membranes
Review
Microscopy and Spectroscopy Techniques for
Characterization of Polymeric Membranes
Yousef Alqaheem *
and Abdulaziz A. Alomair
Petroleum Research Center, Kuwait Institute for Scientific Research, Safat 13109, Kuwait; aomair@kisr.edu.kw
* Correspondence: yqhaeem@kisr.edu.kw; Tel.: +965-2495-6929; Fax: +965-239-0445
Received: 19 December 2019; Accepted: 24 January 2020; Published: 24 February 2020
Abstract: Polymeric membrane is a proven technology for water purification and wastewater treatment.
The membrane is also commercialized for gas separation, mainly for carbon dioxide removal and
hydrogen recovery. Characterization techniques are excellent tools for exploring the membrane
structure and the chemical properties. This information can be then optimized to improve the
membrane for better performance. In this paper, characterization techniques for studying the physical
structure such as scanning electron microscopy (SEM), transmission electron microscopy (TEM), and
atomic force microscopy (AFM) are discussed. Techniques for investigating the crystal structure such
as X-ray diffraction (XRD), small-angle X-ray scattering (SAXS), and wide-angle X-ray scattering
(WAXS) are also considered. Other tools for determining the functional groups such Fourier transform
infrared spectroscopy (FTIR), Raman spectroscopy, and nuclear magnetic resonance (NMR) are
reviewed. Methods for determining the elemental composition such as energy-dispersion X-ray
spectroscopy (EDS), X-ray fluorescent (XRF), and X-ray photoelectron spectroscopy (XPS) are explored.
The paper also gives general guidelines for sample preparation and data interpretation for each
characterization technique.
Keywords: gas-separation membrane; membrane characterization; surface analysis; crystal structure;
functional groups; elemental composition
1. Introduction
Characterization is an important field in material science. It refers to determining the physical and
chemical properties of the material for a better understanding. This high level of knowledge can be then
engineered to optimize the material performance. In membrane science, characterization techniques
are widely used to confirm the quality and the purity of the prepared membranes. Furthermore,
characterization techniques are powerful tools for interpreting the membrane performance and studying
membrane degradation. Moreover, intensive research is carried out on preparing membranes from
different materials to overcome the limitation in polymeric membranes. This limitation is due to the
tradeoff between permeability and selectivity. This means that the produced gas can either have high
flowrate or high purity [1]. This limitation was overcome by copolymerization, functionalization, or
adding fillers [2]. Characterization techniques give valuable information on the interaction between the
polymer and the fillers.
This paper reviews common spectroscopy techniques for the characterization of polymeric
membranes. The techniques usually study the morphology data, crystal structure, functional groups, and
chemical composition. For membrane morphology, scanning electron microscopy (SEM), transmission
electron microscopy (TEM), and atomic force microscopy (AFM) are widely implemented. For studying
the crystal structure and its shape and size, X-ray diffraction (XRD) is classically applied. Techniques
such as small-angle X-ray scattering (SAXS) and wide-angle X-ray scattering (WAXS) can provide
crystallography data with added information about the particle size and pore size distribution.
Membranes 2020, 10, 33; doi:10.3390/membranes10020033
www.mdpi.com/journal/membranes
Membranes 2020, 10, 33
Membranes 2020, 10, 33
2 of 37
2 of 36
and pore size distribution. Fourier‐transform infrared (FTIR) spectroscopy, Raman spectroscopy, and
nuclear magnetic resonance spectroscopy (NMR) are generally utilized for determining the
Fourier-transform
infrared
(FTIR) spectroscopy,
spectroscopy,
and nuclear magnetic
resonance
functional
groups. For
measuring
the chemical Raman
composition,
energy‐dispersion
X‐ray (EDS),
X‐ray
spectroscopy
(NMR)
are
generally
utilized
for
determining
the
functional
groups.
For
measuring
fluorescent (XRF), and X‐ray photoelectron spectroscopy (XPS) are commonly employed. In the
this
chemical
X-rayof(EDS),
X-ray sample
fluorescent
(XRF), andand
X-ray
photoelectron
paper,
eachcomposition,
technology energy-dispersion
is discussed in terms
operation,
preparation,
limitations.
The
spectroscopy
are guidelines
commonly for
employed.
In this
paper,
technology
is discussed in terms of
paper
also gives(XPS)
general
interpreting
the
data each
for polymeric
membranes.
operation, sample preparation, and limitations. The paper also gives general guidelines for interpreting
the data for polymeric
2. Morphology
Analysismembranes.
Morphology
Analysis
2.1.2.Scanning
Electron
Microscopy
2.1.Scanning
Scanning electron
Electron Microscopy
microscopy (SEM) is one of the fundamental techniques for membrane
characterization
as
it
gives
the morphology
topography
data of thetechniques
prepared membranes
[3].
Scanning electron microscopy
(SEM) and
is one
of the fundamental
for membrane
Furthermore,
SEM as
canit be
used
determine the
pore
size in thedata
caseofofthe
a porous
membrane
[4]. [3].
For
characterization
gives
thetomorphology
and
topography
prepared
membranes
a dense
membrane,
SEM
is
used
to
measure
the
thickness
of
the
selective
layer
for
calculation
of
Furthermore, SEM can be used to determine the pore size in the case of a porous membrane [4]. Forthe
a
permeability
in Barrer
[5].is used to measure the thickness of the selective layer for calculation of the
dense membrane,
SEM
SEM worksinsimilarly
permeability
Barrer [5].to an optical microscope but in SEM, electrons are used instead of light [6].
A beamSEM
of concentrated
electrons
will bemicroscope
sent to the sample
and electrons
this will cause
a release
works similarly
to an optical
but in SEM,
are used
insteadofofsecondary
light [6].
electrons
[7].
After
that,
the
electrons
are
collected
by
a
detector
and
then
analyzed
to
form
an
image
A beam of concentrated electrons will be sent to the sample and this will cause a release of secondary
as electrons
demonstrated
in Figure
1. Compared
to collected
a light microscope,
which
a maximum
magnification
[7]. After
that, the
electrons are
by a detector
and gives
then analyzed
to form
an image
of as
1500,
the magnification
can reach
to 1microscope,
million [8].which gives a maximum magnification
demonstrated
in Figurein1.SEM
Compared
to aup
light
of 1500, the magnification in SEM can reach up to 1 million [8].
2.1.1. Sample Preparation
2.1.1. Sample Preparation
Most of the membrane samples require preparation before using SEM. First, the samples should
theexceed
membrane
samples
before using
SEM.cutting
First, the
samples
should
be
be solidMost
and of
not
an area
of 10 require
by 4 cmpreparation
[9]. If the sample
is larger,
it into
smaller
pieces
solid required.
and not exceed
an area
of 10 by 4should
cm [9]. be
If the
sample isconductive
larger, cutting
into smaller
then
is then
Second,
the samples
electrically
in it
order
to havepieces
good is
images
required.
Second,
the
samples
should
be
electrically
conductive
in
order
to
have
good
images
[10].
[10]. Due to the weak conductivity in most polymers, the membrane surface needs to be coated Due
with
to the weak conductivity in most polymers, the membrane surface needs to be coated with a conductive
a conductive material such as gold or palladium [11]. Usually, the sample will be coated with a
material such as gold or palladium [11]. Usually, the sample will be coated with a sputter-coater to
sputter‐coater to form a thin film of gold of 10 nm [12]. In addition to improving the quality of the
form a thin film of gold of 10 nm [12]. In addition to improving the quality of the analyzed images, the
analyzed images, the coating also reduces the thermal damage on the sample due to charge build‐up
coating also reduces the thermal damage on the sample due to charge build-up [13].
[13].
FigureFigure
1. Components
of SEM
device
magnification
membranes
1. Components
of SEM
devicefor
forsurface
surface magnification
of of
membranes
[14]. [14].
Membranes 2020, 10, 33
3 of 37
SEM can measure the thickness of the dense layer too by cutting the membrane and analyzing
the cross‐section area. Before coating the sample, it is advised to insert the sample in liquid nitrogen
Membranes 2020, 10, 33
3 of 36
for a few minutes then cut it by a sharp blade. This will result in a better image and eliminates surface
bend [15].
SEM can measure the thickness of the dense layer too by cutting the membrane and analyzing the
cross-section
area. Before coating the sample, it is advised to insert the sample in liquid nitrogen for
2.1.2. Data Interpretation
a few minutes then cut it by a sharp blade. This will result in a better image and eliminates surface
Data[15].
in SEM is given in the form of a surface image. The membrane surface is expected to be
bend
free from defects such as large holes and cracks. For a porous membrane, the surface is expected to
Data Interpretation
have2.1.2.
a uniform
structure of small holes. SEM can be used to calculate the average pore size by
processing
the
a software
such
MATLAB
function
“regionprops”
shown
Data
in image
SEM is by
given
in the form
of a as
surface
image.and
The the
membrane
surface
is expected as
to be
free in
Figure
2.defects such as large holes and cracks. For a porous membrane, the surface is expected to have a
from
SEM
is structure
an excellent
toolholes.
for analyzing
composite
membranes
made from
twobyorprocessing
more polymers.
uniform
of small
SEM can be
used to calculate
the average
pore size
the
imagea by
a software
suchselective
as MATLAB
and the
function “regionprops”
assupport
shown infor
Figure
2. mechanical
Usually,
thin
layer of the
polymer
is deposited
over a porous
better
SEM
is an
excellent tool
analyzing composite
membranes
made fromabout
two orthe
more
polymers.
properties
[16].
Examining
thefor
cross‐section
surface can
give information
structure
of the
a thin
layer
of the selective
is deposited
over of
a porous
support membrane
for better mechanical
twoUsually,
polymers.
For
example,
Figurepolymer
3 shows
SEM images
a composite
made from
properties [16]. Examining the cross-section
surface
can give information
about (MFFK‐1).
the structureSEM
of thewas
membrane
poly(1‐trimethylsilyl‐1‐propyne)
(PTMSP) over
microfiltration
two polymers. For example, Figure 3 shows SEM images of a composite membrane made from
furtherly used to calculate the thickness of the PTMSP layer, which has an average value of 1.75 m
poly(1-trimethylsilyl-1-propyne) (PTMSP) over microfiltration membrane (MFFK-1). SEM was furtherly
[17].
used to calculate the thickness of the PTMSP layer, which has an average value of 1.75 µm [17].
For polymeric gas separation membranes, the transport mechanism is mainly based on the
For polymeric gas separation membranes, the transport mechanism is mainly based on the
solution‐diffusion
model, which states that the gas dissolves through the polymer and then diffuses.
solution-diffusion model, which states that the gas dissolves through the polymer and then diffuses.
In these
membranes,
the
dense and
andfree
freefrom
fromfully
fully
penetrating
holes.
In these membranes,
thesurface
surfaceshould
should be
be fully
fully dense
penetrating
holes.
However,
bybyanalyzing
cross‐section
surface,
layers
of and
dense
andstructures
porous structures
However,
analyzing the
the cross-section
surface,
twotwo
layers
of dense
porous
are usually are
usually
observed
[5].isThis
is because
the membrane
preparation,
solvent
used
and then
observed
[5]. This
because
during during
the membrane
preparation,
a solvent isaused
and is
then
removed
removed
causing
the formation
ofstructure
a porous
as shown
Figurethe
4. gas
To permeability
calculate the
causing
the formation
of a porous
as structure
shown in Figure
4. To in
calculate
in gas
Barrer, the in
thickness
thethickness
dense layer
needed
and
this is
can
be measured
by SEM
as demonstrated
permeability
Barrer,ofthe
ofisthe
dense
layer
needed
and this
can be
measured byinSEM
Figure
4.
as demonstrated in Figure 4.
Figure 2.2.
Image
processing
of SEMofdata
to calculate
pore sizethe
profile
of asize
silicon
film using
[18].
Figure
Image
processing
SEM
data to the
calculate
pore
profile
of a MATLAB
silicon film
using MATLAB [18].
Membranes2020,
2020,10,
10,33
33
Membranes
4 4ofof37
36
Membranes 2020, 10, 33
4 of 37
Figure3. 3.
SEM
image
of:
(a) Non‐modified
microfiltration
membrane
(MFFK‐1),
(b)
Figure
SEM
image
of: (a)of:
Non-modified
microfiltration
membranemembrane
(MFFK-1),
(b)
composite
poly(1Figure 3.
SEM
image
(a) Non‐modified
microfiltration
(MFFK‐1),
(b)
composite
poly(1‐trimethylsilyl‐1‐propyne)
(PTMSPM)
with
MFFK‐1 support [17].
trimethylsilyl-1-propyne)
(PTMSPM) with MFFK-1
supportwith
[17].
composite poly(1‐trimethylsilyl‐1‐propyne)
(PTMSPM)
MFFK‐1 support [17].
Figure 4. SEM image of the cross‐section polyetherimide membrane showing porous and
Figure4.4.SEM
SEMimage
imageofofthe
thecross-section
cross‐section
polyetherimide
membrane
showing
porous
and
Figure
polyetherimide
membrane
showing
porous
and dense
dense structures
(left)
and
of thethe
dense
layer
(right)[19].
dense
structures
andmeasurement
measurement
dense
layer
(right)[19].
structures
(left) and(left)
measurement
of the denseoflayer
(right)
[19].
Copolymerization produces aapolymer
made from two or more
monomers
of different
materials.
Copolymerization
or more
more
monomers
ofdifferent
different
materials.
Copolymerizationproduces
produces a polymer
polymer made from two or
monomers
of
materials.
This can be beneficial in improving the polymer characteristics such as mechanical properties,
This
beneficial
in improving
the polymer
characteristics
such as
mechanical
properties,
thermal
Thiscan
canbebe
beneficial
in improving
the polymer
characteristics
such
as mechanical
properties,
thermal stability, or gas transport [20]. Porous particles made from polystyrene (PS) and polyacrylic
stability,
gas transport
Porous
particles
frommade
polystyrene
(PS) and polyacrylic
acid (PAA)
thermal or
stability,
or gas [20].
transport
[20].
Porousmade
particles
from polystyrene
(PS) and polyacrylic
acid (PAA) were analyzed by SEM to measure the particle size [21]. Figure 5 shows the particles with
were
by analyzed
SEM to measure
particle size
Figure
shows
the particles
with
size ranging
acid analyzed
(PAA) were
by SEMthe
to measure
the [21].
particle
size 5[21].
Figure
5 shows the
particles
with
size ranging from 1 to 7 m. The above reported studies on characterization of polymeric membranes
from
1 to 7 µm.
The
above
studies
on characterization
of polymeric
using SEM
size ranging
from
1 to
7 m.reported
The above
reported
studies on characterization
ofmembranes
polymeric membranes
using SEM are summarized in Table 1.
are
summarized
in33Table 1. in Table 1.
using
SEM2020,
are 10,
summarized
Membranes
5 of 37
2.1.3. Limitations
2.1.3. Limitations
In SEM, the sample should be in the solid form and free from moistures [22]. This is because
SEM, the
sample
should
be in
solid
form evaporation.
and free from
moistures the
[22].
This isneed
because
SEM In
operates
under
vacuum
and this
willthe
cause
sample
Furthermore,
samples
SEM
under
vacuumand
andifthis
causeissample
evaporation.
to be operates
electrically
conductive
not,will
coating
necessary.
It should Furthermore,
be noted that the
for samples
pore‐sizeneed
to
be electrically
conductive
if not,having
coating
is necessary.
be noted
that forforpore‐size
measurement,
SEM
can be usedand
for pores
a diameter
in tensItofshould
micrometers.
However,
gas
measurement,
SEM canSEM
be used
forsuitable
pores having
a diameter
in tens of micrometers.
However,
separation membranes,
is not
for pore
size measurements
because the pore
size is for
in gas
the fractionsmembranes,
of nanometer
or units
Angstrom
measurements
thin
films
of is in
separation
SEM
is notofsuitable
for [23].
poreFurthermore,
size measurements
becauseofthe
pore
size
thickness
less of
than
10 nm is not
recommended
due to
the Furthermore,
lower resolution
of SEM [24]. In
the
fractions
nanometer
or units
of Angstrom
[23].
measurements
of this
thincase,
films of
high‐resolution
techniques
transmission due
electron
are required,
will
becase,
thickness
less than
10 nm issuch
not as
recommended
to themicroscopy
lower resolution
of SEMwhich
[24]. In
this
discussed
in
the
following
section.
high‐resolution techniques such as transmission electron microscopy are required, which will be
discussed in the following section.
Figure
5. Particle
measurements
of polysulfone
(PS)
and polyacrylic
acidco-polymer
(PAA) co‐ by
Figure
5. Particle
size size
measurements
of polysulfone
(PS) and
polyacrylic
acid (PAA)
polymer
by
SEM
[21].
SEM [21].
Table 1. Performed studies using SEM for characterization of polymeric membranes.
Study
Membrane
Dense surface
Polyetherimide
Pore size
Methodology
Cut samples in liquid nitrogen.
Gold coat samples.
Cut samples in liquid nitrogen.
Conclusion
No pores thus a dense
membrane.
Average pore size of 30
Ref.
[15]
Membranes 2020, 10, 33
5 of 36
2.1.3. Limitations
In SEM, the sample should be in the solid form and free from moistures [22]. This is because
SEM operates under vacuum and this will cause sample evaporation. Furthermore, the samples need
to be electrically conductive and if not, coating is necessary. It should be noted that for pore-size
measurement, SEM can be used for pores having a diameter in tens of micrometers. However, for
gas separation membranes, SEM is not suitable for pore size measurements because the pore size is
in the fractions of nanometer or units of Angstrom [23]. Furthermore, measurements of thin films
of thickness less than 10 nm is not recommended due to the lower resolution of SEM [24]. In this
case, high-resolution techniques such as transmission electron microscopy are required, which will be
discussed in the following section.
Table 1. Performed studies using SEM for characterization of polymeric membranes.
Study
Membrane
Methodology
Dense surface
Polyetherimide
Cut samples in liquid nitrogen.
Gold coat samples.
Pore size
measurements
Polyethersulfone
Cut samples in liquid nitrogen.
Used an imaging software to
determine the pore size.
Membrane
thickness
poly(1-trimethylsilyl-1-propyne)
(PTMSP)
Particle size
measurements
Copolymer of polysulfone (PS)
and polyacrylic acid (PAA)
Immersed samples in isopropanol.
Cut samples in liquid nitrogen.
Coated samples with gold.
Freeze-dried samples.
Sputter-coated samples
with platinum.
Conclusion
Ref.
No pores thus a
dense membrane.
[15]
Average pore size
of 30 nm.
[25]
Average PTMSP
thickness of 1.5 µm.
[17]
Particle size
ranging from
1 to 7 µm.
[21]
2.2. Transmission Electron Microscopy
Transmission electron microscopy (TEM) is another technique for observing the membrane
surface. Compared to SEM, the magnification in TEM can reach 50 million, making it more suitable for
measurements in the nanometer scale [26]. In TEM, an electron gun is used, as a source of electrons,
and pointed to the sample. The electrons pass through the sample and collected thereafter. These
transmitted electrons will form an image in a fluorescent screen giving more details about the internal
structure of the membrane [27]. The components of the TEM instrument are shown in Figure 6.
2.2.1. Sample Preparation
Unlike SEM, TEM needs more preparation as the sample needs to be thin and small; usually less
than 2.5 mm in diameter with thickness not exceeding 150 µm [27,28]. Before cutting the sample, most
of the polymers are soft and it is suggested to harden them by immersing them in liquid nitrogen.
To facilitate ion exchange and improve image contrast, the membranes are stained using a liquid
containing positive stains that deposit electrons on the membrane surface such as sodium oxide and
uranyl acetate [29]. After that, the sample is washed with deionized water and then dried. The sample
is thereafter mounted on a TEM grid made of a metallic mesh and ready for analysis.
2.2.2. Data Interpretation
TEM is widely used over SEM especially for analyzing mixed-matrix membranes made by adding
nanoparticles to the polymer. The nanoparticles are expected to be uniformly distributed through
the membrane [2]. However, agglomeration of the particles may occur due to molecular electrostatic
forces [30]. TEM is an excellent instrument for studying the agglomeration of the particles. For example,
Figure 7 shows the nanoparticles of titanium oxides deposited on polybenzimidazole (PBI) membrane
for water vapor/gas separation. TEM revealed that some of the nanoparticles agglomerated but it is
within the acceptable range [31].
Membranes 2020, 10, 33
6 of 37
Membranes 2020, 10, 33
6 of 37
particles.
For example, Figure 7 shows the nanoparticles of titanium oxides deposited
on
polybenzimidazole
(PBI)
membrane
for
water
vapor/gas
separation.
TEM
revealed
that
some
of
the
Membranes
33
particles.2020,
For10, example,
Figure 7 shows the nanoparticles of titanium oxides deposited6 ofon36
nanoparticles
agglomerated
but
it is within
the vapor/gas
acceptable
range [31].
polybenzimidazole (PBI) membrane
for water
separation.
TEM revealed that some of the
nanoparticles agglomerated but it is within the acceptable range [31].
Figure 6. Components of transmission electron microscopy (TEM) instrument [32].
6. Components
of transmissionelectron
electron microscopy
(TEM)
instrument
[32].
Figure 6.Figure
Components
of transmission
microscopy
(TEM)
instrument
[32].
Figure7.7.TEM
TEM
images
of 2TiO
2 nanoparticles
on polybenzimidazole
(PBI) membrane
with
Figure
images
of TiO
nanoparticles
on polybenzimidazole
(PBI) membrane
with the addition
the
addition
of
TiO
2
nanoparticles:
(a)
At
a
magnification
of
500
nm,
(b)
at
a
magnification
of
TiO
nanoparticles:
(a)
At
a
magnification
of
500
nm,
(b)
at
a
magnification
of
50
nm
[31].
2
of 50 nm [31].
Thin-film composite (TFC) membranes are made mainly by depositing a selective layer over
a support.
Thecomposite
film is usually
nanometers
TEM
can be
to measure
the film
thickness.
Thin‐film
(TFC)inmembranes
areand
made
mainly
byused
depositing
a selective
layer
over a
Figure
8 shows
aimages
film
of polyacrylic
acid (PAA)
deposited
on polysulfone
Based
on TEM,
the
Figure
7.
TEM
of in
TiO
2 nanoparticles
on
polybenzimidazole
(PBI)
membrane
with
support.
The
film
is usually
nanometers
and TEM
can
be used
to measure (PSf).
the
film
thickness.
Figure
film
thickness
was
about
20
nm
[29].
The
discussed
characterization
studies
are
summarized
in
Table
8 shows
a film
of polyacrylic
acid (PAA)
deposited
on polysulfone
(PSf).
the film2.
the
addition
of TiO
2 nanoparticles:
(a) At
a magnification
of 500
nm,Based
(b) atona TEM,
magnification
was about 20 nm [29]. The discussed characterization studies are summarized in Table 2.
ofthickness
50 nm [31].
Thin‐film composite (TFC) membranes are made mainly by depositing a selective layer over a
support. The film is usually in nanometers and TEM can be used to measure the film thickness. Figure
8 shows a film of polyacrylic acid (PAA) deposited on polysulfone (PSf). Based on TEM, the film
thickness was about 20 nm [29]. The discussed characterization studies are summarized in Table 2.
Membranes 2020, 10, 33
7 of 36
Membranes 2020, 10, 33
7 of 37
Figure 8.Figure
TEM8.image
of a nano‐film
of of
grafted
polyacrylic
acid
(PAA)
over polysulfone
TEM image
of a nano-film
grafted on
on polyacrylic
acid
(PAA)
over polysulfone
(PSf) [29].(PSf)
[29].
Table 2. Different studies performed by TEM for membrane characterization.
Table
Study
2. Different
studies performed by
TEM for membrane characterization.
Membrane
Methodology
Conclusion
Ref.
Agglomeration of
Study
particles in
membranes of
Agglomeration
Polybenzimidazole
MembraneCut samples in liquid nitrogen.
Methodology
(PBI) with titanium
Gold
coated samples.
Polybenzimidazole
(PBI)
oxides particles
Cut samples in liquid nitrogen.
Low agglomeration
Conclusion
Ref.
[31]
effectLow
particles in membranes
Stained samples
sodium
hydroxide.
Goldbycoated
samples.
Polyacrylicparticles
acid
Immersed samples in uranyl nitrate for 15 min.
(PAA)/polysulfone Washed samples
withsamples
distilledby
water.
Stained
sodium
(PSf)
Cut sampleshydroxide.
to 60–100 nm in thickness
by ultramicrotome.
PAA film of 20 nm.
Film thickness
with titanium oxides
agglomeration
effect
[31]
[29]
Immersed samples in uranyl nitrate
PAA film of
for 15 min.
[29]
20 nm.
Washed samples with distilled water.
2.2.3. Limitations
Cut samples to 60‐100 nm in thickness
Unlike SEM, TEM requires special preparation.
The samples need to be generally no more than
by ultramicrotome.
Film thickness
Polyacrylic acid
(PAA)/polysulfone (PSf)
2.5 mm in diameter. Also, the thickness should be no more than 100 µm, otherwise reduction in
thickness
will be required. Furthermore, TEM gives the image in 2D format and this does not show the
2.2.3.
Limitations
topography data. Moreover, wet or volatile samples cannot be used in TEM due to the high vacuum,
Unlike
SEM,
TEM evaporation
requires special
preparation.
The
samples
need to be generally no more than
which
causes
sample
and this
may damage
the
instrument.
2.5 mm in diameter. Also, the thickness should be no more than 100 m, otherwise reduction in
2.3. Atomic
Microscopy
thickness
will Force
be required.
Furthermore, TEM gives the image in 2D format and this does not show
the topography
data.
Moreover,
wetisor
volatile samples
cannottechnique
be used for
in surface
TEM due
to the high
Atomic force
microscopy
(AFM)
a high-resolution
microscopy
examination.
Figurewhich
9 shows
the scale
difference
between
SEM,
and AFM,
it is clear that AFM has a
vacuum,
causes
sample
evaporation
and
thisTEM,
may damage
theand
instrument.
similar magnification power compared to TEM. Yet, AFM can provide 3D images of the surface that
provide
surface
topography along with surface roughness [33].
2.3. can
Atomic
Force
Microscopy
Unlike SEM and TEM, AFM works by measuring the force between a probe and the sample.
Atomic force microscopy (AFM) is a high‐resolution microscopy technique for surface
The probe, which consists of a cantilever with a sharp tip, will try to touch the sample and this will cause
examination.
shows
scale difference
between
SEM,and
TEM,
and AFM,
it is clear
the probe toFigure
deflect9due
to thethe
attractive
forces between
the probe
the surface
[34]. and
In reality,
there that
AFM
similarformagnification
power
compared
tosemi-contact
TEM. Yet, AFM
provide
3D images
of the
arehas
twoamodes
AFM operation:
Contact
mode and
mode.can
In contact
mode,
the probe
surface
that
can
provide
surface
topography
along
with
surface
roughness
[33].
will constantly touch the sample surface by adjusting the applied force, while in semi-contact mode, the
Unlike
and TEM,
AFM
by[35].
measuring
the force
between
probethe
and
the sample.
probe
willSEM
periodically
touch
theworks
sample
A laser beam
will be
used toadetect
deflection
and The
the reflected
laser will
through with
a photo-detector
form
Because and
a probe
used,
probe,
which consists
of pass
a cantilever
a sharp tip,towill
trythe
toimage
touch[36].
the sample
thisiswill
cause
properties
as stiffness
canforces
also bebetween
measuredthe
by probe
AFM based
indentation
[37].
the mechanical
probe to deflect
duesuch
to the
attractive
and on
thethe
surface
[34]. In
reality,
there are two modes for AFM operation: Contact mode and semi‐contact mode. In contact mode, the
probe will constantly touch the sample surface by adjusting the applied force, while in semi‐contact
mode, the probe will periodically touch the sample [35]. A laser beam will be used to detect the
deflection and the reflected laser will pass through a photo‐detector to form the image [36]. Because
a probe is used, mechanical properties such as stiffness can also be measured by AFM based on the
Membranes 2020, 10, 33
8 of 36
2.3.1. Sample Preparation
In
AFM,2020,
liquid
Membranes
10, 33 samples, in addition to solids, can be analyzed due to the absence
8 of of
37 vacuum.
Compared to SEM and TEM, AFM requires less or no preparation steps. For most polymers, coating
Sample Preparation
with a2.3.1.
conductive
material or staining is not needed. It is advised however to cut the samples no larger
than 10 byIn10
mmliquid
with thickness
to 2 mm
[38]. The
sample
is then
a substrate
made from
AFM,
samples, inup
addition
to solids,
can be
analyzed
due fixed
to the over
absence
of vacuum.
to SEM
and TEM,
requires
a flat Compared
mica (silicate
minerals)
orAFM
a glass
[39].less or no preparation steps. For most polymers, coating
with a conductive material or staining is not needed. It is advised however to cut the samples no
10 by 10 mm with thickness up to 2 mm [38]. The sample is then fixed over a substrate
2.3.2. larger
Data than
Interpretation
made from a flat mica (silicate minerals) or a glass [39].
AFM is capable of displaying the topography data of the membrane surface in the form of a 3D
2.3.2.
Data
Interpretationcan be useful in studying the change in the membrane roughness due to
image.
This
information
fouling. With
the rejected
molecules
will accumulate
onmembrane
the membrane
causing
AFMtime,
is capable
of displaying
the topography
data of the
surface surface
in the form
of a 3Da swelling,
image. This the
information
can roughness
be useful in [40].
studying
the 10
change
in the
roughness
to
thus alternating
membrane
Figure
shows
the membrane
topography
data of due
polyvinylidene
fouling.
With
time,
the
rejected
molecules
will
accumulate
on
the
membrane
surface
causing
a Figure 9
fluoride (PVDF)/polyvinylalcohol (PA) membrane before and after the operation. The data in
swelling,
thus
alternating
the
membrane
roughness
[40].
Figure
10
shows
the
topography
data
of
are presented by two parameters, Ra and Rq. Ra mainly represents the roughness and it is the arithmetic
polyvinylidene fluoride (PVDF)/polyvinylalcohol (PA) membrane before and after the operation. The
average of the absolute values of the surface height deviations [41]. The higher the value of Ra, the
data in Figure 9 are presented by two parameters, Ra and Rq. Ra mainly represents the roughness
rougher
On the
otherofhand,
Rq represents
thesurface
root mean
square
average
of height
andthe
it issurface.
the arithmetic
average
the absolute
values of the
height
deviations
[41]. The
higherdeviation
takenthe
from
theofplane.
value
Ra, the rougher the surface. On the other hand, Rq represents the root mean square
average
height
deviation
taken from the the
plane.
AFM
is of
also
capable
of determining
pore size distribution especially if the pore size of the
AFM
is
also
capable
of
determining
the
pore
distributionfor
especially
if the pore size
of waste-water
the
membrane is larger than 2 nm [42]. Ultrafiltrationsize
membranes
water purification
and
membrane is larger than 2 nm [42]. Ultrafiltration membranes for water purification and waste‐water
treatments usually have pores with a size of 10 nm making AFM a suitable technique for pore size
treatments usually have pores with a size of 10 nm making AFM a suitable technique for pore size
measurement [43]. Similar to SEM, pore size distribution can be determined by measuring the pore
measurement [43]. Similar to SEM, pore size distribution can be determined by measuring the pore
area of
the
images
inFigure
Figure
[44].
However,
provides
better accuracy
area
of generated
the generated
imagesas
asshown
shown in
1111
[44].
However,
AFM AFM
provides
better accuracy
compared
to
SEM
especially
for
mesoporous
structure
with
pore
size
ranging
from
2
to
compared to SEM especially for mesoporous structure with pore size ranging from 2 to 50 nm. 50 nm.
Figure
9. Differences
between
optical
microscopy,
SEM,
TEM,
and
Figure
9. Differences
between
optical
microscopy,
SEM,
TEM,
andatomic
atomicforce
forcemicroscopy
microscopy (AFM) in
resolution
[45].
Membranes
2020,
10,
33
9 of 37
(AFM) in resolution [45].
Figure 10. 3D AFM data of polyvinylidene fluoride (PVDF)/polyvinylalcohol (PA) membrane: (a)
Figure 10. 3D AFM data of polyvinylidene fluoride (PVDF)/polyvinylalcohol (PA)
Originalmembrane:
surface, (b)
after
fouling
to theafter
exposure
natural
organic
matter
(pH3) [46].
(a)surface
Original
surface,
(b)due
surface
foulingtodue
to the
exposure
to natural
organic matter (pH3) [46].
Because AFM works based on a moving probe, some mechanical properties can be measured
by studyingBecause
the indentation
the on
probe
on the
membrane
surface.properties
A graphcan
ofbe
the
force versus
the
AFM worksof
based
a moving
probe,
some mechanical
measured
by
studying the indentation of the probe on the membrane surface. A graph of the force versus the
displacement can be plotted and then converted to a plot of the stress versus the strain to calculate
the stiffness of the membrane (Youngʹs modulus) as shown in Figure 12. In Figure 12b, the linear
relationship between force and displacement indicates an elastic deformation. Young’s modulus
values in Figure 12c were calculated using the Derjaguin–Muller–Toporov (DMT) method and the
Figure 10. 3D AFM data of polyvinylidene fluoride (PVDF)/polyvinylalcohol (PA)
membrane: (a) Original surface, (b) surface after fouling due to the exposure to natural
Membranes 2020,
10, 33matter (pH3) [46].
organic
9 of 36
Because AFM works based on a moving probe, some mechanical properties can be measured by
studying can
the indentation
of thethen
probe
on the membrane
A graph
of thethe
force
versus
displacement
be plotted and
converted
to a plotsurface.
of the stress
versus
strain
to the
calculate
displacement
be plotted(Young’s
and then converted
a plot
of the
versus
calculate
the stiffness
of thecan
membrane
modulus)toas
shown
instress
Figure
12. the
In strain
Figureto12b,
the linear
the stiffness
of the
membrane
(Youngʹs modulus)
as an
shown
in Figure
12. In Figure
12b,modulus
the linearvalues
relationship
between
force
and displacement
indicates
elastic
deformation.
Young’s
relationship between force and displacement indicates an elastic deformation. Young’s modulus
in Figure 12c were calculated using the Derjaguin–Muller–Toporov (DMT) method and the first value
values in Figure 12c were calculated using the Derjaguin–Muller–Toporov (DMT) method and the
of 0.71first
GPavalue
at the
an elastic
deformation
due to thedue
elastic
inner
chains
the polymer.
of surface
0.71 GPashowed
at the surface
showed
an elastic deformation
to the
elastic
inner of
chains
of
The change
of Young’s
modulus
value to
0.39 GPa
indicates
lessindicates
elasticity
along
withalong
the membrane
the polymer.
The change
of Youngʹs
modulus
value
to 0.39 GPa
less
elasticity
with
depth.the
It was
also found
later
thatalso
the found
stiffer the
the the
lower
the gas permeability
membrane
depth.
It was
latermembrane,
that the stiffer
membrane,
the lower the due
gas to the
permeability
due
to
the
reduction
in
free
volumes
[47].
These
volumes
provide
the
path
for
gas
reduction in free volumes [47]. These volumes provide the path for gas transport through the dense
transport
the studies
dense membrane.
The discussed
studiescharacterization
on using AFM are
for given
membrane
membrane.
Thethrough
discussed
on using AFM
for membrane
in Table 3.
characterization are given in Table 3.
2.3.3. Limitations
2.3.3. Limitations
AFM has some benefits over SEM and TEM in terms of sample preparation. However, AFM
AFM has some benefits over SEM and TEM in terms of sample preparation. However, AFM
analysis
requires more processing time. Also, AFM has a lower depth of field compared to SEM [48].
analysis requires more processing time. Also, AFM has a lower depth of field compared to SEM [48].
Nevertheless,
the contrast
in AFM
images
better[49].
[49].
Nevertheless,
the contrast
in AFM
imagesisisgenerally
generally better
11. Determination
pore
size distribution
of polyvinylidene
(PVDF)
FigureFigure
11. Determination
of poreofsize
distribution
of polyvinylidene
fluoride fluoride
(PVDF) membrane
by
membrane
by
analyzing
the
3D
image
generated
by
AFM
[50].
analyzing
the 3D
image generated by AFM [50].
Membranes 2020,
10, 33
10 of 37
12.ofUse
of for
AFM
the measurements
of the stiffness
of polymer
of intrinsic
FigureFigure
12. Use
AFM
thefor
measurements
of the stiffness
of polymer
of intrinsic
microporosity
microporosity
by plotting
the force
versus displacement
and was
(PIM-1)
membrane by(PIM‐1)
plottingmembrane
the force versus
displacement
and indentation.
Young’s modulus
indentation.
modulus was calculated(DMT)
based method
on the Derjaguin–Muller–Toporov
calculated
based onYoung’s
the Derjaguin–Muller–Toporov
[51].
(DMT) method [51].
Table 3. Studies performed by AFM for membrane characterization.
Table 3. Studies performed by AFM for membrane characterization.
Study
Membrane
Methodology
Conclusion
Study
Membrane
Methodology
Conclusion
Ref.
Roughness decreased
Polyvinylidene fluoride
Samples in contact mode with silicon nitride probe.
Roughness
Samples in contact mode with silicon
(PVDF)/polyvinylalcohol
Scan area of 5 µm by 5 µm.
from
54 to 37 nm
Polyvinylidene fluoride
Roughness
decreased
measurements
Roughness
nitride
probe.
(PA) /polyvinylalcohol
Five areas
were
analyzed.
(PVDF)
from indicating
54 to 37 nma fouling.
[46]
measurements
Scan area of 5 m by 5 m.
◦ C overnight.
(PA)
indicating a fouling.
Dried samples
at
40
Five areas were analyzed.
Pore size
Polyvinylidene fluoride
Average pore
Semi-contact mode.
Dried samples at 40C overnight.
distribution
(PVDF)
size of 0.3 µm.
Image processing by NT-MDT software.
Pore size
Polyvinylidene fluoride
Semi‐contact mode.
Average pore size of
[50]
Contact-mode
analysis. by NT‐MDT
distribution
(PVDF)
Image processing
0.3 m.
Membrane
Polymer of intrinsic
Probe radius
of 8 nm.
Elasticity decrease with
software.
stiffness
microporosity (PIM-1)
Derjaguin–Muller–Toporov
(DMT) method to
membrane depth.
Contact‐mode analysis.
Young’s
modulus.
Membrane
Polymer of intrinsiccalculate
Probe
radius
of 8 nm.
Elasticity decrease
[51]
stiffness
microporosity (PIM‐1)
Derjaguin–Muller–Toporov (DMT)
with membrane depth.
method to calculate Young’s modulus.
3. Crystal Structure Analysis
3.1. X‐ray Diffraction
X‐ray diffraction (XRD) is a technique for studying the crystal structure of the membrane. The
Ref.
[46]
[50]
[51]
Membranes 2020, 10, 33
10 of 36
3. Crystal Structure Analysis
3.1. X-ray Diffraction
X-ray diffraction (XRD) is a technique for studying the crystal structure of the membrane.
The technology is unique in determining the crystal structure type and the distance between the polymeric
chains [52]. Moreover, the polymer structure, either glassy or amorphous, can be identified by XRD.
Glassy polymers feature ordered, more crystallized structure with better mechanical properties [53–55].
They are widely used for gas separation due to the high selectivity [56]. This means that they separate
gases with a higher product purity. The high selectivity in glassy polymers is related to the high gas
solubility due to the excess of free volumes [57].
On the other hand, amorphous (rubbery) polymers have a random structure but with lower
mechanical properties [58]. However, high gas permeability is achieved by amorphous membranes
due to the absence of gas diffusion limitation [59]. In addition to crystal structure studies, the chemical
compounds presented in the sample can be identified by XRD. This information is valuable for
confirming sample purity. The formation of secondary phases due to membrane degradation can also
be spotted by XRD.
In XRD, a filament such as tungsten is used to generate an X-ray beam [60]. This beam is directed
to the sample and the scattered X-ray is detected and analyzed. Figure 13 shows how XRD analysis is
performed. Some X-rays will be reflected upon hitting the surface while some will enter the surface
and then reflect as shown in Figure 14. This variation is called diffraction and Bragg’s law can be
applied to calculate the diffraction angle (2θ) [61]:
nλ = 2dsin θ
Membranes 2020,
2020,10,
10, 33
33
Membranes
(1)
11 of
of 37
37
11
where nwhere
is the order
of reflection
(usually
unity),
λ is
is the
thewavelength
wavelength
andis it
is determined
from the
is the
the order
order of
of reflection
reflection (usually
(usually unity),
unity),  is
and itit is
determined from
from the
the
where nn is
the wavelength
and
determined
source source
of
radiation
usedused
(commonly
copper
anddddrepresents
represents
plane
distance
between
source
of radiation
radiation
used
(commonly
copperK-α),
K‐α), and
and
represents
thethe
plane
distance
between
set a set
of
(commonly
copper
K‐α),
the
plane
distance
between
aa set
of atoms.
of
atoms.
of atoms.
Figure 13.
13. XRD
XRD instrumentation
instrumentation diagram
diagram [62].
[62].
Figure
Figure 13. XRD instrumentation diagram [62].
Figure 14.
14. Diffraction
Diffraction of
of an
an X‐ray
X‐ray beam
beam from
from aa sample
sample [63].
[63].
Figure
Figure 14. Diffraction of an X-ray beam from a sample [63].
3.1.1. Sample
Sample Preparation
Preparation
3.1.1.
XRD is
is aa non‐destructive
non‐destructive test
test with
with minimum
minimum preparation.
preparation. ItIt can
can be
be used
used to
to analyze
analyze solids,
solids, as
as
XRD
well as
as powders.
powders. For
For solid
solid membranes,
membranes, the
the samples
samples are
are usually
usually cut
cut into
into small
small sections
sections of
of 20
20 by
by 20
20
well
mm. The
The sample
sample is
is then
then mounted
mounted on
on the
the XRD
XRD chamber.
chamber. For
For polymer
polymer powders,
powders, generally,
generally, 100
100 to
to 200
200
mm.
mg of
of powder
powder is
is needed
needed and
and spread
spread over
over aa carbon
carbon tape.
tape. This
This tape
tape has
has aa very
very low
low XRD
XRD intensity
intensity and
and
mg
Membranes 2020, 10, 33
11 of 36
3.1.1. Sample Preparation
XRD is a non-destructive test with minimum preparation. It can be used to analyze solids, as well
as powders. For solid membranes, the samples are usually cut into small sections of 20 by 20 mm.
The sample is then mounted on the XRD chamber. For polymer powders, generally, 100 to 200 mg of
powder is needed and spread over a carbon tape. This tape has a very low XRD intensity and will not
affect the analysis [64].
3.1.2. Data Interpretation
The data in XRD are represented by intensity (counts) versus diffraction peak angle (2θ). Usually,
Membranes 2020, 10, 33
12 of 37
the angle starts from 10 to 80 and each compound has a defined set of peaks. For example, polyetherimide
is known
to have sharp
peaks at the
21.4mechanical
and 23.8 properties
(2θ) [15]. (Figure
If additional
peaks
were detected
for a pure
crystallized,
which enhanced
15). Some
researchers
also observed
the relationship
between
themay
intensity
and gas
permeation
they concluded
that the higher the
polyetherimide
membrane,
this
indicate
impurities
or and
membrane
degradation.
themembranes,
lower the permeation
This was
interpreted
by the limitation
in gas diffusion
In intensity,
polymeric
there is[67].
a tradeoff
between
permeability
and selectivity
due to the
through the membrane as the structure becomes more crystallized [68].
nature of the polymer [1]. Mixed-matrix membranes can overcome this limitation by adding fillers of
Amorphous and glassy structures can be distinguished in XRD by observing the peak shape
another[62].
material
to the polymer during the membrane preparation [65]. XRD can be beneficial for
Sharp peaks refer to a more crystallized structure while broad peaks refer to an amorphous
studying
the effect
of adding the
polymer
The fillersparameter
are supposed
to be detected
structure
as demonstrated
in fillers
Figure to
16.the
d‐space
(d) is matrix.
another important
for membrane
in the XRD
peak profile
along with
polymer.
example,
a mixed-matrix
membrane
was made
characterization.
In Equation
(1), dthe
represents
the For
distance
between
two planes of atoms.
In polymeric
membranes,
d‐space can
representtothe
distance between the
polymer
chains
and it is calculated
based
by adding
aluminosilicate
particles
polyethersulfone
with
different
concentrations
[66].
XRD data
on
the
highest
peak
of
the
XRD
pattern.
Long‐chain
polymers
are
presumed
to
have
a
higher
d‐space
revealed that the intensity of polyethersulfone peaks increased making the structure more crystallized,
value. In polymer science, it is observed that the longer the chain, the less the structure crystallinity
which enhanced the mechanical properties (Figure 15). Some researchers also observed the relationship
due to the transformation from a glassy to a rubbery structure [52]. In terms of gas permeability, it is
betweeneasier
the intensity
and gas permeation and they concluded that the higher the intensity, the lower
for the gas to pass through an amorphous structure due to the absence of diffusion limitation.
the permeation
[67].
This
was by
interpreted
the limitation
in gas diffusion
through
the membrane
as
This was also confirmed
noticing anbyincrease
in the membrane
permeability
for longer
chain
the structure
becomes
more
crystallized
polymers
[69]. Table
4 summarizes
the[68].
mentioned studies for membrane analysis by XRD.
Figure
XRD
of a mixed‐matrix
membrane
of polyethersulfone
membrane
with
Figure 15.
XRD15.
data
of adata
mixed-matrix
membrane
of polyethersulfone
membrane
with aluminosilicate
aluminosilicate
particles
[66].
particles [66].
Amorphous and glassy structures can be distinguished in XRD by observing the peak shape [62].
Sharp peaks refer to a more crystallized structure while broad peaks refer to an amorphous structure as
demonstrated in Figure 16. d-space (d) is another important parameter for membrane characterization.
In Equation (1), d represents the distance between two planes of atoms. In polymeric membranes,
d-space can represent the distance between the polymer chains and it is calculated based on the
highest peak of the XRD pattern. Long-chain polymers are presumed to have a higher d-space value.
In polymer science, it is observed that the longer the chain, the less the structure crystallinity due to the
transformation from a glassy to a rubbery structure [52]. In terms of gas permeability, it is easier for
Figure 16. Differentiation of XRD peaks for determining the glassy and amorphous
structures [62].
Membranes 2020, 10, 33
12 of 36
the gas to pass through an amorphous structure due to the absence of diffusion limitation. This was
alsoFigure
confirmed
by noticing
in the membrane
for longer membrane
chain polymers
15. XRD
data ofanaincrease
mixed‐matrix
membranepermeability
of polyethersulfone
with[69].
Table
4 summarizesparticles
the mentioned
aluminosilicate
[66]. studies for membrane analysis by XRD.
Figure
of XRD
for determining
theamorphous
glassy and
amorphous
Figure16.
16. Differentiation
Differentiation of XRD
peaks peaks
for determining
the glassy and
structures
[62].
structures [62].
Table 4. XRD studies for characterization of polymeric membrane.
Study
Membrane
Methodology
Conclusion
Ref.
Membrane purity
Polyetherimide
Samples fractured
in liquid nitrogen.
Carbon tape to
hold the samples.
Only polyetherimide peaks were
detected indicating a pure sample.
[15]
Polymer
Crystallinity
Polyethersulfone with
aluminosilicate particles
Intensity increased indicating a
more crystallized structure with
better mechanical properties.
[66]
Polymer chain
distance (d-space)
6H,12H-5,11methanodibenzo[b,f][1,5]diazocine
[PIM-EA(Me2 )-TB]
Increase in d-space indicated
transformation from glassy to
rubbery phase.
[69]
Not reported.
Not reported.
3.1.3. Limitations
X-ray beam does not interact strongly with lighter elements and this may limit the elements
detection in XRD [70]. Also, XRD is more accurate for measuring large crystal structures rather than
small ones [71]. Furthermore, if more compounds are presented in the sample, peak overlap may
complicate the analysis.
3.2. X-ray Scattering
XRD is performed widely for materials with well-ordered (crystalline) structure. However, for
semi-crystalline or non-crystalline (amorphous) materials, X-ray scattering is preferred [72]. In X-ray
scattering, a monochromatic beam of X-rays is sent to the sample in which some of the X-rays will be
scattered. These scattered X-rays are then detected and analyzed. There are two types of X-ray scattering:
Small-angle X-ray scattering (SAXS) and wide-angle X-ray scattering (WAXS). In SAXS, the scattering
angle (2θ) is less than 5◦ and the technique focuses on studying the particles in a system. For instance,
SAXS can determine the nanoparticle size distribution and pore size [73,74]. These properties can still
be measured by microscopic techniques, however SAXS is better in giving the average values for larger
areas [75]. On the other hand, WAXS scans at a scattering angle of over 5◦ with the ability to study the
degree of crystallinity and the chemical composition [76]. The SAXS and WAXS are usually combined
in one instrument and its components are given in Figure 17.
3.2.1. Sample Preparation
SAXS and WAXS barely need sample preparation for polymeric membranes similar to XRD.
Membranes 2020, 10, 33
13 of 36
3.2.2. Data Interpretation
The data in SAXS/WAXS are presented by the intensity and the scattering vector (q) and the latter
is calculated by:
4π
θ
(2)
q=
sin
λ
2
where λ is the wavelength of the incident X-rays and θ is the scattering angle. SAXS was used to study
the preparation method of sulfonated poly(aryl ether ketone) (SPEEK) membranes. The absence of the
ionomer peak (–SO3 H) indicated no nano-phase separation between the hydrophilic segments and the
polymer matrix (Figure 18). Addition of inorganic filler resulted in a significant increase in intensity.
The high intensity revealed that the filler was merged into the mass-fractal structure meaning that
PMoA was successfully crosslinked with the polymer [77]. SAXS was also used to calculate the particle
radius (Lm ) in the solution using Schimdt equation:
Lm =
Membranes 2020, 10, 33
π
14 of 37 (3)
qmin
where 𝑞
is the lower limit of the scattering vector (q) in SAXS graph. PMoA particles were found
where
qmin is the
lower limit of the scattering vector (q) in SAXS graph. PMoA particles were found to
to have an average radius of 524Å. In addition to the above, WAXS was implemented to calculate the
have an average radius of 524Å. In addition to the above, WAXS was implemented to calculate the pore
pore size distribution of PIM membranes. Equation (3) was used to calculate the pore size but with
size distribution of PIM membranes. Equation (3) was used to calculate the pore size but with the lower
the lower and upper limit of scattering vector (Figure 19). The intensity is proportional to the amounts
and upper limit of scattering vector (Figure 19). The intensity is proportional to the amounts of pores
of pores having that pore size. The reported pore size for amidoxime‐grafted PIM membranes ranged
having
pore
The
reported
pore sizethe
forstudies
amidoxime-grafted
membranes ranged
and 3.9
fromthat
3.9 to
5.9 size.
Å [78].
Table
5 summarizes
for membranePIM
characterization
by SAXSfrom
to 5.9
Å
[78].
Table
5
summarizes
the
studies
for
membrane
characterization
by
SAXS
and
WAXS.
WAXS.
Figure
17. Small‐angle
scattering
(SAXS)/wide‐angle
X‐ray (WAXS)
scattering
(WAXS) for
Figure
17. Small-angle
X-ray X‐ray
scattering
(SAXS)/wide-angle
X-ray scattering
instrument
instrument
for
crystal
and
molecular
analysis:
(1)
Rotation
unit,
(2)
arm,
(3)
counterweight,
crystal and molecular analysis: (1) Rotation unit, (2) arm, (3) counterweight, and (4) hot stage [79].
and (4) hot stage [79].
Table 5. SAXS/WAXS studies for characterization of polymeric membrane.
Study
Membrane
Methodology
Conclusion
Ref.
Polymer-filler
interaction
and particle size
measurements
Sulfonated poly(aryl ether
ketone) (SPEEK) with
phosphomolybdic acid
(PMoA) filler
Fixed wavelength of 1.5Å.
Vacuum operation at
room temperature.
Data calibrated using
positron-emission
tomography (PET).
No detection of –SO3 H group
indicated no nano-phase
separation.
Broad peak showed amorphous
SPEEK structure.
Particle radius of 524 Å.
[80]
Pore size
measurements
Polymer of intrinsic
microporosity (PIM) with
amidoxime groups
Not reported.
Pore size distribution
from 3.9 to 5.9 Å.
[78]
Figure 17. Small‐angle X‐ray scattering (SAXS)/wide‐angle X‐ray scattering (WAXS)
instrument for crystal and molecular analysis: (1) Rotation unit, (2) arm, (3) counterweight,
Membranes 2020,
and10,
(4)33
hot stage [79].
14 of 36
FigureFigure
18. SAXS
analysis
of sulfonated
poly(aryl
ether ether
ketone)
(SPEEK)/phosphomolybdic
acid (PMoA)
18.10,SAXS
analysis
of sulfonated
poly(aryl
ketone)
(SPEEK)/phosphomolybdic
Membranes 2020,
33
15 of 37
membranes
(a),
and
pristine
SPEEK
membranes
(b)
[80].
acid (PMoA) membranes (a), and pristine SPEEK membranes (b) [80].
Figure
19. WAXS
analysis
revealed
an amorphous
structure
of SPEEK
due to
the
peak [80].
Figure
19. WAXS
analysis
revealed
an amorphous
structure
of SPEEK
due
tobroad
the broad
peak [80].
3.2.3. Limitations
Table 5. SAXS/WAXS studies for characterization of polymeric membrane.
Sample homogeneity is critical for SAXS analysis as sample aggregation or degradation will
Study
Ref.
result in incorrect
data [81].Membrane
Sample damage by Methodology
radiation is also possibleConclusion
and can distort the
data [82].
No detection of –SO3H group
Fixed
wavelength
of
1.5Å.
Furthermore,
the scattered
intensity
can be weak for some systems
[83]. no nano‐phase
indicated
Polymer‐filler
Sulfonated
poly(aryl
4.
interaction
ether ketone) (SPEEK)
and particle size
with phosphomolybdic
Functional Groups Analysis
measurements
acid (PMoA) filler
Vacuum operation at room
temperature.
Data calibrated using positron‐
emission tomography (PET).
4.1. Fourier-Transform Infrared
PolymerSpectroscopy
of intrinsic
Pore size
microporosity (PIM)
separation.
Broad peak showed
amorphous SPEEK structure.
Particle radius of 524 Å.
[80]
Pore size distribution from
Not reported.
[78]
Fourier-transform
infrared
spectroscopy
(FTIR) is a technique3.9toto determine
the functional
groups.
measurements
with amidoxime
5.9 Å.
groups
Usually, the functional groups give the chemical reaction properties of the compound. Some examples
of the functional
groups are hydroxyl (OH–) and carbonyl (C=O). If the functional groups are known, the
3.2.3. Limitations
Sample homogeneity is critical for SAXS analysis as sample aggregation or degradation will
result in incorrect data [81]. Sample damage by radiation is also possible and can distort the data [82].
Furthermore, the scattered intensity can be weak for some systems [83].
4. Functional Groups Analysis
made mainly from diamond [86]. This will cause creation of evanescent waves. The reflected beam is
then detected and analyzed to generate the FTIR data.
4.1.1. Sample Preparation
Membranes 2020, 10, 33
15 of 36
FTIR spectroscopy usually requires sample preparation before analysis. Generally, the
membrane will be crushed into a powder of 2 to 5 mg then mixed with a solvent commonly potassium
compound
class
can be determined.
For example,
the hydroxyl
group
(OH–) usually
represents
bromide
with
a mixing
ratio of 1 to 100
[87, 88]. Then,
the mixed
powder
is pressed
in a diealcohols
at a load
carboxyl
acids
and FTIR
easily
those compounds
discussed
of 13can
mm
[89].distinguish
The pellet between
is then inserted
in the FTIRaschamber
forlater.
analysis.
ofand
10 tons
to form
a pellet
In FTIR
spectroscopy,
thepotassium
analysis canbromide
be performed
on gases,
liquids,
and solids.material
It is a quantitative
It should
be mentioned
that
is usually
used
as a reference
because it
−1 [90]. Also,
technique
as wellin[84].
works window
based onfrom
sending
radiation
to the itsample
the use
does
not interfere
the It
infrared
400 infrared
to 4000 cm
acts aswith
a carrier
forof
the
a
reference
solvent
[85].
Some
of
the
radiation
will
be
absorbed
by
the
sample
and
others
will
pass
infrared spectrum. Furthermore, the reference material of potassium bromide will be used to plot the
through or will
bein
transmitted.
A mathematical
modeltechniques
(Fourier transform
will be
used
transmittance
axis
the FTIR graph.
Modern FTIR
such as function)
ATR requires
little
ortono
covert
the
data
into
the
spectrum.
Figure
20
shows
the
components
of
the
FTIR
spectrometer.
sample preparation with the ability to analyze solid samples [91].
Figure
20. Components
of of
FTIR
fordetection
detection
functional
groups
Figure
20. Components
FTIRspectroscopy
spectroscopy for
of of
functional
groups
[32]. [32].
Attenuated
total reflection (ATR) is another method for generating FTIR spectrum without the
4.1.2. Data
Interpretation
use of a reference material. In ATR, a beam of infrared light is sent to the sample through ATR crystal
The data in FITR spectroscopy is given by the transmittance versus the wavelength. Zero
made mainly from diamond [86]. This will cause creation of evanescent waves. The reflected beam is
transmittance
means that the sample absorbed all the radiation while 100% transmittance means that
then detected and analyzed to generate the FTIR data.
the sample absorbed the same amounts of radiation as the reference. The generated graph is unique
for
eachSample
molecule
and this can be used in identifying the functional groups. For instance, the hydroxyl
4.1.1.
Preparation
group (OH–) represents both alcohols and carboxylic acids but FTIR can differentiate between them
FTIR spectroscopy usually requires sample preparation before analysis. Generally, the membrane
based on the wavelength. For instance, alcohol’s peak is detected at a wavelength of 3200 to 3600 cm−1
will be crushed into a powder of 2 to 5 mg then mixed with a solvent commonly potassium bromide
while
carboxylic acids peak is detected at 2500 to 3300 cm−1 [92]. Table 6 shows the wavelength of
with a mixing ratio of 1 to 100 [87,88]. Then, the mixed powder is pressed in a die at a load of 10 tons to
different
functional
groups
FTIR
analysis.
form a pellet
of 13 mm
[89].for
The
pellet
is then inserted in the FTIR chamber for analysis. It should be
FTIR
spectroscopy
can
be
very
useful
determining
thematerial
compatibility
theinterfere
polymer
mentioned that potassium bromide is usuallyfor
used
as a reference
becausebetween
it does not
and
theinfrared
solvent.window
The membrane
preparation
involves
useasofa carrier
solventforfor
solvent
in the
from 400 to
4000 cm−1 [90].
Also,the
it acts
thecasting.
infraredThe
spectrum.
should
not
react
with
the
polymer
to
avoid
changing
the
physical
and
chemical
properties
of in
the
Furthermore, the reference material of potassium bromide will be used to plot the transmittance axis
polymer
A Modern
study was
by preparing
a PSf
membrane
different
solventswith
such
the FTIR[93].
graph.
FTIRmade
techniques
such as ATR
requires
little orwith
no sample
preparation
theas
diethylene
glycol
(DEG),
dimethylacetamide
(DMAc),
dimethylformamide
(DMF),
and
n‐
ability to analyze solid samples [91].
methylpyrrolidone (NMP). FTIR data is given in Figure 21 and none of the solvents altered the FTIR
4.1.2. Data
Interpretation
spectrum
indicating
a pure PSf membrane. In mixed matrix membranes, FTIR analysis is widely used
to determine
thein
interaction
between the
fillers for better
compatibility
[94]. This
can
The data
FITR spectroscopy
is polymer
given byand
thethe
transmittance
versus
the wavelength.
Zero
transmittance means that the sample absorbed all the radiation while 100% transmittance means that
the sample absorbed the same amounts of radiation as the reference. The generated graph is unique
for each molecule and this can be used in identifying the functional groups. For instance, the hydroxyl
group (OH–) represents both alcohols and carboxylic acids but FTIR can differentiate between them
based on the wavelength. For instance, alcohol’s peak is detected at a wavelength of 3200 to 3600 cm−1
while carboxylic acids peak is detected at 2500 to 3300 cm−1 [92]. Table 6 shows the wavelength of
different functional groups for FTIR analysis.
FTIR spectroscopy can be very useful for determining the compatibility between the polymer and
the solvent. The membrane preparation involves the use of solvent for casting. The solvent should not
react with the polymer to avoid changing the physical and chemical properties of the polymer [93].
A study was made by preparing a PSf membrane with different solvents such as diethylene glycol
such as oxygen and nitrogen as they do not absorb infrared radiation [96].
Table 6. Identification of chemicals based on the wavelength of the functional groups by
FTIR spectrometer [97].
Membranes 2020, 10, 33
16 of 36
Wavelength (cm−1) Functional Group Chemical Class
3200–3500
OH–
Alcohols
2500–3300
OH–
Carboxylic acids
(DEG), dimethylacetamide2800–3000
(DMAc), dimethylformamide
(DMF),
and n-methylpyrrolidone (NMP).
N–H
Amine salts
FTIR data is given in Figure
21
and
none
of
the
solvents
altered
the
FTIR spectrum indicating a pure
3267–3333
C–H
alkynes
PSf membrane. In mixed matrix
membranes, FTIRC–H
analysis is widelyalkenes
used to determine the interaction
3000–3100
between the polymer and the
fillers
for
better
compatibility
[94].
This
can be studied by detecting the
2840–3000
C–H
Alkanes
functional group of the fillers2349
within the FTIR spectrum
of
the
mixed
matrix
O=C=O
carbon dioxidemembrane. Table 7 sum
up the discussed membrane
characterization studies
1380–1415
S=Oby FTIR.
sulfates
Figure
FTIR
analysis
indicating
significant
change
preparation
polysulfone
Figure
21.21.
FTIR
analysis
indicating
nono
significant
change
in in
thethe
preparation
of of
polysulfone
(PSF)
(PSF) membrane
using
diethylene
glycol (DEG), (DMAc),
dimethylacetamide
(DMAc),
membrane
using diethylene
glycol
(DEG), dimethylacetamide
dimethylformamide
(DMF),
dimethylformamide
n‐methylpyrrolidone
(NMP) as solvents [93].
and
n-methylpyrrolidone(DMF),
(NMP)and
as solvents
[93].
4.1.3. Limitations Table 7. FTIR studies for characterization of polymeric membrane.
Membrane
Ref.
AsStudy
discussed before,
FTIR may requireMethodology
sample preparation but Conclusion
for modern FTIR techniques,
Functional
groups
of
only
aryl
almost no preparation is needed. Aqueous solutions are difficult to analyze by FTIR due to the strong
Polymer‐
ethers, aryl sulfones and methyl
infraredsolvent
absorption of water
[95]. Furthermore,
FTIR cannot detect
of twonoidentical
Polysulfone
Not reported.
weremolecules
detected indicating
[93]atoms
such compatibility
as oxygen and nitrogen as they do not absorb infrared radiation
[96].with the solvent
interaction
(solvent is compatible).
Peaks of (C‐N), (Si‐O), (CO2–), and
Table 6. Identification of
chemicals based
Polyetherimide
with on the wavelength of the functional groups by FTIR spectrometer [97].
Polymer‐filler
metal‐organic framework
interaction Wavelength (cm−1 )
filler (MIL‐53)
3200–3500
2500–3300
2800–3000
3267–3333
3000–3100
2840–3000
2349
1380–1415
FTIR Spectrum recorded
at 4000–500
cm‐1Group
.
Functional
OH–
OH–
N–H
C–H
C–H
C–H
O=C=O
S=O
(Al‐O) indicated that MIL‐53 was
successfully
incorporated
Chemical
Classin the
polymer matrix.
[94]
Alcohols
Carboxylic acids
Amine salts
alkynes
alkenes
Alkanes
carbon dioxide
sulfates
Table 7. FTIR studies for characterization of polymeric membrane.
Study
Membrane
Polymer-solvent
compatibility
Polysulfone
Polymer-filler
interaction
Polyetherimide
with metal-organic
framework filler
(MIL-53)
Methodology
Conclusion
Ref.
Not reported.
Functional groups of only aryl ethers, aryl
sulfones and methyl were detected
indicating no interaction with the solvent
(solvent is compatible).
[93]
FTIR Spectrum
recorded at
4000–500 cm−1 .
Peaks of (C-N), (Si-O), (CO2 –), and (Al-O)
indicated that MIL-53 was successfully
incorporated in the polymer matrix.
[94]
Membranes 2020, 10, 33
Membranes 2020, 10, 33
18 of 37
17 of 36
4.2. Raman Spectroscopy
4.2. Raman Spectroscopy
Raman spectroscopy studies the vibration and rotation modes of molecules [98]. Each
Raman
studies
the vibration
and be
rotation
modes of molecules
compound
compound
hasspectroscopy
its own unique
spectrum
that can
cross‐referenced
with a [98].
set ofEach
known
Raman
has its own
unique
spectrum
can be
with
a set
of cause
known
Raman spectrum.
spectrum.
Basically,
laser
light isthat
focused
on cross-referenced
the sample and this
light
will
molecular
vibration.
Basically,
light iswill
focused
the sample
andthat
thisresults
light will
cause molecular
vibration.
Thisphoton
form
This
form oflaser
excitation
causeonlight
scattering
in shifting
up or down
the laser
of excitation
will cause light
scattering
that results
shifting
up or down
energy
[99]. Components
of Raman
instrument
areingiven
in Figure
22. the laser photon energy [99].
Components
of Raman instrument
are given
Figure 22. technique that gives information about the
Raman spectroscopy
is a quantitative
andinqualitative
Raman
spectroscopy
is a quantitative
andtoqualitative
technique
that
gives information
about
functional
groups
in the polymer.
Compared
FTIR, Raman
is more
sensitive
to the functional
the
functional
groups
in
the
polymer.
Compared
to
FTIR,
Raman
is
more
sensitive
to
the
functional
groups giving sharper peaks [100]. Furthermore, diatomic molecules such as oxygen and nitrogen
giving sharper
peaks
[100]. Furthermore,
diatomic
molecules such
oxygen tool
and nitrogen
can
cangroups
be detected
in Raman
spectroscopy.
Also, Raman
spectrometer
is an as
excellent
for studying
detectedin
in Raman
spectroscopy.
Raman
spectrometer
an the
excellent
tool for
the
thebechanges
the polymer
crystal Also,
structure
due
to changesisin
chemical
andstudying
mechanical
changes
in
the
polymer
crystal
structure
due
to
changes
in
the
chemical
and
mechanical
properties
[101].
properties [101]. Furthermore, polymer chain orientation and interfacial surface properties can be
Furthermore, polymer chain orientation and interfacial surface properties can be obtained by Raman
obtained
by Raman spectrum [32].
spectrum [32].
4.2.1. Sample Preparation
4.2.1. Sample Preparation
Raman
is is
considered
asas
a non‐destructive
test
and
thethe
preparation
is is
much
easier
compared
Raman
considered
a non-destructive
test
and
preparation
much
easier
comparedto
other
methods
such
as
FTIR
where
dissolving
the
sample
with
a
proper
solvent
may
be
needed.
The
to other methods such as FTIR where dissolving the sample with a proper solvent may be needed.
2 with 2thickness less than 2.5 cm placed
sample
area
for
Raman
spectrometer
is
usually
less
than
10
cm
The sample area for Raman spectrometer is usually less than 10 cm with thickness less than 2.5 cm
over
a glass
orasilicon
substrate
[102]. [102].
placed
over
glass or
silicon substrate
Figure
22. Components
Ramanspectroscopy
spectroscopy instrumentation
[103].
Figure
22. Components
of of
Raman
instrumentation
[103].
4.2.2. Data Interpretation
4.2.2. Data Interpretation
The data in Raman spectroscopy is given in the form of intensity over wavelength of the scattered
The data in Raman spectroscopy is given in the form of intensity over wavelength of the
photons (Raman shift). The technique can be used to determine the form of the crystal structure by
scattered
photons
(Raman
shift).
technique
can in
beintensity
used toofdetermine
the indicates
form of athe
crystal
observing
the intensity
peaks.
For The
example,
the change
the same peak
variation
structure
by
observing
the
intensity
peaks.
For
example,
the
change
in
intensity
of
the
same
peak
in the structure asymmetry. For instance, PVDF membrane is known to have three different structures;
indicates
a variation
in the structure
asymmetry.sequence
For instance,
PVDFand
membrane
is known
have
form I (all
trans structure),
form II (trans-Gauche
structure),
form III (three
transto
bonds
three
different
structures;
form
I
(all
trans
structure),
form
II
(trans‐Gauche
sequence
structure),
and
separated by Gauche structure) [101]. Raman spectroscopy distinguished between these structures
−1
form
(threethe
trans
bonds
separated
by Gauche
structure) [101].
spectroscopy
distinguished
by III
noticing
peak
shape
and intensity
at a wavelength
of 793Raman
cm . The
use of various
solvents
−1. The
between
these
structures
by
noticing
the
peak
shape
and
intensity
at
a
wavelength
of
793
cm
such as DMF, NMP, and triethyl phosphate (TEP) for PVDF membrane preparation resulted in different
usestructures
of various
solvents
such as
NMP,
and oftriethyl
phosphate
(TEP)
for additional
PVDF membrane
as shown
in Figure
23.DMF,
Form II
structure
PVDF was
recognized
by the
peak at
preparation
resulted
in different structures as shown in Figure 23. Form II structure of PVDF was
795 cm−1 (Figure
23c).
recognized by the additional peak at 795 cm−1 (Figure 23c).
Raman peaks can also be used to determine the functional groups of the polymer. For example,
ethylene compounds with functional group of C=C can be recognized at a wavelength of 1623 cm‐1.
Membranes 2020, 10, 33
18 of 36
Raman peaks can also be used to determine the functional groups of the polymer. For example,
ethylene compounds with functional group of C=C can be recognized at a wavelength of 1623 cm−1 .
Membranes 2020, 10, 33
19 of 37
Some chemicals have the same functional group such as ketones and aldehydes in which they share the
carbonyl
(C=O) group.
spectroscopy
can differentiate
between
ketones (RCOR’)
aldehydes
Some chemicals
have Raman
the same
functional group
such as ketones
and aldehydes
in whichand
they
share
−1 , respectively [104,105]. Figure 24 demonstrates the
(RCHO)
at
the
wavelength
of
1659
and
1725
cm
the carbonyl (C=O) group. Raman spectroscopy can differentiate between ketones (RCOR’) and
peaks
and wavelengths
different
functional
groups.
aldehydes
(RCHO) atofthe
wavelength
of 1659
and 1725 cm−1, respectively [104, 105]. Figure 24
In
addition
to
the
above,
Raman
spectroscopy
be utilized
to study the aging effect of the
demonstrates the peaks and wavelengths of differentcan
functional
groups.
membrane
due to to
operation.
example,
the presence
new functional
groups
In addition
the above,For
Raman
spectroscopy
can beof
utilized
to study the
aging within
effect ofRaman
the
spectrum
maydue
indicate
degradation.
Another way
to detect of
membrane
degradation
by observing
membrane
to operation.
For example,
the presence
new functional
groupsis within
Ramanthe
spectrum may
degradation.
Another
way
to detect
membrane degradation is by observing
transformation
ofindicate
the main
backbone into
shorter
chains
[106].
theRaman
transformation
of
the
main
backbone
into
shorter
chains
spectroscopy is a quantitative technique as well [106].
and for some mixtures, the intensity can
Raman
spectroscopy
a quantitative
as wellTo
and
for somethe
mixtures,
theofintensity
canin a
be related
to the
amounts ofisthe
compoundtechnique
in the sample.
illustrate,
volume
methanol
−1 volume
in athe
be related
to thelinearly
amountswith
of the
compound
sample.peak
To illustrate,
the
methanol
sample
increased
the
decreasein
ofthe
intensity
at 729 cm
and thisofwas
used for
−1 and this was used for the
sample increased
with the decrease
of Raman
intensityspectroscopy
peak at 729 cm
quantification
[107].linearly
The mentioned
studies by
for membrane
characterization
arequantification
given in Table[107].
8. The mentioned studies by Raman spectroscopy for membrane characterization
are given in Table 8.
4.2.3. Limitation
4.2.3. Limitation
Unlike the FTIR spectrometer, Raman spectroscopy requires a longer time for the analysis. Due to
Unlike the FTIR spectrometer, Raman spectroscopy requires a longer time for the analysis. Due
the use of laser in Raman spectroscopy, some samples may release fluorescent light that interferes with
to the use of laser in Raman spectroscopy, some samples may release fluorescent light that interferes
the spectrum causing an increase in the background noise [108]. Moreover, polar molecules often have
with the spectrum causing an increase in the background noise [108]. Moreover, polar molecules
weak
Raman signals as the atoms hold electrons so closely [109].
often have weak Raman signals as the atoms hold electrons so closely [109].
Figure
Raman
spectroscopy
of PVDF
membrane
different
solvents:
(a) DMF;
(b) (c)
Figure
23. 23.
Raman
spectroscopy
of PVDF
membrane
usingusing
different
solvents:
(a) DMF;
(b) NMP;
NMP;
(c)
triethyl
phosphate
(TEP)
[101].
triethyl phosphate (TEP) [101].
Membranes 2020, 10, 33
19 of 36
Membranes 2020, 10, 33
20 of 37
Figure
24. Functional
identified
byspectroscopy
Raman spectroscopy
and the corresponding
Figure
24. Functional
groups groups
identified
by Raman
and the corresponding
wavelength [110].
wavelength [110].
Table 8. Raman studies for characterization of polymeric membrane.
Study
Table
8. Raman studies for
characterization of polymericConclusion
membrane.
Membrane
Methodology
Study
Membrane
Methodology
Conclusion
Membrane layers
were
Additional peak at 795 cm−1
Membrane in
layers
were
Polyvinylidene
superimposed
a solid
sampler.
Additional
peak at 795
indicated
formation
of cm−1
fluoride
(PVDF)
Spectrum
recordedinata solid sampler.
Polyvinylidene
superimposed
trans-Gauche
sequence
structure.
Crystal structure
indicated
formation
of
trans‐
−1 resolution.
fluoride (PVDF) 2 cmSpectrum
recorded at 2 cm−1
Crystal
structure
Membrane
Membrane
degradation
degradation
resolution.
Samples
mounted vertically.
Samples
mounted
vertically.
Perfluorinated
He-Ne
laser and
Peltier-cooled
He‐Ne laser and Peltier‐cooled
sulfonic-acid
Perfluorinated charge coupled device (CCD) to
charge
coupled
device (CCD) to
(PFSA) (PFSA)detect
Raman
spectrum.
sulfonic‐acid
detect Raman
Resolution
of < 2 spectrum.
cm−1 .
Resolution of < 2 cm−1.
Ref.
Ref.
[101]
[101]
Gauche sequence structure.
Decrease
in intensity
of C–O–C,
Decrease
in intensity
of C–O–
C–S andC,
S–O
indicated
C–Sbonds
and S–O
bonds
membrane
degradation.
indicated
membrane
[111]
[111]
degradation.
4.3.4.3.
Nuclear
Magnetic
Resonance
Nuclear
Magnetic
ResonanceSpectroscopy
Spectroscopy
Nuclear
magnetic
resonance
(NMR)
spectroscopy
is another
technique
for analyzing
the molecular
Nuclear
magnetic
resonance
(NMR)
spectroscopy
is another
technique
for analyzing
the
structure
and
identifying
the
functional
groups.
It
is
also
used
to
study
polymer
blend
miscibility
molecular structure and identifying the functional groups. It is also used to study polymer blendand
polymer
degradation
[112,113].
NMR spectroscopy
is based
on placing
the sample
between
two poles
miscibility
and polymer
degradation
[112,113]. NMR
spectroscopy
is based
on placing
the sample
The radio
sample
will spin
andbroadcast
radio waves
will
be
two
poles of
a powerful
magnet
of abetween
powerful
magnet
[114].
The sample
will[114].
spin and
waves
will be
to the
sample.
broadcast
to excitation
the sample.
will cause
excitation
of the nuclei
and this
will generate
a resonance
This
will cause
of This
the nuclei
and this
will generate
a resonance
frequency
detected
by a radio
frequency
detected
by a radio
receiver as to
shown
in Figure
25. Compared
to Raman,
NMR spectrum
receiver
as shown
in Figure
25. Compared
Raman,
NMR spectrum
has less
background
interference
has
less
background
interference
and
it
is
capable
of
detecting
polar
compounds
[115].
and it is capable of detecting polar compounds [115].
4.3.1.
Sample
Preparation
4.3.1.
Sample
Preparation
Similar
FTIR
spectroscopy,the
thesample
sample in
in NMR
NMR spectroscopy
toto
bebe
in in
thethe
liquid
form.
Similar
to to
FTIR
spectroscopy,
spectroscopyneeds
needs
liquid
form.
Generally,
the
membrane
is
cut
into
small
pieces
having
a
mass
of
20
to
50
mg
[116].
The
sample
Generally, the membrane is cut into small pieces having a mass of 20 to 50 mg [116]. The sample is is
then
then
dissolved
in
a
strong
solvent
such
as
deuterated
chloroform
[117].
The
concentration
of
the
dissolved in a strong solvent such as deuterated chloroform [117]. The concentration of the sample
sample in the mixture should be about 1 wt% [118]. The liquid is then placed in a 5 mm tube and
in the mixture should be about 1 wt% [118]. The liquid is then placed in a 5 mm tube and ready for
ready for analysis [119].
analysis [119].
4.3.2. Data Interpretation
4.3.2. Data Interpretation
The data in NMR spectroscopy is presented by the chemical shift () in parts per million (ppm).
The data in NMR spectroscopy is presented by the chemical shift (δ) in parts per million (ppm).
This property is calculated from the resonance frequency of the sample (v) with a reference:
This property is calculated from the resonance frequency of the sample (v) with a reference:
𝛿
δ=
𝑣
𝑣
vsample
𝑣 − vref
vref
(4)
(4)
Membranes 2020, 10, 33
20 of 36
Membranes 2020, 10, 33
21 of 37
Routinely, tetramethylsilane (TMS) is used as a reference and it has a chemical formula of Si(CH3 )4 .
TMS hasRoutinely,
a sharp resonance
line with(TMS)
its 12 protons
Furthermore,
TMS
with most
samples.
tetramethylsilane
is used [120].
as a reference
and it
hasisainert
chemical
formula
of
TheSi(CH
chemical
shifthas
is aunique
each functional
as given[120].
in Figure
26.
3)4. TMS
sharp for
resonance
line with group
its 12 protons
Furthermore,
TMS is inert with
NMR
spectroscopy
was used
tounique
determine
the functional
functionalgroup
groups
a polyetherimide
most
samples.
The chemical
shift is
for each
as of
given
in Figure 26. thin-film
0 -oxydianiline (ODA) and trimesoyl chloride [19]. The technique
composite
membrane
made
from
4,4
NMR spectroscopy was used to determine the functional groups of a polyetherimide thin‐film
wascomposite
useful in membrane
determining
the molecular
structure of(ODA)
the composite
membrane
by detecting
compounds
made
from 4,4′‐oxydianiline
and trimesoyl
chloride
[19]. The technique
was
in determining
the molecular
structure
of with
the composite
by an
detecting
such
as useful
carboxylic
acid at 172 ppm,
phenyl rings
attached
oxygen at membrane
155 ppm, and
aromatic
compounds
such
as
carboxylic
acid
at
172
ppm,
phenyl
rings
attached
with
oxygen
at
155
ppm,
and
ring attached to amine at 146.5 ppm (Figure 27).
anFunctionalization
aromatic ring attached
to
amine
at
146.5
ppm
(Figure
27).
refers to adding a chemical group into the polymer molecules. It can enhance
Functionalization
refers to adding
chemical group
into theFor
polymer
molecules.
It can enhance
the polymer
physical, mechanical,
anda chemical
properties.
instance,
amino-functionalized
the
polymer
physical,
mechanical,
and
chemical
properties.
For
instance,
amino‐functionalized
carbon nanotubes were added to polysulfone (PSf) during the membrane preparation to improve the
carbonreactivity
nanotubes[121].
were added
to polysulfone
(PSf) during the membrane
to improve
the
polymer
It is known
that the functionalization
reactionpreparation
of amino-carbon
nanotubes
polymer reactivity [121]. It is known that the functionalization reaction of amino‐carbon nanotubes
with the polymer will result in formation of new functional groups. NMR was used to analyze
with the polymer will result in formation of new functional groups. NMR was used to analyze the
the mixed-matrix membrane and functional group such as NH2 protons was detected at 1.845 ppm.
mixed‐matrix membrane and functional group such as NH2 protons was detected at 1.845 ppm. The
The amino-benzo-crown-ether group was measured at 2.975 ppm, which was attributed to the –NH–
amino‐benzo‐crown‐ether group was measured at 2.975 ppm, which was attributed to the –NH–
bond
confirming the functionalization reaction (Figure 28).
bond confirming the functionalization reaction (Figure 28).
Figure
instrument[122].
[122].
Figure25.
25.Components
Components of
of NMR
NMR instrument
addition
studyingthe
thechemical
chemical structure,
structure, NMR
the
polymer
blend
In In
addition
totostudying
NMRcan
canbe
beused
usedtotostudy
study
the
polymer
blend
miscibility.
Two
or
more
polymers
can
be
blended
to
form
a
single
membrane.
NMP
can
be
miscibility. Two or more polymers can be blended to form a single membrane. NMP can be implemented
implemented
to study between
the interaction
between the
observing
in the in
intensity
to study
the interaction
the polymers
by polymers
observingbythe
changesthe
in changes
the intensity
chemical
in chemical shift. For instance, the miscibility between polysulfone and polyvinyl methyl ether
shift. For instance, the miscibility between polysulfone and polyvinyl methyl ether (PVME) were
(PVME) were studied at a weight percentage of 65 wt% and 35 wt%, respectively. NMR was
studied at a weight percentage of 65 wt% and 35 wt%, respectively. NMR was performed for pure
performed for pure polysulfone and PVME for comparison. The significant increase in polysulfone
polysulfone and PVME for comparison. The significant increase in polysulfone intensity in the polymer
intensity in the polymer mixture indicated a good miscibility [123].
mixture indicated a good miscibility [123].
NMR spectroscopy is an excellent tool for monitoring the membrane degradation. With time,
NMR
spectroscopy
is an excellent
monitoring the
time,
the membrane
performance
reduces tool
due for
to accumulation
of membrane
the rejecteddegradation.
molecules orWith
due to
thethe
membrane
due[124].
to accumulation
of the
rejected
or due
toaccelerate
the reaction
reaction performance
with the feed reduces
impurities
It is known that
some
acidicmolecules
environments
can
with
the
feed
impurities
[124].
It
is
known
that
some
acidic
environments
can
accelerate
membrane
membrane degradation and this can be studied by detecting the free radical peaks by NMR
[125].
degradation
andthe
this
can be studied
by detecting
radical
peaks
by NMRdegradation
[125]. Furthermore,
shiftfree
peaks
indicates
membrane
[126].
Furthermore,
disappearance
of some
chemicalthe
theFor
disappearance
of some chemical
shiftmembrane
peaks indicates
membrane
[126].
For instance, a
instance, a perfluorinated
ionomer
(Nafion®
117) has degradation
been examined
for degradation
® 117) has been examined for degradation due to fuel cell
perfluorinated
ionomer
membrane
(Nafion
due to fuel cell
operation.
It was found
that
the C–F bond was decomposed to fluoride and sulfate
ions indicating
a membrane
detection
of hydroxyl
and ions
hydroperoxyl
operation.
It was found
that thedecomposition.
C–F bond wasMoreover,
decomposed
to fluoride
and sulfate
indicating a
radicals also
confirmed Nafion®
degradation
Table 9 summarizes
the mentioned
in this
membrane
decomposition.
Moreover,
detection[125].
of hydroxyl
and hydroperoxyl
radicalsstudies
also confirmed
® for
paper
membrane[125].
analysis
by NMR.
Nafion
degradation
Table
9 summarizes the mentioned studies in this paper for membrane
analysis by NMR.
Membranes 2020, 10, 33
21 of 36
Table 9. Characterization of polymeric membranes by NMR.
Study
Membrane
Methodology
Conclusion
Ref.
Only peaks of carboxylic acid,
carboxylic-amide-carbon,
phenyl-carbon-oxygen, and
carbon-aromatic ring-amine were
detected indicating a pure
polyetherimide.
[19]
Peaks of NH2 protons and
amino-benzo-crown ether
demonstrated the functionalization of
carbon nanotubes in the polymer.
[121]
Increase in polymer intensity in the
mixture indicated a good mixing.
[123]
Detection of F− ,
SO4 −2 , and OH− indicated
membrane decomposition
[125]
Membrane
purity
Polyetherimide
Spinning carbon (13 C)
to generate the
magnetic field.
Polymer-filler
interaction
Polysulfone with
functionalized carbon
nanotubes
Dissolved samples in
deuterated chloroform.
Polymer
miscibility
Polysulfone and
polyvinyl methyl
ether(PVME)
Membrane
degradation
Perfluorinated
ionomer (Nafion® 117)
Membranes 2020, 10, 33
Membranes 2020, 10, 33
Not reported
Solid-state 19 F NMR.
Inserted samples in
rotors filled with
alcohol or water.
22 of 37
22 of 37
Figure 26. NMR chemical shift for various classes and functional groups [127].
Figure 26. NMR chemical shift for various classes and functional groups [127].
Figure 26. NMR chemical shift for various classes and functional groups [127].
Figure
27. Use
of NMR
for NMR
study the
structure
of polyetherimide
film composite membrane
[19].
Figure
27.
Use of
formolecular
study the
molecular
structure thin
of polyetherimide
thin film
Figure
27. membrane
Use of NMR
composite
[19].for study the molecular structure of polyetherimide thin film
composite membrane [19].
Membranes
2020, 10,
33 Use of NMR for study the molecular structure of polyetherimide thin film 22 of 36
Figure
27.
composite membrane [19].
Figure
NMR
analysis
a mixed
matrix
membrane
made
polysulfone
Figure
28. 28.
NMR
analysis
of aofmixed
matrix
membrane
made
fromfrom
polysulfone
(PSf)(PSf)
and and
carbon
carbon
nanotubes
showing
the
detection
of
amino‐benzo‐crown‐ether
group
at
2.975
ppm
nanotubes showing the detection of amino-benzo-crown-ether group at 2.975 ppm [121].
[121].
4.3.3. Limitations
Paramagnetic elements such as oxygen and sodium have less NMR signals [128]. This is because
those elements lose their magnetic properties after the magnet is removed. Furthermore, the quantitative
analysis of heavy elements may take hours to complete [115].
5. Elemental Composition Analysis
5.1. Energy-Dispersion X-Ray Spectroscopy
Energy-dispersion X-ray spectroscopy (EDS) is a method for detecting the elements of a sample.
EDS also gives the mass fraction of each element. EDS is considered as a “bulk” analysis technique
that covers a larger area with a higher depth of generally of 1 µm [129]. The analysis processing time is
also fast and data are displayed within seconds. EDS is commonly integrated with the SEM because
both EDS and SEM share the same electron beam. This beam will excite the atom and will cause the
release of electrons. The electrons from a higher state will move to fill in the vacancies and this will
emit X-rays to balance the energy difference between the electrons (Figure 29). When the X-rays hit the
EDS detector, pulses are created and then converted to a voltage by an analyzer in which the voltage is
proportional to the energy of the X-rays [130].
5.1.1. Sample Preparation
The prepared samples for SEM can be directly analyzed by EDS. However, if the sample was
coated due to poor conductivity, the coating material will be detected in the EDS spectrum. Therefore,
the coating material should be carefully selected so it cannot overlap with the elements found in
the sample.
5.1.2. Data Interpretation
Data of EDS analysis are presented by the intensity in counts and X-ray energy in keV. Each
element has a peak at specific X-ray energy and this information can be utilized as a reference to
pinpoint the element. EDS is a quantitative technique too and the composition can be calculated based
on the intensity. This is useful as well for determining the chemical formula of the sample. Actually,
Membranes 2020, 10, 33
23 of 36
EDS gives the data in atomic composition then the data are used to calculate the chemical formula.
For example, to calculate the formula of a compound containing two elements (A and B), the following
equation can be used:
Ax B y
Membranes 2020, 10, 33
24 of 37 (5)
where x and y are the atomic percentages determined by EDS. Now, the formula should be normalized
𝐴 𝐵
(4)
by the lowest number (assuming x is the lower number):
where x and y are the atomic percentages determined by EDS. Now, the formula should be
AB y/x
(6)
normalized by the lowest number (assuming x is the
lower number):
𝐴𝐵 /
(5)
The mass percentage can be then calculated using the atomic weight and the atomic number of
The mass percentage can be then calculated using the atomic weight and the atomic number of
thethe
elements:
elements:
x(at%) × m
x(wt%) =
𝑥 𝑤𝑡%
x
𝑎𝑡%
x(at%) 𝑥× m
at%) × m y
x + y(𝑚
𝑥 𝑎𝑡%
𝑚
𝑦 𝑎𝑡%
𝑚
(7)
(5)
where mx and my are the atomic masses of element x and y, respectively. EDS can also be used to
whereany
mx and
my are
masses
of elementdue
x and
y, respectively.
EDS can also
be used
to
monitor
changes
in the
the atomic
membrane
composition
to operation.
For example,
the sulfur
content
monitor
any
changes
in
the
membrane
composition
due
to
operation.
For
example,
the
sulfur
content
in polypiperazine-amide membrane was measured to study the membrane fouling due to coal mine
in polypiperazine‐amide
membrane
was measured
study the
due to
mine
drainage
[131]. EDS assisted
the chemical
cleaningtomethod
to membrane
restore thefouling
membrane
bycoal
measuring
the
chemical
cleaning
method
to
restore
the
membrane
by
measuring
drainage
[131].
EDS
assisted
the sulfur after the treatment. Furthermore, EDS was used to determine the chemical formula of
the sulfur after the treatment. Furthermore, EDS was used to determine the chemical formula of
nanoparticles inserted in a poly(vinyl alcohol) membrane [132]. EDS matched the added filler of
nanoparticles inserted in a poly(vinyl alcohol) membrane [132]. EDS matched the added filler of
bismuth(III) oxide with the formula of Bi2 O3 as given in Figure 30. EDS studies for characterization of
bismuth(III) oxide with the formula of Bi2O3 as given in Figure 30. EDS studies for characterization
polymeric membranes are given in Table 10.
of polymeric membranes are given in Table 10.
Figure
29.29.
Atom
excitation
of -rays
‐raysfor
forEDS
EDSanalysis
analysis[133].
[133].
Figure
Atom
excitationby
byelectron
electronand
and the
the release
release of
4.1.3. Limitations
Table 10. Characterization of polymeric membranes by EDS.
Study
Membrane
Methodology
Conclusion
Ref.
Unfortunately,
EDS cannot
give the full analysis
of polymers due to the
inability to detect
No
additional
elements
to
hydrogen. This is because hydrogen has no electrons
ininthe
k‐shell
[134]. However, EDS can still be
Cut samples
liquid
nitrogen.
Membrane purity
Polyetherimide
polyetherimide were detected
[15]
Coated samples
gold.
useful to detect elements that are usually presented
in by
polymers
such indicating
as carbon,
and
a purenitrogen,
sample.
oxygen.
Nevertheless, the quantitative analysis of EDS is more effective for
elements
with
highdue
atomic
Reduction
in sulfur
content
Restoration of a fouled
Polypiperazine-amide
Not reported
to chemical cleaning showed
[131]
membrane
by chemicalfrom
cleaning20 and above [135]. Because polymers are organic materials with mainly
mass starting
membrane restoration.
hydrogen, carbon, nitrogen,
and oxygen,
it is not suggested to use EDS to
quantify
the polymer.
Poly(vinyl
alcohol) and
The
calculated formula
matched This
Filler chemical formula
Not reported
[132]
oxide fillers
bismuth(III)
oxide.
is because the low atomic bismuth(III)
mass elements
produce weak X‐ray signals and therefore
the
composition
data might be inaccurate. The typical detection limit of EDS is 0.1 wt% or 1000 ppm. This means that
5.1.3.
EDSLimitations
cannot detect elements below that range. The general error expected from EDS is between ‐2 to
2 wt%.
Unfortunately, EDS cannot give the full analysis of polymers due to the inability to detect hydrogen.
This is because hydrogen has no electrons in the k-shell [134]. However, EDS can still be useful to detect
elements that are usually presented in polymers such as carbon, nitrogen, and oxygen. Nevertheless,
Membranes 2020, 10, 33
24 of 36
the quantitative analysis of EDS is more effective for elements with high atomic mass starting from 20
and above [135]. Because polymers are organic materials with mainly hydrogen, carbon, nitrogen,
and oxygen, it is not suggested to use EDS to quantify the polymer. This is because the low atomic
mass elements produce weak X-ray signals and therefore the composition data might be inaccurate.
The typical detection limit of EDS is 0.1 wt% or 1000 ppm. This means that EDS cannot detect elements
below
that range.
Membranes
2020, 10,The
33 general error expected from EDS is between -2 to 2 wt%.
25 of 37
3 particles
deposited
poly(vinyl
alcohol)
membrane,
Figure
30. SEM
(a) SEM
image
of2 O
Bi32O
Figure
30. (a)
image
of Bi
particles
deposited
on on
poly(vinyl
alcohol)
membrane,
(b) (b)
EDS
EDS
analysis
of
Bi
2
O
3
,
(c)
membrane
image
[132].
analysis of Bi2 O3 , (c) membrane image [132].
5.2. X-Ray Fluorescence Table 10. Characterization of polymeric membranes by EDS.
Study
Membrane
Methodology the elementalConclusion
X-ray
fluorescence (XRF)
is a method for determining
composition of a Ref.
sample
additional
elementsto
to the sample.
in wt%. Compared to EDS, XRF is a non-destructive test and can beNo
applied
directly
Membrane
Cut samples in liquid nitrogen.
polyetherimide were
Polyetherimide
[15]
Furthermore,
thesamples
need by
forgold.
a vacuum, detected
liquid samples
purity because XRF operates without
Coated
indicating acan
purebe analyzed.
In XRF, when an X-ray beam is concentrated on the sample, secondary (orsample.
fluorescent) X-rays will
a
emit Restoration
from the ofsample
due to atom excitation as demonstrated. TheReduction
high-energy
in sulfurX-ray
content source will
fouled
due
to
chemical
cleaning
cause membrane
dislodging
of an
electron from the inner
orbital shell of an atom [136]. To fill in this vacancy,
an
[131]
by
Polypiperazine‐amide
Not reported
showed membrane
electronchemical
will drop from another atom with a high-energy orbital shell [137].
This drop will cause the
restoration.
cleaning
release of
secondary or fluorescent X-ray. This emitted X-ray is unique for each element and this will
Filler chemical
Poly(vinyl alcohol) and
The calculated formula
reportedare available and they are intensively used
[132]
be used formula
for identification.
XRF handheldNot
devices
in the
bismuth(III) oxide fillers
matched bismuth(III) oxide.
analysis of metals, soils, and food [138]. Figure 31 shows the components of a portable XRF device and
electron
excitation.
5.2. X‐Ray
Fluorescence
X‐ray Preparation
fluorescence (XRF) is a method for determining the elemental composition of a sample in
5.2.1. Sample
wt%. Compared to EDS, XRF is a non‐destructive test and can be applied directly to the sample.
XRF is a fastbecause
and non-destructive
methodthe
forneed
identifying
and quantifying
elements.
The results
Furthermore,
XRF operates without
for a vacuum,
liquid samples
can be analyzed.
are usually
Most
of the samples
require
preparation.
In XRF, given
when in
an seconds.
X‐ray beam
is concentrated
on do
thenot
sample,
secondary
(or fluorescent) X‐rays will
emit from the sample due to atom excitation as demonstrated. The high‐energy X‐ray source will
5.2.2.
Datadislodging
Interpretation
cause
of an electron from the inner orbital shell of an atom [136]. To fill in this vacancy,
an
electron
willpresented
drop from
with a high‐energy
orbital
shell(keV).
[137].Each
This element
drop willwill
cause
XRF
data are
byanother
plottingatom
the intensity
(counts) with
energy
have
the
release
of
secondary
or
fluorescent
X‐ray.
This
emitted
X‐ray
is
unique
for
each
element
and
this
a unique energy peak for easier identification. Figure 32 shows some elements and their corresponding
willusing
be used
for identification.
XRF handheld
are available
and they
are intensively
energy
cadmium
as an excitation
source. devices
In addition,
the relative
amplitude
of each used
peak in
can
the
analysis
of
metals,
soils,
and
food
[138].
Figure
26
shows
the
components
of
a
portable
XRF
be used to calculate the mass percentage of the elements [139]. XRF was used to analyze adevice
doped
and electron excitation.
membrane made from polycarbonate and iron(III) chloride. The technique detected iron and chloride
with5.2.1.
the chemical
formula of FeCl3 as given in Figure 33. Furthermore, a study was performed by
Sample Preparation
XRF to determine the capability of trioctylmethylammonium thiosalicylate (TOMATS) membrane in
XRF
is a fastfrom
and non‐destructive
for identifying
and
quantifying
elements.
The results
extracting
mercury
natural waters.method
The membranes
were
immersed
inside
the contaminated
are usually given in seconds. Most of the samples do not require preparation.
5.2.2. Data Interpretation
XRF data are presented by plotting the intensity (counts) with energy (keV). Each element will
have a unique energy peak for easier identification. Figure 32 shows some elements and their
Membranes
Membranes 2020,
2020, 10,
10, 33
33
26
26 of
of 37
37
each peak can be used to calculate the mass percentage of the elements [139]. XRF was used to analyze
a
membrane
made from polycarbonate and iron(III) chloride. The technique detected iron and 25 of 36
Membranesdoped
2020, 10,
33
chloride with the chemical formula of FeCl33 as given in Figure 33. Furthermore, a study was
performed by XRF to determine the capability of trioctylmethylammonium thiosalicylate (TOMATS)
membrane
in extracting
mercuryusing
from natural
Theamounts
membranes
immersed
inside the with
water and
mercury
was measured
in-situ waters.
XRF. The
of were
mercury
was monitored
contaminated water and mercury was measured using in‐situ XRF. The amounts of mercury was
time to determine the removal efficiency [140]. Table 11 summarized the above studies for the use of
monitored with time to determine the removal efficiency [140]. Table 11 summarized the above
XRF forstudies
membrane
analysis.
for the use of XRF for membrane analysis.
Figure
31. Components
a handheld
XRF
device
measuring
the
elementalcomposition
composition[141].
Figure
31. Components
of a of
handheld
XRF
device
forfor
measuring
the
elemental
[141].
Figure 32. Identification of compounds using XRF by radioactive cadmium [139].
Figure 32. Identification of compounds using XRF by radioactive cadmium [139].
Table 11. XRF studies for membrane characterization.
Study
Chemical formula
of filler
Mercury extraction
from natural
waters
Membrane
Polycarbonate and iron
chloride filler
Trioctylmethylammonium
thiosalicylate (TOMATS)
Methodology
Conclusion
Ref.
Not reported
Elemental composition of iron
and chloride gives chemical
formula of FeCl3 that matched
the added filler.
[139]
Cut sample to disks of 1 to 3 cm
in diameter.
Palladium target X-ray tube
with beryllium window.
SPECTRA EDX software for
intensity measurements.
Mercury extraction by the
membrane was monitored by
detecting the amounts of
mercury in the polymer.
[140]
Membranes 2020, 10, 33
26 of 36
Membranes 2020, 10, 33
27 of 37
Figure
Detection
FeCl
a polycarbonatemembrane
membranefor
for hydrogen
hydrogen separation
3 in
a polycarbonate
separation[142].
[142].
Figure
33.33.
Detection
ofof
FeCl
3 in
5.2.3. Limitations
Table 11. XRF studies for membrane characterization.
XRF
is a very useful
tool for detecting the Methodology
elements and their weightConclusion
percentages. However,
the
Study
Membrane
Ref.
Elemental
composition
of
quantification data will be given for the total element, not the ions. For example, if iron is measured,
iron and chloride gives
Chemical
Polycarbonate
ironiron, not the mass percentage of each ion such as Fe+2 and Fe+3 .
XRF will
give the data
for the and
total
Not reported
chemical formula of FeCl3
[139]
of filler the difference
chloride filler
Thisformula
is because
in characteristic energy of ions is so small
that thetheXRD
detector cannot
that matched
added
filler.
distinguish [143]. Moreover, some elements cannot be detected by XRF due
to its weak fluorescent
Cut sample to disks of 1 to 3
X-ray [142]. Some examples are hydrogen, carbon, and oxygen. However, polymers are organic
cm in diameter.
Mercury extraction by the
Mercury containing mainly hydrocarbons and therefore, and XRF cannot be used to analyze the
compounds
Trioctylmethylammonium Palladium target X‐ray tube
membrane was monitored
extraction from
[140]
thiosalicylate the
(TOMATS)
withpenetrates
beryllium window.
detecting
the for
amounts
of mm making
purenatural
polymer.
Furthermore,
XRF beam
deeply insidebythe
sample
a few
waters
SPECTRA EDX software for
mercury in the polymer.
it unsuitable for measurements of thin films
of micrometers.
intensity measurements.
5.3. X-ray Photoelectron Spectroscopy
5.2.3. Limitations
X-ray photoelectron spectroscopy (XPS) is a tool for identifying and quantifying the elements.
XRF is a very useful tool for detecting the elements and their weight percentages. However, the
XPS
is
considered
a be
sensitive
surface-characterization
as it gives
theis profile
of the
quantification
dataas
will
given for
the total element, not thetechnique
ions. For example,
if iron
measured,
elemental
with
theiron,
depth
1 tomass
10 nm
[48]. Thisofiseach
favorable
for as
better
XRF willcomposition
give the dataalong
for the
total
notofthe
percentage
ion such
Fe+2 and
and accurate
Fe+3.
measurements
of
the
surface
especially
for
thin-film
membranes
[144].
Furthermore,
XPS
can
easily
This is because the difference in characteristic energy of ions is so small that the XRD detector cannot
detect
elements
thatMoreover,
EDS and some
XRF cannot
detect.
distinguish
[143].
elements
cannot be detected by XRF due to its weak fluorescent
XPS[142].
works
by sending
a beam
of X-ray to
the sample
and thisHowever,
will cause
the release
electrons
X‐ray
Some
examples
are hydrogen,
carbon,
and oxygen.
polymers
are of
organic
along
with kinetic
energymainly
[145]. A
detector is used
to measure
theXRF
number
of be
electrons
the the
kinetic
compounds
containing
hydrocarbons
and therefore,
and
cannot
used towith
analyze
pure polymer.
Furthermore,
the XRF
beamelement
penetrates
inside
the sample
forand
a few
energy
as demonstrated
in Figure
34. Each
hasdeeply
a specific
binding
energy
thismm
canmaking
be used to
it unsuitable
for measurements
of thin for
films
of micrometers.
identify
the element.
XPS is also useful
determining
the oxidation state of the element. For instance,
sulfur compounds can have different forms such as sulfide (S−2 ), sulfite (SO3)−2 , and sulfate (SO4)−2 and
5.3.can
X‐ray
Photoelectron
Spectroscopy
XPS
distinguish
between
these forms due to the difference in binding energy [146]. Furthermore,
the detection
limit in XPS isspectroscopy
similar to EDS
of 1000
ppm
X‐ray photoelectron
(XPS)
is a tool
for[147].
identifying and quantifying the elements.
XPS is considered as a sensitive surface‐characterization technique as it gives the profile of the
5.3.1.
Samplecomposition
Preparationalong with the depth of 1 to 10 nm [48]. This is favorable for better and
elemental
accurate
of thetest
surface
for thin‐film
membranes
Furthermore,
XPS
XPS ismeasurements
a non-destructive
andespecially
usually does
not require
special [144].
preparation
for polymeric
can
easily
detect
elements
that
EDS
and
XRF
cannot
detect.
2
membranes. Generally, the membrane is cut into small pieces with an area of 5 to 10 mm with sample
XPS works by sending a beam of X‐ray to the sample and this will cause the release of electrons
depth not exceeding 4 mm to fit inside the chamber.
along with kinetic energy [145]. A detector is used to measure the number of electrons with the kinetic
(SO4)‐2 and XPS can distinguish between these forms due to the difference in binding energy [146].
Furthermore, the detection limit in XPS is similar to EDS of 1000 ppm [147].
5.3.1. Sample Preparation
XPS
is10,
a 33
non‐destructive test and usually does not require special preparation for polymeric
Membranes
2020,
27 of 36
membranes. Generally, the membrane is cut into small pieces with an area of 5 to 10 mm2 with sample
depth not exceeding 4 mm to fit inside the chamber.
5.3.2. Data Interpretation
5.3.2. Data Interpretation
The data in XPS are presented by the intensity (counts) and the binding energy (eV). First, a
Theisdata
in XPS are
by the intensity
thethat,
binding
energy
(eV).will
First,
wide
wide scan
performed
topresented
find the elements
in the(counts)
sample.and
After
each
element
beascanned
scan
is
performed
to
find
the
elements
in
the
sample.
After
that,
each
element
will
be
scanned
separately to determine the functional groups and the oxidation state of the element. For instance,
separately to polysulfone
determine themembrane
functional was
groups
and theby
oxidation
state
of the voltage
element.range
For instance,
polyol-grafted
analyzed
XPS with
a wide
from 0 to
polyol‐grafted
polysulfone
membrane
was
analyzed
by
XPS
with
a
wide
voltage
range
from
0 to 700
700 eV and five elements were detected such as oxygen, nitrogen, carbon, sulfur, and chlorine
[148].
eV and five elements were detected such as oxygen, nitrogen, carbon, sulfur, and chlorine [148]. The
The carbon was again analyzed from 280 to 296 eV and a large single peak with asymmetrical shape
carbon was again analyzed from 280 to 296 eV and a large single peak with asymmetrical shape was
was detected. This peak is formed due to the combination of other smaller peaks as shown in Figure 30.
detected. This peak is formed due to the combination of other smaller peaks as shown in Figure 30.
Different chemical classes were found such as ketones (C–O), alkanes (C–H), and amines (C–N) as
Different chemical classes were found such as ketones (C–O), alkanes (C–H), and amines (C–N) as
demonstrated
in Figure 35. Furthermore, the successful grafting of polyol on polysulfone was observed
demonstrated in Figure 35. Furthermore, the successful grafting of polyol on polysulfone was
by observed
the increase
in O1s
intensity
and
the reduction
in C1s intensity
[148]. [148].
by the
increase
in O1s
intensity
and the reduction
in C1s intensity
Membranes 2020, 10, 33
29 of 37
hydrogen and helium still cannot be recognized in XPS due to the absence of core electrons [153]. In
addition,
functional
groups of
analysis
byphotoelectron
XPS couldspectroscopy
be spectroscopy
difficult due
to
peaks
Figure
34.
photoelectron
(XPS)
for
elemental
analysis
[149].
Figure
34. Components
Components
ofX-ray
X‐ray
(XPS)
foroverlap.
elemental
analysis
[149].
XPS is also a very useful tool for studying the membrane degradation. During fuel cell operation,
hydrogen peroxide is usually formed due to diffusion of oxygen across the membrane and its reaction
with hydrogen [150]. Degradation of Nafion® membrane was noticed by the decomposition of (CF2)n
backbone to fluoride and sulfate ions [151].
Quantification of elements is generally performed in XPS based on the intensity peak [152]. The
area under the curve is calculated then used to estimate the weight fraction of each element. This will
also help in determining the chemical formula of the sample similar to EDS. Table 12 summarizes the
discussed studies for membrane characterization by XPS.
5.3.3. Limitations
XPS is an excellent technique for determining the oxidation state of the elements in a sample.
Nevertheless, the analysis is only limited to solids due to the requirement of a high vacuum. XPS
analysis is also time‐consuming as it can take hours to cover a large area of the sample. Unfortunately,
Figure
35. XPS
of: (a) Polyol-grafted
polysulfone
membranes,
(a’) with the (a’)
determination
Figure
35. analysis
XPS analysis
of: (a) Polyol‐grafted
polysulfone
membranes,
with the of
carbon
compounds
[148].
determination of carbon compounds [148].
XPS is also a very useful
for studying
the membrane
degradation.
During fuel cell operation,
Table tool
12. Studies
for membrane
characterization
by XPS.
hydrogen peroxide is usually formed due to diffusion of oxygen across the membrane and its reaction
Study
Membrane
Methodology
Conclusion
Ref.
with hydrogen [150]. Degradation of Nafion® membrane was noticed by the decomposition of (CF2 )n
Monochromatic Al K X‐ray
backbone to fluoride and sulfate ions [151].
source.
O1s intensity increased while
Surface
Measurements
at 45 take‐off
C1s intensity
Quantification
of polyol‐grafted
elements is generally
performed
in XPS based
on thedecreased
intensity peak [152].
chemical
angle.
confirming the grafting of
[148]
The area under the curvepolysulfone
is calculated then used to estimate the weight fraction of each element. This
composition
Survey scan from 0 to 1000 eV
hydroxyl groups on the
will also help in determining the chemicalthen
formula
of thescans
sample
to EDS.surface.
Table 12 summarizes
high‐resolution
of similar
membrane
C1s regions.
the discussed studies for membrane characterization
by XPS.
Membrane
degradation
Nafion® 112
Monochromator with Al K
source.
Step of 0.025 eV with 100ms.
Polymer backbone was
decomposed due to detection
of fluoride and sulfate ions.
[151]
6. Conclusions
Characterization techniques are powerful tools for studying the physical structure and chemical
Membranes 2020, 10, 33
28 of 36
Table 12. Studies for membrane characterization by XPS.
Study
Membrane
Methodology
Conclusion
Ref.
[148]
[151]
Surface
chemical
composition
polyol-grafted
polysulfone
Monochromatic Al Kα X-ray source.
Measurements at 45◦ take-off angle.
Survey scan from 0 to 1000 eV then
high-resolution scans of C1s regions.
O1s intensity increased while
C1s intensity decreased
confirming the grafting of
hydroxyl groups on the
membrane surface.
Membrane
degradation
Nafion® 112
Monochromator with Al Kα source.
Step of 0.025 eV with 100ms.
Polymer backbone was
decomposed due to detection of
fluoride and sulfate ions.
5.3.3. Limitations
XPS is an excellent technique for determining the oxidation state of the elements in a sample.
Nevertheless, the analysis is only limited to solids due to the requirement of a high vacuum. XPS
analysis is also time-consuming as it can take hours to cover a large area of the sample. Unfortunately,
hydrogen and helium still cannot be recognized in XPS due to the absence of core electrons [153].
In addition, functional groups analysis by XPS could be difficult due to peaks overlap.
6. Conclusions
Characterization techniques are powerful tools for studying the physical structure and chemical
properties of the membranes. These data are used thereafter for membrane development to enhance
the performance. The basic characterization technique is the SEM where the membrane surface is
examined. TEM provides a higher resolution compared to SEM for better studies on nanomaterials.
One the other hand, AFM has a similar resolution to TEM but it can also provide some mechanical
properties such as roughness. XRD is another tool for studying the crystal structure with the ability to
name the compounds. SAXS and WAXS are more suitable for semi- and non-crystalline polymers with
more statistics on the pore and particle sizes. FTIR, on the other hand, focuses on determining the
functional groups, which can be used to identify the chemicals. Nevertheless, Raman spectroscopy
is more sensitive in detecting these functional groups and requires almost no preparation. NMR
spectroscopy provides another route for identifying the functional groups with minimum background
interference. For the elemental analysis, EDS provides a basic elemental scan with a depth of 10 µm.
XRF is a less sensitive technique for elemental analysis as the scan can reach up to 3 mm of depth,
however the analysis can be directly applied by a handheld device with no sample preparation. On the
other hand, XPS is a more advanced method for determining the elements and it is more of a surface
technique due to the scanning depth of 10 nm. XPS can also give the weight composition along with
the oxidation state of the elements.
It is clear that the selection of the characterization technique depends on the membrane nature and
the required study. However, techniques such as SEM, XRD, and EDS are considered as fundamentals
for membrane characterization. More advanced techniques can give additional information on the
interfacial surface properties, molecular structure, and chemical properties. Comparison between the
techniques in term of performed studies, advantages, and limitations is given Table 13.
Membranes 2020, 10, 33
29 of 36
Table 13. Comparison of various characterization techniques for analysis of polymeric membranes.
Technique
SEM
Performed studies
Surface topography.
Pore size.
Particle size
Membrane Thickness
Advantages
Magnification of up to 1 million.
Samples needs to be conductive.
Not accurate for measurements less 10 nm.
Images does not show topography data.
Staining the sample may be required.
Thick samples (> 100µm) are difficult to be analyzed in TEM.
TEM
Nanoparticles.
Nanofilms.
Nanopores.
Membrane thickness.
Higher resolution than SEM.
AFM
Nano-profiling.
Surface roughness.
Pore size distribution.
Membrane stiffness.
No sample preparation.
Similar resolution to TEM.
Measurements of mechanical properties.
Membrane purity.
Compounds chemical formula.
Crystal structure.
Polymer chain distance.
Limitations
Requires more processing time.
Lower depth of field.
No sample preparation.
Detection of wide range of crystalline compounds.
Heavy elements are less sensitive to XRD.
Less accuracy for small crystals.
Peaks overlap for some compounds.
Crystal structure.
Polymer-filler interaction.
Particle size distribution.
Pore size measurements.
No sample preparation.
Suitable for semi- and non-crystalline materials.
More accurate average measurements for particle size and pore size.
Scattering intensity can be weak for some systems.
Functional groups.
Polymer-solvent compatibility.
Polymer-filler interaction.
Miscibility of polymer blends.
Membrane degradation.
Detection of variety of compounds.
High sensitivity of parts per million (ppm).
Fast analysis time (in seconds).
Raman
Functional groups.
Crystal structure.
Polymer chain orientation.
Polymer blends.
Membrane fouling.
No sample preparation.
More sensitive to functional groups with better intensity peaks.
Release of fluorescent light of some samples may cause
background noise.
Polar molecules have lower Raman signal.
NMR
Functional groups.
Polymer-blend miscibility.
Membrane decomposition.
Less background interference.
Detection of polar molecules.
Liquid samples.
Paramagnetic elements have less NMR signal.
EDS
Elemental composition.
Chemical formula of fillers.
Membrane fouling.
Fast analysis time.
Samples needs to be conductive.
Limitation in detecting light elements.
Cannot quantify ions.
No sample preparation.
In-situ analysis.
Very low sensitivity to hydrogen, carbon and oxygen.
Cannot quantify ions.
Unsuitable for thin film measurements.
High sensitivity.
Quantification of ions.
Measurements of thin films of nm.
Solid samples.
Cannot detect hydrogen and helium.
Peaks Overlap for some elements.
Long processing time.
XRD
SAXS
WAXS
FTIR
XRF
XPS
Elemental composition.
Chemical formula of fillers.
Membrane degradation.
Elemental composition.
Functional groups.
Formula of chemical compounds.
Membrane degradation.
Cannot analyze aqueous samples.
Cannot detect molecules of two identical atoms.
Membranes 2020, 10, 33
30 of 36
Author Contributions: Conceptualization, data collection, and writing by Y.A.; data discussion and interpretation
by A.A.A. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
Freeman, B. Basis of permeability/selectivity tradeoff relations in polymeric gas separation membranes.
Macromolecules 1999, 32, 375–380. [CrossRef]
Goh, P.; Ismail, A.; Sanip, S.; Ng, B.; Aziz, M. Recent advances of inorganic fillers in mixed matrix membrane
for gas separation. Sep. Purif. Technol. 2011, 81, 243–264. [CrossRef]
Czanderna, A.; Madey, T.; Powell, C. Beam Effects, Surface Topography, and Depth Profiling in Surface Analysis;
Springer: New York, NY, USA, 1998.
Ziel, R.; Haus, A.; Tulke, A. Quantification of the pore size distribution (porosity profiles) in microfiltration
membranes by SEM, TEM and computer image analysis. J. Membr. Sci. 2008, 323, 241–246. [CrossRef]
Mohamad, M.; Fong, Y. Preparation of defect—Free Polysulfone membrane: Optimization of fabrication
method. J. Sci. Res. Dev. 2016, 3, 126–131.
Ghorbanpour, M.; Wani, S. Advances in Phytonanotechnology; Academic Press: Cambridge, MA, USA, 2019;
pp. 45–121.
Hawkes, P.; Reimer, L. Scanning Electron Microscopy: Physics of Image Formation and Microanalysis; Springer:
Berlin, Germany, 2013.
Fails, A.; Magee, C. Anatomy and Physiology of Farm Animals; Wiley: Hoboken, NJ, USA, 2018.
Klapetek, P. Quantitative Data Processing in Scanning Probe Microscopy: SPM Applications for Nanometrology;
Elsevier Science: Amsterdam, The Netherlands, 2013.
Aharinejad, S.; Lametschwandtner, A. Microvascular Corrosion Casting in Scanning Electron Microscopy:
Techniques and Applications; Springer: Berlin, Germany, 2012.
Sawyer, L.; Grubb, D.; Meyers, G. Polymer Microscopy; Springer: New York, NY, USA, 2008.
Morozov, O.; Bulgakov, B.; Ivanchenko, A.; Shachneva, S.; Nechausov, S.; Bermeshev, M.; Kepman, A. Data
on synthesis and characterization of sulfonated poly(phenylnorbornene) and polymer electrolyte membranes
based on it. Data Brief 2019, 27, 104626. [CrossRef]
Fleck, R.; Humbel, B. Biological Field Emission Scanning Electron Microscopy; Wiley: Hoboken, NJ, USA, 2019.
Choudhary, O.; Choudhary, P. Scanning electron microscope: Advantages and disadvantages in imaging
components. Int. J. Curr. Microbiol. Appl. Sci. 2017, 6, 1877–1882. [CrossRef]
Alqaheem, Y.; Alomair, A.; Alhendi, A.; Alkandari, S.; Tanoli, N.; Alnajdi, N.; Quesada-Peréz, A. Preparation
of polyetherimide membrane from non-toxic solvents for the separation of hydrogen from methane.
Chem. Cent. J. 2018, 12, 80. [CrossRef]
Ren, Y.; Zhu, J.; Cong, S.; Wang, J.; Van der Bruggen, B.; Liu, J.; Zhang, Y. High flux thin film nanocomposite
membranes based on porous organic polymers for nanofiltration. J. Membr. Sci. 2019, 585, 19–28. [CrossRef]
Bazhenov, S.; Borisov, I.; Bakhtin, D.; Rybakova, A.; Khotimskiy, V.; Molchanov, S.; Volkov, V. High-permeance
crosslinked PTMSP thin-film composite membranes as supports for CO2 selective layer formation.
Green Energy Environ. 2016, 1, 235–245. [CrossRef]
Gaur, G. Quantum Dot Integrated Silicon Photonic Devices for Optical Sensor Applications. Ph.D. Thesis,
Vanderbilt University, Nashville, TN, USA, 2015.
Almijbilee, M.; Wu, X.; Zhou, A.; Zheng, X.; Cao, X.; Li, W. Polyetheramide organic solvent nanofiltration
membrane prepared via an interfacial assembly and polymerization procedure. Sep. Purif. Technol. 2020, 234,
116033. [CrossRef]
Diab, M.; El-Sonbati, A.; El-Bindary, A.; Abd El-Ghany, H. Thermal stability and degradation of poly
(N-phenylpropionamide) homopolymer and copolymer of N-phenylpropionamide with methyl methacrylate.
Arabian J. Chem. 2017, 10, S3732–S3739. [CrossRef]
Alvarez, J.; Saudino, G.; Musteata, V.; Madhavan, P.; Genovese, A.; Behzad, A.; Sougrat, R.; Boi, C.;
Peinemann, K.-V.; Nunes, S. 3D analysis of ordered porous polymeric particles using complementary electron
microscopy methods. Sci. Rep. 2019, 9, 13987. [CrossRef] [PubMed]
Membranes 2020, 10, 33
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
31 of 36
Taheri, S.; Ams, M.; Bustamante, H.; Vorreiter, L.; Withford, M.; Clark, S. A practical methodology to assess
corrosion in concrete sewer pipes. MATEC Web Conf. 2018, 199, 6010. [CrossRef]
Kaufmann, E. Characterization of Materials; Wiley: Hoboken, NJ, USA, 2012.
Brunette, D.; Tengvall, P.; Textor, M.; Thomsen, P. Titanium in Medicine: Material Science, Surface Science,
Engineering, Biological Responses, and Medical Applications; Springer: New York, NY, USA, 2001.
Kellenberger, C.; Pfleiderer, F.; Raso, R.; Burri, C.; Schumacher, C.; Grass, R.; Stark, W. Limestone nanoparticles
as nanopore templates in polymer membranes: Narrow pore size distribution and use as self-wetting dialysis
membranes. RSC Adv. 2014, 4, 61420–61426. [CrossRef]
Bhowmick, A. Current Topics in Elastomers Research; CRC Press: Boca Raton, FL, USA, 2008.
Sharma, S.; Verma, D.; Khan, L.; Kumar, S.; Khan, S. Handbook of Materials Characterization; Springer
International Publishing: New York, NY, USA, 2018.
Li, Z. Industrial Applications of Electron Microscopy; CRC Press: Boca Raton, FL, USA, 2002.
Freger, V.; Gilron, J.; Belfer, S. TFC polyamide membranes modified by grafting of hydrophilic polymers: An
FT-IR/AFM/TEM study. J. Membr. Sci. 2002, 209, 283–292. [CrossRef]
Knez, S.; Stražišar, J.; Golob, J.; Horvat, A. Agglomeration of zeolite in the fluidized bed. Acta Chimica
Slovenica 2001, 48, 487–504.
Akhtar, F.; Kumar, M.; Villalobos, L.; Shevate, R.; Vovusha, H.; Schwingenschlogl, U.; Peinemann, K.-V.
Polybenzimidazole-based mixed membranes with exceptional high water vapor permeability and selectivity.
J. Mater.l Chem. A 2017, 5, 21807–21819. [CrossRef]
Hilal, N.; Ismail, A.; Matsuura, T.; Oatley-Radcliffe, D. Membrane Characterization; Elsevier Science: Amsterdam,
The Netherlands, 2017.
Erinosho, M.; Akinlabi, E.; Johnson, O. Characterization of surface roughness of laser deposited titanium
alloy and copper using AFM. Appl. Surf. Sci. 2018, 435, 393–397. [CrossRef]
Sharpe, W. Springer Handbook of Experimental Solid Mechanics; Springer: New York, NY, USA, 2008.
Kargarzadeh, H.; Ahmad, I.; Thomas, S.; Dufresne, A. Handbook of Nanocellulose and Cellulose Nanocomposites;
Wiley: Hoboken, NJ, USA, 2017.
Giordano, N. College Physics: Reasoning and Relationships; Cengage Learning: Boston, MA, USA, 2009.
Grant, C.; Twigg, P.; Bell, G.; Lu, J. AFM relative stiffness measurement of the plasticising effect of a non-ionic
surfactant on plant leaf wax. J. Colloid Interface Sci. 2008, 321, 360–364. [CrossRef]
Cahn, R.; Lifshitz, E. Concise Encyclopedia of Materials Characterization; Elsevier Science: Amsterdam,
The Netherlands, 2016.
Cohen, S.; Lightbody, M. Atomic Force Microscopy/Scanning Tunneling Microscopy 3; Springer US: New York,
NY, USA, 2007.
Mohammad, A.; Hilal, N.; Lim, Y.; Amin, I.; Raslan, R. Atomic force microscopy as a tool for asymmetric
polymeric membrane characterization. Sains Malays. 2011, 40, 237–244.
Shamohammadi, M.; Hormozi, E.; Moradinezhad, M.; Moradi, M.; Skini, M.; Rakhshan, V. Surface topography
of plain nickel-titanium (NiTi), as-received aesthetic (coated) NiTi, and aesthetic NiTi archwires sterilized by
autoclaving or glutaraldehyde immersion: A profilometry/SEM/AFM study. Int. Orthod. 2019, 17, 60–72.
[CrossRef] [PubMed]
Khulbe, K.; Feng, C.; Matsuura, T. Synthetic Polymeric Membranes: Characterization by Atomic Force Microscopy;
Springer: Berlin, Germany, 2007.
ElHadidy, A.; Peldszus, S.; Van Dyke, M. Development of a pore construction data analysis technique for
investigating pore size distribution of ultrafiltration membranes by atomic force microscopy. J. Membr. Sci.
2013, 429, 373–383. [CrossRef]
Ochoa, N.; Prádanos, P.; Palacio, L.; Pagliero, C.; Marchese, J.; Hernández, A. Pore size distributions based
on AFM imaging and retention of multidisperse polymer solutes: Characterisation of polyethersulfone UF
membranes with dopes containing different PVP. J. Membr. Sci. 2001, 187, 227–237. [CrossRef]
Arita, M.; Sakaguchi, N. Electron Microscopy: Novel Microscopy Trends; IntechOpen: London, UK, 2018.
Meng, X.; Tang, W.; Wang, L.; Wang, X.; Huang, D.; Chen, H.; Zhang, N. Mechanism analysis of membrane
fouling behavior by humic acid using atomic force microscopy: Effect of solution pH and hydrophilicity of
PVDF ultrafiltration membrane interface. J. Membr. Sci. 2015, 487, 180–188. [CrossRef]
Najafi, M.; Sadeghi, M.; Bolverdi, A.; Pourafshari Chenar, M.; Pakizeh, M. Gas permeation properties of
cellulose acetate/silica nanocomposite membrane. Adv. Polym. Technol. 2018, 37, 2043–2052. [CrossRef]
Membranes 2020, 10, 33
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
32 of 36
Ross, R. Microelectronics Failure Analysis Desk Reference, 6th ed.; ASM International: Geauga County, OH,
USA, 2011.
Mittal, V.; Matsko, N. Analytical Imaging Techniques for Soft Matter Characterization; Springer: Berlin, Germany, 2012.
Mistry, R.; Saxena, M.; Ray, P.; Singh, P. Octadecyl-silica—PVDF membrane of superior MD desalination
performance. J. Appl. Polym. Sci. 2018, 135, 46043. [CrossRef]
Yang, Q.; Li, W.; Dong, C.; Ma, Y.; Yin, Y.; Wu, Q.; Xu, Z.; Ma, W.; Fan, C.; Sun, K. PIM-1 as an artificial solid
electrolyte interphase for stable lithium metal anode in high-performance batteries. J. Energy Chem. 2020, 42,
83–90. [CrossRef]
Chowdhury, G.; Kruczek, B.; Matsuura, T. Polyphenylene Oxide and Modified Polyphenylene Oxide Membranes:
Gas, Vapor and Liquid Separation; Springer: New York, NY, USA, 2013.
Li, B.; Wang, B.; Liu, Z.; Qing, G. Synthesis of nanoporous PVDF membranes by controllable crystallization
for selective proton permeation. J. Membr. Sci. 2016, 517, 111–120. [CrossRef]
Chen, Y.; Wang, B.; Zhao, L.; Dutta, P.; Winston Ho, W. New Pebax® /zeolite Y composite membranes for
CO2 capture from flue gas. J. Membr. Sci. 2015, 495, 415–423. [CrossRef]
Powell, C.; Qiao, G. Polymeric CO2 /N2 gas separation membranes for the capture of carbon dioxide from
power plant flue gases. J. Membr. Sci. 2006, 279, 1–49. [CrossRef]
Minelli, M.; Sarti, G. Elementary prediction of gas permeability in glassy polymers. J. Membr. Sci. 2017, 521,
73–83. [CrossRef]
Saberi, M.; Rouhi, P.; Teimoori, M. Estimation of dual mode sorption parameters for CO2 in the glassy
polymers using group contribution approach. J. Membr. Sci. 2020, 595, 117481. [CrossRef]
Swallowe, G. Mechanical Properties and Testing of Polymers: An a–z Reference; Springer: Dordrecht, The Netherlands,
2013.
Ismail, A.; Khulbe, K.; Matsuura, T. Gas Separation Membranes: Polymeric and Inorganic; Springer International
Publishing: New York, NY, USA, 2015.
Weller, M. Inorganic Materials Chemistry; Oxford University Press: Oxford, UK, 1995.
Suryanarayana, C.; Norton, M. X-ray Diffraction: A Practical Approach; Springer: New York, NY, USA, 2013.
Thomas, S.; Rouxel, D.; Ponnamma, D. Spectroscopy of Polymer Nanocomposites; William Andrew Publishing:
Oxford, UK, 2016.
Zakhariev, Z. Polycrystalline Materials: Theoretical and Practical Aspects; IntechOpen: London, UK, 2012.
Shahien, M.; Yamada, M.; Yasui, T.; Fukumoto, M. Reactive atmospheric plasma spraying of AlN coatings:
Influence of aluminum feedstock particle size. J. Therm. Spray Technol. 2011, 20, 580–589. [CrossRef]
Shen, Q.; Cong, S.; He, R.; Wang, Z.; Jin, Y.; Li, H.; Cao, X.; Wang, J.; Van der Bruggen, B.; Zhang, Y.
SIFSIX-3-Zn/PIM-1 mixed matrix membranes with enhanced permeability for propylene/propane separation.
J. Membr. Sci. 2019, 588, 117201. [CrossRef]
Velu, S.; Arthanareeswaran, G.; Lade, H. Removal of organic and inorganic substances from industry
wastewaters using modified aluminosilicate-based polyethersulfone ultrafiltration membranes. Environ.
Prog. Sustain. Energy 2017, 36, 1612–1620. [CrossRef]
Sawada, H.; Takahashi, Y.; Miyata, S.; Kanehashi, S.; Sato, S.; Nagai, K. Gas transport properties and crystalline
structures of poly(lactic acid) membranes. Trans. Mater. Res. Soc. Japan 2010, 35, 241–246. [CrossRef]
Karimi, S.; Firouzfar, E.; Khoshchehreh, M. Assessment of gas separation properties and CO2 plasticization
of polysulfone/polyethylene glycol membranes. J. Pet. Sci. Eng. 2019, 173, 13–19. [CrossRef]
Lasseuguette, E.; Malpass-Evans, R.; Carta, M.; McKeown, N.; Ferrari, M.-C. Temperature and pressure
dependence of gas permeation in a microporous Tröger’s base polymer. Membranes 2018, 8, 132. [CrossRef]
Ali, A. Failure Analysis and Prevention; IntechOpen: London, UK, 2017.
Fendler, J. Nanoparticles and Nanostructured Films: Preparation, Characterization, and Applications; Wiley: Hoboken,
NJ, USA, 2008.
Mishra, S. Fibre Structure; Woodhead Publishing: Cambridge, UK, 2016.
Paradies, H. Particle size distribution and determination of characteristic properties of colloidal
bismuth—silica compounds by small-angle x-ray scattering and inelastic light scattering. Colloids Surf. A
Physicochem. Eng. Asp. 1993, 74, 57–69. [CrossRef]
Radlinski, A.; Mastalerz, M.; Hinde, A.; Hainbuchner, M.; Rauch, H.; Baron, M.; Lin, J.; Fan, L.; Thiyagarajan, P.
Application of SAXS and SANS in evaluation of porosity, pore size distribution and surface area of coal.
Int. J. Coal Geol. 2004, 59, 245–271. [CrossRef]
Membranes 2020, 10, 33
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
33 of 36
Agbabiaka, A.; Wiltfong, M.; Park, C. Small angle X-Ray scattering technique for the particle size distribution
of nonporous nanoparticles. J. Nanopartic. 2013, 2013, 11. [CrossRef]
Schneider, K.; Schöne, A.; Jun, T.-S.; Korsunsky, A. Investigation of changes in crystalline and amorphous
structure during deformation of nano-reinforced semi-crystalline polymers by space-resolved synchrotron
SAXS and WAXS. Procedia Eng. 2009, 1, 159–162. [CrossRef]
Niu, Q.; Pan, J.; Jin, Y.; Wang, H.; Li, M.; Ji, Z.; Wang, K.; Wang, Z. Fractal study of adsorption-pores in
pulverized coals with various metamorphism degrees using N2 adsorption, X-ray scattering and image
analysis methods. J. Pet. Sci. Eng. 2019, 176, 584–593. [CrossRef]
Baran, M.; Braten, M.; Sahu, S.; Baskin, A.; Meckler, S.; Li, L.; Maserati, L.; Carrington, M.; Chiang, Y.-M.;
Prendergast, D.; et al. Design rules for membranes from polymers of intrinsic microporosity for crossover-free
aqueous electrochemical devices. Joule 2019, 3, 2968–2985. [CrossRef]
Rueda, D.; García-Gutiérrez, M.; Nogales, A.; Capitán, M.; Ezquerra, T.; Labrador, A.; Fraga, E.; Beltrán, D.;
Juanhuix, J.; Herranz, J.; et al. Versatile wide angle diffraction setup for simultaneous wide and small angle
x-ray scattering measurements with synchrotron radiation. Rev. Sci. Instru. 2006, 77, 033904. [CrossRef]
De Prado, S.A.L.; Ponce, M.; Funari, S.; Schulte, K.; Garamus, V.; Willumeit, R.; Nunes, S. SAXS/WAXS
characterization of proton-conducting polymer membranes containing phosphomolybdic acid. J. Non-Cryst.
Solids 2005, 351, 2194–2199. [CrossRef]
Mrozowich, T.; McLennan, S.; Overduin, M.; Patel, T. Structural studies of macromolecules in solution using
small angle x-ray scattering. JoVE 2018, 141, e58538. [CrossRef]
Kikhney, A.; Svergun, D. A practical guide to small angle X-ray scattering (SAXS) of flexible and intrinsically
disordered proteins. FEBS Lett. 2015, 589, 2570–2577. [CrossRef]
Tant, M.; Mauritz, K.; Wilkes, G. Ionomers: Synthesis, Structure, Properties and Applications; Springer: Dordrecht,
The Netherlands, 1997.
Xiao, T.; Yuan, H.; Ma, Q.; Guo, X.; Wu, Y. An approach for in situ qualitative and quantitative analysis of
moisture adsorption in nanogram-scaled lignin by using micro-FTIR spectroscopy and partial least squares
regression. Int. J. Biol. Macromol. 2019, 132, 1106–1111. [CrossRef]
Wypych, G. Handbook of Solvents; Noyes Publications: Park Ridge, NJ, USA, 2001.
Sun, D. Modern Techniques for food Authentication; Elsevier Science: Amsterdam, The Netherlands, 2008.
Bouis, P. Reagent Chemicals: Specifications and Procedures; American Chemical Society Specifications: Washington,
DC, USA, 2006.
Mishra, M. Encyclopedia of Polymer Applications; CRC Press: Boca Raton, FL, USA, 2018.
Townshend, A. Encyclopedia of Analytical Science: Gast-Lip; Academic Press: Cambridge, MA, USA, 1995.
Dalton, D. Foundations of Organic Chemistry: Unity and Diversity of Structures, Pathways, and Reactions; Wiley:
Hoboken, NJ, USA, 2011.
Bonilla, J.; Srivatsa, G. Handbook of Analysis of Oligonucleotides and Related Products; CRC Press: Boca Raton,
FL, USA, 2011.
Hansen, S.; Pedersen-Bjergaard, S.; Rasmussen, K. Introduction to Pharmaceutical Chemical Analysis; Wiley:
Hoboken, NJ, USA, 2011.
Adewole, J.; Ahmad, A.; Ismail, S.; Leo, C.; Sultan, A. Comparative studies on the effects of casting solvent on
physico-chemical and gas transport properties of dense polysulfone membrane used for CO2 /CH4 separation.
J. Appl. Polym. Sci. 2015, 132, 42205. [CrossRef]
Zhu, H.; Wang, L.; Jie, X.; Liu, D.; Cao, Y. Improved interfacial affinity and CO2 separation performance of
asymmetric mixed matrix membranes by incorporating postmodified MIL-53(Al). ACS Appl. Mater. Interfaces
2016, 8, 22696–22704. [CrossRef] [PubMed]
Grdadolnik, J. ATR-FTIR spectroscopy: Its advantages and limitations. Acta Chimica Slovenica 2002, 49,
631–642.
Smith, B. Fundamentals of Fourier Transform Infrared Spectroscopy; Taylor & Francis: Abingdon-on-Thames,
UK, 1995.
Winter, A. Organic Chemistry I for Dummies; Wiley: Hoboken, NJ, USA, 2014.
Zhang, S. Raman Spectroscopy and its Application in Nanostructures; Wiley: Hoboken, NJ, USA, 2012.
Korb, C. Differential Absorption Lidars for Remote Sensing of Atmospheric Pressure and Temperature Profiles;
National Aeronautics and Space Administration: Washington, DC, USA, 1995.
Membranes 2020, 10, 33
34 of 36
100. Correia, J.; Detrich, H. Biophysical Tools for Biologists: In Vivo Techniques; Elsevier Science: Amsterdam,
The Netherlands, 2009.
101. Boccaccio, T.; Bottino, A.; Capannelli, G.; Piaggio, P. Characterization of PVDF membranes by vibrational
spectroscopy. J. Membr. Sci. 2002, 210, 315–329. [CrossRef]
102. Andrews, D.; Nann, T.; Lipson, R. Comprehensive Nanoscience and Nanotechnology; Elsevier Science: Amsterdam,
The Netherlands, 2019.
103. Amer, M. Raman Spectroscopy for Soft Matter Applications; Wiley: Hoboken, NJ, USA, 2009.
104. Katoh, T.; Imamura, G.; Obata, S.; Bhanuchandra, M.; Copley, G.; Yorimitsu, H.; Saiki, K. The influence of source
molecule structure on the low temperature growth of nitrogen-doped graphene. Phys. Chem. Chem. Phys. 2015,
17, 14115–14121. [CrossRef] [PubMed]
105. Jacox, M. Vibrational and electronic energy levels of polyatomic transient molecules. Supplement B. J. Phys.
Chem. Ref. Data 2003, 32, 1–441. [CrossRef]
106. Grumezescu, A. Nanostructures for the Engineering of Cells, Tissues and Organs: From Design to Applications;
Elsevier Science: Amsterdam, The Netherlands, 2018.
107. Asundi, A.K.; Rao, V.V. Measurement of Methanol Proportion in Methanol-Gasoline Mixtures—An Application of
Fiber Optic Raman Spectroscopy; SAE International: Warrendale, PA, USA, 2000.
108. Wild, D. The Immunoassay Handbook: Theory and Applications of Ligand Binding, ELISA and Related Techniques;
Elsevier Science: Amsterdam, The Netherlands, 2013.
109. Pathak, Y.; Lokhande, J. Handbook of Metallonutraceuticals; Taylor & Francis: Abingdon, UK, 2014.
110. Schreiner, M.; Strlič, M.; Salimbeni, R. Handbook of the Use of Lasers in Conservation and Conservation Science;
COST Office: Brussels, Belgium, 2008.
111. Holber, M.; Johansson, P.; Jacobsson, P. Raman spectroscopy of an aged low temperature polymer electrolyte
fuel cell membrane. Fuel Cells 2011, 11, 459–464. [CrossRef]
112. Sundarapandiyan, S.; Renitha, T.; Sridevi, J.; Saravanan, P.; Chandrasekaran, B.; Raju, G. Photocatalytic
degradation of highly refractive phenolic polymer—mechanistic insights as revealed by electron spin
resonance (ESR) and solid-state 13 C NMR spectroscopy. Chem. Eng. J. 2017, 313, 1112–1121. [CrossRef]
113. Sangroniz, A.; Gonzalez, A.; Martin, L.; Irusta, L.; Iriarte, M.; Etxeberria, A. Miscibility and degradation of
polymer blends based on biodegradable poly(butylene adipate-co-terephthalate). Polym. Degrad. Stab. 2018,
151, 25–35. [CrossRef]
114. Ebbing, D.; Gammon, S. General Chemistry, 10th ed.; Cengage Learning: Boston, MA, USA, 2012.
115. Emwas, A.; Roy, R.; McKay, R.; Tenori, L.; Saccenti, E.; Gowda, G.; Raftery, D.; Alahmari, F.; Jaremko, L.;
Jaremko, M.; et al. NMR spectroscopy for metabolomics research. Metabolites 2019, 9, 7. [CrossRef]
116. Claridge, T.; Claridge, N. High-Resolution NMR Techniques in Organic Chemistry; Elsevier Science: Amsterdam,
The Netherlands, 2009.
117. Pavia, D.; Kriz, G.; Lampman, G.; Engel, R. A Microscale Approach to Organic Laboratory Techniques; Cengage
Learning: Boston, MA, USA, 2016.
118. Kobayashi, Y.; Nakamitsu, Y.; Zheng, Y.; Takashima, Y.; Yamaguchi, H.; Harada, A. Preparation of cyclodextrinbased porous polymeric membrane by bulk polymerization of ethyl acrylate in the presence of cyclodextrin.
Polymer 2019, 177, 208–213. [CrossRef]
119. Mueller, R.; Zhang, S.; Zhang, C.; Lively, R.; Vasenkov, S. Relationship between long-range diffusion and
diffusion in the ZIF-8 and polymer phases of a mixed-matrix membrane by high field NMR diffusometry.
J. Membr. Sci. 2015, 477, 123–130. [CrossRef]
120. Becker, E. High Resolution NMR: Theory and Chemical Applications; Elsevier Science: Amsterdam, The Netherlands,
1999.
121. Nechifor, A.; Panait, V.; Naftanaila, L.; Batalu, D.; Voicu, Ş. Symmetrically polysulfone membranes obtained
by solvent evaporation using carbon nanotubes as additives. Synthesis, characterization and applications.
Dig. J. Nanomater. Biostruc. 2013, 8, 875–884.
122. Reusch, W. Nuclear Magnetic Resonance Spectroscopy; LibreTexts: Davis, CA, USA, 2019.
123. Parmer, J.; Dickinson, L.; Chien, J.; Porter, R. Polymer-polymer miscibility determination via CP-MAS NMR
in blends of deuteriated and protonated polymers. Macromol. 1987, 20, 2308–2310. [CrossRef]
124. Zatoń, M.; Rozière, J.; Jones, D.J. Current understanding of chemical degradation mechanisms of
perfluorosulfonic acid membranes and their mitigation strategies: A review. Sustain. Energy Fuels 2017, 1,
409–438. [CrossRef]
Membranes 2020, 10, 33
35 of 36
125. Kinumoto, T.; Inaba, M.; Nakayama, Y.; Ogata, K.; Umebayashi, R.; Tasaka, A.; Iriyama, Y.; Abe, T.; Ogumi, Z.
Durability of perfluorinated ionomer membrane against hydrogen peroxide. J. Power Sources 2006, 158,
1222–1228. [CrossRef]
126. Parrondo, J.; Arges, C.; Niedzwiecki, M.; Anderson, E.; Ayers, K.; Ramani, V. Degradation of anion exchange
membranes used for hydrogen production by ultrapure water electrolysis. RSC Adv. 2014, 4, 9875–9879.
[CrossRef]
127. McMurry, J. Organic Chemistry; Cengage Learning: Boston, MA, USA, 2011.
128. Kong, X.; Terskikh, V.; Khade, R.; Yang, L.; Rorick, A.; Zhang, Y.; He, P.; Huang, Y.; Wu, G. Solid-state 17 O
NMR spectroscopy of paramagnetic coordination compounds. Angew. Chem. 2015, 54, 4753–4757. [CrossRef]
129. Moore, T. Characterization of Integrated Circuit Packaging Materials; Elsevier Science: Amsterdam, The Netherlands,
2013.
130. Xiong, Y.; Lu, X. Metallic Nanostructures: From Controlled Synthesis to Applications; Springer International
Publishing: New York, NY, USA, 2014.
131. Wadekar, S.; Vidic, R. Comparison of ceramic and polymeric nanofiltration membranes for treatment of
abandoned coal mine drainage. Desalination 2018, 440, 135–145. [CrossRef]
132. Ayesh, A.; Salah, B.; Al-Sulaiti, L. Production and characterization of flexible semiconducting polymernanoparticle composites for x-ray sensors. Radiat. Phys. Chem. 2019, 167, 108233. [CrossRef]
133. Thermo Fisher Scientific. EDX Analysis with SEM: How Does it Work? Available online: https://www.
thermofisher.com/blog/microscopy/edx-analysis-with-sem-how-does-it-work/ (accessed on 1 November 2019).
134. Saxena, P. IIT Chemistry-I; Krishna Prakashan: Uttar Pradesh, India, 1995.
135. Russ, J.; Frs, M.; Kiessling, R.; Charles, J. Fundamentals of Energy Dispersive X-ray Analysis: Butterworths
Monographs in Materials; Elsevier Science: Amsterdam, The Netherlands, 2013.
136. Taylor, P.; Yurdakul, E.; Ceylan, H. Concrete Pavement Mixtures Design and Analysis: The Application of Portable
X-ray Fluorescence Technique to Assess Concrete Mix Proportions; National Concrete Pavement Technology
Center: Ames, IA, USA, 2012.
137. Verma, H. Atomic and Nuclear Analytical Methods: XRF, Mössbauer, XPS, NAA and Ion-Beam Spectroscopic
Techniques; Springer: Berlin, Germany, 2007.
138. Shugar, A.; Mass, J. Handheld XRF for Art and Archaeology; Leuven University Press: Leuven, Belgium, 2012.
139. Crain, E. Crain’s Petrophysical Handbook; Spectrum 2000 Mindware: Rocky Mountain House, AB, Canada, 2010.
140. Elias, G.; Marguí, E.; Díez, S.; Fontàs, C. Polymer inclusion membrane as an effective sorbent to facilitate
mercury storage and detection by x-ray fluorescence in natural waters. Anal. Chem. 2018, 90, 4756–4763.
[CrossRef]
141. Thermo Fisher Scientific. The Power Behind the Analyzer: X-ray Tubes. Available online: https://www.
thermofisher.com/blog/metals/the-power-behind-the-analyzer-x-ray-tubes (accessed on 24 January 2020).
142. Acharya, N.; Kulshrestha, V.; Awasthi, K.; Jain, A.; Singh, M.; Vijay, Y. Hydrogen separation in doped and
blend polymer membranes. Int. J. Hydrogen Energy 2008, 33, 327–331. [CrossRef]
143. Panchuk, V.; Rabdano, N.; Goidenko, A.; Grebenyuk, A.; Irkaev, S.; Semenov, V. Determination of the
oxidation state of iron by x-ray fluorescence spectroscopy using chemometric approaches. J. Anal. Chem.
2017, 72, 662–670. [CrossRef]
144. Weidner, J. Separators and Membranes for Batteries, Capacitors, Fuel Cells, and Other Electrochemical Systems;
Electrochemical Society: Pennington, NJ, USA, 2009.
145. Hüfner, S. Photoelectron Spectroscopy: Principles and Applications; Springer: New York, NY, USA, 2003.
146. Aulisa, E.; Gilliam, D. A Practical Guide to Geometric Regulation for Distributed Parameter Systems; CRC Press:
Boca Raton, FL, USA, 2015.
147. Shard, A. Detection limits in XPS for more than 6000 binary systems using Al and Mg Kα X-rays. Surf. Interface
Anal. 2014, 46, 175–185. [CrossRef]
148. Du, X.; Meng, J.; Xu, R.; Shi, Q.; Zhang, Y. Polyol-grafted polysulfone membranes for boron removal: Effects
of the ligand structure. J. Membr. Sci. 2015, 476, 205–215. [CrossRef]
149. Van der Heide, P. X-ray Photoelectron Spectroscopy: An Introduction to Principles and Practices; Wiley: Hoboken,
NJ, USA, 2011.
150. Futter, G.A.; Latz, A.; Jahnke, T. Physical modeling of chemical membrane degradation in polymer electrolyte
membrane fuel cells: Influence of pressure, relative humidity and cell voltage. J. Power Sources 2019, 410–411,
78–90. [CrossRef]
Membranes 2020, 10, 33
36 of 36
151. Chen, C.; Levitin, G.; Hess, D.; Fuller, T. XPS investigation of Nafion® membrane degradation. J. Power Sources
2007, 169, 288–295. [CrossRef]
152. Riviere, J.; Myhra, S. Handbook of Surface and Interface Analysis: Methods for Problem-Solving; CRC Press:
Boca Raton, FL, USA, 2009.
153. Stojilovic, N. Why can’t we see hydrogen in x-ray photoelectron spectroscopy? J. Chem. Edu. 2012, 89, 1331–1332.
[CrossRef]
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).