the fouling film, that is those from which
by the following expression:
the contributions of the PES substrate have seen removed and only
the vibration lines of the fouling film per se remain, can be calcu-
1
AðnÞ ¼ APES ðnÞ þ Aff ðnÞ (3) lated using Equation (5):
1 þ dapp
ff
v
1
This expression signifies that for a two-film system with a Aff ðnÞ ¼ AðnÞ
app APES ðnÞ (5)
fouling film deposited on the PES substrate, the contribution of the 1 þ dff v
former will decrease with both increasing thicknesses of the fouling
Results of these calculations for the ATR spectra are shown in
film and IR light wavenumbers. Practical application of the above
Fig. 5 and Fig. S5 for experiments carried out for 0.001 and 0.01 M
expression requires that the value of dappff
be determined. As we
ionic strengths, respectively. The differential ATR spectra for the
argue the sections that follow, experimental determination of the
samples generated in experiments with a 0.1 M ionic strength
dapp
ff
values can be made based on the comparison of the change of
alginate solution were much weaker than those for 0.001 M and
indicative lines in the ATR spectrum of PES as a function of IR light
0.01 M ionic strengths, in agreement with the minimal changes of
wavenumber.
TMPs in the presence of the high salt concentration.
We demonstrate here that determination of dappff
can be done via
The data indicate the nearly complete absence of the PES vi-
the quantitation of the attenuation of the characteristic IR lines of
brations in the differential spectra albeit some minor residual
PES polymer at increasing thicknesses of the fouling layer. To
contributions of these bands, for instance nearly the characteristic
suppress the influence of the background IR absorption present in
line at 560 cm
1 can still be observed. Their presence is caused by
the vicinity of the characteristic PES lines, we calculated the de- ! !
rivatives of the ATR spectra shown in Fig. 2. Changes of the intensity dAPES dAffv
the deviations from a perfectly linear fit of the v
dv
= dv
data.
of these derivatives located in the ranges of a number of IR lines
characteristic for PES (e.g., 560 cm
1 assigned to SO2 vibration, It is pertinent to mention here that while the apparent thickness
720 cm
1, 1075 cm
1, and 1485 cm
1 assigned to AreSO2eAr and other morphological properties of the hydrogel alginate fouling
symmetric stretching vibrations (Pihlajamaki et al., 1998)) at layers may be affected by drying process, a necessary preparation
 H. Guo et al. / Water Research 161 (2019) 27e34
Fig. 3. First derivatives of the ATR FTIR spectra of PES membranes after filtration of 10 mg/L alginate solution. pH 7, ionic strength 0.001 M. Wavenumber ranges of the shown
derivatives of the ATR spectra correspond to the locations of characteristic PES vibrations at 560 cm
1 (SO2 vibration), 720 cm
1, 1075 cm
1, and 1485 cm
1 (AreSO2eAr symmetric
stretching vibration).
step for ex situ analyses, and the compression during optical contact
in ATR FTIR measurements (Wang and Waite, 2008; Zhang et al.,
2018), these steps does not affect the point that the decrease of
the substrate contribution to the overall ATR spectrum is expected
to be inversely proportional to the apparent thickness of the ATR
film.
The data shown in Fig. 5 and Fig. S5 demonstrate that the pro-
files of the vibration lines associated with the alginate fouling film
do not undergo visually prominent changes as apparent thickness
increased although some changes were observed that will be
described elsewhere. Attributions of the observed main lines are
shown in Fig. S6, which largely correspond those discussed in prior
research (Papageorgiou et al., 2010; Zhang et al., 2017). Some dif-
ferences in the positions of these lines are likely to be associated
with the structural arrangements of alginate molecules related
with their incorporation in the fouling hydrogel layer on the surface
of the PES membranes. Further interpretation of these differences
as well as that of the evolution of these lines may be useful for
gaining more microscopic-level insight into the molecular nature of
the fouling films, but this option was outside of the scope of this
paper.
The systems of the broad FTIR bands located in the wavenumber
region 2400 to 3600 cm
1 had only minor contributions of vibra-
tions attributable to anomeric carbon of the alginate. In general,
this band that had local maxima at ca. 3200 and 3350 cm
1 is
31
associated with water molecules in the hydrogel surface layer
(Belfer et al., 2000; Dias et al., 1998; Drazevic et al., 2014). The
evolution of the broad bands located in the region 2400 to
3600 cm
1 and associated with the formation of the fouling algi-
nate hydrogel can be explored further to examine the intrinsic
properties of such systems, which will be pursued elsewhere (Tam
et al., 2005; Xie et al., 2015).
Given that the intensity of the differential background-corrected
ATR FTIR spectra of the surface film and its apparent thickness in-
crease with runtime, it was deemed useful to compare the behavior
of these spectroscopic indicators with those of the ATR. These
comparisons were carried out for two parameters that included the
app
apparent thickness of the surface layer dff and the intensity of the
ATR FTIR spectra at 3200 nm
1. The latter parameter may be
deemed to be proportional, similarly to the independently devel-
oped set of dapp
ff
values, to the apparent thickness of the hydrogel
surface films.
These comparisons showed that the dapp ff
values determined for
filtration runs of different duration were reasonably well correlated
with TMP increases (Fig. 6) although the correlation for the entire
app
dataset of relevant TMP and dff values is not linear. It has been well
accepted that TMP variation might be caused by both cake layers
fouling and membrane pores blocking (Meng et al., 2018), however,
our study here aimed at correlating the overall TMP performance
32 H. Guo et al. / Water Research 161 (2019) 27e34
Fig. 4. Correlation between ratios of the intensities of the first derivatives of ATR
spectra vs. IR wavenumber. PES membrane, 10 mg/L alginate, pH 7, ionic strength 0.001
and 0.01 M.
and taking into the various operational conditions (like ionic
strength), directly with the ATR FTIR results, instead of dis-
tinguishing the intrinsic cause for the two fouling types, in order to
propose a visual response for membrane fouling. The data that
show the relationship between TMP and dapp ff
values for each used
ionic strength are presented in Fig. S7. The latter figure also dem-
onstrates the relationship between the values of ATR absorbance at
3200 cm
1 (defined mostly by water molecules present in the
hydrogel layer) and TMP increases. In this case, the correlation is
clearly strongly non-linear, albeit it deviates from linearity pri-
marily due the presence of the point that correspond to the rapid
increase of TMP in the end of filtration of 0.001 M ionic strength
alginate solution, while the value of ATR 3200 cm
1 absorbance
does not change proportionally in this case. If this point is excluded,
however, the correlation between 3200 cm
1 absorbance and TMP
values becomes practically perfectly linear (Fig. 6).
The occurrence of the linear correlation between 3200 cm
1
absorbance value of the membrane and TMP values in their range
below the onset of rapid fouling, and the pronounced deviation
from this simple relationship when the hydraulic permeability of
the surface film undergoes a steep decrease, necessitate a more
detailed analysis of the structure of the broad bands located in the
2800 to 3200 cm
1 wavenumber range. Given the presence of
several local maxima in the band (e.g., 2940, 3180, 3354 cm
1) and
the noticeable emergence of the broad structure in the 3050 to
3250 wavenumber range (Fig. 5 and Fig. S6), it is clear that this band
comprises contribution of several types of bound water, including
those that may be associated with the structural changes of the
Fig. 5. PES background-corrected ATR FTIR spectra of fouling films formed on PES
membrane at varying filtration run times. Ionic strength 0.001 M, pH 7, alginate con-
centration 10 mg/L.
fouling film, when it reaches the point of rapidly developing failure
of the hydraulic permeability of the system.
4. Conclusions
This study examined mechanisms of fouling films formation
deposited on the surface of a PES membrane during the filtration of
alginate solutions that had ionic strengths varying from 0.001 to
0.1 M. The ATR spectra that comprise the vibration bands charac-
teristic of the PES substrate and the deposited film were processed
using a quantitative approach, via which the intensity of PES vi-
bration bands in the presence of fouling was found to depend on
app
the wavenumber of IR radiation and apparent thickness (dff ) of
the deposited layer. The results presented above showed the exis-
tence of a strong correlation between TMP development and
dapp
ff
values of the fouling layers. The data demonstrate the possi-
bility of unambiguously separating the contributions of the system
components and determining both the thickness of the deposited
layers and the FTIR spectra specific to them. While more research is
needed to further explore the performance of this method applied
to other membrane materials, alternative model foulants and
membranes exposed to actual surface waters or wastewater, it can
also be argued that the presented method broadens the
 H. Guo et al. / Water Research 161 (2019) 27e34
Fig. 6. Correlations between (A) changes of the apparent thickness of the fouling layer
and (B) differential ATR FTIR absorbance at 3200 cm
1 and the development of
transmembrane pressure. PES membrane, alginate concentration 10 mg/L, pH 7, ionic
strengths 0.001, 0.01 and 0.1 M.
interpretability of ATR FTIR data generated in studies concerned
with membrane fouling, allowing for gaining new insights into its
mechanisms.
Declaration of interests
The authors declare that they have no known competing
financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
Acknowledgements
This study was partially supported by MEMPREX project funded
by the Research Council of Norway (INTPART program, grant
261607). The authors wish to thank Prof. Harsha Ratnaweera
(Norwegian University of Life Sciences) who initiated MEMPREX
project. Hongguang Guo would like to thank the Science and
Technology Projects of Sichuan Province (2018 HH0104) and In-
ternational Visiting Program for Excellent Young Scholars of SCU for
supporting his work in University of Washington.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.watres.2019.05.086.
33
References
Ang, W.S., Lee, S., Elimelech, M., 2005. Chemical and physical aspects of chemical
cleaning on organic-fouled reverse osmosis membranes. Abstr. Pap. Am. Chem.
Soc. 229. U633-U633.
Ao, L., Liu, W., Zhao, L., Wang, X., 2016. Membrane fouling in ultrafiltration of
natural water after pretreatment to different extents. J. Environ. Sci. 43,
234e243.
Bass, M., Freger, V., 2015. Facile evaluation of coating thickness on membranes using
ATR-FTIR. J. Membr. Sci. 492, 348e354.
Belfer, S., Fainchtain, R., Purinson, Y., Kedem, O., 2000. Surface characterization by
FTIR-ATR spectroscopy of polyethersulfone membranes-unmodified, modified
and protein fouled. J. Membr. Sci. 172 (1e2), 113e124.
Boccaccio, T., Bottino, A., Capannelli, G., Piaggio, P., 2002. Characterization of PVDF
membranes by vibrational spectroscopy. J. Membr. Sci. 210 (2), 315e329.
Cui, L., Chen, P., Zhang, B., Zhang, D., Li, J., Martin, F.L., Zhang, K., 2015. Interrogating
chemical variation via layer-by-layer SERS during biofouling and cleaning of
nanofiltration membranes with further investigations into cleaning efficiency.
Water Res. 87, 282e291.
Cui, L., Yao, M., Ren, B., Zhang, K.-S., 2011. Sensitive and versatile detection of the
fouling process and fouling propensity of proteins on polyvinylidene fluoride
membranes via surface-enhanced Raman spectroscopy. Anal. Chem. 83 (5),
1709e1716.
Dias, C.R., Rosa, M.J., de Pinho, M.N., 1998. Structure of water in asymmetric cel-
lulose ester membranes - an ATR-FTIR study. J. Membr. Sci. 138 (2), 259e267.
Doumeche, B., Galas, L., Vaudry, H., Di Martino, P., 2007. Membrane foulants char-
acterization in a drinking water production unit. Food Bioprod. Process. 85 (C1),
42e48.
Drazevic, E., Kosutic, K., Freger, V., 2014. Permeability and selectivity of reverse
osmosis membranes: correlation to swelling revisited. Water Res. 49, 444e452.
Dreszer, C., Wexler, A.D., Drusova, S., Overdijk, T., Zwijnenburg, A., Flemming, H.C.,
Kruithof, J.C., Vrouwenvelder, J.S., 2014. In-situ biofilm characterization in
membrane systems using Optical Coherence Tomography: formation, structure,
detachment and impact of flux change. Water Res. 67, 243e254.
Fortunato, L., Bucs, S., Linares, R.V., Cali, C., Vrouwenvelder, J.S., Leiknes, T., 2017.
Spatially-resolved in-situ quantification of biofouling using optical coherence
tomography (OCT) and 3D image analysis in a spacer filled channel. J. Membr.
Sci. 524, 673e681.
Freger, V., Gilron, J., Belfer, S., 2002. TFC polyamide membranes modified by grafting
of hydrophilic polymers: an FT-IR/AFM/TEM study. J. Membr. Sci. 209 (1),
283e292.
Gamage, N.P., Chellam, S., 2014. Mechanisms of physically irreversible fouling
during surface water microfiltration and mitigation by aluminum electro-
flotation pretreatment. Environ. Sci. Technol. 48 (2), 1148e1157.
Gao, Y., Haavisto, S., Li, W., Tang, C.Y., Salmela, J., Fane, A.G., 2014. Novel approach to
characterizing the growth of a fouling layer during membrane filtration via
optical coherence tomography. Environ. Sci. Technol. 48 (24), 14273e14281.
Hart, S.D., Maskaly, G.R., Temelkuran, B., Prideaux, P.H., Joannopoulos, J.D., Fink, Y.,
2002. External reflection from omnidirectional dielectric mirror fibers. Science
296 (5567), 510e513.
Henderson, R.K., Subhi, N., Antony, A., Khan, S.J., Murphy, K.R., Leslie, G.L., Chen, V.,
Stuetz, R.M., Le-Clech, P., 2011. Evaluation of effluent organic matter fouling in
ultrafiltration treatment using advanced organic characterisation techniques.
J. Membr. Sci. 382 (1e2), 50e59.
Herzberg, M., Kang, S., Elimelech, M., 2009. Role of extracellular polymeric sub-
stances (EPS) in biofouling of reverse osmosis membranes. Environ. Sci. Tech-
nol. 43 (12), 4393e4398.
Howe, K.J., Clark, M.M., 2002. Fouling of microfiltration and ultrafiltration mem-
branes by natural waters. Environ. Sci. Technol. 36 (16), 3571e3576.
Howe, K.J., Ishida, K.P., Clark, M.M., 2002. Use of ATR/FTIR spectrometry to study
fouling of microfiltration membranes by natural waters. Desalination 147,
251e255.
Huang, B.C., Guan, Y.F., Chen, W., Yu, H.Q., 2017. Membrane fouling characteristics
and mitigation in a coagulation-assisted microfiltration process for municipal
wastewater pretreatment. Water Res. 123, 216e223.
Jing, Y., Guo, L., Chaplin, B.P., 2016. Electrochemical impedance spectroscopy study
of membrane fouling and electrochemical regeneration at a sub-stoichiometric
TiO2 reactive electrochemical membrane. J. Membr. Sci. 510, 510e523.
Kane, S.R., Ashby, P.D., Pruitt, L.A., 2009. ATR-FTIR as a thickness measurement
technique for hydrated polymer-on-polymer coatings. J. Biomed. Mater. Res. B
Appl. Biomater. 91B (2), 613e620.
Kim, H.C., Dempsey, B.A., 2013. Membrane fouling due to alginate, SMP, EfOM,
humic acid, and NOM. J. Membr. Sci. 428, 190e197.
Kogler, M., Zhang, B., Cui, L., Shi, Y., Yliperttula, M., Laaksonen, T., Viitala, T.,
Zhang, K., 2016. Real-time Raman based approach for identification of
biofouling. Sensor. Actuator. B Chem. 230, 411e421.
Le-Clech, P., Marselina, Y., Ye, Y., Stuetz, R.A., Chen, V., 2007. Visualisation of poly-
saccharide fouling on microporous membrane using different characterisation
techniques. J. Membr. Sci. 290 (1e2), 36e45.
Lee, J., Jeong, S., Ye, Y., Chen, V., Vigneswaran, S., Leiknes, T., Liu, Z.W., 2017. Protein
fouling in carbon nanotubes enhanced ultrafiltration membrane: fouling
mechanism as a function of pH and ionic strength. Separ. Purif. Technol. 176,
323e334.
Lee, N.H., Amy, G., Croue, J.P., Buisson, H., 2004. Identification and understanding of
34 H. Guo et al. / Water Research 161 (2019) 27e34
fouling in low-pressure membrane (MF/UF) filtration by natural organic matter
(NOM). Water Res. 38 (20), 4511e4523.
Lee, S., Elimelech, M., 2006. Relating organic fouling of reverse osmosis membranes
to intermolecular adhesion forces. Environ. Sci. Technol. 40 (3), 980e987.
Li, J.X., Hallbauer, D.K., Sanderson, R.D., 2003. Direct monitoring of membrane
fouling and cleaning during ultrafiltration using a non-invasive ultrasonic
technique. J. Membr. Sci. 215 (1e2), 33e52.
Li, J.X., Sanderson, R.D., Chai, G.Y., Hallbauer, D.K., 2005. Development of an ultra-
sonic technique for in situ investigating the properties of deposited protein
during crossflow ultrafiltration. J. Colloid Interface Sci. 284 (1), 228e238.
Listiarini, K., Chun, W., Sun, D.D., Leckie, J.O., 2009. Fouling mechanism and resis-
tance analyses of systems containing sodium alginate, calcium, alum and their
combination in dead-end fouling of nanofiltration membranes. J. Membr. Sci.
344 (1e2), 244e251.
Lu, X.L., Castrillon, S.R.V., Shaffer, D.L., Ma, J., Elimelech, M., 2013. In situ surface
chemical modification of thin-film composite forward osmosis membranes for
enhanced organic fouling resistance. Environ. Sci. Technol. 47 (21),
12219e12228.
Lujan-Facundo, M.J., Mendoza-Roca, J.A., Cuartas-Uribe, B., Alvarez-Blanco, S., 2015.
Evaluation of cleaning efficiency of ultrafiltration membranes fouled by BSA
using FTIR-ATR as a tool. J. Food Eng. 163, 1e8.
Marselina, Y., Lifia, Le-Clech, P., Stuetz, R.M., Chen, V., 2009. Characterisation of
membrane fouling deposition and removal by direct observation technique.
J. Membr. Sci. 341 (1e2), 163e171.
Mendret, J., Guigui, C., Schmitz, P., Cabassud, C., Duru, P., 2007. An optical method
for in situ characterization of fouling during filtration. AIChE J. 53 (9),
2265e2274.
Meng, F., Liao, B., Liang, S., Yang, F., Zhang, H., Song, L., 2010. Morphological visu-
alization, componential characterization and microbiological identification of
membrane fouling in membrane bioreactors (MBRs). J. Membr. Sci. 361 (1e2),
1e14.
Meng, S.J., Fan, W.H., Li, X.M., Liu, Y., Liang, D.W., Liu, X.X., 2018. Intermolecular
interactions of polysaccharides in membrane fouling during microfiltration.
Water Res. 143, 38e46.
Mi, B.X., Elimelech, M., 2010. Organic fouling of forward osmosis membranes:
fouling reversibility and cleaning without chemical reagents. J. Membr. Sci. 348
(1e2), 337e345.
Mohr, J.M., Paul, D.R., Taru, Y., Mlsna, T.E., Lagow, R.J., 1991. Surface fluorination of
composite membranes .2. Characterization of the fluorinated layer. J. Membr.
Sci. 55 (1e2), 149e171.
Mores, W.D., Davis, R.H., 2001. Direct visual observation of yeast deposition and
removal during microfiltration. J. Membr. Sci. 189 (2), 217e230.
Papageorgiou, S.K., Kouvelos, E.P., Favvas, E.P., Sapalidis, A.A., Romanos, G.E.,
Katsaros, F.K., 2010. Metal-carboxylate interactions in metal-alginate complexes
studied with FTIR spectroscopy. Carbohydr. Res. 345 (4), 469e473.
Pawlowski, S., Galinha, C.F., Crespo, J.G., Velizarov, S., 2016. 2D fluorescence spec-
troscopy for monitoring ion-exchange membrane based technologies - reverse
electrodialysis (RED). Water Res. 88, 184e198.
Pihlajamaki, A., Vaisanen, P., Nystrom, M., 1998. Characterization of clean and fouled
polymeric ultrafiltration membranes by Fourier transform IR spectroscopy-
attenuated total reflection. Colloid. Surf. Physicochem. Eng. Asp. 138 (2e3),
323e333.
Rao, A.P., Joshi, S.V., Trivedi, J.J., Devmurari, C.V., Shah, V.J., 2003. Structure-per-
formance correlation of polyamide thin film composite membranes: effect of
coating conditions on film formation. J. Membr. Sci. 211 (1), 13e24.
Shannon, M.A., Bohn, P.W., Elimelech, M., Georgiadis, J.G., Marinas, B.J., Mayes, A.M.,
2008. Science and technology for water purification in the coming decades.
Nature 452 (7185), 301e310.
Sim, L.N., Suwarno, S.R., Lee, D.Y.S., Cornelissen, E.R., Fane, A.G., Chong, T.H., 2018.
Online monitoring of transparent exopolymer particles (TEP) by a novel
membrane-based spectrophotometric method. Chemosphere 220, 107e115.
Sioutopoulos, D.C., Karabelas, A.J., 2012. Correlation of organic fouling resistances in
RO and UF membrane filtration under constant flux and constant pressure.
J. Membr. Sci. 407, 34e46.
Sioutopoulos, D.C., Karabelas, A.J., 2016. Evolution of organic gel fouling resistance
in constant pressure and constant flux dead-end ultrafiltration: differences and
similarities. J. Membr. Sci. 511, 265e277.
Tam, S.K., Dusseault, J., Polizu, S., Menard, M., Halle, J.P., Yahia, L., 2005. Physico-
chemical model of alginate-poly-L-lysine microcapsules defined at the micro-
metric/nanometric scale using ATR-FTIR, XPS, and ToF-SIMS. Biomaterials 26
(34), 6950e6961.
Thibault, Y., McEvoy, J.G., Mortazavi, S., Smith, D., Doiron, A., 2017. Characterization
of fouling processes in ceramic membranes used for the recovery and recycle of
oil sands produced water. J. Membr. Sci. 540, 307e320.
van den Brink, P., Zwijnenburg, A., Smith, G., Temmink, H., van Loosdrecht, M., 2009.
Effect of free calcium concentration and ionic strength on alginate fouling in
cross-flow membrane filtration. J. Membr. Sci. 345 (1e2), 207e216.
Wang, L.F., He, D.Q., Chen, W., Yu, H.Q., 2015. Probing the roles of Ca2þ and Mg2þ in
humic acids-induced ultrafitltration membrane fouling using an integrated
approach. Water Res. 81, 325e332.
Wang, X.M., Waite, T.D., 2008. Gel layer formation and hollow fiber membrane
filterability of polysaccharide dispersions. J. Membr. Sci. 322 (1), 204e213.
Xie, M., Lee, J., Nghiem, L.D., Elimelech, M., 2015. Role of pressure in organic fouling
in forward osmosis and reverse osmosis. J. Membr. Sci. 493, 748e754.
Xie, M., Luo, W.H., Gray, S.R., 2017. Synchrotron Fourier transform infrared mapping:
a novel approach for membrane fouling characterization. Water Res. 111,
375e381.
Yamamura, H., Okimoto, K., Kimura, K., Watanabe, Y., 2014. Hydrophilic fraction of
natural organic matter causing irreversible fouling of microfiltration and ul-
trafiltration membranes. Water Res. 54, 123e136.
Yao, S., Costello, M., Fane, A.G., Pope, J.M., 1995. Noninvasive observation of flow
profiles and polarization layers in hollow-fiber membrane filtration modules
using Nmr microimaging. J. Membr. Sci. 99 (3), 207e216.
Yu, H., Wu, Z., Zhang, X., Qu, F., Wang, P., Liang, H., 2019. Characterization of fluo-
rescence foulants on ultrafiltration membrane FA using front-face excitation-
emission matrix (FF-EEM) spectroscopy: fouling evolution and mechanism
analysis. Water Res. 148, 546e555.
Zhang, M.J., Hong, H.C., Lin, H.J., Shen, L.G., Yu, H.Y., Ma, G.C., Chen, J.R., Liao, B.Q.,
2018. Mechanistic insights into alginate fouling caused by calcium ions based
on terahertz time-domain spectra analyses and DFT calculations. Water Res.
129, 337e346.
Zhang, M.J., Lin, H.J., Shen, L.G., Liao, B.Q., Wu, X.L., Li, R.J., 2017. Effect of calcium
ions on fouling properties of alginate solution and its mechanisms. J. Membr.
Sci. 525, 320e329.