Polymer hybrid thin-film composites with tailored permeability and anti-fouling performance. Juha Nikkola. VTT Science 66
Economic pressures and competition have created a need for superior performance and multifunctional properties of materials. The ever-increasing demand for new product properties makes it challenging to achieve the required features when using traditional materials. Therefore, new material combinations (i.e. hybrid or composite materials) are intended to meet these challenges. Composites and hybrid materials are new material combinations which can provide added value for existing products or create novel multifunctional properties. This thesis aimed at fabricating and modifying thin-film composites (TFC) by using various coating technologies. Moreover, the target was to tailor the permeability or to create anti-fouling performance. Inorganic, inorganic-organic and organic coating layers were made by atmospheric plasma deposition, sol-gel, atomic-layerdeposition (ALD) or polyvinyl alcohol (PVA) dispersion coating methods. Coatings were deposited using either roll-to-roll or batch process. The scientifically novel contribution of this thesis is as follows. The thesis contributes to the development of novel TFC structures using various coating methods. The novel TFC structures provide new alternatives or improvements for the existing barrier materials in food packaging. In addition, the developed anti-fouling coatings tackle the critical challenge of membrane fouling and introduce new low-fouling TFC membranes for water treatment.
Published on: Mar 4, 2016
Transcripts - Polymer hybrid thin-film composites with tailored permeability and anti-fouling performance. Juha Nikkola. VTT Science 66
VISIONS• S CI ENCE •TECHNOLOGY•RESE ARCH HIGHLIGHTS
Polymer hybrid thin-film
composites with tailored
permeability and anti-fouling
VTT SCIENCE 66
Polymer hybrid thin-film
composites with tailored
permeability and anti-fouling
Thesis for the degree of Doctor of Science to be presented with due
permission for public examination and criticism in Auditorium RG202,
at Tampere University of Technology (Tampere, Finland), on the 31st
October, 2014, at 12 noon.
ISBN 978-951-38-8163-4 (Soft back ed.)
ISBN 978-951-38-8164-1 (URL: http://www.vtt.fi/publications/index.jsp)
VTT Science 66
ISSN 2242-119X (Print)
ISSN 2242-1203 (Online)
Copyright © VTT 2014
JULKAISIJA – UTGIVARE – PUBLISHER
PL 1000 (Tekniikantie 4 A, Espoo)
Puh. 020 722 111, faksi 020 722 7001
PB 1000 (Teknikvägen 4 A, Esbo)
Tfn +358 20 722 111, telefax +358 20 722 7001
VTT Technical Research Centre of Finland
P.O. Box 1000 (Tekniikantie 4 A, Espoo)
FI-02044 VTT, Finland
Tel. +358 20 722 111, fax +358 20 722 7001
Cover image: Marjo Ketonen and Mari Raulio
Grano Oy, Kuopio 2014
This thesis was produced at VTT Technical Research Centre of Finland. Some of
the experimental parts were done by using the facilities at Department of the Ma-terial
Science at Tampere University of Technology in Finland, at Nanyang Tech-nological
University in Singapore and at Singapore Membrane Technology Centre
The work was part of research projects PLASTEK1 (2005–2007), PLASTEK2
(2007–2009) and FRONTWATER (2010–2015) funded by Tekes – the Finnish
funding agency for technology and innovation, VTT and Finnish industry.
I am sincerely grateful to my supervisor, Professor Jyrki Vuorinen from the De-partment
of Materials Science at Tampere University of Technology for the enor-mous
time and effort spent in guidance and advice. I also want to express my
gratitude to my instructor at VTT, Professor Chuyang Y. Tang, for valuable guid-ance
and introduction to membrane technologies. My sincere thanks go to the pre-examiners
for their valuable comments and advice.
I am deeply grateful to my current and former colleagues at VTT and especially,
to Dr. Kalle Nättinen, Mr. Juha Mannila, Ms. Riitta Mahlberg, Ms. Marjo Ketonen,
Ms. Sini Eskonniemi, Mrs. Ulla Kanerva, Mrs Lisa Wikström, Mr. Jarmo Siivinen,
Dr. Hanna-Leena Alakomi, Dr. Mari Raulio, Dr. Hanna Kyllönen and Mr. Pekka
Taskinen. Furthermore, I am grateful to my current and former team leaders Dr.
Marke Kallio, Mr. Amar Mahiout, Mr. Tomi Lindroos, Mr. Janne Hulkko and tech-nology
managers Dr. Erja Turunen, Dr. Aino Helle, Dr. Tarja Laitinen, Dr. Tuula-mari
Helaja at VTT. Also, sincere thanks go to Dr. Jing Wei, Dr. Yining Wang and
Mr. Xin Liu at Nanyang Technological University. I would also like to thank all my
co-authors who have made contributions to the included publications.
Finally, I would like to thank my parents Maria-Lea and Heikki Nikkola as well
as my sister Kirsi for all the support and care needed to undertake this journey.
Most of all, I would like to thank my wife Jonna for giving all the strength, faith and
love, which was really needed. My lovely son Jaki, you are the reason I go on.
Espoo, October 2014
Supervisor Professor Jyrki Vuorinen
Department of Materials Science,
Tampere University of Technology, Finland
Instructor Professor Chuyang Y. Tang
VTT Technical Research Centre of Finland, Finland
Reviewers Professor Mika Mänttäri
Lappeenranta University of Technology, Finland
Dr. Dirk Vangeneugden
Opponents Professor Mika Mänttäri
Lappeenranta University of Technology, Finland
Professor José María Kenny
University of Perugia, Italy
List of publications
This thesis is based on the following original publications which are referred to in
the text as [P1–P6].The publications are reproduced with kind permission from the
[P1] K. Nättinen, J. Nikkola, H. Minkkinen, P. Heikkilä, J. Lavonen, M. Tuominen,
Reel-to-reel inline atmospheric plasma deposition of hydrophobic coatings,
Journal of Coatings Technology and Research, 8 (2011) 2, 237–245.
[P2] J. Nikkola, Raising the barriers, European Coatings Journal, 1 (2011) 41–47.
[P3] T. Hirvikorpi, M. Vähä-Nissi, J. Nikkola, A. Harlin, M. Karppinen, Thin Al2O3
barrier coatings onto temperature-sensitive packaging materials by atomic
layer deposition, Surface and Coatings Technology, 205 (2011) 21–22,
[P4] J. Nikkola, H.-L. Alakomi, C. Y. Tang, Bacterial anti-adhesion of coated and
uncoated thin film composite (TFC) polyamide (PA) membranes, Journal of
Coating Science and Technology, 1 (2014) 1–7.
[P5] J. Nikkola, X. Liu, Y. Li, M. Raulio, H.-L. Alakomi, J. Wei, C. Y. Tang, Surface
modification of thin film composite RO membrane for enhanced anti-biofouling
performance, Journal of Membrane Science, 444 (2013) 192–200.
[P6] J. Nikkola, J. Sievänen, M. Raulio, J. Wei, J. Vuorinen, C. Y. Tang, Surface
modification of thin film composite polyamide membrane using atomic layer
deposition method, Journal of Membrane Science, 450 (2014) 174–180.
[P1] Contributed to the coating trials, conducted the AFM characterisation, su-pervised
and analysed the characterisation results (SEM and FTIR) and
was the co-author in the manuscript.
[P2] Designed and synthesised the sol-gel coatings, planned and conducted the
coating trials, applied the coatings on the selected substrates, conducted
the AFM characterisation, supervised and analysed the other characterisa-tion
results (SEM, contact angle of oleic acid and OTR) and was the main
author of the manuscript.
[P3] Designed and conducted the synthesis of sol-gel hybrid coatings, planned
the lab-scale sol-gel coating trials, applied the sol-gel coatings on the se-lected
substrates, conducted the AFM characterisation, analysed the char-acterisation
results (water contact angle, OTR, WVTR) and was the co-author
of the manuscript.
[P4] Conducted the AFM characterisation, supervised and analysed the other
characterisation results (water contact angle, surface energy and microbio-logical
studies) and was the main author of the manuscript.
[P5] Designed and fabricated the PVA coatings, planned the lab-scale coating
trials, applied the coatings on the selected substrates, conducted the AFM
and FTIR characterisation, supervised and analysed the other characteri-sation
results (SEM, water contact angle, surface energy measurements,
microbiological studies, separation studies) and was the main author of the
[P6] Designed and supervised the fabrication of ALD coatings, planned the lab-scale
coating trials, conducted the AFM and FTIR characterisation, super-vised
and analysed the other characterisation results (SEM, water contact
angle, surface energy measurements, microbiological studies, separation
studies) and was the main author of the manuscript.
Preface .................................................................................................................. 3
Academic dissertation ......................................................................................... 4
List of publications .............................................................................................. 5
Author’s contributions ........................................................................................ 6
List of abbreviations and symbols ..................................................................... 9
1. Introduction ................................................................................................. 13
2. Advanced surface modification of polymeric substrates ........................ 16
2.1 Atmospheric plasma pre-treatment of polymers ................................... 16
2.2 Atmospheric plasma deposition of thin coatings on polymers .............. 18
2.3 Hybrid coatings synthesised using the sol-gel route ............................ 19
2.4 Atomic-layer-deposition of inorganic thin coatings on polymers ........... 21
3. Non-permeable packaging films ................................................................ 23
3.1 Permeability ......................................................................................... 23
3.2 Traditional packaging materials ........................................................... 24
3.3 Developments of barrier films for packaging application ...................... 25
4. Semi-permeable thin-film composite membranes ................................... 27
4.1 Reverse osmosis (RO) membranes ..................................................... 27
4.2 Permeability of reverse osmosis (RO) membranes .............................. 27
4.3 Fouling of reverse osmosis (RO) membranes ...................................... 29
4.4 Development of low-fouling thin-film composite membranes ............... 30
5. Aims of the study ........................................................................................ 33
6. Experimental procedures ........................................................................... 35
6.1 Substrate materials .............................................................................. 35
6.2 Atmospheric plasma deposition of SiOx coating .................................. 35
6.3 Synthesis and processing of sol-gel hybrid coatings ............................ 36
6.3.1 Sol-gel-coated LDPE-board ...................................................... 37
6.3.2 Sol-gel-coated PLA-board ......................................................... 37
6.4 ALD processing of Al2O3 coatings ........................................................ 38
6.4.1 ALD deposition on sol-gel-coated PLA-board ........................... 38
6.4.2 ALD deposition on TFC-PA ....................................................... 38
6.5 Preparation and processing of PVA and PVA-PHMG coatings ............ 38
6.6 Surface characterisation ...................................................................... 39
6.7 Anti-fouling performance ...................................................................... 40
6.8 Barrier performance and permeability .................................................. 41
7. Results and discussion .............................................................................. 43
7.1 Fabrication of thin-film composites with tailored permeability .............. 43
7.1.1 Atmospheric plasma deposition of SiOx film on LDPE-board ...... 43
7.1.2 Atmospheric plasma-assisted sol-gel hybrid coating on
LDPE-board .............................................................................. 45
7.1.3 ALD and sol-gel based thin-film composite on PLA-board ........ 50
7.2 Surface modification of thin-film composites with anti-fouling
performance ......................................................................................... 52
7.2.1 Characterisation of uncoated and coated thin-film composites ...... 52
7.2.2 PVA- and PVA-PHMG-coated thin-film composites .................. 54
7.2.3 ALD-coated thin-film composites .............................................. 64
8. Conclusions................................................................................................. 71
References ......................................................................................................... 74
List of abbreviations and symbols
Al2O3 Aluminium oxide
APD Atmospheric plasma deposition
BO Biaxial orientation
BOPP Biaxially oriented polypropylene
CA Cellulose acetate
CFU Colony-forming unit
DBD Dielectric barrier discharge
DIZ Diffusion inhibition zone
EPS Extracellular polymer substances
EVOH Ethylene vinyl alcohol
MIP Microwave induced plasma
NaCl Sodium chloride
LBL Layer-by-layer deposition
LDPE Low-density polyethylene
OTR Oxygen transmission rate
PECVD Plasma enhanced chemical vapour deposition
PET Polyethylene terephthalate
PEG Polyethylene glycol
PHMG Polyhexamethylene guanidine hydrochloride
PLA Polylactic acid
PVA Polyvinyl alcohol
PVDC Polyvinylidene chloride
RRMS Root-mean-square roughness
RO Reverse osmosis
SiOx Silicon oxide
TFC Thin-film composite
WVTR Water vapour transmission rate
WV Water vapour
P Permeability coefficient
S Solubility coefficent
D Diffusion coefficient
q Quantity of permeant
Δp Partial pressure difference
Jv Water flux
Qp Volumetric flow rate of permeate
Am. Membrane area
P Applied pressure
π Osmotic pressure
Rm Hydraulic resistance of the membrane
Rg Universal gas constant (Rg = 8.31 J/mol K)
C Molar concentration
Cf Concentration of feed solution
Cp Concentration of permeate solution
R Rejection of a membrane
A Water permeability coefficient
B Solute permeability coefficent
Rf Hydraulic resistance caused by fouling
Rc Hydraulic resistance caused by coating
Economic pressures and competition have created a need for superior perfor-mance
and multifunctional properties of materials. The ever-increasing demand for
new product properties makes it challenging to achieve the required features when
using traditional materials. Therefore, new material combinations (i.e. hybrid or
composite materials) are intended to meet these challenges, providing added
value for the existing products or creating novel spinoffs.
The difference between hybrid and composite materials is narrow, and in some
cases the terms overlap. Hybrid material can be defined as a combination of mate-rials,
which are composed of at least two different materials, usually inorganic and
organic . Indeed, hybrid material is typically a blend, multilayer or nanostruc-tured
material. Moreover, composite is a consolidation of materials, in which the
matrix material is further enhanced using particles, fibres or coatings.
In recent years, significant progress has been made in the choice of materials
and in the manufacturing routes for nanocomposites. Totally new properties can
be obtained using only nanoscale fillers . For example, carbon nanotubes and
fibres have been studied for reinforcement in nanocomposites . Among the
nanocomposite research, the modification of bulk material has gained most atten-tion.
Nevertheless, the fabrication of such materials in the form of thin films can
also provide special features. For example, a thin-film composite (TFC) structure
can be expected to tailor the permeability of polymeric films applied in various
industrial applications, ranging from non-permeable packaging to semi-permeable
membranes. TFCs usually have a three layer structure, containing a support, a
core and a skin layer.
Permeation is described as diffusion of a permeate (e.g. liquid, gas, ions or par-ticles)
from a high concentration to a low concentration . Permeability is the
measured material permeation property, describing the penetration of permeate
through a solid interface . Tailored permeability of a material usually indicates
the creation of a barrier against a permeate . These barrier layers can be semi-or
non-permeable structures. The former is a mesoporous bulk material or coating
layer, which rejects the desired gases, vapours, liquids, ions or particles, and is
permeable for the targeted substance, while the latter is a nonporous bulk material
or a coating layer, which rejects the targeted substances.
Food and pharmaceutical packaging is a typical example, where non-permeable
features are desired. Food packaging is needed to protect the items of food from
various exposures. New developments aim to improve the barrier properties or to
create antimicrobial activity to prevent the microbiological spoilage of food .
Furthermore, recyclable and renewable materials are considered to increase the
sustainability of packaging by replacing the existing metal or polymer coatings.
Therefore, various new polymers, biopolymers, nanoparticles and coatings have
been studied . Another recent trend in packaging material development is leading
towards high barrier performance with a minimal amount of material . Therefore,
there is a need for functional coatings on paper, paperboard and plastic laminates
for special or enhanced properties in the field of paper converting.
A water treatment membrane is used in various filtration processes to provide
purified water or other liquids. A membrane is a typical example of a semi-permeable
structure . The membranes used in reverse osmosis (RO) and nano-filtration
(NF) have a nearly nonporous structure . The commercial RO and NF
membranes are based on either cellulose acetate (CA) or polyamide (PA) TFC.
The latter is composed of polyethersulfone (PES) cast on non-woven as a core
and a support layer, which is further coated with PA as a skin layer. The fouling of
membranes, such as colloidal fouling (inorganic and organic) or biofouling, is a
critical issue in pressure-driven membrane processes . The routes for low-fouling
membranes include new polymers or their combinations, and incorporation
of nanoparticles, biocides or pore-forming agents . In addition, various anti-fouling
coating approaches have been presented in the literature . The optimal
surface properties have, however, remained undetermined. Further research is,
therefore, needed to identify the connection between membrane surface properties
and the accumulation of biofilm.
Figure 1 presents the materials and processes of this thesis. Firstly, the aim was
to fabricate a TFC with tailored permeability. Secondly, the aim was to enhance the
anti-fouling performance of a TFC-PA. Industrially applicable and novel coating
methods were used so as to fabricate TFCs by using roll-to-roll or batch processes.
Inorganic, inorganic-organic and organic surface chemistries were created using
coating methods such as sol-gel (SG), atmospheric plasma deposition (APD), atomic-layer-
deposition (ALD) and polyvinyl alcohol (PVA) based dispersion. These were
deposited either as skin layers on a low-density polyethylene (LDPE) or a polylactic
acid (PLA)-coated board or as a top-coating on a TFC-PA membrane.
The scientifically novel contribution of this thesis is as follows. The thesis con-tributes
to the development of novel TFC structures using various coating meth-ods.
The novel TFC structures provide new alternatives or improvements for the
existing barrier materials in food packaging. In addition, the developed anti-fouling
coatings tackle the critical challenge of membrane fouling and introduce new low-fouling
TFC-PA membranes for water treatment.
The effects of coating chemistry and composition of the multilayer structure
were studied in order to find optimal permeability. In particular, the difference
between 2- and 3-dimensional oligomeric sol-gel coating layers was of interest.
Non-polar polymers (e.g. LDPE) are known for their low-adhesion to coatings or
inks. Therefore, it was essential to study the effect of plasma-assisted deposition
and pre-treatment on the adhesion between the coating layers and the substrate.
The polymer surfaces (e.g. PLA) can be relatively rough and therefore challenges
can be expected when applying a uniform and nano-sized coating layer. Conse-quently,
the effect of a primer layer was also needed to study the deposition of an
extremely thin ALD layer.
In membrane technologies for water treatment, the permeability is targeted to
remain unmodified after applying a top-coating. For example, the permeability of a
membrane may be reduced when an additional anti-fouling coating layer is used.
In this thesis, organic and inorganic anti-fouling coatings were developed for TFC-PA
membranes with bacterial anti-adhesion and antimicrobial performances. Firstly,
the effect of the surface physiochemical properties on the attachment of bacteria
was of interest. The correlation between surface polarity and bacteria repellence
was demonstrated by using PVA and ALD coatings. In addition, for the first time the
ALD method was studied as a surface modification of TFC-PA membranes. Secondly,
the anti-fouling performance was further enhanced by the combination of a biocide
effect with the desired surface properties. The PVA-based antimicrobial coatings
were investigated in order to overcome an uncontrolled leaching of biocides.
Figure 1. The schematic presentation describes the materials and processes of
the PhD thesis, which investigates various coating technologies in order to tailor
the permeability and anti-fouling performance of thin-film composites.
2. Advanced surface modification of
The surface modification of a polymeric material can occur as a result of a chemical
or a physical reaction . In addition, a coating layer may attach to a substrate by
The chemical reaction can occur through grafting-to or grafting-from methods,
when appropriate chemically reactive groups can be found on the surface. The
linkage or coupling directly to the existing end groups of the surface is called the
grafting-to method . This does not cause changes in the bulk material. In some
cases, surface activation is needed prior to the coating deposition. The grafting-from
method involves a chemical reaction of the coating to the solid surface, after
it has been physically activated . In general, the surface activation involves
controlled degradation of the surface, using high energy radiation, such as electron
beam, plasma or ultraviolet (UV) radiation.
The chemical bonding of the coating on the substrate is not necessary in every
case. A coating application is also possible through physical adsorption, where the
coating is attached by electrostatic forces . Layer-by-layer deposition (LBL) is
a common method to produce an electrostatically adhered coating. In addition to
chemical and physical bonding, the coating can be attached by mechanical inter-locking.
In this case, the coating is attached on a rough or porous structure of the
substrate . Moreover, the attachment of the coating may be a combination of a
chemical or physical reaction with mechanical interlocking.
2.1 Atmospheric plasma pre-treatment of polymers
Corona and plasma treatments of polymers are known to enhance their wettability
and adhesion to coatings, inks and adhesives . Moreover, a hydrophilic sur-face
tends to provide better adhesion than hydrophobic ones . The surface
energy can be tailored according to the need using different treatment gases or
chemicals. Figure 2 presents schematically the difference of plasma-activation and
Figure 2. Plasma activation and plasma deposition [P1].
Corona treatment is one of the oldest industrial methods of polymer surface acti-vation.
Various plasma technologies have been developed in order to overcome
some of the limitations of the corona technology . In comparison to plasma
technologies, surface-activation effect of the corona is less stable, and tailoring of
plasma chemistry with different monomers or gases is limited. Dielectric barrier
discharge (DBD), radio frequency (RF) and microwave induced plasma (MIP)
technologies have been developed so as to generate plasma treatment for batch
application . In recent years, cold or non-equilibrium atmospheric pressure
plasmas have been developed to produce new functionalities of heat-sensitive
substrates [16,17]. Figure 3 shows an example of atmospheric plasma pre-treatment
unit. In comparison to vacuum plasma treatments, less effort has been
put into atmospheric pressure plasma developments as pre-treatments for coat-ings
and deposition methods.
Both vacuum and atmospheric plasma technologies have been used for sur-face
activation of various polymers . In particular, the modification of surface
energy and hydrophobicity by plasma or corona pre-treatment is necessary, when
using non-polar plastic substrates, such as polypropylene (PP) or polyethylene
(PE). Comparative studies have been carried out for screening the different pre-treatment
methods to enhance the adhesion of a sol-gel coating on polyolefin .
The plasma pre-treatment showed a stronger effect on the adhesion of the sol-gel
coating in comparison to chemical acid etching. When the pre-treatment was not
used, the sol-gel coating delaminated during the curing process. Besides the in-creased
surface hydrophilicity, a short-term (e.g. 15–30 s.) plasma treatment may
reduce the surface roughness of polymers . This may be attributed to cleaning
of the surface by plasma flame. On the other hand, a long-term (e.g. 5 min.) plas-ma
treatment time has been shown to increase the surface roughness . Fur-thermore,
high-temperature vacuum plasma treatment may already degrade the
polymer surface, also causing undesired changes in the bulk material. In some
cases, the increased surface roughness may enhance the adhesion of a coating.
However, the surface roughness should not exceed the thickness of the coating
layer. Low-temperature atmospheric plasma treatment may overcome these chal-lenges.
Moreover, atmospheric plasma has resulted in a strong increase in the
surface energy, while the surface roughness was not affected . Atmospheric
plasma experiments have been shown to create durable hydrophilicity for im-proved
adhesion between polymer surfaces and additional coating layers .
Figure 3. An example of atmospheric plasma pre-treatment unit.
2.2 Atmospheric plasma deposition of thin coatings on
Recently, atmospheric plasma deposition (APD) of functional coatings on plastic
films has gained increasing attention . The main advantages of atmospheric
plasma systems are the possibility of using different treatment gases and gas
mixtures as well as deposition of chemicals . In addition, the processing is
more flexible and feasible, because there is no need to use a vacuum reactor or
other low pressure systems.
An example of a plasma deposition unit and its configuration is described in
Figure 4. Plasma deposition or plasma enhanced chemical vapour deposition
(PECVD) is typically based on dielectric barrier discharge (DBD) plasma . The
two general types of PECVD methods are direct (glow) and remote (afterglow). In
the direct method, the monomer is fed together with the plasma gas to the dis-charge
area, resulting in reaction of the monomer. In the remote method, the
monomer is fed and activated in the downstream of the discharge and, thus, only
the carrier gas is fed through the discharge. Deposition of various molecules or
compounds is allowed. Furthermore, the substrate has a enough long distance
from the discharge, enabling treatments of heat-sensitive materials, such as polymers.
Carrier gas feed
Figure 4. An example of atmospheric plasma deposition (APD) unit configuration [P1].
Fabrication of a hydrophobic silicon oxide (SiOx) coating layer on polyester ter-ephthalate
(PET) film has been the target in most of the studies . For example,
hexamethyldisiloxane (HMDSO) has been used as an SiOx source. In addition,
glass-like coatings have been produced using tetraethoxysilane (TEOS) monomer
with oxygen-helium carrier gas . The TEOS monomer was used to enhance
the abrasion resistance properties of the polymer surface. Furthermore, inorganic-organic
hybrid coatings have been produced on PET to replace polyvinylidene
chloride (PVdC) barrier coatings with atmospheric plasma deposited hybrid coat-ings
. In addition, polymeric coatings have been deposited on PP using the
atmospheric plasma method .
The effects of APD conditions have been studied by analysing the surface
properties of a coating . For example, the increase of monomer flow rate has
improved the hydrophobicity of coatings deposited on PET substrate. The use of
lower flow rates may have led to a thinner or a discontinuous coating. In addition,
the experiments have shown that the SiOx coating became more silica-like (SiO2),
when the air concentration was increased in the argon carrier gas .
2.3 Hybrid coatings synthesised using the sol-gel route
The sol-gel process involves the evolution of inorganic networks in a continuous
liquid phase through the formation of colloidal suspension and following gelation of
the sol . The sol-gel process can be used to manufacture various materials,
including coatings, powders, monoliths, capsules, fibres or aerogels. The main
advantages of the sol-gel thin films are the homogeneity and purity of the end-products
formed at relatively low temperatures.
Metal or non-metal alkoxides are used as monomers in a typical sol-gel synthe-sis
for coatings. The sol-gel synthesis is based on controlled hydrolysis and con-densation
reactions. Reaction 1 represents the hydrolysis, where M is usually
silicon, zirconium or titanium and n generally equals four . Hydrolysis acts as a
rapid initial reaction of sol-gel processes, where reactive alkoxide groups (-OR,
where R is e.g. CH3, CH3CH2, CH3(CH2)2) react with water molecules to form
hydroxyl groups (-OH). After the initiation of the reaction, the hydrolysed alkoxides
Pre-treatment Deposition Post-treatment
easily react with each other, forming dimers through the condensation reaction.
Water or alcohol is obtained as a by-product, depending on the reaction mecha-nism.
The condensation reactions are described with reactions 2 and 3.
M(OR)n + H2O M(OH) (OR) n-1 + ROH (1)
M(OH)(OR)n-1 + M(OR)n (OR)n-1MOM(OR)n-1 + ROH (2)
2 M(OH)(OR)n-1 (OR)n-1MOM(OR)n-1 + H2O (3)
The alkoxide monomers have a different reactivity, which is related to the partial
charge of the metal or the non-metal alkoxide . The reactivity of the monomers
in hydrolysis and condensation reactions can be accelerated or hindered by using
catalysts or by increasing or decreasing the reaction temperature.
Sol-gel coatings can be applied on various surfaces to provide functional prop-erties.
Easy-to-clean, self-cleaning as well as anti-fouling coatings have been
developed by using sol-gel chemistry. The surface chemistry (e.g. polarity) and
topography (smooth or rough) can be modified based on the performance re-quirements.
Furthermore, active agents such as biocides can be incorporated to
the sol-gel coating matrix.
Figure 5 presents schematically the basis of sol-gel processing of thin coatings.
A variety of different types of thin films and coatings can be synthesised by using
the sol-gel technology. In general, the sol-gel thin films can either be pure inorganic
or inorganic-organic hybrid materials. Different wet chemical coating methods can
be used depending on the targeted approach. Drying of the applied coating is an
important step, where solvents and water are evaporated in order to avoid cracking
in the final curing stage. Typically, the thickness of inorganic layers is from 20 nm
to 200 nm. The film thickness of the inorganic sol-gel thin film as a one-layer sys-tem
cannot be raised much more, due to possible cracking behaviour.
Figure 5. Schematic presentation of sol-gel processing of thin coatings.
From the late 1980s, there has been an increasing interest in preparing inorganic-organic
hybrid coatings using sol-gel syntheses . The basic idea is to introduce
organic components into the nanostructured glass-like (e.g. SiO2) or ceramic (e.g.
ZrO2, Al2O3) network. Thus, the hybrid coating is not as fragile as the pure inor-ganic
coatings. Therefore, the coating thickness can be increased from tens of
nanometres up to tens of micrometres. The hybrid coatings are typically two or
three dimensional (2D or 3D) oligomer structures . For example, TEOS can be
used to form the inorganic Si-O-Si backbone. TEOS bonds via hydrolysis and
condensation reactions to organically modified alkoxysilanes, which can contain
variety of different functional groups that can be used to provide hydrophilic, hy-drophobic
or even antimicrobial behaviour. This route can be used to create a 2D
oligomeric structure. The 3D oligomeric hybrid structure is synthesised quite simi-larly
as the 2D structure. The main difference is the use of silane coupling agents,
which are organically modified alkoxysilanes or polymeric monomers, containing
reactive groups (e.g. amino, isocyanate, vinyl, acryl, mercapto, methacryl or
epoxy). These are able to polymerise, when induced using initiators, catalysts,
cross-linkers, heat or radiation (e.g. UV or plasma) . The 3D oligomer structure
may improve the mechanical properties, elasticity and plasticity of the end-product.
In addition, the highly cross-linked hybrid structure may be favourable for
creating enhanced barrier performance.
Recently, the sol-gel technology has evolved to the next level and a few at-temptions
to produce sol-gel coatings in inline process have been presented
[16,23,P2]. Furthermore, the development of inorganic-organic hybrid structures
makes it a potential choice to apply the sol-gel coatings on flexible paper, board or
plastic film surfaces. The possibility to include functional groups and tailor-made
the surface chemistry makes the sol-gel technology even more desirable.
2.4 Atomic-layer-deposition of inorganic thin coatings on
Atomic-layer-deposition (ALD) is a chemical vapour deposition (CVD) technique
based on cyclic and self-terminating gas–solid reactions. ALD can be considered
as an advanced version of CVD technology. ALD technology enables the manu-facturing
of several different kinds of inorganic materials such as oxides, nitrides,
sulphides and others with thickness down to the nanometre range . Moreover,
polymeric and inorganic-organic hybrid films have been manufactured by using
ALD technology .
One of the most studied ALD processes is deposition of Al2O3 thin films, which
are prepared using trimethylaluminium (TMA, Al(CH3)3) and water or ozone as
monomers . The self-terminating reaction of TMA with water is also considered
to be an ideal example of ALD process. The TMA monomer is highly reactive as
well as thermally stable. The Al(CH3)3-H2O process involves switches from methyl-to
hydroxyl-terminated aluminium as presented by the side reactions in equations
4 and 5. Methane is obtained as a gaseous reaction product and it does not dimin-ish
the growth-per-cycle (GPC) of the Al2O3 thin film. Furthermore, the overall
reaction of Al(CH3)3-H2O process is presented in equation 6.
Al-OH + Al(CH3)3 (g) Al-O-Al(CH3)2 + CH4 (g) (4)
Al-CH3 + H2O (g) Al-OH + CH4 (g) (5)
2 Al(CH3)3 (g) + 3 H2O (g) Al2O3 (s) + 6 CH4 (g) (6)
Figure 6 presents the ALD device used in this thesis. In a typical Al2O3 ALD pro-cess,
vaporized TMA and H2O monomers are fed as pulses one-by-one into a
vacuum chamber. Between the monomer pulses, an inert gas purge (e.g. N2) is
used so as to remove unreacted monomers and by-products. After the gas purge,
the saturated surface layer of the TMA can react with H2O, which is again subject-ed
to a gas purge in order to complete the first ALD cycle . These cycles can
be repeated according to the desired film thickness or coating functionality.
Recent advances in ALD technology have increased its potential to produce
functional thin films on flexible and temperature-sensitive polymeric materials .
ALD has also been applied to coat various fibrous polymers [30,31,32,33]. Alt-hough
the ALD process is mainly carried out in batches in its current state, the
development of a slow-speed roll-to-roll ALD technology is underway .
Figure 6. An example of atomic-layer-deposition (ALD) device.
3. Non-permeable packaging films
Permeates (e.g. liquid, gas, ions or particles) diffuse from high concentration to
low concentration . The phenomenon is called permeation. The penetration of a
permeate through a solid interface is described by permeability. Permeability is the
material permeation property . The transfer of permeate molecules through a
material is a process that includes sorption, desorption and migration . In gen-eral,
the permeate molecules attach onto a material surface by a sorption mecha-nism.
Migration occurs, when the attached molecules diffuse through the material.
Finally, diffused molecules detach to the environment by desorption on the other
side of the material.
Diffusion, solubility and permeability are the main parameters used to describe
the mass transfer for a material . The mass transfer of molecules can be de-scribed
by the permeability coefficient (P) of the material . The permeability
coefficient, P, can be calculated using a solubility coefficient (S, solubility) and a
diffusion coefficient (D, diffusivity). The permeability coefficient, P, is the volume of
vapour (q) passing through a unit area of the polymer in a time unit, with a pres-sure
difference unit across the polymer:
( ) ( )
( ) ( ) ( ) (7)
where, S is a thermodynamic term describing the volume of vapour per unit vol-ume
of polymer per unit pressure, D is a kinetic term indicating the velocity of the
molecule in the specific material. Respectively, q is the quantity of the permeant
transferred by a unit of area (A) in a specific time (t), l is the thickness of the material,
and Δp is the partial pressure difference.
The barrier performance of packaging material can be determined using various
barrier measurements . In general, the molecules diffusing into and through
the packaging material are detected. For example, the measurement of gas and
water vapour transmission is of interest due to their effect on the shelf-life of the
packaged product. The oxygen transmission rate (OTR) is the quantity of oxygen
(O2) molecules passing through an area in a certain time period under specified
conditions of temperature, humidity and pressure . The OTR measurement is
based on the O2 flux diffusing through the sample, which is detected using an
electrochemical (non-coulometric) or a coulometric oxygen sensor. In the literature
various units have been used to report permeability of gases . The most commonly
used unit for permeability of gases in barrier polymers is (cm3 mil) / (m2 day atm).
Respectively, the water vapour transmission rate (WVTR) determines the amount
of water vapour transmitted through an area in a certain time under specified con-ditions
of temperature and humidity . There are many standard procedures for
measuring WVTR using gravimetric methods [37,38]. The most commonly used
unit for permeability of water vapour in barrier polymers is (g mil) / (m2 day). OTR
and WVTR values are reported in the literature as normalised to a layer thickness
of 1 mil (1 mil = 25.4 μm).
3.2 Traditional packaging materials
Microbiological contamination is one of the major causes of food losses in the
world . Food packaging has been part of the food supply chain for decades.
Among the traditional preservation systems, packaging technologies have an
important role in securing food safety . Furthermore, barrier films are needed
in pharmaceutical packaging applications.
Packaging has to protect the substance from various environmental exposures,
including oxygen, moisture, light, microbes and dust. Moreover, packaging material
has certain requirements in terms of its usability (e.g. fresh outlook or decorative),
technical aspects (e.g. sealability, easy opening, applicability for food contact,
recyclability or refilling) and reasonable price .
Paper and paperboard (i.e. board) are so-called traditional packaging materials.
They have several advantages, which make their use as food packaging extremely
favourable. Paper and board meet the needs, because they have the required
usability and technical properties, which enable their cost-effective use in a wide
range of packaging structures. Furthermore, they can be coated, laminated or
impregnated to improve their barrier properties and to extend their functional per-formance,
such as heat sealability, heat resistance, barrier, grease resistance or
product release. For example, board can be extrusion-coated using polymeric
coatings made of polyethylene (PE).
Other traditional polymeric packaging materials include polypropylene (PP),
polyamide (PA), polyethylene terephthalate (PET) and ethylene vinyl alcohol
(EVOH) . Furthermore, packaging performance can be further enhanced using
surface treatments such as, polyvinylidene chloride (PVdC), wax, silicone or fluo-rocarbon
based coating. Coatings can be applied by flexo, soft bar, foam, curtain
and spray coating methods, including optional pre-treatments by plasma or corona
and post-treatments/curing by cold or hot air drying, IR-drying or UV-curing.
Recently, bio-based and renewable coating materials have been introduced as
a sustainable alternative to the petroleum-based polymeric materials . For
example, polylactic acid (PLA) has been applied to some extent in commercial
products. In addition, other bio-based materials, such as starch, xylan, chitosan
and cellulose, have been investigated .
3.3 Developments of barrier films for packaging application
The polymers may have a variety of oxygen (O2) and water vapour (WV) barrier
properties, which are affected by their chemical structure, crystallinity and orienta-tion
. Concerning the O2 permeability, the presence of polar groups in polymer
chains often increases the chain rigidity and decreases the specific free volume.
This can further increase its packing density and therefore, increase crystallinity in
the polymer. As a result, the permeability of gaseous molecules is reduced
through the polymer . For example, a PLA contains more polar groups than an
LDPE, and lower O2 permeability has been measured for PLA in comparison to
LDPE. Respectively, PET and PA have lower O2 permeability in comparison to PP
. On the other hand, the polymer chain flexibility affects on the gas permeabil-ity.
The non-polar methyl (-CH3) attached PP structure decreases the mobility of
chains and decreases the gas permeability. Thus, PP has somewhat lower per-meability
compared LDPE . Concerning the WV permeability, the hydrophobicity
of a polymer has an effect . The non-polar structure shows repellence of polar
water molecules. This partly explains the lower water vapour permeability of PP
and PE in comparison to the other substances.
The stretching of amorphous and semicrystalline polymer films can improve
their barrier properties. For instance, biaxial orientation (BO) of PET, PP and PA
reduces both the O2 and WV permeability . Furthermore, lamination and extru-sion
coating are two additional industrial processes for producing multilayer barrier
structures . Several polymeric layers can be coated to form a high barrier
material. For example, EVOH is typically extrusion-coated as one layer in the 3- or
5-layer system. Use of EVOH has led to a slight improvement in the barrier prop-erties
. The drawback of multilayer lamination is the increase in the coating
thickness and, therefore, the increase of net weight of the packaging.
Active packaging materials have been widely studied and patented in recent
years . Active packaging materials include various types of functional systems,
for example, oxygen, carbon dioxide and ethylene scavengers, carbon dioxide and
ethylene emitters, moisture scavengers, ethanol emitters and absorbing systems
as well as antimicrobial packaging materials . Currently, oxygen and moisture
scavengers are commercially the most significant, but other techniques have been
predicted to play an important role in the near future.
Another recent trend in packaging materials is the change towards new func-tionalities
using nanosized materials, such as nanoclays . In addition, new sur-face
treatments and coating technologies have been introduced to traditional
packaging materials. The novel thin coating technologies may provide thinner
barrier films. The replacement of conventional coating materials such as polyole-fins,
EVOH, fluorocarbons and silicones, typically several microns thick, by na-noscale
coatings has the evident benefits of both a reduction of consumed materi-
als and negative effects on recyclability. Furthermore, these new coating methods
may replace the existing metallisation treatments in the future.
Recently, sol-gel hybrid coating and ALD technologies have shown promising
barrier performance enhancement, when deposited on polymeric film or polymer
(e.g. PLA or LDPE) coated board [44,45,P2,P3]. Among the ALD-coated PLA-board
specimens, Al2O3 ALD coating showed better barrier performance than SiO2
. Concerning the LDPE-board, the studies revealed that ALD-coating did not
show as good barrier performance as shown for PLA-board . This can be
explained by the difference between the reactivity of the PLA and LDPE surfaces.
Therefore, corona or plasma pre-treatment or primer layers could be used to im-prove
the film growth or adhesion.
Some of the commercially available approaches include deposition of silicon
oxide (SiOx) or aluminium oxide (AlOx)-based inorganic coatings on a polymer
film for barrier property improvements . The SiOx coating has been shown to
reduce O2 and WV permeability of PET . Furthermore, the oxygen and water
vapour barrier of a BOPP film can be enhanced by deposition of nanometre thick
inorganic AlOx layer . However, the inorganic coating layers may be fragile
and delaminate from the flexible polymer surface.
To overcome the delamination issue, thin-film composite (TFC) concepts have
been developed, where thin organic and inorganic coating layers are separately
deposited on polymers. The TFC concept for enhanced O2 barrier properties has
already been presented in the late 1990s by Haas et al . They deposited hy-brid
sol-gel on a PP film, which was followed by the deposition of a SiOx layer. In
addition, enhancement of barrier properties has been obtained with TFC structure
of an AlOx and a few micrometre thick acrylate coating. Recently, a combination of
sol-gel hybrid coating and ALD technologies has also shown promising barrier
performance enhancement, when deposited on polymeric films or polymer coated
4. Semi-permeable thin-film composite
4.1 Reverse osmosis (RO) membranes
Membrane filtration technology is used in various industries to provide purified
water from sources including, mining, biorefinery, brackish, processing, or sea-water
. Moreover, the membrane filtration technology is utilised in microfiltration
(MF), ultrafiltration (UF), nanofiltration (NF) and reverse osmosis (RO) . These
methods have been applied in water and wastewater treatment, desalination and
water re-use since the late 1960s.
Reverse osmosis (RO) membranes are nonporous systems which retain parti-cles
and low molecular weight species such as salt ions, micropollutants, pesti-cides
and pharmaceuticals . The commercial RO membrane materials are ei-ther
cellulose acetate (CA) or thin-film composite (TFC) membrane, which consists
of polyamide (PA) as a selective layer and polyethersulfone (PES) cast on non-woven
as a support layer . The traditional CA membranes were developed in
the late 1950s and commercialized in the early 1960s. CA has formerly been widely
used, but nowadays TFC-PA membranes are more common. The TFC-PA mem-branes
offer several advantages over the CA membranes, such as improved re-tention,
increased productivity at lower operating pressures, great structural stability,
and the ability to produce two to three times more purified water per unit area than
CA membranes .
Both of the TFC-PA and CA membranes have one common challenge restricting
their usage in industrial processes: fouling. Membrane fouling has an unwanted
effect on membrane separation efficiency as well as on energy consumption .
Membrane permeability declines, resulting in unstable water flux behaviour, and
therefore feed pressure should be increased in order to stabilize the process .
4.2 Permeability of reverse osmosis (RO) membranes
Permeability of a membrane is important for applications in water and wastewater
treatment. In general, there is a strong trade-off between water permeability and
salt rejection. The more water permeable membrane tends to have lower salt
rejection and higher solute permeability .
An adequate water flux is essential because of its relationship to the membrane
productivity and process economics. Water flux is defined as a mass (or a volume)
passing through a membrane unit within a given duration of time . The pure
water flux Jv is typically defined as:
where Qp is the volumetric flow rate that permeates through the defined mem-brane
area Am. In the literature the typical unit for water flux is L/m2h.
The effect of the applied pressure (P) and the osmotic pressure (π) of the
solution on the permeate flux should be taken into account in the pressure-driven
membrane systems . Therefore, Darcy’s law can be commonly determined
according to the following equation:
where Rm is the hydraulic resistance of the membrane and η is the viscosity of the
permeating water. Rm can be determined for a membrane according to its water
The applied pressure P, i.e. the trans-membrane pressure (TMP), is the differ-ence
between the applied pressure on the feed water side and the pressure of
permeate water . The osmotic pressure, π, of the solutions can be determined
by the van’t Hoff equation:
where Rg is the universal gas constant (Rg = 8.31 J/mol K), T is the absolute tem-perature
given in Kelvin degrees, and C is the molar concentration of dissolved
substances. In addition to the water flux, solute retention is one of the most im-portant
performance parameters for a membrane process. The retention ability of
a membrane in water applications is expressed by the membrane rejection. Mem-brane
rejection (R) can be determined by:
where Cf and Cp are the concentrations (via conductivity measurements) of the
feed and permeate solutions, respectively. Based on the literature, the solution-diffusion
model is the most widely used transport model for RO membranes . It
is assumed that both solvent and solute absorbs into the rejection layer and dif-fuses
through the nonporous layer . Therefore, the water flux, Jv, through an RO
membrane corresponds to the applied pressure (P-π). Respectively, solute flux,
Js, through an RO membrane corresponds to the concentration difference (C)
across the membrane.
( ) (13)
where A is the effective water permeability of an RO membrane.
( ) (14)
where B is the effective solute permeability of an RO membrane. A more funda-mental
property describing the real salt permeability of an RO membrane is the
solute permeability coefficient (B) . The solute permeability coefficient can be
determined from the water flux and rejection test results via:
4.3 Fouling of reverse osmosis (RO) membranes
Membrane fouling occurs when particles and solutes accumulate on the mass
transfer boundary layer of the membrane and they begin to impede the solvent
flow through the membrane . Fundamentally, the fouling layer increases the
hydraulic resistance on the membrane surface and, therefore, Darcy’s law for
water flux (Jv) can be determined according to :
where, Rf is the hydraulic resistance caused by fouling. After the initiation of the
fouling macromolecules, proteins, gels or particles will continue adsorbing on the
membrane surface . The types of membrane fouling are classified into colloidal
(inorganic and organic) and biological fouling (i.e. biofouling). Inorganic fouling (i.e.
scaling) is occurred by deposition of major ions on membrane surface including
calcium, magnesium, barium, bicarbonate, sulphate and inorganic salts. This
causes pore-blocking and reduction of the water flux. The scaling is mainly avoid-ed
chemically by using antiscalants and pH control to pre-treat the feed solution
before the separation process. Organic fouling typically involves attachment of
proteins and humic acid on the membrane surfaces. Several researchers have
studied the effects of the surface properties on the organic fouling tendency of the
commercial RO membranes [51,52,53].
Biofouling is generally initiated by adhesion and accumulation of the microor-ganisms,
which is followed by primary colonization and logarithmical growth . In
addition, a number of the microorganisms, such as bacteria, can be detected within
the network of extracellular polymer substances (EPS). It is difficult to completely
eliminate biofouling and, therefore, it is problematic for all membranes, especially in
RO and NF processes . The biofouling tendency of commercial membranes has
been studied by several research groups [55,56,57]. Moreover, a few studies have
concentrated on the interaction between bacteria and the surface properties of a
variety membranes [58,59,60,61]. These previous studies have revealed that bacterial
anti-adhesive behaviour varies between the different membranes.
The membrane surface properties have been shown to play an important role in
affecting biofilm formation . Studies have shown that membrane surface prop-erties
such as hydrophobicity, surface roughness and surface charge have an
effect on the reduction in fouling . Most of the commercial RO membranes are
hydrophilic rather than hydrophobic. For example, it has been shown that TFC-PA
and CA membranes have water contact angle values between 20–60º . The
TFC-PA membranes for RO processes are typically aromatic or semi-aromatic. Fully
aromatic PA membranes are formed by 1,3-benzenediamine (m-phenylenediamine)
and trimesoyl chloride (1,3,5-benzenetricarbonyl chloride) . Respectively, the
semi-aromatic poly(piperazinamide) membranes are formed by piperazine and
trimesoyl chloride . However, it is difficult to conclude whether the RO mem-brane
has a hydrophilic nature based on chemical structure. Some of the com-mercial
TFC-PA membranes are coated with an additional thin polyvinyl alcohol
(PVA) layer in order to introduce a hydrophilic and smooth surface . The water
contact angle value of a fully aromatic PA membrane is reduced significantly by
applying the PVA coating. On the other hand, the PVA coating layer has only a
minor effect on the semi-aromatic membranes .
4.4 Development of low-fouling thin-film composite
The development of new anti-fouling RO membranes has gained increasing atten-tion
. In many cases, an anti-fouling membrane is achieved by combining the
surface physiochemical properties, such as increased hydrophilicity, lowered sur-face
roughness and neutralised surface charge . These physiochemical anti-fouling
properties are commonly referred to as anti-adhesion surfaces. In addition,
antimicrobial surfaces have been presented as an active anti-fouling approach.
Several different ways have been used to modify a membrane material itself or its
surface by an additional coating layer.
The membrane material has usually been modified using polymer blends or
composites, which consist of another polymer or some additives including titanium
dioxide (TiO2) or silica (SiO2) nanoparticles and biocides (e.g. silver, copper, zinc
or quaternary ammonium) . In several approaches, improved water flux
through a membrane is achieved by incorporating pore-forming agents, such as
silica, zeolites, carbon nanotubes and aquaporin protein into the skin or mem-brane
Various different surface treatments have been developed for RO membranes.
These include: 1) surface grafting of antimicrobial groups, 2) contact active am-phiphilic,
microbe-repelling or anti-adhesive polymers, 3) functional coating meth-ods
(LBL, casting, dipping, soaking or spraying) or 4) reactive coating methods
(CVD, sputtering, UV and plasma immobilization) . An ideal fouling-resistant
coating would be an ultrathin, highly water-permeable surface layer that does not
significantly increase the resistance to water flux . The additional coating layer
may indeed increase the hydraulic resistance of the membrane surface and the
flux can thus be determined using Darcy’s law:
where, Rc is the hydraulic resistance caused by the coating. For example, the
water flux through the membrane may be reduced, if the top-coating layer is much
more hydrophobic, denser or thicker in comparison to the PA rejection layer.
The surface modification occurs mostly on the outermost surface of a mem-brane.
For example, the CA membrane surface bears several hydroxyl groups for
the chemical bonding. Furthermore, low concentrations of the amino or carboxylic
groups can be found in TFC-PA membranes. Most of the polymeric membranes
are, however, chemically rather stable and thus the controlled activation of the
membrane is complicated or even impossible without changing the polymer bulk
material morphology. Therefore, grafting-from approaches have been used in
order to increase the reactive area of the surface and, furthermore, to increase the
coverage of applied surface modification . For example, the silylation of sur-faces
by using functional alkoxy- or chlorosilanes is one of the well-known meth-ods
for introducing reactive groups (amino, isocyanate, aldehyde, epoxide or car-boxyl)
onto a solid surface. The silylation layer is typically only molecular with a
thickness of about 1 nm. In addition, physical activations (ion, plasma or UV) have
been used to improve the adsorption of the coupling agents on the membrane
Concerning the development in low-fouling TFC-PA membranes, the research
groups have applied their surface treatments to either commercial or experimental
TFC-PA membrane. The different treatments can be divided into passive (e.g.
hydrophilication and lowering surface roughness) and active (e.g. antimicrobial)
fouling prevention approaches as well as methods focusing on flux improvement
. Most of the studies have focused on the deposition of hydrophilic polymer
brushes, which have either anionic or non-ionic behaviour [67,68,69]. For example,
polyethyleneglycol (PEG), methacrylate acid (MA), polyethyleneglycolmethacrylate
(PEGMA) and copolymers of N-isopropylacrylamide and acrylic acid (P(NIPAm-co-
AAc)) have been applied on TFC RO membranes for repellence of bacteria and
protein [70,71,72]. Some of the approaches have focused on plasma-assisted
grafting of molecular layers. Also, plasma deposition has been used to fabricate
the coatings [73,74]. Silane coupling agents have recently been used to create
hydrophilic and hydrophobic surfaces on TFC-PA membrane. This approach
showed a flux decline of the TFC-PA membrane, because of increased hydropho-bicity
by the silane coupling agents. On the other hand, the use of silane coupling
agents significantly increased the salt rejection of the TFC-PA membrane
[75,76,77]. A few research groups have investigated atomic-layer-deposition
(ALD) technology as surface modification of porous polymeric and ceramic mem-branes,
which are used in microfiltration and ultrafiltration [78,79,80,81,82]. These
studies suggested that ALD processing parameters, such as temperature and the
number of ALD cycles, have a clear impact on membrane performance. For ex-ample,
they have shown an effect on the pore size of membranes, which has
affected solute permeability and rejection properties [80,81]. Furthermore, hydro-philic
and low surface roughness polymeric membranes have been obtained by
using TMA based ALD technology . This could increase the potentiality of the
ALD method to create low-fouling membrane surfaces. However, hydrophilicity
and low surface roughness have typically been obtained using a relatively thick
ALD coating, such as 300–500 ALD cycles [78,80]. In addition, change in the ALD
processing temperature has an effect on the reactivity of TMA . For example,
the most hydrophilic membrane surface has been obtained at 150ºC .
5. Aims of the study
The aims of this research were to fabricate and modify thin-film composite struc-tures
with tailored permeability and anti-fouling performance. In particular, the
interest was in studying the effects of pre-treatment, coating chemistry and multi-layer
coating structure on the permeability of a thin-film composite (TFC). In addition,
the thesis aimed to gain further understanding of the anti-fouling phenomena of
coated TFC membranes.
This study was limited to industrially applicable coating methods by studying
both roll-to-roll and batch processes. Inorganic, organic and inorganic-organic
surface chemistries were created using coatings based on sol-gel (SG), atmos-pheric
plasma deposition (APD), polyvinyl alcohol (PVA) and atomic-layer-deposition
(ALD). These were deposited either as a skin layer on low-density
polyethylene (LDPE) or polylactic acid (PLA)-coated board, or as a top-coating on
a polyamide (PA)-based TFC membrane. Because of the variety of substrate
materials, the coating application processes were investigated together with the
surface properties obtained by a coating. The end-applications were targeted for
food packaging [P1–P3] and for water treatment membranes [P4–P6]. The re-search
questions of this study were as follows:
Can TFC structure be fabricated with tailored permeability?
o Does SiOx coating form on LDPE-board using atmospheric plasma
o Does atmospheric plasma pre-treatment enhance the wettability and
adhesion of sol-gel coatings deposited on LDPE-board? [P2]
o Does 2-dimensional or 3-dimensional sol-gel network have different ef-fect
on the gas permeability of LDPE-board? [P2]
o Do TFC structure-based sol-gel and ALD coatings reduce the gas and
water vapour permeability of PLA-board? [P3]
Does top-coating enhance the anti-fouling performance of TFC membrane?
o Does the increase in surface polarity reduce the attachment of model
o Can an organic or inorganic coating layer create an anti-fouling effect
on a TFC membrane without sacrificing the permeability of the sub-strate?
o Can the anti-fouling effect be enhanced by combining hydrophilicity and
low surface roughness with antimicrobial additives? [P5]
o What is the effect of ALD processing temperature and number of ALD
cycles (i.e. film thickness) on hydrophilicity, surface roughness and
permeability of TFC membrane? [P6]
6. Experimental procedures
6.1 Substrate materials
A variety of substrate materials were used in this thesis. These were polyamide
(PA)-based thin-film composite (TFC) membranes and low-density polyethylene
(LDPE) or polylactic acid (PLA) coated board. Table 1 summarises the substrate
materials for the characterisation and coating deposition.
Table 1. The substrate materials used for the characterisation and coatings.
ID Substrate material Provider
LDPE-board Low-density polyethylene coated
(LDPE 15 g/m2) board (210 g/m2)
Stora Enso Oyj
PLA-board Polylactide-coated (PLA 35 g/m2) board
Stora Enso Oyj
BW30 PVA-coated thin-film composite; polyamide (PA),
polyethersulfone (PES), non-woven.
Dow FilmTec ltd.
SW30HR PVA-coated thin-film composite; polyamide (PA),
polyethersulfone (PES), non-woven.
Dow FilmTec ltd.
LE Thin-film composite polyamide (PA),
polyethersulfone (PES), non-woven.
Dow FilmTec ltd.
XLE Thin-film composite polyamide (PA),
polyethersulfone (PES), non-woven.
Dow FilmTec ltd.
NF90 Thin-film composite polyamide (PA),
polyethersulfone (PES), non-woven.
Dow FilmTec ltd.
6.2 Atmospheric plasma deposition of SiOx coating
The plasma deposition was carried out in three continuous steps: pre-treatment,
deposition and post-treatment, as described in Figure 4. The power of the plasma
deposition unit was varied between 0.4–2.0 kW. The treatment width was 380 mm.
Helium and argon were used as the carrier gas in the three carrier gas feeds of
the equipment. The gas feed was 90 litres/min in the middle slit and 60 litres/min
in both side slits. In a typical plasma deposition experiment, the precursors were
vaporized in the pre-heated carrier gas prior to the deposition. The SiOx source
was hexamethyldisiloxane (HMDSO) precursor and it was used as such. The
precursor (HMDSO) feed was 980 g/h. With the target of developing an industrially
applicable plasma deposition method, the line speeds of 2, 5, 50, 100 and 200
m/min were used.
6.3 Synthesis and processing of sol-gel hybrid coatings
Three different kinds of sol-gel hybrid coatings were synthesised in this thesis. For
the purpose of this study the coatings were named SG200, SG204 and SG206.
The following alkoxysilane monomers (Figure 7) were used in this study: tetra-ethoxysilane
(Si(OC2H5)4, TEOS), octyltriethoxysilane (C14H32O3Si, C8TES),
3-glycidyloxypropyltrimethoxysilane (C9H20O5Si, GLYMO) and 3-(2-aminoethyl-amino)
propyltrimethoxysilane (C8H22N2O3Si, NAMS). All the monomers were or-dered
from ABCR and used as such. SG200 (TEOS+GLYMO) was aimed to cre-ate
a smooth primer layer for ALD layer. SG204 (TEOS+C8TES) was aimed to
form a hydro- and oleophobic coating layer. SG206 (NAMS+GLYMO) was aimed
to form a tight barrier coating layer. Table 2 summarises the sol-gel formulations
synthesised and used in this thesis. All the coatings were prepared by dissolving
the alkoxysilane monomers with a certain amount of ethanol (EtOH). The ethanol-silane
mixture was vigorously mixed for 30 minutes using a magnetic stirrer. Then
a solution of water (H2O) and nitric acid (HNO3) was prepared and added to the
ethanol-silane mixture during mixing. The sol-gel solutions were mixed for 24
Table 2. The sol-gel (SG) formulations synthesised and used in the thesis.
Coating ID /publication
SG200 / [P3]
SG204 / [P2]
SG206 / [P2]
TEOS 13.0 26.4 0
C8TES 0 10.5 0
GLYMO 29.5 0 7.8
NAMS 0 0 24.6
EtOH 45.9 46.8 50.9
H2O (with HNO3) 11.2 (0.4) 16 (0.3) 16.0 (0.7)
Figure 7. The chemical structures of the monomers used in the sol-gel hybrid
6.3.1 Sol-gel-coated LDPE-board
The sol-gel coatings, named SG204 and SG206 for this study, were spray-coated
onto LDPE-board. The atmospheric plasma equipment with argon or helium
treatment gases was used to activate the LDPE surface prior to deposition of the
sol-gel coatings. The following parameters were used for the plasma-activation:
treatment power 2 kW, treatment width 370 mm, air gap 1–2 mm and speed 100
m/min. The sol-gel hybrid coatings were spray-coated onto the plasma-activated
LDPE surface using the line speed of 60 m/min. The coatings were thermally
cured in a pilot coating line by using infrared and suspended heaters.
6.3.2 Sol-gel-coated PLA-board
The sol-gel coating, named SG200 for this study, was sprayed onto corona-treated
A4-size PLA-board substrates and heat-treated at 120°C for 10 mins.
Traditional corona treatment unit (ET1 from Vetaphone, Denmark) with a treat-ment
time of 60 s was used to ensure wetting and adhesion properties between
the coating and the PLA-board substrate.
6.4 ALD processing of Al2O3 coatings
6.4.1 ALD deposition on sol-gel-coated PLA-board
The ALD-Al2O3 depositions were carried out on the sol-gel coated PLA-board at
80°C using a Picosun SUNALE™ reactor (Picosun, Espoo, Finland). Trimethyl-aluminum
(TMA, electronic grade purity, SAFC Hitech) and H2O were used as
precursors. The nominal thicknesses of the coating were estimated based on a
growth rate of 0.1 nm/cycle [44,45]. In comparison to a silicon wafer, the ALD
coatings have shown a minor variation in the actual film thickness, because of the
difference in surface chemistry and roughness of different polymers.
6.4.2 ALD deposition on TFC-PA
The ALD-Al2O3 depositions were carried out in a Picosun SUNALE™ reactor on
substrates that were 10×10 cm2 in size. TMA and H2O were used as precursors.
Two different processing temperatures (70ºC and 100ºC) and three different ALD
cycles (10, 50 and 100) were used to deposit the ALD coatings on the commercial
LE-400 TFC-PA membranes. The coated membranes were named according to
the coating cycles and temperature (e.g. c10_t70 denotes a membrane with 10
coating cycles at 70ºC). The precursor pulsing sequence was: 0.1 s TMA pulse,
10 s N2 purge, 0.1 s H2O, 10 s N2 purge and the number of ALD cycles was ad-justed
according to the targeted Al2O3 coating thickness (0.1 nm/cycle).
6.5 Preparation and processing of PVA and PVA-PHMG
The following chemicals were used for the membrane surface modification. PVA
(molar weight 9000-10,000 g/mol, 80 % hydrolysed) polyvinyl alcohol was pur-chased
from Sigma-Aldrich. PHMG, polyhexa-methylene guanidine hydrochloride
(C7N3H15HCl, Mw 8000-10,000 g/mol) was provided by Soft Protector Oy (Fin-land).
The chemical structures are presented in Figure 8.
Polyvinyl alcohol (PVA)
Figure 8. Chemical structure of polyvinyl alcohol (PVA) and polyhexamethylene
guanidine hydrochloride (PHMG).
The preparation of the PVA coatings was carried out by mixing 2 wt-% of PVA into
milliQ water at 60ºC. After complete mixing, the solution was cooled down to room
temperature. The PVA-PHMG solutions were similarly prepared as described
above, by adding PHMG to the PVA solution after the cooling step. PVA-PHMG
solutions were prepared in ratios 95:5 and 99:1 (PVA:PHMG) with a total polymer
content of 2 wt-%. The pure PHMG solution (2 wt-%) was prepared by mixing
PHMG with milliQ water to reach the targeted concentration. PVA, PVA-PHMG
and PHMG solutions were applied on the commercial LE-400 TFC-PA membrane
coupons using the dispersion coating method with a K Control Coater K202 device
(RK Print-Coat Instruments Ltd., United Kingdom). The device was equipped with
a glass bed, and K202 coating bar No. 0 was used to cast the wet films with a
thickness of 4 μm. The coatings applied were heat-treated at 110ºC for 2 minutes
in order to evaporate the water and to increase the coating stability using thermal
6.6 Surface characterisation
A Perkin Elmer spectrum BX II FTIR (Fourier transform IR) system (USA) was
equipped with vertical-ATR (Attenuated total reflectance) and KRS-5 (Thallium
Bromoiodide) crystal [P3–P6]. ATR-FTIR measurements were carried out in order
to analyse the surface chemistry. In a typical analysis 50 scans were collected
from 500 to 4000 cm-1 at 4 cm-1 resolution. A background spectrum of pure KRS-5
was collected before running the samples.
The microscopic imaging was conducted using a scanning electron microscope
(SEM) JEOL JSM-6360 LV equipment (USA) with 11 kV on high vacuum mode
[P3–P6]. All SEM samples were sputter-coated with gold before imaging.
The surface topographies were characterised using non-contact mode atomic
force microscopy (NC-AFM) [P1–P6]. The NC-AFM analysis was performed using
Park Systems XE-100 AFM equipment (Suwon, South-Korea), with cantilever
905M-ACTA (AppNano Inc., USA). Typically, the scan rate was 0.4–0.6 Hz, and
the measured area was 5x5 μm2. Six replicate measurements were taken in order
to determine the roughness values, average roughness Ra or root-mean-square
roughness value RRMS.
The contact angle measurements were taken using an Optical Tensiometer
Theta T200 device (Biolin Scientific, Sweden) [P3–P6]. The measurements were
taken in a controlled atmosphere (RH 50%, temperature 23ºC) and the results are
given as an average of five parallel measurements. The water contact angle val-ues
are presented at the time point of 30 s from the moment the drop contacts the
surface. The surface energy values were obtained by measuring the contact angle
of three different probe liquids, namely water (H2O, γ = 72.80 mN/m), di-iodo-methane
(CH2I2, γ = 50.80 mN/m) and formamide (CH3NO, γ = 58.20 mN/m). The
total surface energy values, as a summary of polar and dispersive surface energies,
were determined from the measured contact angle data using Fowkes’ theory .
6.7 Anti-fouling performance
Biofilm formation was demonstrated by analysing the attachment of Pseudomonas
aeruginosa (P. aeruginosa) on uncoated and coated membranes [P3–P6].
P. aeruginosa is Gram-negative, aerobic, rod-shaped bacterium, which is widely
used as a model microbe for biofilm formation studies [55,58,89]. Therefore,
P. aeruginosa was selected as model bacterium in this study so as to provide
comparable data related to the attachment of bacteria.
A bacteria attachment test was carried out by submerging the membranes (Ø
4.7 cm) in a bacterial suspension consisted of standard seawater ASTM D1141-98
(2008) . The suspension was inoculated overnight with a culture of P. aeru-ginosa
(VTT E-96726) cultivated in 37°C Trypticase soy agar (TSA) broth solution,
harvested by centrifugation (3000 rpm, 10 mins) and washed with phosphate-buffered
saline (PBS) solution (10 mM). The cell density was approximately 1×108
CFU mL-1 determined by plate count on a TSA (37°C, 1 day). The 1 day exposure
of membranes was conducted in a rotary shaker (75 rpm) at room temperature.
The number of adhered cells on the membranes was determined after swabbing
by plate count on TSA (37°C, 1 day), and it was intended to describe the anti-adhesion
performance. Results are presented as colony-forming units per mem-brane
area (CFU cm-2). Three replicate membrane samples were examined for
each membrane type.
The antimicrobial performance of uncoated and coated membranes was analysed
using the colony-forming unit (CFU) test for the two rod-shaped model microor-ganisms,
Gram-positive bacteria Bacillus subtilis (ATCC 6633) and Gram-negative
bacteria Escherichia coli (ATCC 8739) [P5]. These bacteria are extensively used
to investigate surface antimicrobiality [91,92]. Bacteria cultivation was conducted
by following the instruction of the producer, ATCC (24 hour cultivation for
B. subtilis in 30°C nutrient broth solution and 12 hour cultivation for E. coli in a
37°C nutrient broth solution, respectively). Bacterial cells were harvested by cen-trifugation
(3,000 rpm, 10 mins) and washed twice with 0.9 % NaCl solution. Har-vested
bacterial cells were suspended in a 0.9 % NaCl solution to achieve a cell
density of 1×107 CFU mL-1. The membrane samples (Ø 1.27 cm) were immersed
in Milli-Q water for 24 hours before the actual exposure to bacteria suspension. The
respective membranes specimens were in contact with the bacterial suspensions for
24 hours. A serial dilution of bacterial suspensions was performed for the CFU test
according to CLSI M07-A9 . In the CFU test, 0.1 mL volume of diluents was
inoculated and spread on a nutrient agar plate, and incubated for another 24 hours.
CFU results were obtained from 3 measurements of 3 independent membranes, and
each measurement was based on the results of 3 replicate agar plates.
A diffusion inhibition zone (DIZ) method was conducted to analyse the antimi-crobial
performance of the coatings in more detail [P5]. DIZ test of surface modi-fied
membranes were performed by following the CLSI document M2-A9 26:1 .
100 μL bacterial inoculums were spread on the agar plates. Punched membrane
discs (Ø 12.7 mm) were placed on the agar plates and incubated at 24 hour culti-vation
for B. subtilis at 30°C and 12 hour cultivation for E. coli at 37°C. The diffu-sion
inhibition zones formed around the membranes were determined by visual
6.8 Barrier performance and permeability
The grease barrier performance of the specimens was studied by contact angle
measurements with oleic acid using an Optical Tensiometer Theta T200 device [P2].
The contact angle was followed for 25 seconds to estimate the grease barrier effect.
The durability of the grease barrier effect and thus the adhesion of the coatings
to the substrate were studied by measuring the contact angle of oleic acid before
and after an abrasive wearing [P2]. The wearing test was carried out using a modi-fied
standard paint washing tester apparatus (modified standard DIN 53778),
where a brush was used as an abrasive material (1000 abrasion cycles).
OTR and WVTR transmission
The oxygen and water vapour transmission rate (OTR and WVTR, respectively)
values were measured for determination of the gas and vapour permeability of the
barrier films [P3].
The OTR values were measured (Systech M8001or Mocon Oxtran 2/20) from
two to three parallel samples using humid gases at 23°C and in 50% relative hu-midity.
The size of samples was either 50 cm2 or with masking 5 cm2. 100% oxygen
was used as the test gas.
The WVTR values were measured from two to three parallel samples according
to the modified gravimetric methods ISO 2528:1995 and SCAN P 22:68 with a
moisture gradient of 75% relative humidity (at 23°C).
Water permeability and salt rejection
Uncoated and coated membranes were tested in a lab-scale crossflow filtration
setup in order to evaluate their water permeability and salt (NaCl) rejection accord-ing
to Wang et al. . Briefly, membranes were tested at a feed pressure of 27.6
bar (400 psi) using a feed water containing 10 mM NaCl. The temperature of the
feed water was maintained at 23°C. Water flux of the membranes was determined
by the gravimetric method.
The data obtained from the water flux (Jv) and the salt rejection (R) measure-ments
was used to calculate water (A) and solute (B) permeability. These were
used to analyse the effect of coatings on the membrane separation performance.
Water permeability (A) was calculated using the equation A = Jv / ΔP. Solute per-meability
(B) was correspondingly calculated according to the equation B = A(ΔP-
Δπ) * (1/R−1).
7. Results and discussion
7.1 Fabrication of thin-film composites with tailored
7.1.1 Atmospheric plasma deposition of SiOx film on LDPE-board
Recently, several authors have presented their studies on novel atmospheric
plasma equipment and surface activation or deposition [16–23]. However, the use
of atmospheric plasma deposition in roll-to-roll process has not been widely re-ported
. In this work, the SiOx coating was plasma-deposited on LDPE-board
using a pilot extrusion coating line equipped with an atmospheric plasma deposi-tion
unit [P1]. In general, the aim was to create SiOx coating on LDPE-board and
to study its surface properties.
The appearance of the SiOx coating on the LDPE surface was confirmed using
FTIR [P1]. The comparative analysis of the SEM images of both the plain LDPE
surface and the plasma deposited LDPE surface showed that a coating of notice-able
thickness is obtained. The processing marks were typically covered by the
coating (Figure 9). However, when running at lower power or at higher line
speeds, thinner coatings were obtained. The thinner coatings were visible on the
surface, but these were not thick enough to cover the processing marks.
(b) Plasma-deposited SiOx
Figure 9. SEM images of the LDPE surface without (a) and with (b) plasma de-posited
SiOx coating (modified from [P1]).
In addition to the SEM characterisation, the surface morphology of the plasma-deposited
film was examined with AFM. Figure 10 presents the AFM images for
uncoated (a) and SiOx-coated LDPE (b). The comparison of the AFM topography
images showed covering of the LDPE surface by a SiOx layer. The plain LDPE
showed typical crystallized and oriented morphology of LDPE . The observed
SiOx film did not show such a columnar well-defined growth structure of a HMDSO
film obtained using vacuum RF plasma deposition . However, the plasma-deposited
SiOx film seemed to be a quite similar compared to previous achieve-ments
obtained by using atmospheric plasma deposition .
(b) SiOx-coated LDPE-board
Figure 10. AFM topography of uncoated LDPE-board (a) and plasma deposited
SiOx on LDPE-board (b).
Table 3 presents the surface roughness values of both the plain LDPE surface
and the plasma-deposited LDPE surface. The decrease in Ra and RRMS (=Rq)
values showed that plasma deposition smoothed the average roughness of the
Table 3. Surface roughness values of uncoated LDPE and SiOx plasma deposited
LDPE (modified from [P1]).
Specimen Ra (nm) RRMS (nm)
LDPE-board 12 ± 4 16 ± 5
SiOx Plasma deposited LDPE-board 8 ± 1 10 ± 2
SEM and AFM results indicated some non-uniform deposition of the SiOx coating
layer. In this case, the water vapour transport may occur through the non-coated
areas of the coating. This can partly explain why the WVTR measurements did not
reveal considerable changes in water vapour permeability [P1]. The differences
between untreated, plasma-treated and plasma-deposited samples were within
the limits of the standard deviation.
Further development of the plasma deposition process may improve the barrier
performances. For example, more powerful plasma pre-treatment may be neces-sary
to ensure better film formation. On the other hand, the LDPE surface is rather
rough for deposition of uniform and dense thin films. Therefore, further research
should focus on the effect of pre-treatment or primer layer in smoothing the LDPE
surface prior to the plasma deposition. Moreover, a thicker SiOx coating layer
might improve the barrier performance as well. However, it may lead to unwanted
cracking of the inorganic coating layer. Therefore, a protective coating layer is
possibly needed to improve the durability of the inorganic layer.
7.1.2 Atmospheric plasma-assisted sol-gel hybrid coating on LDPE-board
In relation to work on plasma-deposition of SiOx coatings, a pilot extrusion coating
line was equipped with an atmospheric plasma pre-treatment unit. The main goal
was to obtain high barrier effects for oxygen and grease by deposition of plasma-assisted
sol-gel hybrid coatings on the LDPE-board [P2].
The sol-gel hybrid coatings consisted of either a 2-dimensional (SG204) or 3-
dimensional (SG206) oligomer hybrid network. The SG204 coating was based on
the 2D oligomer hybrid structure synthesised according to Figure 11, which illus-trates
the reaction between metal alkoxide (e.g. tetraethoxysilane, TEOS) and
organically modified alkoxysilane (e.g. octyltriethoxysilane, C8-TES). Typically,
some of the hydroxyl groups remain unreacted or react more slowly depending on
the catalysis of the reaction . These hydroxyl groups can enhance the bonding
of the synthesised hybrid coating on substrates bearing hydroxyl or carboxyl
groups. Therefore, the hydrophobic functional end-groups of the organically modi-fied
alkoxysilanes should align out from the substrate.
Figure 11. Synthesis of 2D oligomeric hybrid structure using sol-gel route.
The SG206 coating consisted of 3-glycidyloxypropyltrimethoxysilane and 3-(2-
aminoethylamino) propyltrimethoxysilane. Figure 12 describes the synthesis route
for formation of the 3D hybrid network, where an epoxy coupling reaction occurs
between amino and epoxy groups, followed by methoxy group reaction through
hydrolysis and condensation reactions to form the polysiloxane Si-O-Si bonds.
Figure 12. Synthesis of 3D oligomeric hybrid structure using epoxy coupling reac-tion
and the sol-gel route.
The SEM analysis was used to characterise the structure and film thickness of the
coated specimens. Figure 13 presents an SEM image of a cross-section of the
SG206 coating deposited on the LDPE. The existence of the coating layer can be
seen as a light grey colour. The coating thickness determined from the SEM im-age
was approximately 1.3–1.5 μm. The SEM image indicated that the coating
was uniform and dense. In addition, no film cracking was observed.
Figure 13. Cross-section SEM image of the SG206 coated LDPE-board.
Figure 14 presents the AFM topography images of (a) the uncoated LDPE-board
and (b) the SG206-coated LDPE surface. Based on AFM images, the sol-gel coat-ing
formed a smooth and dense layer on the LDPE surface. Furthermore, the
average roughness (Ra) value for the sol-gel-coated LDPE (0.4 nm) was evidently
lower in comparison to that of the uncoated LDPE-board (5.1 nm).
(b) SG206-coated LDPE-board
Figure 14. AFM topography images of (a) the uncoated and (b) the SG206-coated
The oxygen barrier performance of the two different hybrid sol-gel coatings was
studied by comparing their oxygen transmission rate (OTR) values. The aim was to
demonstrate the effect of an additional coating layer on the oxygen barrier perfor-mance
of LDPE. The coating chemistry was expected to have an effect on the
permeability properties of the coatings. In addition, the effect of plasma pre-treatment
on spreadability and uniformity of the film was studied by comparing
OTR values between the coatings deposited with and without plasma pre-treatments.
Figure 15 presents the OTR of the uncoated LDPE and sol-gel coatings (SG204
and SG206) with and without plasma pre-treatment. The oxygen transmission rate of
the LDPE-board was approximately 8700 cm3/(m2*atm*day). All the coated specimens
had lower oxygen transmission in comparison to the uncoated LDPE-board. The
comparison between the non-treated and the plasma-treated coatings showed
some differences. The OTR values of the non-treated SG204 and SG206 coatings
were 7100 and 6900 cm3/(m2*atm*day). The OTR values of the plasma-pretreated
SG204 and SG206 coatings were 7100 and 900 cm3/(m2*atm*day), respectively.
Figure 15. The oxygen transmission rates (OTR) of LDPE-board and sol-gel coat-ings
SG204 and SG206 with and without plasma pre-treatment on the LDPE sur-face
(modified from [P2]).
The differences in the barrier performance between the coatings can be explained
by the coating chemistry. The SG206 coating formed a denser organic network,
where epoxy groups reacted with the amino groups, forming the 3-dimensional
network. The SG204 coating did not form a similar network, due to the lack of
reactive groups for the cross-linking. The results indicate that the plasma pre-treatment
did not affect the barrier performance of the SG204 coating, whereas
the plasma pre-treatment improved the oxygen barrier performance of the SG206
coating. The enhancement after the plasma pre-treatment was most likely due to
the uniformity of the SG206 coating, resulting from the improved wetting.
Oleophobicity and grease barrier
Contact angles of oleic acid on the specimens were measured in order to investi-gate
the oil repellence behaviour. In addition, a decrease in the contact angle
values during the measurement time was expected to demonstrate the grease
barrier performance. Figure 16 presents the contact angle of oleic acid for the
uncoated and SG204-coated LDPE surface. The oleic acid droplets spread rapidly
on the uncoated LDPE surface. The contact angle value was 10º for the uncoated
LDPE after 25 s measurement time. In fact, the contact angle value decreased
constantly during the measurement time, indicating possible absorption of the oil
through or into the LDPE layer. In contrast, the contact angle value was a steady
45º for all the SG204-coated LDPE surfaces with or without plasma pre-treatment.
In addition, the droplets were stable during the measurement time, indicating no
absorption through the sol-gel coated surface.
Figure 16. Contact angle values of oleic acid for the uncoated and the sol-gel
(SG204)-coated LDPE surface with and without plasma pre-treatment (modified
The improved oil repellence performance might be attributed to the formation of
long-tail hydrocarbon groups on the LDPE by the deposition of the SG204 coating.
In addition, the smooth and dense surface structure of the sol-gel coating might
have contributed to the oil repellence. The plasma pre-treatment did not affect the
contact angle value of the coatings. This indicates that the coating had spread
sufficiently on the LDPE even without the pre-treatment.
The effect of the atmospheric plasma pre-treatment on adhesion of the SG204
coating was studied, comparing the oil repellence behaviour before and after
abrasive wearing test. Figure 17 presents the contact angles of oleic acid after the
abrasive wearing for the uncoated LDPE and the SG204-coated LDPE with and
without plasma pre-treatment. The contact angle values of the uncoated LDPE
were remarkably lower after the wearing test. The contact angle values could not
be measured after 3 seconds, indicating the rapid absorption of the oil through the
abraded LDPE layer. In contrast, the sol-gel coated LDPE specimen revealed only
a minor decrease in the contact angles. As expected, the contact angle values
measured after the abrasive wearing showed some differences between the un-treated
and plasma pre-treated LDPE. The contact angle values of non-treated
sol-gel coating decreased down to 20º. However, the contact angle value of 35º
was measured for the plasma-treated sol-gel coating. The comparison between
the abraded sol-gel coated samples with and without plasma pre-treatment sug-gested
that the plasma pre-treatment enhanced the adhesion of the coating.
Figure 17. Contact angle values of oleic acid for uncoated LDPE and sol-gel coated
LDPE specimens after the abrasive wearing (modified from [P2]).
7.1.3 ALD and sol-gel based thin-film composite on PLA-board
Thin-film composite (TFC) structures were aimed to enhance gas and water va-pour
barrier performance of PLA-coated board. The sol-gel (SG200)-based hybrid
coating was intended to form a hydrophilic pre-barrier layer and additionally to
smooth the surface roughness of the PLA surface. An inorganic ALD-Al2O3 layer
was applied on the sol-gel layer in order to make a dense and nanosized barrier
Figure 18 presents AFM topography images of (a) PLA-board and TFC struc-tures,
(b) SG200/PLA-board, and (c) ALD/SG200/PLA-board. The TFC structure
was formed by applying the SG200 coating on the PLA-board, which was followed
by a coating of ALD-Al2O3 layer. Indeed, the SG200 coating, as a pre-barrier layer
covered the larger valleys, pinholes and pores of the PLA surface, making it denser
and smoother. Therefore, the surface is more favourable for the ALD based Al2O3
barrier layers, which was intended to give further densification of the surface.
(a) Uncoated (b) SG200-coated (c) ALD/SG200-coated
Figure 18. AFM images of (a) uncoated PLA-board, and coated PLA-boards: (b)
SG200 and (c) ALD/SG200.
Table 4 summarises the values for average surface roughness (Ra), water contact
angle, OTR and the WVTR of uncoated, SG200-coated and ALD/SG200-coated
PLA-boards. The Ra roughness results indicated that the SG200 coating reduced
the Ra value making the PLA surface smoother. In addition, the SG200 coating
reduced the water contact angle value indicating a more hydrophilic surface. The
ALD-Al2O3 layer on top of the SG200 coating made the surface even smoother.
However, some particle-like ALD growth as noticed, which has also been seen in
other type of ALD experiments [78,79]. The total decrease in Ra was from 54 to
15 nm. Moreover, the water contact angle value dropped even further with the
additional Al2O3 coating.
The high-barrier performance of the TFC structure was demonstrated using
OTR and WVTR measurements. A minor improvement in the OTR and the WVTR
values was obtained with the SG200 coating in comparison to plain PLA-board. A
remarkable decrease in OTR and WVTR values was noticed after applying the
ALD layer on top of the sol-gel layer. The lowest OTR and WVTR values of
2 cm3/(m2*atm*day) and 2 g/(m2 *day) were reached with the ALD/SG200 struc-ture.
Apparently, the obtained OTR and WVTR values of the ALD/SG200 were
lower than those achieved for the ALD coating without the sol-gel (OTR:
6 cm3/(m2*atm*day) and WVTR: 3 g/(m2 *day)).
Table 4. Values of water contact angle (WCA), average roughness (Ra) OTR and
WVTR transmission of plain PLA-board and coated specimens SG200/PLA-board
and ALD/SG200/PLA-board (modified from [P3]).
Uncoated PLA-board 54 71 400±9 75±2
SG200-coated PLA-board 27 58 310±2 44±2
ALD/SG200-coated PLA-board 15 52 2±1 2±1
7.2 Surface modification of thin-film composites with anti-fouling
7.2.1 Characterisation of uncoated and coated thin-film composites
The surface properties and their effect on bacterial anti-adhesion performance
were studied using uncoated and coated polyamide (PA) thin-film composite
(TFC) membranes [P4]. All the membranes were commercially available fully-aromatic
TFC-PA membranes. Two of the TFC-PA membranes (SW30 and
BW30) were coated using polyvinyl alcohol (PVA) coating, while the other three
membranes (LE, XLE and NF90) were uncoated. Table 5 summarises the results
on root-mean-square surface roughness (RRMS), water contact angle and surface
energies with total, polar and dispersive values. The PVA coated membranes
(SW30HR and BW30) had a lower water contact angle and higher surface energy
values in comparison to uncoated membranes (LE, XLE and NF90). Indeed, the
most hydrophilic and highest surface energy was measured for the SW30HR
membrane. In addition, SW30HR showed the highest polar surface energy value,
among the characterised membranes. BW30 membrane had a slightly higher
water contact angle and respectively, lower surface energy. Moreover, the polar
surface energy of BW30 was somewhat lower than SW30HR. Based on the results,
the PVA coating increased the polarity of the TFC-PA membrane surface.
Among the uncoated TFC-PA membranes, the LE membrane clearly had the
lowest values for water contact angle and the highest for the surface energy re-spectively.
The surface energy value for LE was almost at the same level with the
coated membranes. However, comparison of polar surface energy values re-vealed
that the LE membrane surface had a lower polarity. In contrast, among the
characterised membranes the XLE and NF90 membranes showed most hydropho-bic
and non-polar surface properties. In fact, the highest value for water contact
angle and respectively, the lowest value for surface energy were measured for the
XLE membrane. Similarly, the NF90 membrane showed hydrophobic surface prop-erties.
The NF90 membrane had a slightly lower value for water contact angle value
and a higher value for surface energy respectively in comparison to the XLE. How-ever,
the polarity of XLE membrane was higher compared to NF90 membrane.
The RRMS roughness values for the coated membranes SW30HR and BW30
were somewhat lower in comparison to the uncoated membranes (LE, XLE and
NF90). The slightly decreased roughness is most likely due to deposition of a very
thin coating layer on the BW30 and SW30HR membranes. The coating layer might
have filled the valleys of the PA selective layer. The high deviation in roughness
may be explained by the uneven deposition of the coating. Among the uncoated
TFC-PA membranes, LE and NF90 had a lower surface roughness than XLE. The
uncoated membranes showed similarly high deviation of the roughness values.
This can be explained by the differences in the formation of the PA selective layer,
resulting in an uneven peak-and-valley surface structure.
Table 5. Summary of the results on water contact angle and surface energies.
SW30HR 28 ± 5 66 24 42 70 ± 12
BW30 43 ± 4 56 19 37 57 ± 5
LE 53 ± 9 54 11 43 75 ± 23
XLE 76 ± 5 38 4 34 96 ± 21
NF90 70 ± 4 42 2 40 76 ± 12
Bacterial anti-adhesion performance
Figure 19 presents the results of the bacteria viability analysis of the membrane
samples after a 24 hour exposure to P. aeruginosa. The attachment of the bacte-ria
was evidently different between the membranes. Indeed, the decrease in the
number of viable cells can be observed simultaneously with the increase of hydro-philicity
and surface polarity (Table 5). The lowest amount of bacteria was at-tached
to the most hydrophilic and polar SW30HR and BW30 membranes. Among
the uncoated membranes, the LE membrane had a somewhat lower amount of
attached bacteria in comparison to the XLE and NF90 membranes. Indeed, the
amount of bacteria found on the uncoated membrane surface increased in the
order: LE, XLE and NF90.
The correlation between the polar surface energy and the attachment of bacte-ria
was investigated in order to gain further knowledge of the effect of the surface
energy on the attachment of bacteria. Apparently, the polar surface energy values
clearly correlated with the attachment of the bacteria. The amount of bacteria
decreased identically with the increase of polar surface energy value. In addition,
the type of surface roughness may somehow contribute to bacteria repellence.
However, an exact conclusion cannot be presented due to the heterogeneous
surface topography of the membranes.