Corrosion Resistance and Formability of Ultra-thin
Plasma Polymer Films on Galvanised Steel
Dissertation
Zur Erlangung des akademischen Grades eines
Doktors der Naturwissenschaften (Dr. rer. nat.)
der Fakultät für Naturwissenschaften der
Universität Paderborn
vorgelegt im Juni 2009 von
Dipl.-Phys. Tobias Titz
Die vorliegende Arbeit wurde angefertigt am
Max-Planck Institut für Eisenforschung GmbH
in Düsseldorf
Referent: Prof. Dr.-Ing. G. Grundmeier
Department Chemie
Fachgebiet für Technische und Makromolekulare Chemie
der Universität Paderborn
Korreferent: Prof. Dr. C. Schmidt
Department Chemie
Fachgebiet für Physikalische Chemie und Makromolekulare Chemie
der Universität Paderborn
´
Tag der mündlichen Prüfung: 2. Juli 2009
Danksagung – Acknowledgement
Die vorliegende Arbeit wurde während meiner Tätigkeit als wissenschaftlicher Angestellter
der Max-Planck-Institut für Eisenforschung GmbH in Düsseldorf angefertigt.
An erster Stelle möchte ich mich bei meinem Doktorvater Herrn Prof. Dr.-Ing. G. Grundmeier
sehr herzlich bedanken für die Vergabe meines jederzeit spannenden Themas, für die
intensive und konstruktive Betreuung meiner Arbeit und die Möglichkeit, meine
wissenschaftlichen Ergebnisse auf verschiedenen Konferenzen und Tagungen zu präsentieren.
Frau Prof. Dr. C. Schmidt danke ich sehr herzlich für die freundliche Übernahme des
Koreferats.
Herrn Prof. Dr. M. Stratmann gilt mein Dank für die Möglichkeit der Durchführung meiner
experimentellen Arbeiten am MPIE in der Abteilung Grenzflächen und Oberflächentechnik.
Allen Mitarbeitern der Abteilung danke ich sehr herzlich für die angenehme Atmosphäre und
die stete Hilfsbereitschaft bei allen Anliegen.
Der Firma OCAS danke ich für die Bereitstellung der finanziellen Mittel sowie Herrn Dr. F.
Hörzenberger und Frau K. Van den Bergh für die zahlreichen und anregenden Diskussionen
während der Projekttreffen.
Ein ganz ganz herzliches Dankeschön geht an Fr. P. Ebbinghaus, Dr. N. Fink, Dr. M. Giza,
Dr. P. Keil, G. Klimow, Dr. I. Klüppel, Ö. Ozcan, R. Posner, Dr. J. Raacke, M. Santa, Dr. M.
Valtiner, R. Vlasak und Dr. K. Wapner und vielen anderen, die meine Zeit am MPIE mit
interessanten Messungen und Diskussionen sowie zahlreichen Anekdoten einfach
unvergesslich gemacht haben.
Mein besonderer Dank gilt meiner gesamten Familie sowie allen Freunden für ihr Verständnis
und ihre tatkräftig Unterstützung.
Bei meiner geliebten Frau Jolante werde ich mich niemals genug für ihr Verständnis, ihre
Unterstützung und ihre unermüdliche Geduld bedanken können. Unserem Sonnschein Sophie
Amelie danke ich vor allem für ihr bezauberndes Gemüt, das sonnige Lächeln und vor allem
für ihren gesunden Schlaf während des Schreibens meiner Dissertation.
Index 1
Index
Index ..........................................................................................................................................1
Symbols and abbreviations........................................................................................................4
1 State of research and motivation ......................................................................................7
2 Fundamentals....................................................................................................................9
2.1 Corrosion and corrosion protection of galvanised steel ................................................... 9
2.1.1 Degradation of organically coated zinc substrates .......................................................................... 9
2.1.2 Corrosion protection of zinc by thin plasma polymer films.......................................................... 12
2.1.3 The Kelvin probe – a unique tool for in-situ corrosion studies ..................................................... 13
2.2 Cold plasmas....................................................................................................................... 15
2.2.1 Plasma modification of metal surfaces.......................................................................................... 15
2.2.2 Plasma polymerization of organosilanes....................................................................................... 16
2.3 Forming of coated metals.................................................................................................. 18
2.3.1 Fundamentals of forming of metal sheets ..................................................................................... 18
2.3.2 Forming of zinc and zinc coated metals........................................................................................ 22
2.3.3 Forming of plasma polymer films.................................................................................................26
3 Experimental measurement techniques .........................................................................29
3.1 Electrochemical measurement techniques....................................................................... 29
3.1.1 In-situ and scanning Kelvin probe................................................................................................. 29
3.1.2 Electrochemical impedance spectroscopy..................................................................................... 30
3.1.3 Cyclic voltammetry....................................................................................................................... 32
3.2 Spectroscopic measurement techniques........................................................................... 33
3.2.1 Infrared spectroscopy.................................................................................................................... 33
3.2.2 X-ray photoelectron spectroscopy................................................................................................. 33
Index 2
3.2.3 Spectroscopic ellipsometry ........................................................................................................... 34
3.3 Microscopic techniques ..................................................................................................... 35
3.3.1 Force microscopy.......................................................................................................................... 35
3.3.2 Scanning electron and Auger microscopy..................................................................................... 35
4 Experimental set-up and sample treatment....................................................................37
4.1 Materials............................................................................................................................. 37
4.1.1 Used substrate material ................................................................................................................. 37
4.1.2 Substrate cleaning and pre-treatment ............................................................................................37
4.1.3 Application of organic top coat..................................................................................................... 38
4.2 Plasma modification........................................................................................................... 39
4.2.1 In-situ plasma treatment of metal surfaces.................................................................................... 39
4.2.2 PE-CVD film deposition ............................................................................................................... 40
4.3 Forming of samples............................................................................................................ 41
4.3.1 Miniature tensile forming device .................................................................................................. 41
4.3.2 In-situ cyclic voltammetry and forming set-up ............................................................................. 41
4.3.3 Tensile forming device for ex-situ experiments............................................................................ 43
5 Corrosion resistance of plasma modified HDG steel.....................................................45
5.1 Stability and chemical composition of oxygen plasma modified zinc surfaces............. 46
5.1.1 In-situ FT-IR analysis during plasma treatment............................................................................ 46
5.1.2 Chemical surface change after plasma treatment .......................................................................... 47
5.1.3 Influence of plasma treatment on Volta potential ......................................................................... 49
5.1.4 Stability of zinc surfaces after plasma modification ..................................................................... 50
5.2 Deposition of ultra-thin plasma polymer films................................................................ 51
5.2.1 In-situ FT-IR analysis after plasma deposition.............................................................................. 52
5.2.2 Chemical composition of plasma polymer films........................................................................... 53
5.2.3 Change of surface structures after film deposition........................................................................ 54
5.3 Barrier properties of SiO2-like films for different film thickness values...................... 55
5.3.1 Surface coverage for different film thickness values .................................................................... 55
5.3.2 Barrier properties - pore and film resistance ................................................................................. 56
5.4 Influence of intact SiO2-like films on unformed zinc surfaces....................................... 58
5.4.1 Kelvin probe analysis in vacuum and humid atmospheres............................................................ 58
5.4.2 Corrosion processes and kinetics during the de-adhesion processes in corrosive environments .. 62
5.4.3 Influence of SiO2-like films on the cathodic protection of iron by zinc coatings.......................... 64
5.5 Conclusions and model...................................................................................................... 66
Index 3
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel...............69
6.1 Forming of uncoated and coated HDG steel.................................................................... 70
6.1.1 Forming of uncoated HDG steel ...................................................................................................70
6.1.2 Forming of thin SiO2-like films on HDG steel.............................................................................. 71
6.1.3 Forming of ultra-thin SiO2-like films on HDG steel ..................................................................... 74
6.2 Barrier properties and corrosion resistance of SiO2-like films after forming.............. 75
6.2.1 In-situ cyclic voltammetry during stretch forming of thin film coated substrates......................... 75
6.2.2 Micro- and nanoscopic Kelvin probe studies of defects and interfacial corrosive de-adhesion.... 79
6.3 Conclusions and model...................................................................................................... 84
7 General conclusions........................................................................................................87
8 Outlook.............................................................................................................................89
9 Literature.........................................................................................................................91
10 Publications related to this work...............................................................................103
Symbols and abbreviations 4
Symbols and abbreviations
Latin symbols
a edge size of hexagonal cell
+
Zn
a Zn+ activity
A surface area
A0 initial sample cross-sectional area
c height of hexagonal cell
C capacitance
CC coating capacitance
CDL double layer capacitance
CE counter electrode
d film thickness
D spring constant
E applied voltage
Ecorr electrode potential
Eg band gap
f frequency
F Faraday constant
Fext external force
I applied current
IAC alternating current
l final gauge length
l0 initial gauge length
P applied load
R universal gas constant
Symbols and abbreviations 5
RC ohmic resistor
RCt pore resistance
Re elastic limit
Rohm ohmic resistance
RM ultimate tensile strength
RU uncompensated electrolyte resistance
RE reference electrode
Uext applied external voltage
WRef work function of reference metal
WE working electrode
Z complex impedance
Greek symbols
ε0 dielectric vacuum constant
ε½ half-cell potential of reference metal
εr dielectric material constant
εt technical deformation
θ phase
μe chemical potential
σ true stress
φi forming degree (i=1, 2, 3)
χEl surface dipole potential of electrolyte
χPol surface dipole potential of polymer surface
ω radial frequency
Δ phase difference
Δl elongation
ΔΦ Galvani potential difference
ΔΦD Galvani potential difference (Donnan potential)
ΔΨ Volta potential difference
Θfilm film surface area
Ψ amplitude ratio
Symbols and abbreviations 6
Abbreviations
AFM atomic force microscopy
AES scanning Auger electron spectroscopy
CV cyclic voltammetry
EBSD electron backscattered diffraction
EIS electrochemical impedance spectroscopy
FT-IRRAS Fourier transformed infrared reflection absorption spectroscopy
hcp hexagonal closed-packed
HDG hot-dip galvanised steel
HMDSO hexamethyldisiloxane
HR-SKP height-regulated scanning Kelvin probe
MCT mercury cadmium telluride
PE-CVD plasma enhanced chemical vapour deposition
PET polyethylene terephthalate
SE spectroscopic ellipsometry
SEM scanning electron microscopy
SHE standard hydrogen electrode
SKP scanning Kelvin probe
SKP-FM scanning Kelvin probe force microscopy
TEOS tetramethoxysilane
THF Tetrahydrofuran
XPS x-ray photoelectron spectroscopy
1 State of research and motivation 7
1 State of research and motivation
Corrosion resistant polymer/zinc interfaces are of high importance for organically coated or
adhesively bonded galvanised steel substrates [1-4]. Moreover, zinc oxides are applied in the
construction of photovoltaic cells and also in this case have to be protected by thin films [5,
6]. Plasma technology has become an interesting and environmentally friendly dry process at
atmospheric or reduced pressures to modify oxides and oxide covered metal surfaces [7-9].
In the past, several studies concerning the corrosion resistance of metal substrates covered
with thin plasma polymer films were carried out [10-13]. However, the mechanism of
corrosion protection of organically coated plasma modified surfaces is still poorly understood.
Grundmeier et al. showed that a plasma oxidation of iron oxide films on iron leads to a
change in their electronic properties [14]. However, the highly oxidised state of the passive
film was not stable and changed again in humid atmospheres to the state which is determined
by the oxygen reduction kinetics at the outer oxide surface. This process is also determined by
the kinetics of the metal dissolution into the passive film at the metal/oxide interface.
Barranco et al. showed that for cathodically deposited plasma polymer films the interfacial
electrode potential on iron could be effectively shifted cathodically which indicate a strong
inhibition of interfacial oxygen reduction kinetics [15]. However, a reductive plasma
treatment had to be followed by a sufficiently thick plasma polymer film deposition to prevent
the plasma polymer/metal interface from re-oxidation. Based on this work electrochemical
methods could be established for an improved understanding of how plasma modifications
influence the corrosion mechanism of thin film coated metals.
From the viewpoint of thin film engineering, it is the aim to achieve functional properties at
extremely low thickness values. With regard to interfacial corrosive de-adhesion of polymer
coated zinc and iron substrates, the interfacial oxygen reduction kinetics should be strongly
inhibited by the respective plasma polymer film [15].
1 State of research and motivation 8
Forming of corrosion protected metal sheets is one of the most important applications in many
industries for economic reasons. The formation of forming induced defects is not only limited
on defects in the protective inorganic [16-19] or organic [20] layers but also on substrate
defects [21-24]. The formation of defects in brittle films on ductile substrates has been
intensively studied both experimentally and theoretically [17, 25]. Usually multiple sequential
crack formation is observed for more or less brittle films on ductile or high elongation
substrates. By increasing the strain the number of cracks increases as well. This observation is
explained by the so-called shear lag approximation presented in detail by Wojciechowski and
Mendolia [16, 17].
Depending on the applied stress and strain different defect mechanism can be observed on
pure zinc coated steel sheets [21-23]. Slip, twinning and cracking are the major deformation
modes on zinc grains depending on the applied stress. Due to the hcp structure of zinc, basal
slip is the easiest deformation mode in zinc and is therefore the predominant defect mode [26-
28]. Twinning can be found on zinc grains if compressive and tension stress is applied
simultaneously [21].
As it is well known from literature ceramic films can only be formed for less than 1%
elongation. Plasma polymer films are intended to follow the deformation of the substrate due
to their high cross-linking during plasma deposition [9-11, 29]. B. Baumert et al. showed that
plasma polymer films on steel show substrate induced defects for high forming degrees [24].
The aim of the present work is to investigate the corrosion protection of ultra-thin barrier
plasma polymer films on a complex alloy surface. Therefore, the investigation focuses on hot-
dip galvanised steel substrates with an aluminium zinc alloy coating.
Low-pressure plasma polymerization coatings are used as they are formed by an economically
advantageous dry process which uses a minimum amount of energy and substances compared
to wet chemical corrosion protection coatings.
This work focuses especially on the understanding of the corrosion protection, film thickness
dependence and the influence of the formability of these ultra-thin plasma polymer films.
The correlation of nanoscopic defect structures and macroscopic performance of the corrosion
protective system should be evaluated by electrochemical analysis and corrosion studies of
the organically coated composites in humid and corrosive environments.
2 Fundamentals 9
2 Fundamentals
2.1 Corrosion and corrosion protection of galvanised steel
Zinc and zinc alloy coatings on steel are some of the commercially most used methods to
protect steel components which are exposed to corrosive environments. The use of
galvanisation techniques like hot-dip galvanising to deposit protective zinc coatings are well
established since centuries [30].
Zinc coatings protect the underlying steel in several ways. As a barrier layer, the zinc coating
separates the steel from the corrosive environment. Zinc itself acts as a sacrificial anode to
galvanically protect the underlying steel at defects, scratches and cut edges of the zinc coating
[30-32].
However, because of new requirements of the industry, additional protective coatings are
necessary to increase the corrosion protection. Besides the conventional surface treatments for
reactive metals like wet chemical phosphatisation, atmospheric and vacuum plasma
deposition of inorganic coatings offer an environmentally friendly alternative [8-10, 12, 13,
24, 33-35].
2.1.1 Degradation of organically coated zinc substrates
For intact organic coatings the metal surface is effectively protected against corrosion. If a
polymer/metal composite is exposed to environments with high water activites, the interfacial
adhesion force between the metal surface and the organic coating is lowered. This happens
because of the strongly negative Gibbs energy of the adsorption of water molecules on the
oxide or hydroxide covered metallic surface according to
zyxaqzyxg OHOMeOHOHOMeOH
⋅
⋅
⋅
→+ )(2)(2 (1)
2 Fundamentals 10
In any case water molecules are able to penetrate through the polymer to the interface and
adsorb at the polymer/oxide interface. Here, the water molecules adsorb immediately at free
adsorption sites and lead to a displacement of physisorbed organic functionalities. In most
cases the adhesion forces of the molecules are still high enough that the coating adheres. A
very detailed description of the adhesion and de-adhesion mechanisms at polymer/metal
interfaces is published in the review by Grundmeier and Stratmann [36].
The de-adhesion of the polymer from the metallic or inorganic substrate can be described by
the picture of breaking bonds at the interface. This can occur by chemical displacement
reactions (e.g. physisorbed water molecules), by mechanical force or by chemical reactions
which alter the chemical bonds at the interface. Organically coated metals like iron or zinc
degrade at the interface by the oxygen reduction process [37-40]. During oxygen reduction a
strong alkaline electrolyte is formed which can stabilise or destabilise the metal oxide. The
delamination of the organic coating originates only from the bond breaking within the organic
coating interface. Intermediate radicals such as OH2
-, OH and −
2
O are formed during
oxygen reduction and participate in the degradation of the organic coating. Wroblowa [39]
proposed an initial step of the reaction mechanism for the oxidative degradation of the
polymer as follows
−•− +→−+ OHRHRO 22 (2)
with the subsequent reactions:
•• →+ ROOOR 2 (3)
•• +→−+ RROOHHRROO (4)
Additionally to the chemical reaction of the oxygen reduction products the change of the pH
at the interface due to fact that zinc oxide is not stable in alkaline environments can contribute
to the delamination process [41].
If a defect occurs in the polymer coating and ions and water molecules can directly access the
bare zinc or iron surface electrochemical reactions will take place at the metal electrolyte
interface. Due to these reactions delamination of the organic polymer coating will occur and
the intact substrate/polymer interface will be replaced by two new interfaces
substrate/electrolyte and electrolyte/polymer. The electrochemical reactions will therefore
expand from the defect below the organic coating and the change in the electrochemical
reactivity of this interface will be rejected in a local change of the free electrode potential as it
2 Fundamentals 11
Zinc
Na
+
,
Cl
-
oxide
organic coating
O
2
, H
2
O
i
Δφ
i
Δφ
+
i
Δφ
..
..
I · R
O
2
+ 4e
-
+ 2H
2
O →4OH
-
2Zn →2Zn
2+
+ 4e
-
Δφ‘
Δφ
Zinc
Na
+
,
Cl
-
oxide
organic coating
O
2
, H
2
O
i
Δφ
i
Δφ
+
i
Δφ
..
..
I · R
O
2
+ 4e
-
+ 2H
2
O →4OH
-
2Zn →2Zn
2+
+ 4e
-
Δφ‘
Δφ
Figure 1: Principle schematic of the corrosion model explaining the formation of a galvanic element
on zinc. Central picture: cross-section with defect filled with electrolyte (Na+ ,Cl- ions) and an intact
oxide/organic coating interface. Polarization curves at the defect, the intact interface and the result
after galvanic coupling of both elements for zinc (in accordance to [40]).
is shown in Figure 1. The resulting principle schematic of the corrosion model is
comprehensively summarized by Fürbeth et al. for galvanised steel and for steel by Leng et al.
[40, 41].
Due to the metallic substrate, a galvanic element can be formed between the defect with the
electrolyte and the intact interface of the polymer coating and the zinc surface. Zinc
dissolution occurs at the defect area when the metal is in contact with the electrolyte. At the
intact interface oxygen reduction of molecular oxygen in the presence of water molecules
occur. The rate of the metal dissolution is limited as the same amount of electrons has to be
provided by the oxygen reduction at the intact interface. Within this galvanic element cations
2 Fundamentals 12
have to be transported from the local anode (defect) to the local cathode (delaminated area). If
the oxygen reduction is reduced by a barrier layer in top of the zinc oxide, the metal
dissolution and also the delamination is effectively inhibited.
2.1.2 Corrosion protection of zinc by thin plasma polymer films
Plasma polymerisation, as a process technology for corrosion resistant thin film deposition
has been explored during the last 20 years [7, 42]. Recent studies reveal the good corrosion
protection properties of organosilicon based plasma polymers on steel substrates and the
crucial influence of the pre-treatment process on the stability of the resulting interface [13,
34]. The pre-treatments for trimethylsilane based films consist of an oxidative step (O2-
plasma) to remove organic contaminations from the substrate and a second reductive step
(Ar/H2-plasma) to remove the metal oxide layer. While the successive application of both
steps provides the best corrosion protection of various plasma treatments for steel in
combination with a cathodic electrocoat, little is known about the chemical structure of the
interface. Yasuda and van Ooij in particular have shown that the deposition of plasma
polymers on steel and galvanised steel might even substitute the chromatation process [13, 34,
43, 44].
Very little knowledge is gained about the corrosion protection of these coatings when stress is
applied on the coated steel substrates. Most studies focus on the formation of defects in the
plasma polymer coating during forming, but no structure-relation between corrosion
protection and crack formation was studied in detail [24, 45].
2 Fundamentals 13
2.1.3 The Kelvin probe – a unique tool for in-situ corrosion studies
The Kelvin probe (KP) is a unique tool for corrosion studies and can be used in vacuum,
corrosive environments like electrolytes and high humidity and also in combination with
atomic force microscopy [40, 41, 46-48]. Stratmann et al. studied polymer coated metals on
iron and zinc as the Kelvin probe can measure the Volta potential at the buried interface
between the coating and the metal surface [36, 40, 41, 46, 47]. The principle measurement
set-up is shown in Figure 2. If two metals are brought into contact, the Fermi level of both
equalizes. In case of differing work functions of both metals, charging of one material in
respect to the other materials occurs. By using a capacitor and a vibrating needle as shown in
the schematic, the charge transfer can be measured as the formed capacitor (needle surface to
metal surface) is loaded and unloaded periodically.
By applying an external voltage to the capacitor, the charge transfer can be zeroed. In this
case, the applied external voltage is identical to the Volta potential difference between the two
metals. The Volta potential difference directly correlates with the Fermi level difference. If
one Fermi level is known or the value can be calibrated, than the missing Fermi level of the
unknown material can be calculated.
metal
coating U
ext
I
ac
vibrating needle
d
0
Δd
Δd
ref
sample
ΔΨ
x
metal
coating U
ext
I
ac
vibrating needle
d
0
Δd
Δd
ref
sample
ΔΨ
x
Figure 2: Schematic of the measurements of electrochemical potentials at buried interfaces with the
Kelvin probe
2 Fundamentals 14
The article from Grundmeier and Stratmann summarizes the theoretical basis of the Kelvin
probe [36]. The work function of a metal is defined as the work to remove an electron from
the metal to a point far away from the surface. Under high vacuum conditions with clean
surfaces the work function of the metals is well defined.
If the metal is covered with aqueous electrolytes or organic polymers, additional interfaces
have to be passed and the measured Volta potential difference is changed by the additional
potential differences across the interfaces. The identical behaviour can be found for the
electrode potential which is determined by the potential difference across the electrified
interface. Therefore, a simple relation exists between the Volta potential difference f
Pol
Re
ΔΨ
and the electrode potential Corr
E of the buried interface. For liquid phases covering one metal
surface, the following expression can be used
f
Pol
f
El
f
Corr F
W
EReRe
2
1
Re ΔΨ+
⎩
⎨
⎧
⎭
⎬
⎫
−−=
εχ
(5)
where F is the Faraday constant, f
WRe the work function of the reference metal, El
χ
the
surface dipole potential of the electrolyte and fRe
2
1
ε
is the half-cell potential of the reference
metal. If less oriented polymers with small dipole moments are used, a similar expression can
be used
f
Pol
f
Pol
f
Corr F
W
EReRe
2
1
Re ΔΨ+
⎩
⎨
⎧
⎭
⎬
⎫
−−=
εχ
(6)
with the surface dipole Pol
χ
of the polymer surface. In this case, the Kelvin probe monitors
the buried interface without touching the surface. For reactive metals, the metal/electrolyte
interface is not stable. However, the metal surface is covered by a thin native oxide layer.
Therefore, the Volta potential difference must be expressed by the sum of the potential
differences over all interfaces:
Pol
f
Me
e
Ox
PolOx
f
Ox
f
Pol F
W
F
χμ
+−−ΔΦ+ΔΦ+ΔΦ=ΔΨ Re
ReRe 1 (7)
whereas ΔΦ is the respective Galvani potential difference and e
μ
is the chemical potential of
the electron in the metal. By replacing the Galvani potential across the oxide layer by the
corresponding change in chemical composition, then the Volta potential difference represents
2 Fundamentals 15
the oxidation level within the oxide at the metal/polymer interface, as e.g. for an oxide
covered iron interface:
[
]
[]
+
+
++−ΔΦ+
Δ
−=ΔΨ ++
2
3
Re
0
/
Re ln
23
Fe
Fe
F
RT
F
W
FPol
f
Ox
Pol
FeFe
f
Pol
χ
μ
(8)
During de-adhesion an additional liquid phase is formed between the substrate and the
organic phase. Then the metal/electrolyte interface can be treated as a conventional
electrochemical interface but with an additional Galvani Potential difference D
ΔΦ (Donnan
potential):
f
PolD
f
Pol
f
Corr F
W
EReRe
2
1
Re ΔΨ+ΔΦ+−−=
εχ
(9)
Therefore, the Volta potential difference allows the measurement of the corrosion potential at
the buried interface between metal and polymer only if the Donnan potential is know or small.
Simple lacquers for corrosion protection have only some fixed ionic groups and as their
concentration is very small, the Donnan potential can be neglected.
2.2 Cold plasmas
2.2.1 Plasma modification of metal surfaces
Plasma processing is well known in industry and science as many applications like film
deposition, etching or surface modifications use a quasi neutral gas to modify surfaces [8, 42].
The physical properties of plasmas are mainly influenced by the amount of free charge
carriers, free electrons and ionized atoms or molecules. As conventional plasmas are
macroscopically quasi neutral by comparing the amount of positive and negative charge
carriers, the movement of the charge carrier is limited due to Coulomb forces.
Different types of plasmas can be distinguished, isothermal and low temperature (cold)
plasmas. In the case of the cold plasmas, the temperature of the electrons and therefore their
kinetic energy is much higher than the temperature of the positive charge carriers. In this case
the ionisation of the plasma occurs due to free electrons with high energy which are powered
by an external energy field. Low temperature plasmas have the advantage that the substrate
material is only moderately heated during plasma interaction in comparison to isothermal
plasmas with high ion energies of the activated plasma species.
2 Fundamentals 16
Nevertheless the formation and modifications of thin oxide films by low temperature plasmas
is used by several groups to tailor the surface reactivity of metal oxides. Grundmeier et al.
have studied e.g. the modification of native oxide films on iron and zinc by oxidizing and
reducing plasmas to adjust the surface chemistry [14, 15, 49]. The consequences of low
temperature plasma modification of zinc oxide surfaces based on the work of Grundmeier et
al. will be discussed in detail in chapter 5.1.3.
2.2.2 Plasma polymerization of organosilanes
The plasma polymerization of organosilane precursors under various conditions is a topic that
has received a great deal of attention by many researchers in recent years and has been
extensively reported in various literatures. A first summarizing article about plasma
polymerizations in glow discharges was published by Yasuda in 1981 summarizing the work
of different authors, which have described plasma polymer films as by-products of
phenomena associated with electric discharge in the early 20th century. By recognition of
various advantages of plasma polymer films more and more work was focused on the
deposition and characterization of these films. However many authors, especially those trying
to deposit SiO2 like films, didn’t classify the plasma polymerised siloxane films as organically
modified ceramics despite the presence of carbon and hydrogen.
Substrate
Plasma polymer
Starting material
(precursor)
Non-polymer
forming fragments/gas
Polymer forming
intermediates
Plasma or glow dis-
charge polymerization
Plasma induced
polymerization
Fragmentation
Ablation
Film deposition
Substrate
Plasma polymer
Starting material
(precursor)
Starting material
(precursor)
Non-polymer
forming fragments/gas
Non-polymer
forming fragments/gas
Polymer forming
intermediates
Polymer forming
intermediates
Plasma or glow dis-
charge polymerization
Plasma induced
polymerization
Fragmentation
Ablation
Film deposition
Figure 3: Schematic representation of the plasma polymerization mechanism according to Yasuda [8].
2 Fundamentals 17
The plasma deposition processes can be separated into two classes, direct and remote plasma
deposition by using different power supplies and experimental set-ups. When an organosilane
precursor is introduced into the plasma zone, it is ionized and fragmented by high-energy
electrons and ions inside the plasma. At low pressure and therefore a long mean free path,
ionized molecular fragments and ions are more likely to collide with the chamber walls then
with other gas phase species and hence adsorb and react on surfaces within the reactor to
grow thin films of similar chemical composition and structure in comparison with the injected
precursor monomer. The schematic film forming processes are shown in Figure 3.
This plasma polymerization process has been used to deposit hybrid films from a variety of
organosilane precursors with applications ranging from corrosion protection [50-52],
biomedical films for implants [53], barrier coatings for packaging [54-56] to dielectric layers
in integrated electronic circuits [57].
For the deposition of organically modified ceramics, the most popular precursors are
hexamethyldisiloxane (HMDSO, see Figure 4) and tetramethoxysilane (TEOS) often in
combination with oxygen and/or a variety of dopants. The chemical and physical properties of
the resulting films depend strongly on the process parameters, such as the way of coupling the
power into the gas, injection of the precursor and the amount of oxidants.
Coatings produced from siloxane-based precursors have the general composition SiOxCyHz
where x ≤ 2 and y and z can vary up to the stoichiometry of the used precursor.
The chemistry of the siloxane plasma process is complicated by the wide variety of
compounds formed by both carbon and silicon. However, some general observations can be
made.
Si O Si
CH3
CH3
CH3
CH3
CH3
CH3
Figure 4: Chemical structure of HMDSO.
2 Fundamentals 18
Plasma polymers are generally characterized by an extremely tight and three-dimensional
network. At relatively low discharge power and without addition of oxygen the deposited
films generally have a similar elemental composition as the precursor. Frascassi et al. showed
for higher discharge power that the concentration specific molecular groups (e.g. C-H, C-O)
and the functionality of is different while the elemental composition is still identical to that of
the precursor [58].
Several studies compared the film composition to process parameters and precursors [13, 59-
63]. The addition of oxygen to the siloxane plasma to deposit SiOx-like films further
complicates the plasma chemistry and has been also observed in films even when no gaseous
oxygen was directly added to the plasma. Many authors have attributed the oxygen to
irremovable water molecules which are adsorbed on the surface of the chamber and the
substrate surface. However, the hydroscopic nature of the precursor itself can be the source
for oxygen which is incorporated in the deposited film.
Molecular oxygen in the plasma zone is readily fragmented by the plasma state and forms
highly reactive atomic oxygen and ozone which react with the siloxane precursor. At low
oxygen concentrations, the atomic oxygen highly increases the film deposition rate of the
plasma but also steadily etches the deposited film surface and by removing carbon and retards
the deposition rate at the film surface. Increasing the fraction of oxygen in the plasma zone,
the resulting deposited films change their composition from being similar to the precursor to
SiO2-like films. However, high purity SiO2 films are difficult to obtain as the film network
strongly depend on the process parameters. As the SiO2 films are deposited from small
particles with nanometre size, voids inside the coating cannot be avoided without using high
energy ion bombardment or high substrate temperatures.
2.3 Forming of coated metals
2.3.1 Fundamentals of forming of metal sheets
The theoretical descriptions of metals based on the view that metal atoms are orientated
regularly and form a closed packed lattice. In the ideal case single crystals are formed. In
most cases technical materials are polycrystalline materials. During cooling of the metal
different orientated single crystals (grains) are formed. The sizes of these grains depend on the
cooling rate and the composition of the metal or alloy. During forming of metals by an
2 Fundamentals 19
external force atoms are moved against each other on an atomistic length scale.
Macroscopically the metal is deformed either reversible and elastic or irreversible and plastic.
To plastically deform a polycrystalline material the direct stress must be higher than on a
single crystalline material. This is based on the fact, that the applied stress on an individual
grain can also change neighbouring grains with their individual stress levels [64, 65].
From a technical point of view, the mechanical properties of metal sheets are evaluated by so-
called elongation-strain diagrams. An extensive introduction of sheet metal forming is given
by Marciniak et al. and Lange and Döge et al. [66-68]. The elongation of a sample is recorded
versus the applied strain. By normalizing the measured strain with the cross-sectional area of
the sample the so-called engineering stress-strain curve is calculated. Thereby the material
properties are independent of the initial dimensions of the sample. To overcome the fact, that
the strain is always measured on the original sample length, the study of the forming process
is based on the true stress and true strain measurement.
For metal forming a constant volume of the test sample can be assumed and therefore the true
stress is defined as
00 l
l
A
P
=
σ
(10)
where 0
A is the initial sample cross-sectional area and 0
l is the initial gauge length of the
sample. l defines the final gauge length and P is the applied load.
The deformation degree of a sample is normally written as technical deformation t
ε
given in
percent. A second definition for the forming degree is the logarithmic forming degree
ϕ
,
which change is defined as the length change dl of the actual length l of the test specimen:
l
dl
d=
ϕ
(11)
For very small strains, the strain increment is very similar to the engineering strain, but for
larger strains there is a significant difference. If the straining process continues uniformly in
one direction till 1
l is reached, as it does in the tensile test, the strain increment can be
integrated to represent the true strain:
0
ln
1
0l
l
l
dl
d
l
l
=== ∫∫
ϕϕ
(12)
2 Fundamentals 20
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
0
50
100
150
200
250
300
True stress / MPa
True strain (ϕ)
Figure 5: True stress-strain diagram of a tensile tested HDG steel versus logarithmic deformation
degree.
The use of the logarithmic forming degree has the advantage, that forming steps can easily be
summed up. The typical true stress versus true strain diagram of the used HDG steel is shown
in Figure 5.
Comparing the derived stress – strain curves, the forming of metal sheets can be divided in
different mechanism. Depending on the external applied force reversible, elastic and
irreversible plastic deformation takes place.
The elastic tension of the metal sheet is described by the Hook’s straight where the material
behaves like a spring and returns to the original form according to Hook’s law with the
applied external force Fext, the spring constant D and the elongation l
Δ
:
lDFext
Δ
⋅
=
(13)
As most metals do not show a defined elastic limit, typically the elastic limit for metals is
below 0.2% elongation of the initial length of the test specimen. This point at the end of the
Hook’s straight is called elastic limit Re. From this point the plastic deformation starts and the
material will be formed irreversible.
2 Fundamentals 21
The maximum true stress is called the ultimate tensile strength RM and is calculated as
0
max
A
P
RM= (14)
Until RM the sample will be formed without constricting and this region is called uniform
strain as the sample keeps its shape. Beyond RM the sample constricts in the central part till
further forming will break the sample in this area.
The forming of metal sheets can be done in different ways depending of the form of the
sample and the tooling machine. Four major groups can be separated uniaxial, biaxial, plane
strain and deep drawing. The forming process can be described by three individual forming
degrees 3,2,1
ϕ
, which are parallel to the direction in space. 1
ϕ
represents the major strain with
the largest change of the dimension of the test specimen. 2
ϕ
indicates the change of the width
of the metal sheet and 3
ϕ
represents the change of the sheet thickness.
For all forming mechanisms the volume of the specimen is constant during forming and
therefore the sum of the three forming degrees is zero:
0
321
=
+
+
ϕ
ϕ
ϕ
(15)
All forming operations related to this thesis are done during tensile testing. Therefore, all
samples were formed by uniaxial deformation of the metal samples. This means, that the
samples were elongated in the direction of 1
ϕ
, whereas the width of the samples ( 2
ϕ
)
decreases. The correlation between the major and minor strains is given in equation 16.
21 2
ϕ
ϕ
−
= (16)
The comparison of the uniaxial forming and further forming operations like plane strain, deep
drawing and biaxial forming is shown in the forming limit diagram in Figure 6.
2 Fundamentals 22
-0.3 -0.2 -0.1 0.0 0.1 0.2 0.3
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
forming limit
ϕ1 = ϕ2
Major forming degree ϕ1
Minor forming degree ϕ2
deep drawing
ϕ1 = -ϕ2
ϕ1 = -2ϕ2
uniaxial
ϕ2 = 0
plane strain
biaxial
Figure 6: Forming limiting diagram for sheet metal forming indicating the major forming modes
(according to [66]).
2.3.2 Forming of zinc and zinc coated metals
Macroscopically the forming of metal sheets was described above. On an atomistic level, the
deformation of the polycrystalline zinc can be understood on closer examination of the
deformation behaviour of single crystals [67, 68].
In the case of small deformations, the applied stress leads to a linear translation of the lattice
which shows elastic behaviour due to Hook’s law (Figure 7). Releasing the applied stress the
atoms are forced to return to their origin position.
2 Fundamentals 23
τ
τ
τ
τ
γ
τ
τ
τ
τ
γ
Figure 7: Elastic deformation of a crystal lattice due to shear stress
τ
for an angle
γ
[68].
If the applied stress is larger than the elastic part the crystal is plastically deformed.
Hexagonal close-packed (hcp) metal single crystals, e.g. zinc crystals, are characterized by the
ratio (c/a) where c and a are the height and the edge size of the hexagonal cell, respectively.
They are also known to exhibit a very anisotropic mechanical behaviour like elasticity and
plasticity. For zinc, c/a = 1:856 > 1:633 which is the value for the maximal density of hcp
metals. Therefore, basal slip is the easiest deformation mode in zinc [27, 28]. Non-basal slip,
twinning and strain relaxation at grain boundaries have been experimentally observed for zinc
[21, 69].
For zinc, twinning is favoured by compression parallel and tension normal to the c-axis. The
formation of twins is shown in Figure 8. They appear along the twin plane, where a part of the
lattice will be transformed into a mirror image. Thus, unlike slip, twinning is very dependent
on the sense of the stress applied on the crystal [27]. One effect of twinning is that grains
unfavourably oriented for slip will be reoriented into a more favourable position.
τ
τ
γ
τ
τ
γ
twinning
plane
τ
τ
γ
τ
τ
γ
twinning
plane
Figure 8: Formation of twins due to shear stress
τ
[68].
2 Fundamentals 24
1 2 3 4 5 6 1 2 3 4 5 6
τ
τ
τ
τ
b
(a)
(b)
1 2 3 4 5 6 1 2 3 4 5 6
τ
τ
τ
τ
b
τ
τ
τ
τ
b
(a)
(b)
Figure 9: (a) Fixed sliding along the sliding plane for a length of the lattice constant b due to shear
stress
τ
for a perfect crystal. (b) Single dislocation for a none ideal crystal with the migration of
defects between different lattice planes [68].
Plastic deformation by sliding is separated in two mechanisms as shown in Figure 9. Fixed
sliding can be found on ideal crystals if all atoms from one lattice plane are moved
simultaneously against the atoms from the neighbouring lattice plane. In technical metals
defects inside the lattice plane allows the migration between the lattice planes. Typically
atomic surface steps are caused by single dislocations within the crystals [70].
2 Fundamentals 25
slip
lines
slip
band
slip
steps
F
r
F
r
slip
lines
slip
band
slip
steps
F
r
F
r
Figure 10: Schematic picture of plastic deformation by large crystallographic slip steps created by a
set of collectively gliding dislocations [68].
Figure 10 shows the schematic picture of plastic deformation which can be found on zinc
crystals. Atomic slip steps are caused by single dislocations leaving the bulk, larger
crystallographic slip steps (so-called slip bands) created by sets of collectively gliding
dislocations on parallel or identical glide planes [70].
In most cases technical samples are polycrystalline materials, which are formed from single
crystals, also called grains, with different crystallographic orientation. The applied stress for
the plastic deformation of polycrystals with different grain orientations must be higher than
the stress of the individual singe grains. A single grain cannot be formed individually as long
as the critical stress in the neighbouring grains is high enough to also plastically form the
neighbouring grains. Therefore, the deformation of neighbouring grains show identical
forming behaviour at the grain boundaries [68, 71].
(a) (b)(a) (b)
Figure 11: (a) Orange peel phenomena. (b) Ridging and roping phenomena. [70]
2 Fundamentals 26
Besides the nanoscopic defects on the single grains, macroscopic phenomena occur during
forming which is described in detail by Raabe et al. [70]. The grain-scale roughening in
homophase alloys can be grouped into orange peel and banding phenomena which are
illustrated in Figure 11. Both are characterized by the out-of-plane displacement fields
(negative or positive), which roughly map the grain shape of the material. Orange peel occurs
when different crystals produce individual out-of-plane grain-scale surface displacements due
to their different orientation factors and the resulting shape changes. Banding is commonly
referred to as ridging or roping. It occurs in the form of banded surface undulations.
2.3.3 Forming of plasma polymer films
The mechanical behaviour of thin films on metallic or deformable substrates was studied
intensively in several publications [17-19, 72]. Most of the works focused on the formation
and the propagation of cracks in the coatings on ductile or deformable substrates and were
studied experimentally e.g. for zirconia thin films on stainless steel or inorganic films on
polymeric substrates like SiOx films on PET.
In the case of thin ductile films on elastic substrates the crack formation and the crack
propagation can be described within the framework by the so-called shear lag approximation
explained in detail by Wojciechowski and Mendolia [17]. Usually, multiple sequential crack
formation in the coating for these cases is observed with an increase of the number of cracks
with increasing strain (see Figure 12). If a certain crack density is reached further strain
results only in an opening of the cracks but the number of cracks stays constant. The
formation of equidistant cracks in the coating is based on the non-localised distribution of
strain over the coating. For metallic surfaces this behaviour is limited on small forming
degrees as the plastic deformation of the substrate cannot be neglected for larger strains.
The mechanical behaviour of the coating strongly depends on the adhesion between the
coating and the substrate surface. In the case of weak adhesion or strong intrinsic stress inside
the coating, the coating will delaminate easily if further stress is applied due to straining of
the sample. In the case of strong adhesion of the coating during forming, the surface defects
as described in the previous chapter like atomic slip steps caused by single dislocations
leaving the bulk, larger slip bands created by collectively gliding dislocations, surface twins
and macroscopic phenomena like orange peel and ridging and roping have strong impact on
the behaviour of the coating.
2 Fundamentals 27
Increasing strain
crack
Stress Crack
density
Strain
Saturation
Increasing strain
crack
Stress Crack
density
Strain
Saturation
Figure 12: Schematic picture of sequential or so-called mode-I crack formation during uniaxial
forming [17].
Kirk and Pilliar reported about slip planes emerging the surface of TiN coatings on steel [72].
All cracks in the TiN coating were observed to be located near the slip bands.
Baumert et al. reported similar behaviour of the forming of thin plasma polymer films on
electrolytically galvanised and chemically polished steel [24].
Stress
Zinc
SiO
2
SiO
2
Zinc
Shearing
Defect
Zinc
SiO
2
SiO
2
Zinc
Shearing
Defect
Loss of
adhesion
SiO
2
Zinc
Shearing
Defect
No loss of
adhesion
Stress
(a)
(b)
Stress
Zinc
SiO
2
SiO
2
Zinc
Shearing
Defect
Zinc
SiO
2
SiO
2
Zinc
Shearing
Defect
Loss of
adhesion
SiO
2
Zinc
Shearing
Defect
No loss of
adhesion
Stress
(a)
(b)
Figure 13: Substrate induced crack formation in SiO2 coatings for (a) thick SiO2 coatings and (b) for
ultra-thin coatings (20-50 nm) according to [24].
2 Fundamentals 28
They reported, that unlike the multiple sequential cracking with equidistant cracks for thin
coatings on strained substrates, most of the cracks within the SiCxOy-like coating are observed
near to slip bands of the steel. On the slip bands, however, the crack morphology is quite
similar to the usual multiple cracking on deformable substrates. For thicker coatings (e.g.
thickness > 100nm) the localized strain and stress concentrations within the coating seem to
be dissipated beyond the slip bands. Figure 13 shows the substrate induced cracks, whereas
the stress within the thick coating can be dissipated over defects in the substrate without
formation of cracks within the plasma polymer coating. On fine gliding grains real transverse
crack formation is observed. However, at high film thickness values the complete de-adhesion
of the SiOx-like coating can be observed.
3 Experimental measurement techniques 29
3 Experimental measurement techniques
3.1 Electrochemical measurement techniques
3.1.1 In-situ and scanning Kelvin probe
Height regulated scanning Kelvin probe (HR-SKP) measurements were performed for
corrosion studies by using a self-designed height regulated probe which is described explicitly
in [73, 74]. The samples were measured in a cleaned gas atmosphere by using air with 95%
relative humidity. A plane-ended, electrochemical etched Ni/Cr cylindrical wire (diameter
approximately 80 µm) was used for the scanning of the sample surface in x, y and z direction.
The calibration of the NiCr probe was done by measuring and referring to Cu/CuSO4. For
corrosive de-adhesion studies, the defect area was filled with 0.5M NaCl solution.
The work function of the samples in vacuum was detected in-situ by means of a Kelvin probe
in the plasma cell described in Chapter 4.2.1. A heat treated graphite needle (Staedler 2B
marsmicro) of 0.5 mm diameter was used for the in-situ vacuum measurements during each
plasma treatment step. The graphite tip was measured versus HOPG as reference as described
in [75]. For in-situ Volta potential relaxation measurements by switching from dry to
humidified air a plane-ended, electrochemical etched Ni/Cr cylindrical wire (diameter
approximately 500µm) was used.
All potentials measured by the KP and the HR-SKP are given with respect to standard
hydrogen electrode (SHE).
3 Experimental measurement techniques 30
3.1.2 Electrochemical impedance spectroscopy
Electrochemical impedance spectroscopy (EIS) is a very common technique used in material
and surface science to investigate the barrier properties of films e.g. organic coatings on
metallic substrates [11, 15, 76]. To test the barrier properties of a coating the electrochemical
impedance is measured by applying a sinusoidal potential excitation with a small amplitude
over a range of frequencies to an electrochemical cell. The frequency dependent current
answer of the electrochemical cell is described by the complex impedance, which also takes
into account the phase shift of the current answer and the applied potential.
If the interface can be described by an ideal resistor, no phase shift takes place and the
impedance is identical with the ohmic law, where E is the applied voltage, I the applied
current and Rohm the ohmic resistance.
IRE ohm ⋅= (17)
In non-ideal systems, the complex impedance measurements can distinguish between ohmic
parts (e.g. interface and electrolyte resistance) and capacitive parts (e.g. double layer
capacitance or coating capacitance) by simulating the system as a network of resistors and
capacitors. The complex impedance is described by the capacitance and the radial frequency
ω
:
Ci
Z
ω
1
|| = (18)
The capacitance of a perfect barrier coating can be described by a plate condenser. The
condenser consists of two plates formed by the surface or the electrode and the electrolyte.
Between these two plates the coatings exists as a dielectric. The capacity C of the condenser is
proportional to the plate area A, the dielectric constant of the vacuum 0
ε
, the dielectric
constant of the material between the plates r
ε
and inverse proportional to the thickness of the
coating d.
d
A
Cr
⋅
⋅
=
ε
ε
0 (19)
3 Experimental measurement techniques 31
log f
log |Z|
phase
-90
0
C
DL
R
Ct
R
C
(a) (b)
C
C
R
U
C
C
log f
log |Z|
phase
-90
0
C
DL
R
Ct
R
C
(a) (b)
C
C
R
U
C
C
Figure 14: (a) Bode plot and (b) equivalent circuit of a non ideal barrier coating in contact with an
electrolyte.
Figure 14 shows the Bode plot and the respective equivalent circuit for a non ideal barrier
coating. Defects occur by pores within the barrier coating and the electrolyte is able to
penetrate the coating and reaches the surface of the metal electrode. The penetrating
electrolyte is reflected similar to the uncompensated electrolyte resistance RU and by an
ohmic resistor RC of the coating which is connected in parallel with the coating capacitance
CC. If the electrolyte has reached the metal surface, the system can be described by a normal
metal electrode equivalent circuit with the pore resistance RCt and the double layer
capacitance CDL. The coating capacitance can be distinguished from other capacitive parts, as
the phase is quite close to -90° for ideal barrier systems.
The electrochemical measurements were performed using a Zahner IM6d potentiostat
(Zahner-Elektrik Gmbh & CoKG, Kronach, Germany). A three electrode setup with the
working electrode comprising of a reproducible, exposed area of 2.26 cm2, a platinum counter
electrode and a saturated Ag/AgCl reference electrode (+198mV versus SHE) was used. All
electrochemical experiments were carried out in quiescent borate buffer solution (0.05M
Na2B4O7·10H2O + 0.05M Na2SO4 + 0.2M H3BO3) prepared using millipore water with pH 8.3
at 22 ± 1°C. The electrochemical impedance spectra were measured, at open circuit potential,
in the range of 0.1 Hz up to 100 kHz with 10 points per decade and with an amplitude of 10
mV rms, using the above described three electrode set-up.
3 Experimental measurement techniques 32
3.1.3 Cyclic voltammetry
Cyclic voltammetry (CV) can be described as a potentiodynamic electrochemical
measurement as described in detail in [77]. The working electrode potential is ramped linearly
versus time and the flowing current density is measured. This ramping is known as the
experiment's scan rate.
If a material can be reduced by the electrolyte, than the current density will increase as the
ramped potential reaches the reduction potential of the material, but decreases as the
concentration of the material is depleted close to the electrode surface. If the redox couple is
reversible than the former reduced material can be reoxidized when the oxidation potential is
reached. For ideal systems the current densities between the oxidation and reduction process
are equal and the oxidation and reduction peaks in the voltage versus current density plot have
a similar shape. The amount of charge which is transferred to the electrolyte when the
material is reduced or oxidized is directly proportional to the surface area which is in contact
with the electrolyte. If parts of the surface are covered with a barrier film which cannot be
oxidized or reduced at potentials when the metal substrate can be oxidized or reduced, than
the protected area can be directly calculated by the difference of the charge transfer on an
unprotected and a partially protected surface [78, 79].
Cyclic voltammograms were obtained with a Zahner IM6d potentiostat and the same set-up as
described above over a voltage range of -750 to -1150 mVSHE at ambient temperature and a
scan rate of 50 mV/s.
Cyclovoltammetry was also measured in-situ during forming by a unique combination of a
tensile testing device with a home-made three electrode electrochemical micro capillary cell
[80]. The layout of the bone shaped tensile samples and the setup of the in-situ
cyclovoltammetry measurements is described in Chapter 4.3.2. All ex-situ and in-situ CV
measurements were carried out in quiescent borate buffer solution (0.05M Na2B4O7·10H2O +
0.05M Na2SO4 + 0.2M H3BO3) prepared using distilled water with pH 8.3 at 22 ± 1°C. The
cyclic voltammograms were obtained over a voltage range of -525 to -1275 mVSHE at ambient
temperature and a scan rate of 100 mV/s with a gold reference electrode.
3 Experimental measurement techniques 33
3.2 Spectroscopic measurement techniques
3.2.1 Infrared spectroscopy
Fourier transformed infrared reflection absorption spectroscopy (FT-IRRAS) in combination
with polarization modulation can be used for the characterization of thin films or monolayers
on metallic or highly doped semiconducting surfaces. The advantage of this method is the
high surface sensitivity due to the grazing angle of incidence and the surface selection rule
and that the modulated reflectivity is independent of the isotropic adsorption from gas or bulk
water. Therefore, the interfering effects of gas or bulk water and carbon dioxide in the beam
path can be eliminated [49, 81].
The infrared spectra were measured between 4000 and 750 cm-1 by means of a Biorad
Excalibur 3000 (Digilab, Randolph, USA) equipped with a mercury cadmium telluride
detector (MCT) cooled with liquid nitrogen. Each spectrum represents a co-addition of 256 or
512 scans at a resolution of 4.0 cm–1. The spectra were water and baseline corrected using the
Digilab resolution pro software package (version 4.0). The infrared beam was transmitted
through ZnSe windows on the sample in the plasma cell and reflected under 80°. A discrete
polarization modulation was used to enhance surface sensitivity and reduce gas phase
absorption of residual water and carbon dioxide in the optical path [49]. The set-up is
described in detail in chapter 4.2.1.
3.2.2 X-ray photoelectron spectroscopy
X-ray photoelectron spectroscopy (XPS) is a surface analysis technique to quantitatively
measure the elemental composition, chemical and electronic state of the elements which can
be found at the outermost 1 to 10 nm of the material. The primary excitation is accomplished
by the irradiation of the surface by a source of monochromatic X-rays. The X-rays leads to the
photoionisation of the atoms and the response of the atoms (photoemission) is detected by
measuring the energy spectra of the emitted photoelectrons. The presence of peaks at
particular binding energies therefore indicates the presence of a specific element in the sample
under study. Furthermore, the intensities of the peaks are related to the concentration of the
element within the sample region. The exact binding energy of the electrons depends not only
upon the level from the photoemission, but also on the formal oxidation state of the atoms and
the local chemical and physical environment. Therefore, small variations in the former
3 Experimental measurement techniques 34
parameters result in a shift of the binding energy, the so-called chemical shift. These chemical
shifts can be corrected by literature values, e.g. by comparing the binding energy of the C1s
peak of the aliphatic groups with literature values of 285 eV. In combination with a sputter
gun, depth profiling of the sample is possible with the determination of the elements and also
with the binding state by using fitting algorithms [82].
XPS-spectra were measured by means of a Quantum 2000 ESCA Microprobe (Physical
Electronics Inc., Chanhassen, USA) with a take-out angle of 45° and a spot size of 100 µm
using monochromatic Al Kα-radiation. The element spectra were obtained by using a pass
energy of 23.5 eV and an energy resolution of 0.2 eV. Sputtering was performed by ion beam
sputtering with an ion beam of 2 kV on a 2 x 2 mm2 spot. The sputter rate was 2.6 nm/min as
calibrated with a SiO2 standard sample. For data evaluation CASA XPS (version 2.3.12) was
used.
3.2.3 Spectroscopic ellipsometry
Spectroscopic ellipsometry (SE) is widely used as standard for measuring the thickness and
the optical constants of thin films. The key feature of ellipsometry is that it measures the
change in polarized light upon light reflection on a surface or interface. Under certain
conditions, linearly polarized light is reflected from a sample and the elliptically polarized
light is analyzed. The amplitude ratio
Ψ
and the phase difference
Δ
between the p- and s-
polarized light waves are measured and fitted by a certain model which can describe the
theoretical film system. From the fitting results, the thickness and the complex refractive
indices can be determined [83].
A spectroscopic ellipsometer SE 800 from Sentech (Berlin, Germany) was used for
ellipsometry measurements. The samples were measured at three different angle of incidence
(50°, 60° and 70°) for a spectral range of 300 nm to 820 nm. The measurements were
evaluated with SpectraRay 2 from Sentech. Therefore, a Cauchy-layer model was used to fit
the ellipsometry data.
3 Experimental measurement techniques 35
3.3 Microscopic techniques
3.3.1 Force microscopy
Atomic force microscopy (AFM) was developed to image surfaces in three dimensions on
nanoscopic scales. A laser beam deflection system is used to detect the bending of the
cantilever when the tip, which is located on the front of the cantilever spring, interacts with
the sample surface by mechanical, magnetical or electrical forces. Scanning Kelvin probe
force microscopy (SKP-FM) combines the high-resolution of force microscopy with the
Kelvin probe technique to determine the potential or work function distribution of the sample
surface. In a first step the sample surface is determined in contact mode. With the
topographical information the cantilever is than moved over the surface with a certain
distance. The potential or work function of the surface is than measured by analysis of the
modulated voltage signal which is applied on the capacitor which is formed by the tip and the
sample surface [48].
Surface topography and Kelvin probe potential mappings were investigated by means of AFM
with a NanoWizard® AFM from JPK Instruments AG (Berlin, Germany). Measurements were
carried out in tapping or contact mode for SKP-FM by using silicon sensors (NHC, obtained
from NanoWorld, k=285 kHz, typical tip radius less than 10 nm). AFM topography and SKP-
FM images were measured in dry nitrogen atmosphere with a typical tip-to-surface distance of
50 nm for the SKP-FM measurements.
3.3.2 Scanning electron and Auger microscopy
The scanning electron microscope (SEM) allows the study of morphology, composition and
crystallography of conductive or metallised surfaces by scanning the surface with a high
energy beam of electrons[78]. The resolution of a SEM is much higher in comparison to light
microscopy as the wavelength of the electrons is much smaller than the wavelength used for
optical microscopy. The most common mode used for imaging is detecting the secondary
electrons which are ejected due to inelastic scattering with the beam electrons [84]. By using
the backscattered electrons from the elastic scattered beam electrons, electron backscattered
diffraction (EBSD) can be measured. EBSD is used to examine the crystallographic
orientation of crystalline or polycrystalline materials [85].
3 Experimental measurement techniques 36
SEM pictures were obtained on a Leo 1550 VP (Leo Elektronenmikroskopie GmbH,
Oberkochen, Germany) for different forming degrees of the samples. SEM pictures were
taken with an annular inlens SE detector, typical acceleration voltages between 2 and 5 kV
and a working distance of 10 mm. The crystallographic orientation of the zinc grains was
measured with a TSL EBSD analyzer by measuring the electron back scatter diffraction
(EBSD) patterns and evaluation of the resulting Kikuchi lines.
Scanning Auger electron spectroscopy (AES) surface mappings were measured with a PHI
700 Scanning Auger Nanoprobe (Physical Electronics Inc., Chanhassen, USA). In comparison
to SEM the subsequent relaxation of the ionized atoms leads to the emission of Auger
electrons which are characteristic of the elements present in this part of the sample [84].
4 Experimental set-up and sample treatment 37
4 Experimental set-up and sample treatment
4.1 Materials
4.1.1 Used substrate material
Hot-dip galvanised steel sheets (HDG) were used as samples of technical interest. The
interstitial free HDG steel (steel type DX54D, sheet thickness 0.85 mm) was supplied by
OCAS N.V. (Zelzate, Belgium) in the non skin passed state, i.e. without being temper rolled
after the application of the zinc coating. The thickness of the zinc coating (alloy composition:
Zn + <0.3% Al) after hot-dip galvanising was 7 µm.
Additionally pure zinc foil (99.99%, Goodfellow, Bad Nauheim, Germany) with a thickness
of 3 mm and a size of 50 x 50 mm2 was used especially for the plasma modification of zinc
surfaces.
Polished and solvent cleaned silicon wafers were used for thickness control measurements by
spectroscopic ellipsometry.
4.1.2 Substrate cleaning and pre-treatment
Smaller samples were cut from the supplied material in a size of 12 x 4 cm2 for the HR-SKP
measurements. For the forming experiments with the miniature tensile testing device samples
(2 x 5.1 cm2) were cut in a bone like shape. All samples were cleaned in three different
solvents in an ultra-sonic bath for 20 minutes to ensure complete removal of any oils or
surface contaminations. As solvents Tetrahydrofuran (THF), iso-propanol and ethanol with
analytical grade (Merck KGaA, Darmstadt, Germany) were used. After every solvent
treatment the samples were dried under flow of dry nitrogen. Subsequently, the samples were
cleaned by an additional alkaline cleaning step to remove the aluminium oxyhydroxide
surface layer.
4 Experimental set-up and sample treatment 38
Figure 15: AFM topography of (a) solvent cleaned HDG and (b) after alkaline cleaning of HDG.
The samples were immersed for 30 seconds in a magnetically stirred aqueous solution of 30
g/l Ridoline 1570, 3 g/l Ridosol (Henkel KGaA, Düsseldorf, Germany) at 55 °C. Finally,
samples were rinsed thoroughly with purified water and immediately dried in a stream of dry
nitrogen.
Table 1: Atomic percent of the surface composition after cleaning steps measured by XPS.
Zn O Al C
Solvent cleaned 16 % 49 % 14 % 21 %
5 s alkaline cleaned 31 % 47 % 8 % 14 %
30 s alkaline cleaned 33 % 46 % <0.1 % 21 %
The topography of the surface before and after alkaline cleaning is shown in Figure 15. The
surface of the solvent cleaned sample shows a large scale waviness due to the drying process
during hot-dip galvanising. The alkaline cleaning, which is an etching of the aluminium
surface layer leads to nanoscopic roughening of the surface. The waviness of the surface is
still present after alkaline cleaning. Table 1 shows the atomic concentration for the different
cleaning steps with the reduction of the aluminium concentration due to the alkaline cleaning.
4.1.3 Application of organic top coat
For the corrosive de-adhesion studies obtained by HR-SKP, a styrene/n-butylacrylate latex
binder from BASF AG (BASF AG, Ludwigshafen, Germany) with a particle size of 140 nm
was bar coated on the samples (film thickness 2 µm) and dried for 12 hours at 100 °C. This
4 Experimental set-up and sample treatment 39
water-based coating is free from pigments or additives. The organic coating was applied after
forming of the SiO2 coated samples to avoid additional effects from forming induced defects
of the organic coating.
4.2 Plasma modification
4.2.1 In-situ plasma treatment of metal surfaces
For the in-situ studies of the plasma surface modification and thin film deposition by low-
temperature plasmas a special plasma cell was used. The set-up of the cell is shown in Figure
16 and was described and published in detail by Raacke et al. [75]. The chamber is equipped
with a sample holder which can be moved by a linear stage between the plasma modification
position and the FT-IRRAS and Kelvin probe (KP) measurement position. Different
configurations of the electrodes allow the generation of low-pressure direct or remote
plasmas.
The chamber was evacuated to a base pressure of 10-4 mbar before plasma treatment of the
samples by using rotary vane pump in combination with a turbo molecular pump. A liquid
nitrogen cooling trap was used during polymer plasma deposition to protect the pumping
system. In this work the remote plasma was generated by using a commercial high voltage
power supply with a frequency of 30.7 kHz and an adjustable voltage.
PI: M-632
Steppermotor
Vacuumchamber
Kelvin Probe head
Micrometer screw
Gas inlet & HV power supply
IR windows (ZnSe)
Sample holder
Electrode for remote
or direct plasma treatment
Kelvin Probe needle
PI: M-632
Steppermotor
Vacuumchamber
Kelvin Probe head
Micrometer screw
Gas inlet & HV power supply
IR windows (ZnSe)
Sample holder
Electrode for remote
or direct plasma treatment
Kelvin Probe needle
Figure 16: Experimental set-up for the plasma modification of metal surfaces and the in-situ
measurement of FT-IRRAS and work function by a vacuum Kelvin probe.
4 Experimental set-up and sample treatment 40
The samples were moved through the modification area of the remote plasma in an
argon/hydrogen plasma (1.5 sccm flow, 5:1 argon/hydrogen (purities 5.0/5.0), p = 0.3 mbar,
30 V) or an oxygen plasma (with 1.5 sccm O2 (purity 4.5), p = 0.3 mbar, 30 V) within 6
minutes. After this oxygen plasma cleaning the optional SiO2 film deposition was performed
in a mixture of oxygen and hexamethyldisiloxane (HMDSO, (CH3)3SiOSi(CH3)3) from Fluka
(purity 99.5 %) with a flow ratio of 20:1 (p = 0.3 mbar, 30 V) and different deposition times
for varying film thickness values (typical deposition rate 0.5 nm/s). FT-IRRAS and KP
measurements were performed before and after each plasma treatment step to correlate the
surface chemistry with the chemical composition and the electronic structure.
4.2.2 PE-CVD film deposition
Besides the in-situ measurement described above, ultra-thin plasma polymer films were
deposited by plasma enhanced chemical vapour deposition (PE-CVD). A plasma chamber
from Roth & Rau (Hohenstein-Ernstthal, Germany) was used with a linear microwave plasma
source and this enables the film deposition on samples up to DIN A4 in size. The chamber is
shown in Figure 17 and described in detail elsewhere [86].
Before film deposition the chamber was pumped down to a base pressure below 10-3 mbar by
using a roots pump and flushing the chamber three times with argon. The samples were
moved with a constant speed through the plasma zone and cleaned by an oxygen plasma (400
W, 80 sccm/min O2, v = 2 mm/s) at a pressure of 0.2 mbar. Immediately after this cleaning
and activation step, the plasma polymer was deposited from a gas mixture of oxygen and
HMDSO with a flow ratio of 20:1 (100 sccm : 5 sccm, p = 0.2 mbar, 300 W).
Plasma chamber
Roots pump
(p<10
-3
mbar)
Movable stage with
sample holder
Quartz tube
Shielding
Plasma zone
(400W, 2,4GHz)
Fast entry
door
O
2
, N
2
, Ar,…
Precursor, e.g. HMDSO
Plasma chamber
Roots pump
(p<10
-3
mbar)
Movable stage with
sample holder
Quartz tube
Shielding
Plasma zone
(400W, 2,4GHz)
Fast entry
door
O
2
, N
2
, Ar,…
Precursor, e.g. HMDSO
Figure 17: Experimental set-up for large area PE-CVD.
4 Experimental set-up and sample treatment 41
The substrates were moved through the plasma zone at a constant speed of 15, 5, 2 and 1
mm/s to obtain film thickness values of 10, 30, 50 and 100 nm. The thickness of the films was
controlled with measurements on the samples as well as on silicon wafers by means of UV–
VIS spectral ellipsometry (SE 800, Sentech, Berlin, Germany). The silicon wafers were
attached near to the steel substrates during film deposition.
4.3 Forming of samples
4.3.1 Miniature tensile forming device
A custom built linear tensile testing device from Kammrath & Weise (Dortmund, Germany)
with dog-bone shaped specimens (51 x 20 mm2) was used for uniaxial straining along the long
axis of the samples. The tensile forming device allows the uniaxial forming of samples with a
maximum load of 5 kN. The clamps of the sample holder are electrical insulated and allow
thereby electrochemical measurements without special shielding.
In this work, the forming degree for strained samples is given as the major strain
ϕ
in a range
of 0.00 to 0.25 which is directly correlated to the technical deformation usually expressed in
percent.
4.3.2 In-situ cyclic voltammetry and forming set-up
A unique system for measuring impedance of organic coating on zinc coating steel during
tensile forming was published by Klüppel et al [80]. This system allows the formability
analysis of organic coatings on coated metal substrates during each separate forming step. In
this work cyclic voltammetric measurements were used to follow the change of the
unprotected zinc area during forming by adapting this set-up as shown in Figure 18.
4 Experimental set-up and sample treatment 42
F
2
1
r
F
2
1
r
WE
CE
RE
Ø~1mm
20mm
51mm
(a)
(d)
(b)
500µm
(c)
F
2
1
r
F
2
1
r
WE
CE
RE
Ø~1mm
20mm
51mm
(a)
(d)
(b)
500µm
(c)
Figure 18: Schematic set-up of the micro-capillary cell for in-situ cyclic voltammetry measurements
during forming. (a) Side view of the sample with capillary cell. (b) Top view of the sample with central
measurement spot. (c) Averaging of different grain sizes and grain. (d) Experimental set-up.
4 Experimental set-up and sample treatment 43
A unique combination of the tensile testing device with a home-made three electrode
electrochemical micro capillary cell with an Au-counter (CE) and Au-pseudo reference
electrode (RE) was used. The sample as working electrode (WE) is clamped by an electrical
insulated mounting in the tensile testing device. A 1 mm in diameter glass capillary was
positioned in the middle of the forming area and filled with borate buffer solution. The glass
capillary is many times larger than the typical grain size. This allows the measurement to be
independent from different grain sizes and texture orientations as the results are averaged over
many grains. A flexible silicon ring served as sealing and stayed in contact with the substrate
surface during forming. This procedure ensured that always nearly the same sample area was
electrochemically investigated; the contact pressure of the glass capillary on the surface is
measured and controlled by a force gauge.
4.3.3 Tensile forming device for ex-situ experiments
Most experimental set-ups require a certain minimum size of the substrate material. For EIS,
ex-situ CV and HR-SKP measurements larger substrates (120 x 40 mm2) were uniaxially
strained along the long axis on a Z100 from Zwick & Roell (Ulm, Germany). The resulting
strain values and observed surface defects are comparable to the results obtained by the
miniature tensile testing device.
4 Experimental set-up and sample treatment 44
5 Corrosion resistance of plasma modified HDG steel 45
5 Corrosion resistance of plasma modified HDG steel
In the past, several studies concerning the corrosion resistance of metal substrates covered
with thin plasma polymer films were carried out [10-13]. However, the mechanism of
corrosion protection of organically coated plasma modified surfaces is still poorly understood.
Grundmeier et al. showed that a plasma oxidation of iron oxides on iron leads to change in the
electronic properties [14]. However, the highly oxidised state of the passive film is not stable
and changed again in humid atmospheres to the state which is determined by the oxygen
reduction kinetics at the outer oxide surface. This process is also determined by the kinetics of
the metal dissolution into the passive film at the metal/oxide interface.
Barranco et al. showed that for cathodically deposited plasma polymer films the interfacial
electrode potential on iron could be effectively lowered which indicates a strong inhibition of
interfacial oxygen reduction kinetics [15]. However, a reductive plasma treatment had to be
followed by a sufficiently thick plasma polymer film deposition to prevent the plasma
polymer/metal interface from re-oxidation. Based on this work electrochemical methods
could be established for an improved understanding how plasma modifications influence the
corrosion mechanism of thin film coated metals.
From the viewpoint of thin film engineering, it is the aim to achieve functional properties at
extremely low thickness values. With regard to interfacial corrosive de-adhesion of polymer
coated zinc and iron substrates, the interfacial oxygen reduction kinetics should be strongly
inhibited by the respective plasma polymer film [15].
Therefore, the aim of this chapter is to show the modification of the passive film on zinc by
ultra-thin barrier plasma polymer films and to correlate the structure and physical properties
of the surface layer with the inhibition of the interfacial oxygen reduction kinetics in humid
and corrosive environments.
5 Corrosion resistance of plasma modified HDG steel 46
5.1 Stability and chemical composition of oxygen plasma modified
zinc surfaces
5.1.1 In-situ FT-IR analysis during plasma treatment
Pre-cleaned zinc coatings were argon/hydrogen and oxygen plasma treated. The
corresponding in-situ FT-IRRAS difference spectra are shown in Figure 19. All peak
assignments are listed in Table 2. After the Ar/H2 plasma treatment, a removal of zinc-
oxyhydroxides, carbonates, adsorbed hydrocarbons and surface hydroxyls could be detected
as indicated by the negative peaks in the difference spectra. [35, 87].
Table 2: Assignment of the FT-IRRAS peaks of the plasma modified zinc surface.
Wavenumber (cm-1) Group Assignment Ref.
945 Zn-OH δ(OH) [35, 87]
1170 Zn-OH ν (Zn-OH) [35, 87]
1440-1475 CH2, CH3 δ(CHx) [88]
1450 R-COO- νs(R-COO-) [35, 87]
1570 R-COO- νas(R-COO-) [35, 87]
1600 H2O δ(H2O) [87, 89]
2850-2900 CH2, CH3 νas,s(CHx) [88]
3300 H2O νs(OH) [87, 89]
3520 H2O νas(OH) [87, 89]
The subsequent oxygen plasma results in a measurable growth of the zinc-oxyhydroxide peak
at 1170 cm-1 with the composition 2/)2(
)(
δδ
−
OOHZn and further removal of surface
hydrocarbons (CHx). An increase in the surface hydroxyl density was not observed [35].
5 Corrosion resistance of plasma modified HDG steel 47
2500275030003250350037504000
-0.75
-0.50
-0.25
0.00
0.25
0.50
0.75
1.00
νas,s(CHx)
Ar/H2 plasma
subsequent O2 plasma
Absorbance / 10-3
Wavenumber / cm-1
νs(OH)
νas(OH)
80010001200140016001800
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
νs(R-COO_)
ν(Zn-OH)
δ(H2O) δ(CHx)
νas(R-COO_)
Ar/H2 plasma
subsequent O2 plasma
Absorbance / 10-3
Wavenumber / c
m
-1
δ(OH)
(a)
(b)
2500275030003250350037504000
-0.75
-0.50
-0.25
0.00
0.25
0.50
0.75
1.00
νas,s(CHx)
Ar/H2 plasma
subsequent O2 plasma
Absorbance / 10-3
Wavenumber / cm-1
νs(OH)
νas(OH)
80010001200140016001800
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
νs(R-COO_)
ν(Zn-OH)
δ(H2O) δ(CHx)
νas(R-COO_)
Ar/H2 plasma
subsequent O2 plasma
Absorbance / 10-3
Wavenumber / c
m
-1
δ(OH)
(a)
(b)
Figure 19: In-situ FT-IRRAS difference spectra after Ar/H2-plasma (reference: state before Ar/H2-
plasma) and subsequent O2-plasma modification (reference: state after Ar/H2-plasma) of alkaline
cleaned hot-dip galvanised (HDG) steel. (a) 4000 - 2500 cm-1 and (b) 1800 – 800 cm-1.
5.1.2 Chemical surface change after plasma treatment
Changes in the chemical composition of the passive film were observed by means of XPS
sputter profiling. Sputter profiles of the Zn2p and O1s peaks for the alkaline cleaned HDG
sample before and after oxygen plasma treatment are shown in Figure 20. The Zn2p and O1s
peaks were fitted to calculate the different amounts of oxidized and metallic zinc as well as
the different amounts of zinc hydroxide and oxide of the passive surface layer. The results
indicate a thickening of the zinc hydroxide/oxide passive film due to the oxygen plasma
5 Corrosion resistance of plasma modified HDG steel 48
treatment in comparison to the non-plasma treated surface. The growth of the oxide thickness
is most probably due to the increased density of adsorbed surface hydroxides and occurs via
the high field mechanism [90]. Electrons are transferred from the zinc/zinc oxide interface to
the outer zinc oxide surface and reduce the adsorbed oxygen atoms or hydroxyls. Zn ions are
formed at the oxide/zinc interface and migrate to the surface to neutralize the adsorbed
negative hydroxide ions.
012345678
0
10
20
30
40
50
60
70
80
90
100
Sputter depth / nm
Zn2pZn2p
Sputter depth / nm
Reference 5.9nm
SiO2 in 60 sec
012345678
0
10
20
30
40
50
60
70
80
90
100 after O2-plasma
alkaline cleaned HDG steel
Zn_red
Zn_ox
Zn_sum
Concentration / %
012345678
0
10
20
30
40
50
60
70
80
90
100 alkaline cleaned HDG steel
Reference 5.9nm
SiO2 in 60 sec
Sputter depth / nm
Sputter depth / nm
012345678
0
10
20
30
40
50
60
70
80
90
100
O1s
O2-_oxide
O2-_hydroxide
O2-_sum
Concentration / %
after O2-plasma
O1s
(a)
(b)
012345678
0
10
20
30
40
50
60
70
80
90
100
Sputter depth / nm
Zn2pZn2p
Sputter depth / nm
Reference 5.9nm
SiO2 in 60 sec
012345678
0
10
20
30
40
50
60
70
80
90
100 after O2-plasma
alkaline cleaned HDG steel
Zn_red
Zn_ox
Zn_sum
Concentration / %
012345678
0
10
20
30
40
50
60
70
80
90
100 alkaline cleaned HDG steel
Reference 5.9nm
SiO2 in 60 sec
Sputter depth / nm
Sputter depth / nm
012345678
0
10
20
30
40
50
60
70
80
90
100
O1s
O2-_oxide
O2-_hydroxide
O2-_sum
Concentration / %
after O2-plasma
O1s
(a)
(b)
Figure 20: Ex-situ XPS analysis of thickness change of oxide layer after oxygen plasma treatment.
5 Corrosion resistance of plasma modified HDG steel 49
5.1.3 Influence of plasma treatment on Volta potential
In Figure 21 the change in the Volta potential of the argon/hydrogen cleaned and oxygen
plasma modified surface in comparison to the untreated HDG surface is illustrated as it was
measured in-situ at 10-4 mbar by means of the Kelvin probe. The reduction of the zinc surface
by the argon/hydrogen plasma shifts the potential more cathodic (-150 mV). A very strong
anodic shift of +600 mV after oxygen plasma treatment can be observed.
-900
-800
-700
-600
-500
-400
-300
-200
-100
0
100
after
O2 plasma
Zn/Zn2+
after
Ar/H2 plasma
VSHE / mV
native
surface
Zn
Surface Potential
Figure 21: Mean surface potentials of a HDG surface after alkaline cleaning, argon/hydrogen-plasma
treatment and after oxygen plasma treatment measured in vacuum by means of Kelvin probe.
The formation of the insulating zinc oxide passive films by the oxygen plasma treatment
results in a further oxidation of Zn and Zn+ donor states to Zn2+ in the passive film as shown
in Figure 22.
5 Corrosion resistance of plasma modified HDG steel 50
Oxidation of Zn+-donor states to Zn2+
O*, OH*, O2-
O2- O2- O2-
Zn
Zn+
Zn2+ Zn2+
O2- O2- O2-
Zn2+ Zn2+
O2- O2- O2-
Zn2+ Zn2+
Zn2+
O2- O2-
Zn2+ Zn2+
Zn2+
O2- O2-
Zn2+ Zn2+
Zn2+
Zn+
Zn+
Zn2+
Zn
O2- O2- O2-
Zn
Zn+
Zn2+ Zn2+
O2- O2- O2-
Zn2+ Zn2+
O2- O2- O2-
Zn2+ Zn2+
Zn2+
O2- O2-
Zn2+ Zn2+
Zn2+
O2- O2-
Zn2+ Zn2+
Zn2+
Zn2+
Zn
Zn2+
Zn2+
Oxidation of Zn+-donor states to Zn2+
O*, OH*, O2-
O2- O2- O2-
Zn
Zn+
Zn2+ Zn2+
O2- O2- O2-
Zn2+ Zn2+
O2- O2- O2-
Zn2+ Zn2+
Zn2+
O2- O2-
Zn2+ Zn2+
Zn2+
O2- O2-
Zn2+ Zn2+
Zn2+
Zn+
Zn+
Zn2+
Zn
O2- O2- O2-
Zn
Zn+
Zn2+ Zn2+
O2- O2- O2-
Zn2+ Zn2+
O2- O2- O2-
Zn2+ Zn2+
Zn2+
O2- O2-
Zn2+ Zn2+
Zn2+
O2- O2-
Zn2+ Zn2+
Zn2+
Zn+
Zn+
Zn2+
Zn
O2- O2- O2-
Zn
Zn+
Zn2+ Zn2+
O2- O2- O2-
Zn2+ Zn2+
O2- O2- O2-
Zn2+ Zn2+
Zn2+
O2- O2-
Zn2+ Zn2+
Zn2+
O2- O2-
Zn2+ Zn2+
Zn2+
Zn2+
Zn
Zn2+
Zn2+
Figure 22: Schematic lattice structure of zinc oxide for the oxidation of interstitial Zn+ ions due to
plasma oxidation.
This further oxidation leads to the observed anodic shift of the Volta potential according to
the interpretation of plasma oxidation of iron surfaces by Grundmeier and Stratmann [14].
Equation 20 suggests an increase of the Volta potential difference between the non-modified
and plasma modified state due to the lowering of the Zn+-density in the passive film.
()
(
)
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
⋅=
+
+
ox.plasmaa
surface nativea
ln
F
RT
ΔΨΔ
Zn
Zn
Ref
Ox (20)
(
)
Ref
Ox
ΔΨ is the Volta potential difference between the vibrating reference electrode and the
oxide surface and +
Zn
ais the Zn+ activity respectively in the native or plasma oxidized zinc
oxide surface.
The corresponding chemical reaction can be described as follows:
−+•+ +→+ ads
2
ZnO
ads
ZnO OHZnOHZn (21)
The change of the surface potential due to a change in the surface dipole density and
orientation is neglected in this consideration since the surface of the passive film is anyhow
hydroxide terminated as shown by XPS measurements [1].
5.1.4 Stability of zinc surfaces after plasma modification
The electronic properties of the passive film after plasma modification strongly change in
humid atmospheres. Figure 23 shows the Volta potential transient of an oxygen plasma
treated zinc surface in humidified air (95% r.h.). The Volta potential shifts cathodically within
minutes due to the change of the electronic properties of the passive layer. The adsorbed ultra-
thin electrolyte layer promotes the dissolution of Zn ions into the aqueous phase.
5 Corrosion resistance of plasma modified HDG steel 51
Figure 23: In-situ Kelvin probe measurement of the relaxation of the Volta potential after oxygen
plasma measured in dry air and switching from 0 to 95% relative humidity.
The increased electric field in the passive film then leads to a further oxidation of Zn to Zn+ at
the oxide/metal interface. The passive film thereby readjusts its chemical composition to the
state which is given under the condition of atmospheric conditions with lower oxygen activity
in comparison to the oxygen plasma.
These results show that zinc oxide films with low donor densities prepared by oxygen plasma
treatment cannot be stabilised in the presence of atmospheric water activities.
5.2 Deposition of ultra-thin plasma polymer films
Since highly oxidised zinc oxides are reduced again in humid atmospheres leading to
increased donor densities, an insulating film is required to inhibit the transfer of electrons to
surface adsorbed oxygen molecules.
This concept takes into account that the donor density of the oxide and thereby its electronic
conductivity is high and that the insulating properties of the oxide/plasma polymer are
5 Corrosion resistance of plasma modified HDG steel 52
provided by the ultra-thin plasma polymer film. These ultra-thin SiO2-like plasma polymer
films there are well known due to their barrier properties [50, 52, 86].
5.2.1 In-situ FT-IR analysis after plasma deposition
Figure 24a and b show a FT-IRRAS spectrum of a SiO2-like film after 20 s film deposition
from a mixture of HMDSO and oxygen. Both regions show the carbon free deposition of the
films indicated by the negative vibration band for CH2,3 at 2850-2900 cm-1 as the negative
peaks for CH2,3 occur due to the removal of organic contaminations on the reference sample.
The asymmetric Si-O-Si stretching vibration with the main peak at 1232 cm-1 is a mixture of
transverse optical and different longitudinal optical modes which originate from the angle of
incidence of the polarized laser light. Silica-like structures are represented by the Si-O-Si
peak at 1170 cm-1 and a small shoulder at 1070 cm-1. The free dangling OH-groups can be
attributed to the peak at 3600 cm-1 whereas water which is adsorbed at the surface and
incorporated in voids of the film during plasma deposition gives rise to the broad absorption
at 3300 and 3520 cm-1 [91].
All peaks assignments are listed in Table 3.
Table 3: Assignment of the FT-IRRAS peaks after deposition of SiO2-like film.
Wavenumber (cm-1) Group Assignment Ref.
820 Si-O-Si δ(Si-O-Si) [52]
921 Si-OH νs(Si-OH) [52, 62]
1070 Si-OH νas(Si-OH) [62]
1170 Si-O-Si νas(Si-O-Si) [52]
1232 Si-O-Si νas(Si-O-Si) [52, 62]
2850-2900 CH2, CH3 νas,s(CHx) [88]
3300 H2O νs(OH) [87, 89]
3520 H2O νas(OH) [87, 89]
3600 Si-OH ν(OH) [91]
5 Corrosion resistance of plasma modified HDG steel 53
2500275030003250350037504000
0
1
2
3
ν(OH)
(a)
νs(OH)
νas,s(CHx)
νas(OH)
Absorbance / 10-3
Wavenumber / cm-1
70080090010001100120013001400
0
10
20
30
40
(b)
νas(Si-OH)
νas(Si-O-Si)
Absorbance / 10-3
δ(Si-O-Si)
νs(Si-OH)
Wavenumber / c
m
-1
νas(Si-O-Si)*
2500275030003250350037504000
0
1
2
3
ν(OH)
(a)
νs(OH)
νas,s(CHx)
νas(OH)
Absorbance / 10-3
Wavenumber / cm-1
70080090010001100120013001400
0
10
20
30
40
(b)
νas(Si-OH)
νas(Si-O-Si)
Absorbance / 10-3
δ(Si-O-Si)
νs(Si-OH)
Wavenumber / c
m
-1
νas(Si-O-Si)*
Figure 24: FT-IRRAS difference spectrum of an ultra-thin insulating SiO2-like film on an alkaline
cleaned HDG after oxygen plasma cleaning and film deposition for two wavenumber regions: (a)
4000 - 2500 cm-1 and (b) 1400 - 700 cm-1 (reference: state after alkaline cleaning).
5.2.2 Chemical composition of plasma polymer films
The XPS depth profiling in Figure 25 shows in detail, that the chemical composition of such
inorganic plasma deposited films differs from pure SiO2 layers. The oxygen excess which can
be observed over the whole layer thickness can be explained by the formation of OH-dangling
groups and the adsorption of water molecules at the outer film surface and at interfaces within
the film due to the network structure with voids and defects. The depth profile also proofs the
carbon free film deposition as well as the existence of the zinc oxide/hydroxide layer between
the zinc and the SiO2-like film.
5 Corrosion resistance of plasma modified HDG steel 54
0 2 4 6 8 101214161820
0
10
20
30
40
50
60
70
80
90
100
Reference SiO2
6.4nm in 60sec
Atomic concentration / %
Sputter depth / nm
C
O
Si
Zn
Figure 25: XPS sputter profile of a 10 nm thin SiO2 film on alkaline cleaned HDG.
5.2.3 Change of surface structures after film deposition
To get information on the film morphology AFM images were measured on selected single
grains of the Zn-alloy after the alkaline cleaning (Figure 26a) and after deposition of a 10 nm
SiO2-like film (Figure 26b). The wavy structure (nanometre scale) on the alkaline cleaned
HDG steel is formed during the cooling process after the hot-dip galvanising process [1]. The
alkaline cleaning results in the removal of the surface aluminium layer and leads to etch
grooves with nanometre depth [1]. The deposited 10 nm thin SiO2-like film perfectly imitates
and covers these etch grooves and moreover partly levels the original wavy structure.
However, nanometre and sub-nanometre sized defects within the SiO2-like film cannot be
observed by such an AFM-measurement.
5 Corrosion resistance of plasma modified HDG steel 55
300 nm
4 nm
0 nm
Topography
(a)
300 nm
Topography
6 nm
0 nm
(b)
300 nm
4 nm
0 nm
Topography
(a)
300 nm
Topography
6 nm
0 nm
(b)
Figure 26: AFM measurements of the topography of (a) an alkaline cleaned HDG surface and (b) the
same surface covered with a 10 nm thin SiO2 film.
5.3 Barrier properties of SiO2-like films for different film thickness
values
While the FT-IRRAS, XPS and AFM measurements reveal the composition and morphology
of the SiO2-like film, the barrier properties of this ultra-thin layer has to be proven by
electrochemical methods as it not possible to detect nanosized defects in the layer directly
with spectroscopic and microscopic measurements.
5.3.1 Surface coverage for different film thickness values
Hence, additional electrochemical measurements were carried out to determine the ratio
between covered and uncovered surface. Cyclic voltammetry (CV) allows the measurement of
the Zn oxidation and reduction current densities which can be related to the coverage of the
oxidised zinc surface [78, 79]. Results of the corresponding CV measurements are shown in
Figure 27 for a bare oxide covered zinc surface, as well as for three different film thickness
values of the SiO2-like film. For the bare zinc surface the area of the Zn reduction peak (-1.04
V versus SHE) and the Zn oxidation peak (-0.80 V versus SHE) are clearly visible. Already a
10 nm thin SiO2-like film inhibits the zinc reduction and oxidation almost completely as it can
be seen in the inset of Figure 27 (small inset: y-axis magnified by 2 orders of magnitude). The
cyclic voltammograms could be repeated several times without a measurable increase in the
measured current densities. The resulting current densities are further reduced with increased
film thickness which shows that small defects at 10 nm film thickness are diminishing at film
thickness of several ten nanometres.
5 Corrosion resistance of plasma modified HDG steel 56
-1.2 -1.1 -1.0 -0.9 -0.8 -0.7 -0.6
-0.80
-0.60
-0.40
-0.20
0.00
0.20
0.40
-1.2 -1.0 -0.8 -0.6
-2
-1
0
1
without SiO2
10 nm SiO2
50 nm SiO2
100 nm SiO2
i / mAcm-2
U / V (SHE)
× 1E-2
Figure 27: Cyclic voltammograms on alkaline cleaned HDG (continuous line) and with plasma
polymers on top with thickness values of 10 (dashed line), 50 (doted line) and 100 nm (dashed-doted
line).
For the plasma polymer film covered surfaces no CV peak analysis could be performed.
However, the estimated uncovered area is below 0.5% already at 10 nm film thickness.
The non-vanishing slope of the curve for the 10 nm thin film in the cathodic region shows that
very low oxygen and hydrogen reduction current densities are still possible in nanoscopic
pinholes. However, in comparison to the bare oxide covered zinc surface the electron transfer
reactions are significantly inhibited.
5.3.2 Barrier properties - pore and film resistance
Electrochemical impedance spectroscopy (EIS) measurements proved the existence of
nanoscopic pores in the deposited films as shown in Figure 28a and b. For ultra-thin film
coated surfaces, two different time constants can be observed at 10 and 5000 Hz. The
decrease of the film capacitance values with increasing film thickness confirms the barrier
properties of the insulating films. The pore resistance below 1 Hz is increasing with film
thickness, but even the value for the 10 nm film is one order of magnitude higher than for the
5 Corrosion resistance of plasma modified HDG steel 57
-1012345
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
without SiO2
10nm SiO2
50nm SiO2
100nm SiO2
log (f / Hz)
log (|Z| / (Ohm cm
2
) )
(a)
-1 0 1 2 3 4 5
-90
-75
-60
-45
-30
-15
0
(b)
without SiO
2
10nm SiO
2
50nm SiO
2
100nm SiO
2
log (f / Hz)
θ
/ °
-1012345
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
without SiO
2
10nm SiO
2
50nm SiO
2
100nm SiO
2
log (f / Hz)
log (|Z| / (Ohm cm
2
) )
(a)
-1 0 1 2 3 4 5
-90
-75
-60
-45
-30
-15
0
(b)
without SiO
2
10nm SiO
2
50nm SiO
2
100nm SiO
2
log (f / Hz)
θ
/ °
Figure 28: Electrochemical impedance spectroscopy on alkaline cleaned HDG and with plasma
polymers on top with thickness values of 10, 50 and 100 nm: (a) impedance and (b) phase.
uncovered HDG steel which indicates that these ultra-thin films form an efficient barrier for
the aqueous electrolyte. For all films the existence of nanoscopic pinholes were confirmed by
EIS measurements as the phase
ϑ
would be around ~|90°| for perfect barrier coatings and the
impedance curve could be expressed by a straight line of slope -1 in the log(|Z|)/log(f) plot.
However, for the application as interfacial films between the metal and an organic coating
these few pinholes are not influencing the corrosion resistance.
5 Corrosion resistance of plasma modified HDG steel 58
5.4 Influence of intact SiO2-like films on unformed zinc surfaces
Klimow et al. observed a correlation between the inhibition of the oxygen reduction at the
interface and the cathodic shift of the surface potential on conversion chemistry treated zinc
surfaces of galvanised steel [2]. This result could be also connected to a decreased progress of
cathodic de-adhesion processes at the interface. To reveal, if similar mechanisms are effective
for SiO2-like plasma deposited ultra-thin layers, the influence of SiO2-plasma polymers on the
interfacial ion transport kinetics and their effects on the interface potential has been
investigated by the SKP.
5.4.1 Kelvin probe analysis in vacuum and humid atmospheres
In Figure 29 the change in the Volta potentials directly after film deposition, previous oxygen
plasma cleaning and the untreated HDG surface are shown as measured in-situ by means of
the Kelvin probe in vacuum. A strong anodic shift after the oxygen plasma treatment of +600
mV was observed and assigned to the oxidation of the interstitial Zn and Zn+ donor states to
Zn2+ [14].
-900
-800
-700
-600
-500
-400
-300
-200
-100
0
100
SiO2 like
film
after
O2 plasma
VSHE / mV
native
surface
Zn
Surface Potential
Zn/Zn2+
Figure 29: Mean surface potentials of a HDG surface after alkaline cleaning, oxygen plasma
treatment and 10 nm SiO2 film deposition measured in-situ in vacuum by means of Kelvin probe.
5 Corrosion resistance of plasma modified HDG steel 59
Zinc-oxyhydroxide
10 nm SiO2
50 nm SiO2
Zinc
0 5 10 15 20 25 30 35
-750
-700
-650
-600
-550
-500
-450
-400
-350
VSHE / mV
Distance / mm
ΔE ~ 220mV
50 nm
SiO2
Zinc-oxyhydroxide
10 nm SiO2
50 nm SiO2
Zinc
0 5 10 15 20 25 30 35
-750
-700
-650
-600
-550
-500
-450
-400
-350
VSHE / mV
Distance / mm
ΔE ~ 220mV
50 nm
SiO2
Figure 30: Ex-situ height regulated scanning Kelvin probe line scan of HDG surface with thickness
step-gradient of SiO2-like film measured in O2 atmosphere with high humidity (95% r.h.).
After 10 nm SiO2-like film deposition, the surface potential was shifted 175 mV cathodically
in comparison to the native surface and thereby 675 mV cathodically in comparison to the just
plasma oxidised surface. By increasing the SiO2 film thickness to 60 or 110 nm, no further
decrease of the surface potential could be observed as shown in Figure 30, as the potential
stayed constant for film thickness values of 10, 60 and 110 nm.
5 Corrosion resistance of plasma modified HDG steel 60
Figure 31: Schematic band structure of a Zn surface covered with the corresponding n-
semiconducting oxide (a) or an insulating SiO2-like film (b) (following [92, 93]).
The oxygen reduction at the surface of the zinc oxide leads to a band bending of the valence
band due to the depletion of the electrons at the surface in oxygen containing environments.
This depletion is already completely inhibited by the insulating properties of the 10 nm SiO2-
like film and leads to the cathodic shift of the Volta potential.
Figure 31 illustrates the experimentally observed potential distributions and the particular
Galvani potential differences at the interfaces of the metal/metal oxide structures. The
electronic properties of the oxides are either n-semiconducting for the ZnO layer or insulating
for the SiO2-like film. In the case of the n-semiconducting ZnO layer and for the electronic
equilibrium between the metallic zinc and its oxide, the Galvani potential differences at the
interfaces between metal/metal oxide and metal oxide/humid atmosphere are strongly
determined by the presence of molecular oxygen [92]. In the case for the zinc oxide, the
relative position of the energy levels of the conduction band and the electronic terms of
oxygen allows an electron transfer to reduce oxygen at the surface and to form surface states
which shifts the potential distribution at the outermost interface.
5 Corrosion resistance of plasma modified HDG steel 61
0
4
8
12
16
20
0
4
8
12
16
-1050
-900
-750
-600
-450
E(mV SHE)
Potential / mVSHE
y-axis / mm
y-axis / mm
-1000
-900
-800
-700
-600
-500
Zn(OH)
SiO
2
(d=10nm)
Polymer (d=2µm)
Zn
Zn(OH)
Zn
Zn(OH)
Polymer (d=2µm)
Zn
Zn(OH)
SiO
2
(d=10nm)
Zn
0
4
8
12
16
20
0
4
8
12
16
-1050
-900
-750
-600
-450
E(mV SHE)
Potential / mVSHE
y-axis / mm
y-axis / mm
-1000
-900
-800
-700
-600
-500
Zn(OH)
SiO
2
(d=10nm)
Polymer (d=2µm)
Zn
Zn(OH)
SiO
2
(d=10nm)
Polymer (d=2µm)
Zn
Zn(OH)
Zn
Zn(OH)
Zn
Zn(OH)
Polymer (d=2µm)
Zn
Zn(OH)
Polymer (d=2µm)
Zn
Zn(OH)
SiO
2
(d=10nm)
Zn
Zn(OH)
SiO
2
(d=10nm)
Zn
Figure 32: Ex-situ height regulated scanning Kelvin probe potential mapping of an alkaline cleaned
HDG sample with four different surface configurations in oxygen atmosphere (95% r.h.).
SiO2 is an insulating material with a band gap Eg of about 9.0eV [94] which inhibits the
electron transfer to the oxygen states located at the SiO2 surface. For perfect barrier films in
the ideal case, the calibrated Volta potential difference 2
fRe
SiO
ΔΨ between the sample covered
with the insulating SiO2-like film and the reference probe is expected to be independent from
the prevailing atmosphere and very close to the electrode potential of the Zn/Zn2+ electrode.
According to the hypothesis that the applied interfacial potential, due to the electrolytic
connection between the polymer/metal interface and the corroding defect, drives the
interfacial oxygen reduction kinetics and thereby the cathodic de-adhesion process [2] it is
interesting to measure the adjusted interfacial potentials after the coverage of the modified
surfaces with an organic film.
Therefore, a sample with four different surface configurations was prepared. The Volta
potential mapping of the surface of measurement of such a sample in humid oxygen
containing atmosphere (95% relative humidity) by means of the height regulated scanning
Kelvin probe is displayed in Figure 32. The respective potentials as a function of the interface
chemistry can be directly observed by comparing the four quadrants of the sample.
5 Corrosion resistance of plasma modified HDG steel 62
By applying the organic polymer film, the resulting interface potential is shifted cathodically.
As the possibility to reduce oxygen correlates with the measured Volta potential, the
occupation of possible oxygen reduction sites by polymer chains results in a decreased
oxygen reduction at the zinc oxide/polymer interface. Moreover, the micrometer thick
polymer film decreases the activity of oxygen at the interface. This leads to the cathodic shift
of the Volta potential in comparison to the bare zinc oxide surface. Due to the insulating
SiO2-like film, the combination of the SiO2-like film with the organic polymer system leads to
the strongest cathodic shift of the Volta potential as the oxygen reduction is synergistically
minimized.
5.4.2 Corrosion processes and kinetics during the de-adhesion
processes in corrosive environments
Cathodic delamination in oxygen containing humid atmosphere is based on the de-adhesion of
polymer films from iron or zinc substrates [46, 95-99]. Oxygen reduction at the
polymer/oxide interface leads to an oxidative degradation of the polymer and increased
interfacial pH-values. This local cathode is electrolytically connected to the corresponding
anodic metal dissolution in the defect area, which is the local anode. To compensate the
generated negative interfacial charge, hydroxide ions are transported to the defect whereas
cations from the defect electrolyte enter the polymer/oxide/metal interface and are transported
to the front of de-adhesion [100, 101].
Figure 33a shows the averaged line scan of the cathodic delamination for the zinc surface of
the HDG samples coated with a 2 µm polymeric coating and a defect filled with 0.5M NaCl
as electrolyte. The x-axis represents the distance to the filled defect. After 3 hours the
delamination front is detected in the line scans as a potential difference between the intact
area (interface potential = -650 mV) and the defect coupled potential (-1100 mV) and
propagates along the zinc oxide/polymer interface. The speed of this propagating
delamination front is about 800 µm/h and slows down with increasing distance from the
defect.
5 Corrosion resistance of plasma modified HDG steel 63
0 2000 4000 6000 8000 10000
-1100
-1000
-900
-800
-700
-600
2 hours
3 hours
4 hours
6 hours
8 hours
10 hours
12 hours
14 hours
16 hours
Volta potential / mVSHE
Distance / µm
0 2000 4000 6000 8000 10000 12000
-1000
-900
-800
-700
-600
-500 Zn+10nm SiO2 2 hours
4 hours
6 hours
8 hours
10 hours
12 hours
14 hours
16 hours
18 hours
20 hours
Volta Potential / mVSHE
Distance / µm
Zn
(a)
(b)
Zn(OH)
Polymer (d=2µm)
Zn
Na
+
Cl
-
Zn(OH)
Polymer (d=2µm)
Zn
Na
+
Cl
-
SiO
2
(d=10nm)
0 2000 4000 6000 8000 10000
-1100
-1000
-900
-800
-700
-600
2 hours
3 hours
4 hours
6 hours
8 hours
10 hours
12 hours
14 hours
16 hours
Volta potential / mVSHE
Distance / µm
0 2000 4000 6000 8000 10000 12000
-1000
-900
-800
-700
-600
-500 Zn+10nm SiO2 2 hours
4 hours
6 hours
8 hours
10 hours
12 hours
14 hours
16 hours
18 hours
20 hours
Volta Potential / mVSHE
Distance / µm
Zn
(a)
(b)
Zn(OH)
Polymer (d=2µm)
Zn
Na
+
Cl
-
Zn(OH)
Polymer (d=2µm)
Zn
Na
+
Cl
-
Zn(OH)
Polymer (d=2µm)
Zn
Na
+
Cl
-
SiO
2
(d=10nm)
Zn(OH)
Polymer (d=2µm)
Zn
Na
+
Cl
-
SiO
2
(d=10nm)
Figure 33: Height regulated scanning Kelvin probe line profiles of corrosive de-adhesion on (a)
alkaline cleaned HDG and (b) with 10 nm SiO2 coating for x > 1600 µm.
In Figure 33b the averaged line scans for a sample with a 10 nm interfacial SiO2-like film is
shown. The SiO2-like film was deposited on the sample for distances to the defect larger than
1600 µm. The anodic shift (-350 mV) caused by the deposition of the SiO2-like film is clearly
visible in comparison to the uncovered zinc surface for distances to the defect smaller than
1600 µm. When the cathodic delamination on zinc reaches the SiO2-like film the de-adhesion
process stopped and the potential step vanished, as the potential difference between the defect
and the SiO2 coated area becomes nearly zero. As shown by the HR-SKP measurement, the
effective inhibition of the oxygen reduction at the intact oxide/SiO2/polymer interfaces leads
to a strong inhibition of the cathodic de-adhesion process and to a diminishing potential
difference between the intact area and the corroding defect which could act as a driving force
for the de-adhesion process.
5 Corrosion resistance of plasma modified HDG steel 64
Defect
with NaCl
Defect
with NaCl
Figure 34: Microscopic picture of an alkaline cleaned HDG sample partially covered with 10 nm SiO2
after 38 hours delamination.
Figure 34 shows the microscopic picture of an alkaline cleaned HDG sample with partial
coverage of a 10 nm SiO2-like film in the lower half and a polymeric top coating on the whole
sample after 38 hours with activated defect (left part). The corrosive de-adhesion propagating
away from the defect in the upper part on the pure zinc surface can be clearly seen due to the
darkening of the surface due to the formation of zinc corrosion products. In the lower part the
step between the pure zinc surface and the SiO2 covered part is clearly visible as sharp edge,
with no colour change on the intact SiO2 surface as no zinc corrosion products are formed.
Moreover, peel tests proved that no de-adhesion of the polymer film from the SiO2-surface
occurred.
5.4.3 Influence of SiO2-like films on the cathodic protection of iron by
zinc coatings
A very fast corrosive de-adhesion of the polymer was observed for zinc coated steel in the
previous chapter. In that case the local anode for the corrosive de-adhesion was formed by the
dissolution of zinc in the defect area. On HDG steel not only the zinc surface but also the iron
of the steel can be exposed to corrosive environments.
The cathodic protection of iron is based on the formation of a galvanic element between zinc
and iron if the surfaces are exposed to corrosive environments [31]. The iron is protected by
5 Corrosion resistance of plasma modified HDG steel 65
(a)
(b)
0 2000 4000 6000 8000 10000
-1100
-1000
-900
-800
-700
-600
-500
-400
-300
-200
-100
0
100
1 hours 20 hours
2 hours 30 hours
4 hours 40 hours
6 hours 50 hours
10 hours
Zn+10nm SiO2
Volta Potential / mVSHE
Distance / µm
Fe
0 2000 4000 6000 8000 10000
-1100
-1000
-900
-800
-700
-600
-500
-400
-300
-200
-100
0
100
1 hours 20 hours
2 hours 30 hours
4 hours 40 hours
6 hours 50 hours
10 hours
Volta Potential / mVSHE
Distance / µm
Zn
Fe
Polymer (d=2µm)
Fe
Na
+
Cl
-
Zn(OH)
Zn
Polymer (d=2µm)
Fe
Na
+
Cl
-
Zn(OH)
Zn
SiO2 (d=10nm)
(a)
(b)
0 2000 4000 6000 8000 10000
-1100
-1000
-900
-800
-700
-600
-500
-400
-300
-200
-100
0
100
1 hours 20 hours
2 hours 30 hours
4 hours 40 hours
6 hours 50 hours
10 hours
Zn+10nm SiO2
Volta Potential / mVSHE
Distance / µm
Fe
0 2000 4000 6000 8000 10000
-1100
-1000
-900
-800
-700
-600
-500
-400
-300
-200
-100
0
100
1 hours 20 hours
2 hours 30 hours
4 hours 40 hours
6 hours 50 hours
10 hours
Volta Potential / mVSHE
Distance / µm
Zn
Fe
Polymer (d=2µm)
Fe
Na
+
Cl
-
Zn(OH)
Zn
Polymer (d=2µm)
Fe
Na
+
Cl
-
Zn(OH)
Zn
Polymer (d=2µm)
Fe
Na
+
Cl
-
Zn(OH)
Zn
SiO2 (d=10nm)
Figure 35: Height regulated scanning Kelvin probe line profiles of corrosive de-adhesion on (a)
alkaline cleaned HDG with the zinc coating etched down to steel and (b) with 10 nm SiO2 coating on
the remaining zinc surface for x > 1900 µm.
the zinc coating due to a barrier effect (intact zinc coating) and the galvanic element in which
zinc serves as sacrificial anode while steel acts as cathode.
A defect in the zinc coating was created by etching the zinc in the defect area with
hydrochloric acid. Similar to the measurement in Chapter 5.4.2 the sample was coated with
the organic polymer and partially coated with a 10 nm SiO2-like film. Figure 35 shows the
scanning Kelvin probe line profiles for both cases. In both cases, a galvanic element is formed
between zinc and iron as the Volta potential shifts anodic (-800mV) when the corrosive de-
adhesion front has reached the zinc coated surface. For the unprotected zinc surface a slow
corrosive de-adhesion can be observed, whereas the SiO2 covered zinc surfaces shows a stable
interface between the SiO2 and the organic layer.
Therefore, the barrier properties of the insulating SiO2-like film stabilise the interface and
slows the corrosive de-adhesion also for steel surfaces which have defects in the protective
zinc coating.
5 Corrosion resistance of plasma modified HDG steel 66
5.5 Conclusions and model
Based on the presented results in this chapter, it became possible to establish a structure-
property relationship that explains why highly corrosion resistant interfaces between
polymeric films and zinc substrates can be achieved by means of plasma processes.
The barrier properties of the PE-CVD deposited SiO2-like films correlate with the measured
interface electrode potential, as the kinetics of oxygen reduction on the oxide surface
determine the oxidation state of the oxide itself. Moreover, it is the Volta potential difference
between an intact film and an active defect which reflects the driving force and therefore the
rate for corrosive de-adhesion at the interphase of polymer coated SiO2 film covered zinc
surfaces.
The principal behaviour of the interfacial SiO2-like film can be assumed to be representative
for different highly crosslinked insulating films, e.g. such as Zr-oxide films.
Figure 36: Schematic principle of the oxygen reduction curve for the pure zinc and the SiO2 covered
zinc surface with a strongly reduced oxygen reduction current.
5 Corrosion resistance of plasma modified HDG steel 67
Figure 36 illustrates the two different situations on the pure zinc surface and with the ultra-
thin insulating 10 nm SiO2-like film on top. Plotted are the current density curves for the
oxygen reduction kinetics on zinc and the SiO2-like film. The steady state for the oxygen
reduction and oxide oxidation for the intact zinc oxide/polymer interface results in a large
anodic overpotential in comparison to the potential of the active defect where zinc ions are
dissolved. This overpotential represents the driving force for the cathodic delamination.
The insulating property of the SiO2-like film effectively inhibits the oxygen reduction at the
interface and therefore decreases the resulting current density which results in a very small
overpotential. The measured potential difference between defect and intact area therefore
gives direct information about the electron transfer reactions and as a consequence about the
stability of the interface in corrosive environments.
5 Corrosion resistance of plasma modified HDG steel 68
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 69
6 Corrosion resistance of tensile formed plasma polymer
films on HDG steel
In chapter 5 a correlation between the barrier properties of ultra-thin plasma polymer films
and the corrosion resistance of hot-dip galvanised steel was established. Chapter 6 focuses on
the influence of forming processes on the structure and functionality of interfacial ultra-thin
SiO2-like plasma polymer films on galvanised substrates.
Forming of thin film or organically coated metal sheets is one of the most important steps to
generate functional products. The generation of forming induced defects is not only limited to
defects in the protective inorganic [16-19] or organic [20] layers but also comprises defects in
the galvanised substrates [21-24]. The formation of defects of brittle films on ductile
substrates has been intensively studied both experimentally and theoretically [17, 25].
Usually, multiple sequential crack formation is observed for rather brittle films on ductile or
high elongation substrates. By increasing the strain the number of cracks increases as well.
This observation is explained by the so-called shear lag approximation presented in detail by
Wojciechowski and Mendolia [16, 17].
Depending on the applied stress and strain different mechanism of the formed defects can be
observed on pure zinc coated steel sheets [21-23]. Slip, twinning and cracking are the major
deformation modes on zinc grains depending on the applied stress. Due to the hcp structure of
zinc, basal slip is the easiest deformation mode in zinc and is therefore the predominant
source for defects [26-28]. Twinning can be found on zinc grains if compressive and tension
stress is applied simultaneously [21].
As it is well known from literature ceramic films can only be formed for less than 1%
elongation. Plasma polymer films are intended to follow the deformation of the substrate due
to their high cross-linking during plasma deposition [9-11, 29]. B. Baumert et al. reported that
plasma polymer films show substrate induced defects for high forming degrees [24].
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 70
To acquire a deeper understanding of forming induced defects and to correlate the barrier
properties of plasma polymer SiO2-like films with the overall corrosion resistance, ex-situ and
in-situ electrochemical measurements were combined with a miniature tensile testing device.
6.1 Forming of uncoated and coated HDG steel
6.1.1 Forming of uncoated HDG steel
Stretch forming of hot-dip galvanised steel leads to a grain orientation dependant change in
the structure and orientation of the metal surface region. According to Lazik et al. slip,
twinning and crack formation are the predominant modes to accommodate the strain states
during forming of the zinc crystals and determine the structural changes of the crystal surfaces
[21].
500 µm
φ=0.00 φ=0.10 φ=0.25
500 µm 500 µm
011
2
0101
0112
0101
0001
500 µm
φ=0.00 φ=0.10 φ=0.25
500 µm 500 µm500 µm
φ=0.00 φ=0.10 φ=0.25
500 µm 500 µm
011
2
0101
0112
0101
0001 011
2011
2
0101 0101
0112
0101
0001
Figure 37: EBSD grain orientation mapping before and after 0.1 and 0.25 uniaxial elongation of a
HDG coated steel sheet. The uniaxial forming direction is indicated by the arrow. The grain
orientation is indicated by the colours of the inverse pole figure. The lateral analysis spot size is 5µm.
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 71
The EBSD analysis of the galvanised steel surface illustrates the different forming behaviour
of the individual zinc grains (Figure 37). The typical grain size of the zinc coating is about
100 µm. Based on the hcp structure of the zinc grains, mechanical twinning and major basal
slip planes could be identified as main forming effects for the grains with changed texture [21,
26-28]. The texture change of these grains is mainly perpendicular orientated to the strain
direction. Further increase of the strain led to an enhanced turn over of the zinc crystal
orientation.
6.1.2 Forming of thin SiO2-like films on HDG steel
It can be expected from the structural change of the grains that a highly crosslinked thin SiO2-
like film is not able to follow elastically or plastically such a deformation but will lead to
crack formation as shown by B. Baumert et al. for plasma polymer coatings on hot-dip
galvanised steel according to the mechanism proposed by Wojciechoswski et al. [16, 17].
In Figure 38 the FE-SEM observation of the thin film coated substrate (film thickness: 50 nm)
illustrates the typical cracking behaviour of the SiO2 film on the surface of different grains
and in the area of grain boundaries for different uniaxial forming levels. The orientation of the
cracks inside the SiO2 coating follows the orientation of slip bands.
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 72
10 µm
φ=0.00
φ=0.10
φ=0.25
10 µm
10 µm
10 µm
φ=0.00
φ=0.10
φ=0.25
10 µm
10 µm
Figure 38: FE-SEM pictures of a 50 nm SiO2-like film on an alkaline cleaned HDG surface before and
after 0.1 and 0.25 uniaxial forming. The uniaxial forming direction is indicated by the arrow.
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 73
φ=0.25
2 µm
φ=0.25
2 µm
Figure 39: FE-SEM pictures of a 50 nm SiO2-like film on an alkaline cleaned HDG surface with 0.25
uniaxial forming. The uniaxial forming direction is indicated by the arrow.
Close to the grain boundaries cracks tend to grow perpendicular to the grain boundary while
the orientation changes to a common perpendicular orientation with regard to the direction of
strain on the grain surface in several micrometer distance to the grain boundary. To
compensate the strain between different grains, strain adjustment by activation of non-major
slip planes takes place along the grain boundaries indicated by cracks with differing
orientation [102].
According to the mechanism proposed by Kirk and Pilliar [72] the initial crack formation is
strongly located at slip bands with an increase in crack density and an increase of the width of
the cracks with increased strain. On fine gliding zinc grains with 50 nm SiO2-like films,
highly localised strains are distributed over the whole film thickness.
Figure 39 illustrates the situation in which the film with several tens of nanometres thickness
only partly keeps adhered to the oxide covered substrate (dark areas) while de-adhesion
occurs locally due to its cohesion strength (brighter areas). The approach of two neighbouring
cracks stops at a certain distance which is in good agreement of the common crack theory
because the local energy is not high enough to support both crack dissipations [17, 103].
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 74
6.1.3 Forming of ultra-thin SiO2-like films on HDG steel
For smaller film thickness values of about 10 nm the de-adhesion process is less favourable
compared to 50 or 100 nm thick SiO2-like films, since less stress is built up inside the films
before cracking occurs.
Green: Si Red: Zn
Zn-line scanSi-line scan
0.5 µm
φ=0.15
0.5 µm
(a) (b)
Green: Si Red: Zn
Zn-line scanSi-line scan
0.5 µm
φ=0.15
0.5 µm
(a) (b)
Figure 40: (a) FE-SEM picture of a 10 nm SiO2-like film after 0.15 uniaxial forming with a scanning
Auger line scan overlay for Si (green) and Zn (red) and (b) with a scanning Auger colour overlay for
Si (green) and Zn (red). The uniaxial forming direction is indicated by the arrow.
For ultra-thin plasma polymer films Baumert et al. could show that small cracks are located at
atomic scale slip lines [24]. Slip bands can be seen in Figure 40 for 10 nm thick SiO2-like
films with a spacing of 500 µm. Atomic scale slip lines can be observed between these major
slip bands as thin parallel cracks inside the coating. The free zinc surface inside these cracks
can be indentified by means of the scanning Auger electron spectroscopy surface mapping
with colour overlay for Zn and Si. However, no de-adhered areas could be identified as
contrast difference for the 10 nm film thickness by means of FE-SEM studies.
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 75
6.2 Barrier properties and corrosion resistance of SiO2-like films
after forming
6.2.1 In-situ cyclic voltammetry during stretch forming of thin film
coated substrates
Ultra-thin plasma polymer films with film thickness of 10 nm in the unformed state strongly
reduce the oxidation and reduction of the zinc surfaces as shown by cyclic voltammetric
measurements in Chapter 5.3.1. Measurements which can provide information on the
influence of forming on the electrochemical properties of ultra-thin films on HDG steel are
usually very time consuming as each sample has to be formed individually.
By measuring in-situ cyclic voltammograms during the stretch forming process the full range
of forming degrees can be covered in several ten minutes. Furthermore, statistical and
measurement set-up errors are reduced since the measurement is performed on one single
spot. The initial steps and the evolution of crack formation in ultra-thin plasma polymer films
can be followed by calculating the free zinc surface as a function of the applied strain. The
evaluation of the cyclic voltammograms was done based on the work of Schultze et al. [78,
79].
Figure 41a shows selected cyclic voltammetric measurements at different forming levels
during stretching of the uncoated zinc surface. The zinc oxidation and reduction peak
increases with increasing strain. As hydrogen evolution cannot be neglected for the zinc
reduction peak, only the values of the maximum current density of the zinc oxidation peak
were evaluated [78]. The measurement area itself was limited due to the capillary diameter
and by this averages several grains and grain boundaries. The maximum current density of the
zinc oxidation peak ( )|(max ZnOZniox ) reached saturation values for
ϕ
> 0.05 as shown in
Figure 41b.
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 76
-1.3 -1.2 -1.1 -1.0 -0.9 -0.8 -0.7 -0.6 -0.5
-1.20
-1.05
-0.90
-0.75
-0.60
-0.45
-0.30
-0.15
0.00
0.15
0.30
Zn reduction
Zn oxidation
ϕ=0.00
ϕ=0.05
ϕ=0.10
ϕ=0.15
ϕ=0.20
ϕ=0.25
i / mAcm-2
U / V (SHE)
(a)
(b)
0.00 0.05 0.10 0.15 0.20 0.25 0.30
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
iox / mAcm-2
ϕ
max iox(Zn|ZnO)
fit max iox(Zn|ZnO)
-1.3 -1.2 -1.1 -1.0 -0.9 -0.8 -0.7 -0.6 -0.5
-1.20
-1.05
-0.90
-0.75
-0.60
-0.45
-0.30
-0.15
0.00
0.15
0.30
Zn reduction
Zn oxidation
ϕ=0.00
ϕ=0.05
ϕ=0.10
ϕ=0.15
ϕ=0.20
ϕ=0.25
i / mAcm-2
U / V (SHE)
(a)
(b)
0.00 0.05 0.10 0.15 0.20 0.25 0.30
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
iox / mAcm-2
ϕ
max iox(Zn|ZnO)
fit max iox(Zn|ZnO)
Figure 41: (a) In-situ CV measurement of uncovered alkaline cleaned HDG steel during uniaxial
forming. (b) Maximum oxidation current densities of zinc plotted versus the forming level.
The increase of the maximum oxidation current density for the uncovered zinc is based on the
creation of fresh zinc surface due to the forming induced increase of the surface. The large
increase can be explained by the change of the smooth and planar surface in a 3-dimensional
surface with atomic height steps, surface roughening and morphological changes especially at
the grain boundaries.
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 77
Figure 42a shows selected cyclic voltammetric measurements at different forming levels
during stretching of a 30 nm SiO2-like film on the zinc surface of the HDG steel. Without
forming (
ϕ
= 0) no oxidation or reduction peak of zinc is measured due to the blocking
character of the SiO2-like film. The zinc oxidation and reduction peak of the cyclic
voltammograms increases with increasing forming level. Forming induced cracks in the SiO2
film at slip bands and morphologic changes especially at the grain boundaries increase the
free zinc surface which is directly correlated with the current densities of the oxidation and
reduction of zinc. The maximum current densities for the uncovered zinc oxide surface
()|(max ZnOZniox ) and for a 30 nm SiO2-like film surface ( )||(max 2
SiOZnOZniox ) are
plotted in Figure 42b.
By comparing the maximum oxidation current densities of the uncovered and the film covered
surface, the film surface area film
Θ was calculated as follows:
For the calculation the curve of the maximum oxidation current densities of the uncovered
zinc surface is idealized by a fit.
A strong reduction of the SiO2 film covered surface of the insulating ultra-thin 30 nm thick
SiO2-like film can be measured already at small forming degrees with only 40% film
coverage at ϕ = 0.10 (Figure 42c). The film coverage further decreases with increasing strain.
For high strains due to the large area of uncovered zinc after forming, the barrier function of
the SiO2-like film seems to be almost reduced to the value of the unprotected surface. Only
20% of the surface is still blocked by the SiO2 film.
)(max
)(max
1)(
)|(
)||( 2
ϕ
ϕ
ϕ
ZnOZn
ox
SiOZnOZn
ox
film ifit
i
−=Θ (22)
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 78
-1.3 -1.2 -1.1 -1.0 -0.9 -0.8 -0.7 -0.6 -0.5
-0.75
-0.60
-0.45
-0.30
-0.15
0.00
0.15
0.30
Zn reduction
Zn oxidation
ϕ=0.00
ϕ=0.05
ϕ=0.10
ϕ=0.15
ϕ=0.20
ϕ=0.25
i / mAcm-2
U / V (SHE)
(a)
(b)
(c)
0.00 0.05 0.10 0.15 0.20 0.25 0.30
0
20
40
60
80
100
Θfilm / %
ϕ
SiO2 film coverage
0.00 0.05 0.10 0.15 0.20 0.25 0.30
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
iox / mAcm-2
ϕ
max iox(Zn|ZnO)
fit max iox(Zn|ZnO)
max iox(Zn|ZnO|SiO2)
-1.3 -1.2 -1.1 -1.0 -0.9 -0.8 -0.7 -0.6 -0.5
-0.75
-0.60
-0.45
-0.30
-0.15
0.00
0.15
0.30
Zn reduction
Zn oxidation
ϕ=0.00
ϕ=0.05
ϕ=0.10
ϕ=0.15
ϕ=0.20
ϕ=0.25
i / mAcm-2
U / V (SHE)
(a)
(b)
(c)
0.00 0.05 0.10 0.15 0.20 0.25 0.30
0
20
40
60
80
100
Θfilm / %
ϕ
SiO2 film coverage
0.00 0.05 0.10 0.15 0.20 0.25 0.30
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
iox / mAcm-2
ϕ
max iox(Zn|ZnO)
fit max iox(Zn|ZnO)
max iox(Zn|ZnO|SiO2)
Figure 42: (a) In-situ CV measurement of alkaline cleaned HDG steel with 30nm SiO2-like film during
uniaxial forming. (b) Maximum oxidation current densities for HDG steel without and with 30nm
SiO2-like film plotted versus the forming level. (c) Decrease of the SiO2-like film coverage with
increasing forming level.
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 79
6.2.2 Micro- and nanoscopic Kelvin probe studies of defects and
interfacial corrosive de-adhesion
In the past 20 years the interfacial corrosive de-adhesion processes of thin and thick polymer
films on iron and zinc substrates was studied by integral and local electrochemical methods
[104-106]. Especially the scanning Kelvin probe was established as a method to measure non-
destructively the corrosion potential at the buried polymer/metal interface [2, 100, 101, 107].
In Chapter 5 it is shown that the corrosive de-adhesion on organically coated zinc is almost
completely inhibited by interfacial 10 nm SiO2-like plasma polymer films. The effective
inhibition of the oxygen reduction at the intact oxide/SiO2/polymer interfaces leads to a strong
inhibition of the cathodic de-adhesion process. Moreover, the potential difference between the
intact area and the corroding defect which could act as a driving force for the de-adhesion
process also diminishes. The typical line profiles obtained by the HR-SKP are shown in
Figure 43 with no detectable corrosive de-adhesion for the intact SiO2-like film and with the
fast propagating delamination front for non-modified polymer/zinc interfaces. The corrosive
de-adhesion of the polymer film starts at the defect and propagates in the direction of the
intact zinc oxide/SiO2/polymer interface.
0 1500 3000 4500 6000 7500 9000
-1100
-1000
-900
-800
-700
-600
-500
Zn
SiO2 + ϕ=0.00
SiO2 + ϕ=0.05
after 6 hours
ΔE
Volta Potential / mVSHE
Distance / µm
Δx
Polymer
Zn(OH)
Zn
Zn(OH)
Polymer
Zn
Polymer
Zn(OH)
SiO
2
Zn
0 1500 3000 4500 6000 7500 9000
-1100
-1000
-900
-800
-700
-600
-500
Zn
SiO2 + ϕ=0.00
SiO2 + ϕ=0.05
after 6 hours
ΔE
Volta Potential / mVSHE
Distance / µm
Δx
Polymer
Zn(OH)
Zn
Zn(OH)
Polymer
Zn
Zn(OH)
Polymer
Zn
Polymer
Zn(OH)
SiO
2
Zn
Polymer
Zn(OH)
SiO
2
Zn
Figure 43: Typical height regulated scanning Kelvin probe line profiles after 6 hours for bare zinc,
defect free SiO2 coated HDG and defect containing SiO2 coatings after forming (
ϕ
= 0.05). The thick-
ness of the SiO2 like films was 10nm.
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 80
0246810121416182022
0
1000
2000
3000
4000
5000
6000
7000
8000
Front position / µm
Time / h
Zinc
Zinc + 10nm SiO2 + ϕ=0.00
Zinc + 10nm SiO2 + ϕ=0.05
Zinc + 10nm SiO2 + ϕ=0.10
Zinc + 10nm SiO2 + ϕ=0.15
Zinc + 10nm SiO2 + ϕ=0.20
Figure 44: Comparison of the position of the delamination front for zinc without and with 10 nm SiO2-
like film and different uniaxial forming degrees.
The in-situ cyclic voltammetry measurements showed the strong decrease of the SiO2 film
covered zinc surface area during stretch forming of the samples. Therefore, corrosive de-
adhesion is also observed on the stretch formed samples with similar kinetics as for the
samples with no interfacial plasma polymer film.
To compare the delamination kinetics of the different formed samples in detail, the turning
points of the potential-transitions were plotted versus time in Figure 44. For the intact SiO2
film no propagating delamination front could be observed. For the non-modified interface and
the stretch formed samples with interfacial plasma polymer film a fast delamination was
observed. The corresponding delamination rates were not constant for longer times, as the
delamination process is strongly transport limited for large distances between the defect and
the delamination front.
To compare the delamination rates only the linear propagation within the first 7 hours was
evaluated. In Figure 45a the delamination rates are plotted versus the respective forming
degree. In Figure 45b the delamination rates are further plotted versus the potential difference
between the defect potential and the potential of the intact area. In both cases the delamination
rates show a step like increase with the formation of cracks in the plasma polymer film.
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 81
0.00 0.05 0.10 0.15 0.20
0
100
200
300
400
500
600
700
800
900
Front speed / µm/h
ϕ
Zinc
Zinc + 10nm SiO
2
0 75 150 225 300 375 450 525 600
0
100
200
300
400
500
600
700
800
900
ϕ
=0.00
ϕ
=0.15
ϕ
=0.20
ϕ
=0.00
ϕ
=0.05
Zinc
Zinc + 10nm SiO
2
ϕ
=0.10
Front speed / µm/h
|
Δ
potential | / mV
(a)
(b)
0.00 0.05 0.10 0.15 0.20
0
100
200
300
400
500
600
700
800
900
Front speed / µm/h
ϕ
Zinc
Zinc + 10nm SiO
2
0 75 150 225 300 375 450 525 600
0
100
200
300
400
500
600
700
800
900
ϕ
=0.00
ϕ
=0.15
ϕ
=0.20
ϕ
=0.00
ϕ
=0.05
Zinc
Zinc + 10nm SiO
2
ϕ
=0.10
Front speed / µm/h
|
Δ
potential | / mV
(a)
(b)
Figure 45: Comparison of the delamination speed for zinc without and with 10 nm SiO2-like film and
different uniaxial forming degrees in relation to (a) the forming degree and (b) the Volta potential
difference between the delaminated and the intact area.
Microscopic FT-IRRAS measurements after the removal of the delaminated organic coating
still confirm the existence of the SiO2 spectrum of the 10 nm thick film. Therefore, it could be
concluded that the organic coating was de-adhered from the defect containing SiO2 surface
film and the oxide covered zinc areas while the SiO2-like film still stays adhered to the
substrate.
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 82
0.00 0.05 0.10 0.15 0.20
-900
-800
-700
-600
-500
-400
Zinc
Zinc + 10nm SiO2
ϕ
Volta Potential / mVSHE
Figure 46: Mean Volta potential of HDG surface without and with 10 nm SiO2-like film after different
uniaxial forming degrees measured in O2 atmosphere with high humidity (95% r.h.).
The influence of forming on the measured Volta potential is plotted for different forming
degrees in Figure 46. The uncovered zinc surface showed a stable anodic Volta potential, as
the bare zinc surface is immediately oxidised under atmospheric conditions. For the SiO2 film
protected zinc surface the measured Volta potential increased with increasing forming degree.
The strong increase of the potential for small forming values correlates very well with the
strong decrease of the SiO2 film coverage based on the in-situ cyclic voltammetry
measurements.
It should be noted that for standard Kelvin probe measurements the used SKP tip diameter is
500 times larger than typical crack widths in the SiO2 film and therefore the SKP studies
always present a laterally averaged interface potential.
SKP-FM measurements were performed to overcome this averaging effect and to correlate the
microscopically measured potential of the HR-SKP with the nanoscopic defects. On an intact
and insulating 10 nm SiO2 film a constant potential level could be detected without an
indication of topography artefacts (see Figure 47).
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 83
300 nm
3 nm
0 nm
Topography SKP-FM
50 mV
0 nm
Phase
300 nm 300 nm
300 nm
3 nm
0 nm
Topography SKP-FM
50 mV
0 nm
Phase
300 nm 300 nm
Figure 47: AFM topography, phase and SKP-FM measurement of an alkaline cleaned HDG surface
with 10 nm SiO2-like film.
Figure 48 shows for instance a formed sample with cracks in a 50 nm SiO2 film which are
orientated vertical in the picture. Due to the high surface roughening small cracks in the SiO2-
like film are difficult to measure topographically However, distinct potential differences
between the intact and adhering SiO2 film and the uncovered zinc surface areas could be
0246
0
10
20
30
40
50
60
70
Height / nm
Distance / µm
170 nm
0 nm
Topography
0246
0
50
100
150
200
250
300
Height / nm
Distance / µm
SKP-FM
800 mV
0 nm
0246
0
200
400
600
800
Volta potential / mV
Distance / µm
0246
0
200
400
600
800
1000
1200
1400
Volta potential / mV
Distance / µm
(a) (b)
2 µm 2 µm
0246
0
10
20
30
40
50
60
70
Height / nm
Distance / µm
170 nm
0 nm
Topography
0246
0
50
100
150
200
250
300
Height / nm
Distance / µm
SKP-FM
800 mV
0 nm
0246
0
200
400
600
800
Volta potential / mV
Distance / µm
0246
0
200
400
600
800
1000
1200
1400
Volta potential / mV
Distance / µm
(a) (b)
2 µm 2 µm
Figure 48: (a) AFM topography and (b) SKP-FM measurement of an alkaline cleaned HDG surface
with 50 nm SiO2-like film after 0.2 uniaxial forming. In the centre of the pictures a part of the 50 nm
SiO2 film is peeled off after forming.
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 84
detected easily in the centre of the scan area where a part of the 50 nm thick SiO2 film was de-
adhered. A 40-50 nm step in the SiO2 coating could be observed in the topography line scan
as well as the corresponding jump of the potential of 600mV between the bare zinc and the
insulating SiO2-like film. This value is three times larger as it would be expected from
previous Kelvin probe measurements. But it should be noted that the absolute potential
contrast of the SiO2 and free zinc surface does not necessarily represent the true potential
value in SKP-FM measurements but indicates well in this case the existence of intact and non-
intact areas [48].
6.3 Conclusions and model
The microscopic and electrochemical studies illustrates that the formation of line shaped
defects with a distance of some hundred nanometre and a width of several tens of nanometres
after stretch forming leads to an almost complete loss of the barrier properties of the
interfacial 10 nm thick SiO2-like films with regard to the cathodic de-adhesion process. In-situ
cyclic voltammetry during forming is a valuable method to monitor the increase in uncovered
zinc during such a forming process. Due to the formation of the cracks in the SiO2 coating, the
interfacial oxygen reduction densities are significantly increased to values which are
comparable to those of organically coated zinc surfaces without an interfacial plasma polymer
film. Since still a large part of the surface is covered with the adhering SiO2-like film this can
be explained by the formation of a microelectrode array.
SKP-FM and HR-SKP studies show that the cathodic interfacial potential shift caused by the
SiO2 film is increased by the formation of cracks and that the integral potential is a
superposition of the interfacial potentials in those areas where the SiO2-like film is still
adhering and those areas where the surface is uncovered or where the SiO2 film is de-adhered.
In such cases the oxygen adsorption and reduction on the zinc oxide surface is not fully
inhibited any more.
The SKP studies of the cathodic de-adhesion process of the organic primer on the formed thin
film coated substrate show, that as for the electrochemical studies of immersed samples the
interfacial oxygen reduction and thereby the cathodic disbondment process under the organic
coating is comparable to the non thin film coated substrate as soon as cracks are formed in the
ultra-thin film.
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 85
~100nm
SKP needle
diameter ~50-150µm
Na
+
Cl
-
Volta
potential
Effective potential
Current
density
Delaminated area
Micro-electrode array
Transport of ions and water
SiO
2
~100nm
SKP needle
diameter ~50-150µm
Na
+
Cl
-
Volta
potential
Effective potential
Current
density
Delaminated area
Micro-electrode array
Transport of ions and water
SiO
2
Figure 49: Mechanistic picture of the correlation between forming of insulating film, corrosion
protection and SKP Volta potential values.
Figure 49 shows the proposed mechanistic picture of the cathodic de-adhesion process on the
crack containing interfacial thin interfacial films. The oxygen reduction density in the de-
adhered area of the organic coating with an ultra-thin interfacial electrolyte layer is not
limited by the SiO2 film covered areas due to the micro-electrode array. In the intact area the
adhering SiO2 film is leading to a cathodic shift of the effective interfacial potential
depending on the ratio between thin SiO2 film covered and uncovered zinc surface. The
applied organic coating is not able to reduce the interfacial oxygen reduction density and the
correlated potential as effectively as the SiO2 film due to its higher free volume at the
interface to the oxide covered zinc.
At the front of electrochemical de-adhesion, the oxygen reduction process is occurring in the
cracks and the small distance between the defects of few hundred nanometres enables a fast
transport of hydrated ions from the front of the de-adhered area to the next crack. Even for
small interfacial transport rates with an effective mobility of some micrometers per hour it
would take only minutes for hydrated ions to migrate from one defect to the next one [101].
After reaching the next defect the oxygen reduction density in this local area is increasing to
the value comparable to the one for the system without the interfacial SiO2-like film. This
explains why the increase of the corrosive de-adhesion velocity is valid already for quite
small strains of about 0.05.
6 Corrosion resistance of tensile formed plasma polymer films on HDG steel 86
7 General conclusions 87
7 General conclusions
The scope of this thesis was to modify the passive film on zinc with ultra-thin barrier plasma
polymer films and to correlate the structure and physical properties of the surface layer with
the inhibition of the interfacial oxygen reduction kinetics in humid and corrosive
environments. Especially the influence of forming of the ultra-thin barrier plasma polymer
coated substrates and the decrease of the corrosion protection was investigated in detail by
using a combination of a tensile testing device and an electrochemical capillary cell. This set-
up allowed monitoring in-situ the formation of free metallic surface during stretch forming of
the plasma polymer coated HDG substrate and therefore enables the correlation of the
forming degree with the loss of the corrosion resistance during tensile forming. Hot-dip
galvanised steel was used as a technical material of high industrial relevance.
The corrosion resistance of PE-CVD deposited SiO2-like films is based on the barrier
properties of the inorganic coating by inhibition of the oxygen reduction at the metal/polymer
interface in corrosive environments. Kelvin probe measurements reveal, that the Volta
potential difference between an intact film and an active defect on zinc reflects the driving
force for the cathodic corrosive de-adhesion of the polymer top-coat and therefore the rate for
corrosive de-adhesion at the interphase of the polymer coated SiO2 film covered zinc surfaces.
Even the insulating properties of the 10nm thick SiO2-like film effectively inhibit the oxygen
reduction and therefore decrease the resulting current density which results in a very small
overpotential.
Electrochemical impedance and cyclic voltammetry measurements showed that the oxygen
reduction kinetics were strongly inhibited even for the 10nm thick SiO2-like films. From the
7 General conclusions 88
viewpoint of thin film engineering, it is the aim to achieve functional properties at extremely
low thickness values. It can be assumed from the measurements that the principal behaviour
of the interfacial SiO2-like film as corrosion protection is representative for different highly
crosslinked insulating films, e.g. such as Zr-oxide films.
Chapter 6 focussed on the influence of the forming processes on the structure and
functionality of the interfacial ultra-thin SiO2-like plasma polymer films on zinc galvanised
substrates. The low flexibility of such thin films leads to characteristic defects during forming
of coated substrates which lead to a partial loss of the achieved functional properties. Stretch
forming of galvanised steel coated with SiO2-like plasma polymer films with a thickness of 10
to 50 nm was performed to study the formation of defects in the films and their relevance for
the corrosion protection properties of the coated substrate. For ultra-thin films surface induced
defects like atomic slip lines and slip bands on the zinc surface lead to crack formation in the
SiO2-like films, whereas the film adheres very well to the surface. If the film thickness is
larger than 50nm the induced stress by the surface defects of the zinc surface leads to a partial
flaking off of the SiO2 coating.
Microscopic, electrochemical and spectroscopic methods were combined to correlate the size
and distribution of these nanoscopic defects with the corrosive de-adhesion mechanisms and
kinetics for organically coated substrates. Cyclic voltammetric and scanning Kelvin probe
measurements showed that oxygen reduction occurs due to the 3-dimensional increased zinc
surface in the nanoscopic defects in between completely insulating areas of SiO2 film acting
as microelectrode arrays for the cathodic de-adhesion process.
In-situ cyclic voltammetry during forming is a valuable method to monitor the increase of the
uncovered zinc area during such a forming process. SKP-FM and HR-SKP studies showed
that the cathodic interfacial potential shift caused by the SiO2 film is increased by the
formation of cracks and that the integral potential is a superposition of the interfacial
potentials in those areas where the SiO2-like film is still adhering and those areas where the
surface is uncovered or where the SiO2 film is de-adhered. In such cases the oxygen
adsorption and reduction on the zinc oxide surface is not fully inhibited any more.
8 Outlook 89
8 Outlook
In future studies the existence of forming induced defects is the most limiting factor in the
development of formable functional barrier coatings.
Besides the restricted possibility to increase the elastic properties of ceramic or ductile plasma
polymer films further investigations should focus on the possibility to influence the
electrochemical activity of the forming induced nanosized defects within the insulating
plasma polymer films.
Electrochemical measurements used in-situ during forming allows the improved investigation
of functional coatings and the use of this combination should therefore be expanded.
As corrosion protected galvanised steel consists of different layers, plasma polymer films
should be combined with other established and new experimental concepts to form systems
with superior corrosion protection properties.
The galvanised steel could be protected by an enhanced zinc/magnesium coating as base
substrate which is proposed in various publications [108]. But up to now, no knowledge of the
Volta potential after forming of such enhanced coatings and therefore about a possible change
of the driving force of the cathodic delamination is available.
The combination of such enhanced coatings with ultra-thin insulating plasma polymer films
could probably form a very effective corrosion protection.
After forming and creation of defects in these kinds of substrates the cathodic delamination
processes are mainly limited by the strength of the bonding of the organic top coat and the
oxygen reduction in the formed defects at the metal surface. Therefore, adhesion promoting
molecules like gamma-aminopropyltriethoxysilane between the organic lacquer and the
plasma polymer films should increase the strength of the multi layer system and therefore
stabilise the system.
8 Outlook 90
The electrochemical activity of nanosized defects after forming could be influenced by highly
mobile lacquer additives which are able to migrate to the substrate/plasma polymer/lacquer
interfaces and inhibit or reduce the oxygen reduction of water molecules by inhibition of the
electron transfer from the free metallic surface to intermolecular water.
The combination of versatile in-situ tensile testing devices with electrochemical methods like
cyclic voltammetry or even Kelvin probe measurements offer a very effective way to further
test these complex systems.
9 Literature 91
9 Literature
[1] N. Fink, B. Wilson and G. Grundmeier, Formation of ultra-thin amorphous conversion
films on zinc alloy coatings Part 1. Composition and reactivity of native oxides on
ZnAl (0.05%) coatings, Electrochimica Acta 51 (2006) 2956-2963.
[2] G. Klimow, N. Fink and G. Grundmeier, Electrochemical studies of the inhibition of
the cathodic delamination of organically coated galvanised steel by thin conversion
films, Electrochimica Acta 53 (2007) 1290-1299.
[3] C. Stromberg, P. Thissen, I. Klueppel, N. Fink and G. Grundmeier, Synthesis and
characterisation of surface gradient thin conversion films on zinc coated steel,
Electrochimica Acta 52 (2006) 804-815.
[4] B. Wilson, N. Fink and G. Grundmeier, Formation of ultra-thin amorphous conversion
films on zinc alloy coatings - Part 2: Nucleation, growth and properties of inorganic-
organic ultra-thin hybrid films, Electrochimica Acta 51 (2006) 3066-3075.
[5] H. Nagel, A.G. Aberle and R. Hezel, Optimised antireflection coatings for planar
silicon solar cells using remote PECVD silicon nitride and porous silicon dioxide,
Progress in Photovoltaics 7 (1999) 245-260.
[6] B.S. Richards, Comparison of TiO2 and other dielectric coatings for buried-contact
solar cells: a review, Progress in Photovoltaics 12 (2004) 253-281.
[7] H. Yasuda, Glow-Discharge Polymerization, Macromolecular Reviews Part D-Journal
of Polymer Science 16 (1981) 199-293.
[8] H. Yasuda,
Plasma Polymerization. 1985, Orlando: Academic Press Inc.
9 Literature 92
[9] H.K. Yasuda, T.H.F. Wang, D.L. Cho and T.J. Lin, Corrosion protection of cold-
rolled steel by low temperature plasma interface engineering .2. Effects of oxides on
corrosion performance of E-coated steels, Progress in Organic Coatings 30 (1997) 31-
38.
[10] G. Grundmeier and M. Stratmann, Plasma polymerization - a new and promising way
for the corrosion protection of steel, Materials and Corrosion-Werkstoffe Und
Korrosion 49 (1998) 150-160.
[11] G. Grundmeier, P. Thiemann, J. Carpentier and V. Barranco, Tailored thin plasma
polymers for the corrosion protection of metals, Surface & Coatings Technology 174
(2003) 996-1001.
[12] T.J. Lin, J.A. Antonelli, D.J. Yang, H.K. Yasuda and F.T. Wang, Plasma treatment of
automotive steel for corrosion protection - a dry energetic process for coatings,
Progress in Organic Coatings 31 (1997) 351-361.
[13] W.J. Vanooij, D. Surman and H.K. Yasuda, Plasma-Polymerized Coatings of
Trimethylsilane Deposited on Cold-Rolled Steel Substrates .2. Effect of Deposition
Conditions on Corrosion Performance, Progress in Organic Coatings 25 (1995) 319-
337.
[14] G. Grundmeier and M. Stratmann, Influence of oxygen and argon plasma treatments
on the chemical structure and redox state of oxide covered iron, Applied Surface
Science 141 (1999) 43-56.
[15] V. Barranco, P. Thiemann, H.K. Yasuda, M. Stratmann and G. Grundmeier,
Spectroscopic and electrochemical characterisation of thin cathodic plasma polymer
films on iron, Applied Surface Science 229 (2004) 87-96.
[16] P.H. Wojciechowski and M.S. Mendolia, On the Multiple Fracture of Low-Elongation
Thin-Films Deposited on High-Elongation Substrates, Journal of Vacuum Science &
Technology A: Vacuum, Surfaces, and Films 7 (1989) 1282-1288.
[17] P.H. Wojciechowski and M.S. Mendolia, Physics of Thin Films. Thin Films for
Emerging Applications, ed. M.H. Francombe and J.L. Vossen. 1992, New York:
Academic Press.
9 Literature 93
[18] Y. Leterrier, Durability of nanosized oxygen-barrier coatings on polymers, Progress in
Materials Science 48 (2003) 1-55.
[19] A.P. McGuigan, G.A.D. Briggs, V. Burlakov, M. Yanaka and Y. Tsukahara, An
elastic-plastic shear lag model for fracture of layered coatings, Thin Solid Films 424
(2003) 219-223.
[20] A.C. Bastos and A.M.P. Simoes, Effect of uniaxial strain on the protective properties
of coil-coatings, Progress in Organic Coatings 46 (2003) 220-227.
[21] S. Lazik, C. Esling and J. Wegria, Cracking in Zinc Layers on Continous Galvanized
Sheets, Textures and Microstructures 23 (1995) 131-147.
[22] R. Parisot, S. Forest, A. Pineau, F. Grillon, X. Demonet and J.M. Mataigne,
Deformation and damage mechanisms of zinc coatings on hot-dip galvanized steel
sheets: Part 1. Deformation modes, Metallurgical and Materials Transactions A 35A
(2004) 797-811.
[23] R. Parisot, S. Forest, A. Pineau, F. Nguyen, X. Demonet and J.M. Mataigne,
Deformation and damage mechanisms of zinc coatings on hot-dip galvanized steel
sheets: Part II. Damage modes, Metallurgical and Materials Transactions A 35A
(2004) 813-823.
[24] B. Baumert, M. Stratmann and M. Rohwerder, Formability of ultra-thin plasma-
polymer films deposited on metal sheet: mesoscopic and nanoscopic aspects of defect
formation, Zeitschrift für Metallkunde 795 (2004) 257-262.
[25] H.L. Cox, The Elasticity and Strength of Paper and Other Fibrous Materials, British
Journal of Applied Physics 3 (1952) 72-79.
[26] R.L. Bell and R.W. Cahn, The Dynamics of Twinning and the Interrelation of Slip and
Twinning in Zinc Crystals, Proceedings of the Royal Society of London. Series A,
Mathematical and Physical Sciences 239 (1957) 494-521.
[27] P.G. Partridge, The crystallography and deformation modes of hexagonal close-
packed metals, Metallurgical and Materials Transactions A 12 (1957) 169-194.
9 Literature 94
[28] E.J. Stofel and D.S. Wood, Fracture of Solids. Vol. 20. 1963, New York: Interscience.
521-39.
[29] G. Grundmeier, M. Brettmann and P. Thiemann, In situ spectroscopic and corrosion
studies of ultra-thin gradient plasma polymer layers on zinc, Applied Surface Science
217 (2003) 223-232.
[30] A.R. Marder, The metallurgy of zinc-coated steel, Progress in Materials Science 45
(2000) 191-271.
[31] X.G. Zhang, Galvanic corrosion of zinc and its alloys, Journal of the Electrochemical
Society 143 (1996) 1472-1484.
[32] T.H. Muster, A.K. Neufeld and I.S. Cole, The protective nature of passivation films on
zinc: wetting and surface energy, Corrosion Science 46 (2004) 2337-2354.
[33] H. Yasuda, B.H. Chun, D.L. Cho, T.J. Lin, D.J. Yang and J.A. Antonelli, Morphology
of phosphate layer investigated by vacuum depositing a parylene C film, Corrosion 52
(1996) 584-591.
[34] H. Yasuda, B.H. Chun, D.L. Cho, T.J. Lin, D.J. Yang and J.A. Antonelli, Interface-
engineered parylene C coating for corrosion protection of cold-rolled steel, Corrosion
52 (1996) 169-176.
[35] N.J. Shirtcliffe, M. Stratmann and G. Grundmeier, In situ infrared spectroscopic
studies of ultrathin inorganic film growth on zinc in non-polymerizing cold plasmas,
Surface & Interface Analysis 35 (2003) 799-804.
[36] G. Grundmeier and M. Stratmann, Adhesion and De-adhesion mechanisms at
polymer/metal interfaces: Mechanistic understanding based on in situ studies of buried
interfaces, Annual Review of Materials Research 35 (2005) 571-615.
[37] H.S. Wroblowa and S.B. Qaderi, The Mechanism of Oxygen Reduction on Zinc,
Journal of Electroanalytical Chemistry and Interfacial Electrochemistry 295 (1990)
153-161.
9 Literature 95
[38] H.S. Wroblowa and S.B. Qaderi, Mechanism and Kinetics of Oxygen Reduction on
Steel, Journal of Electroanalytical Chemistry and Interfacial Electrochemistry 279
(1990) 231-242.
[39] H.S. Wroblowa, Intermediate Products of Atmospheric Oxygen Reduction and the
Integrity of Metal Organic Coating Interface, Journal of Eectroanalytical Chemistry
and Interfacial Electrochemistry 339 (1992) 31-40.
[40] A. Leng, H. Streckel and M. Stratmann, The delamination of polymeric coatings from
steel. Part 1. Calibration of the Kelvin probe and basic delamination mechanism,
Corrosion Science 41 (1999) 547-578.
[41] W. Fürbeth and M. Stratmann, The delamination of polymeric coatings from
electrogalvanised steel - a mechanistic approach. Part 1: delamination from a defect
with intact zinc layer, Corrosion Science 43 (2001) 207-227.
[42] R. d'Agostino,
Plasma Deposition, Treatment, and Etching of Polymers. 1990, Boston:
Academic Press.
[43] S. Eufinger, W.J. van Ooij and K.D. Conners, DC-plasma polymerization of
hexamethyldisiloxane .2. Surface and interface characterization of films deposited on
stainless-steel substrates, Surface and Interface Analysis 24 (1996) 841-855.
[44] W.J. vanOoij, S. Eufinger and S.Y. Guo, DC plasma polymerization of
hexamethyldisiloxane, Plasma Chemistry and Plasma Processing 17 (1997) 123-154.
[45] W. Michaeli, S. Gobel and R. Dahlmann, The influence of strain on the properties of
plasma polymerized permeation barriers, Journal of Polymer Engineering 24 (2004)
107-122.
[46] M. Stratmann and H. Streckel, On the Atmospheric Corrosion of Metals Which Are
Covered with Thin Electrolyte Layers .1. Verification of the Experimental-Technique,
Corrosion Science 30 (1990) 681-696.
[47] M. Stratmann, K.T. Kim and H. Streckel, New Experimental-Techniques for
Investigating the Atmospheric Corrosion of Metals Covered by Thin Electrolyte
Layers, Zeitschrift für Metallkunde 81 (1990) 715-725.
9 Literature 96
[48] M. Rohwerder and F. Turcu, High-resolution Kelvin probe microscopy in corrosion
science: Scanning Kelvin probe force microscopy (SKPFM) versus classical scanning
Kelvin probe (SKP), Electrochimica Acta 53 (2007) 290-299.
[49] G. Grundmeier, E. Matheisen and M. Stratmann, Formation and stability of ultrathin
organosilane polymers on iron, Journal of Adhesion Science and Technology 10
(1996) 573-588.
[50] T.F. Wang and H.K. Yasuda, Modification of Wettability of a Stainless-Steel Plate by
Cathodic Plasma Polymerization of Trimethylsilane Oxygen Mixtures, Journal of
Applied Polymer Science 55 (1995) 903-909.
[51] N. Shirtcliffe, P. Thiemann, M. Stratmann and G. Grundmeier, Chemical structure and
morphology of thin, organo-silicon plasma-polymer films as a function of process
parameters, Surface and Coatings Technology 142-144 (2001) 1121-1128.
[52] G. Grundmeier, P. Thiemann, J. Carpentier, N. Shirtcliffe and M. Stratmann, Tailoring
of the morphology and chemical composition of thin organosilane microwave plasma
polymer layers on metal substrates, Thin Solid Films 446 (2004) 61-71.
[53] G.R. Prasad, S. Daniels, D.C. Cameron, B.P. McNamara, E. Tully and R. O'Kennedy,
PECVD of biocompatible coatings on 316L stainless steel, Surface and Coatings
Technology 200 (2005) 1031-1035.
[54] H. Chatham, Oxygen diffusion barrier properties of transparent oxide coatings on
polymeric substrates, Surface and Coatings Technology 78 (1996) 1-9.
[55] U. Moosheimer and C. Bichler, Plasma pretreatment of polymer films as a key issue
for high barrier food packagings, Surface and Coatings Technology 116-119 (1999)
812-819.
[56] M. Deilmann, M. Grabowski, S. Theiss, N. Bibinov and P. Awakowicz, Permeation
mechanisms of pulsed microwave plasma deposited silicon oxide films for food
packaging applications, Journal of Physics D: Applied Physics 41 (2008) 135207
(7pp).
9 Literature 97
[57] V.S. Nguyen, J. Underhill, S. Fridmann and P. Pan, Plasma Organo-Silicon Polymers -
Deposition, Characterization, and Application in Multilayer Resist, Journal of the
Electrochemical Society 132 (1985) 1925-1932.
[58] F. Fracassi, R. Dagostino and P. Favia, Plasma-Enhanced Chemical Vapor-Deposition
of Organosilicon Thin-Films from Tetraethoxysilane-Oxygen Feeds, Journal of the
Electrochemical Society 139 (1992) 2636-2644.
[59] D. Hegemann, U. Vohrer, C. Oehr and R. Riedel, Deposition of SiOx films from O-
2/HMDSO plasmas, Surface & Coatings Technology 119 (1999) 1033-1036.
[60] S. Sahli, Y. Segui, S. Ramdani and Z. Takkouk, Rf Plasma Deposition from
Hexamethyldisiloxane Oxygen Mixtures, Thin Solid Films 250 (1994) 206-212.
[61] G. Grundmeier and A. Stratmann, Nucleation and Growth of Plasma-Polymerised
Hexamethyldisilazane on Iron -Substrates, Berichte der Bunsen-Gesellschaft für
Physikalische Chemie 99 (1995) 1387-1392.
[62] C. Rau and W. Kulisch, Mechanisms of Plasma Polymerization of Various Silico-
Organic Monomers, Thin Solid Films 249 (1994) 28-37.
[63] C.T. Lin, F. Li and T.D. Mantei, Low-temperature plasma deposition of dielectric
coatings from organosilicon precursors, Journal of Vacuum Science & Technology A:
Vacuum, Surfaces, and Films 17 (1999) 735-740.
[64] A.J. Beaudoin and A. Acharya, A model for rate-dependent flow of metal polycrystals
based on the slip plane lattice incompatibility, Materials Science & Engineering. A,
Structural Materials : Properties, Microstructure and Processing 309 (2001) 411-415.
[65] A.J. Beaudoin, A. Acharya, S.R. Chen, D.A. Korzekwa and M.G. Stout, Consideration
of grain-size effect and kinetics in the plastic deformation of metal polycrystals, Acta
Materialia 48 (2000) 3409-3423.
[66] Z. Marciniak, J.L. Duncan and S.J. Hu, Mechanics of Sheet Metal Forming. Second
edition ed. 2002, Oxford: Butterworth-Heinemann.
[67] K. Lange,
Umformtechnik. Second edition ed. 2002, Berlin Heidelberg New York:
Springer.
9 Literature 98
[68] E. Döge and B.-A. Behrens, Handbuch Umformtechnik: Grundlagen, Technologien,
Maschinen. 2007, Berlin Heidelberg New York: Springer.
[69] R. Parisot, S. Forest, A.F. Gourgues, A. Pineau and D. Mareuse, Modeling the
mechanical behavior of a multicrystalline zinc coating on a hot-dip galvanized steel
sheet, Computational Materials Science 19 (2000) 189-204.
[70] D. Raabe, M. Sachtleber, H. Weiland, G. Scheele and Z.S. Zhao, Grain-scale
micromechanics of polycrystal surfaces during plastic straining, Acta Materialia 51
(2003) 1539-1560.
[71] R. Becker, Effects of strain localization on surface roughening during sheet forming,
Acta Materialia 46 (1998) 1385-1401.
[72] P.B. Kirk and R.M. Pilliar, The deformation response of sol-gel-derived zirconia thin
films on 316L stainless steel substrates using a substrate straining test, Journal of
Materials Science 34 (1999) 3967-3975.
[73] K. Wapner and G. Grundmeier, Scanning Kelvin probe measurements of the stability
of adhesive/metal interfaces in corrosive environments, Advanced Engineering
Materials 6 (2004) 163-167.
[74] K. Wapner, B. Schoenberger, A. Stratmann and G. Grundmeier, Height-regulating
scanning Kelvin probe for simultaneous measurement of surface topology and
electrode potentials at buried polymer/metal interfaces, Journal of the Electrochemical
Society 152 (2005) E114-E122.
[75] J. Raacke, M. Giza and G. Grundmeier, Combination of FTIR reflection absorption
spectroscopy and work function measurement for in-situ studies of plasma
modification of polymer and metal surfaces, Surface & Coatings Technology 200
(2005) 280-283.
[76] F. Mansfeld, Use of electrochemical impedance spectroscopy for the study of
corrosion protection by polymer coatings - Reply, Journal of Applied Electrochemistry
25 (1995) 1145-1145.
[77] C.H. Hammann and W. Vielstich, Elektrochemie. 2003, Weinheim, Germany: Wiley-
VCH.
9 Literature 99
[78] E. Klusmann, U. Konig and J.W. Schultze, Electrochemical Characterization of
Phosphate Layers on Zinc Coated Steel, Werkstoffe Und Korrosion - Materials and
Corrosion 46 (1995) 83-91.
[79] A. Losch, J.W. Schultze and H.D. Speckmann, A New Electrochemical Method for
the Determination of the Free-Surface of Phosphate Layers, Applied Surface Science
52 (1991) 29-38.
[80] I. Klüppel, B. Schinkinger and G. Grundmeier, In-Situ Electrochemical Studies of
Forming Induced Defects of Organic Coatings on galvanised steel, Electrochimica
Acta 54 (2009) 3553-3560.
[81] V.P. Tolstoy, V.I. Chernyshova and S.V. A., Handbook of infrared spectroscopy of
ultrathin films. 2003: John Wiley & Sons.
[82] J.F. Moulder, W.F. Stickle, P.E. Sobol and K.D. Bomben, Handbook of X-ray
Photoelectron Spectroscopy. 1992, Eden Prairie, USA: Perkin-Elmer Corporation,
Physical Electronics Division.
[83] H. Fujiwara,
Spectroscopic ellipsometry: principles and applications. 2007,
Chichester, England: John Wiley and Sons.
[84] S.L. Flegler, J.W. Heckman and K.L. Klomparens, Scanning and Transmission
Electron Microscopy: An Introduction. 1993, Cary, U.S.A: Oxford University Press.
[85] A.J. Schwartz, M. Kumar and A.B. L., Electron backscatter diffraction in materials
science. 2000: Springer.
[86] N. Shirtcliffe, P. Thiemann, M. Stratmann and G. Grundmeier, Chemical structure and
morphology of thin, organo-silicon plasma-polymer films as a function of process
parameters, Surface & Coatings Technology 142 (2001) 1121-1128.
[87] A.K. Neufeld and I.S. Cole, Using Fourier transform infrared analysis to detect
corrosion products on the surface of metals exposed to atmospheric conditions,
Corrosion Science 53 (1997) 788-799.
9 Literature 100
[88] D. Lin-Vien, N.B. Colthup, W.G. Fateley and J.G. Grasselli, The Handbook of
Infrared and Raman Spectroscopy Chracteristic Frequencies of Organic Molecules.
1991, San Diego: Academic Press Inc.
[89] D. Eisenberg and W. Kauzmann, The Structure and Properties of Water. 1969,
London: Oxford University Press.
[90] N. Cabrera and N.F. Mott, Theory of the Oxidation of Metals, Reports on Progress in
Physics 12 (1948) 163-184.
[91] K. Fukui, H. Miyauchi and Y. Iwasawa, Highly sensitive detection of adsorbed species
on a SiO2 surface by reflection-absorption infrared spectroscopy, Chemical Physics
Letters 274 (1997) 133-139.
[92] C. Stromberg, P. Thissen, I. Klüppel, N. Fink and G. Grundmeier, Synthesis and
characterisation of surface gradient thin conversion films on zinc coated steel,
Electrochimica Acta 52 (2006) 804-815.
[93] J.W. Schultze and M.M. Lohrengel, Stability, reactivity and breakdown of passive
films. Problems of recent and future research, Electrochimica Acta 45 (2000) 2499-
2513.
[94] A.J. Bard, M. Stratmann and G.S. Frankel, Encyclopedia of Electrochemistry, Volume
4, Corrosion and Oxide Films. Encyclopedia of Electrochemistry. Vol. 4. 2003,
Weinheim: Wiley-VCH. 755.
[95] A. Leng, H. Streckel and M. Stratmann, The delamination of polymeric coatings from
steel. Part 1. Calibration of the Kelvinprobe and basic delamination mechanism,
Corrosion Science 41 (1999) 547-578.
[96] M. Stratmann, K.T. Kim and H. Streckel, New Experimental-Techniques for
Investigating the Atmospheric Corrosion of Metals Covered by Thin Electrolyte
Layers, Zeitschrift Fur Metallkunde 81 (1990) 715-725.
[97] M. Stratmann, A. Leng, W. Furbeth, H. Streckel, H. Gehmecker and K.H.
GrosseBrinkhaus, The scanning Kelvin probe; A new technique for the in situ analysis
of the delamination of organic coatings, Progress in Organic Coatings 27 (1996) 261-
267.
9 Literature 101
[98] M. Stratmann and H. Streckel, The Investigation of the Corrosion of Metal-Surfaces,
Covered with Thin Electrolyte Layers - a New Experimental-Technique, Berichte Der
Bunsen-Gesellschaft-Physical Chemistry Chemical Physics 92 (1988) 1244-1250.
[99] M. Stratmann, H. Streckel and R. Feser, A New Technique Able to Measure Directly
the Delamination of Organic Polymer-Films, Corrosion Science 32 (1991) 467-470.
[100] R. Posner, K. Wapner, M. Stratmann and G. Grundmeier, Transport processes of
hydrated ions at polymer/oxide/metal interfaces: Part 1. Transport at interfaces of
polymer coated oxide covered iron and zinc substrates, Electrochimica Acta 54 (2009)
891-899.
[101] R. Posner, T. Titz, K. Wapner, M. Stratmann and G. Grundmeier, Transport processes
of hydrated ions at polymer/oxide/metal interfaces: Part 2. Transport on oxide covered
iron and zinc surfaces, Electrochimica Acta 54 (2009) 900-908.
[102] J.G. Sevillano, P. Vanhoutte and E. Aernoudt, Large Strain Work-Hardening and
Textures, Progress in Materials Science 25 (1980) 71-412.
[103] M.H. Zhao, R. Fu, D. Lu and T.Y. Zhang, Critical thickness for cracking of
Pb(Zr0.53Ti0.47)O3 thin films deposited on Pt/Ti/Si(100) substrates, Acta Materialia
50 (2002) 4241-4254.
[104] M. Stratmann, A. Leng, W. Fürbeth, H. Streckel, H. Gehmecker and K.H.
GrosseBrinkhaus, The scanning Kelvin probe; A new technique for the in situ analysis
of the delamination of organic coatings, Progress in Organic Coatings 27 (1996) 261-
267.
[105] K. Ogle, S. Morel and N. Meddahi, An electrochemical study of the delamination of
polymer coatings on galvanized steel, Corrosion Science 47 (2005) 2034-2052.
[106] K. Wapner, M. Stratmann and G. Grundmeier, In situ infrared spectroscopic and
scanning Kelvin probe measurements of water and ion transport at polymer/metal
interfaces, Electrochimica Acta 51 (2006) 3303-3315.
[107] G. Grundmeier, K.-M. Jüttner and M. Stratmann, Novel Electrochemical Techniques
in Corrosion Research. Materials Science and Technologies, ed. R.W. Cahn, P.
Haasen, and E.J. Kramer. 2000, Weinheim: Wiley-VCH. 285–381.
9 Literature 102
[108] R. Hausbrand, M. Stratmann and M. Rohwerder, Delamination resistant zinc alloys:
Simple concept and results on the system zinc-magnesium, Steel Research
International 74 (2003) 453-458.
10 Publications related to this work 103
10 Publications related to this work
Peer-reviewed articles:
G. Grundmeier, M. Giza and T. Titz, Analysis of corrosion resistant plasma polymer films on
metals, Vakuum in Forschung und Praxis 18 (2006) 32–36.
R. Posner, T. Titz, K. Wapner, M. Stratmann and G. Grundmeier, Transport processes of
hydrated ions at polymer/oxide/metal interfaces: Part 2. Transport on oxide covered iron and
zinc surfaces, Electrochimica Acta 54 (2009) 900-908.
T. Titz, F. Hörzenberger, K. Van den Bergh and G. Grundmeier, Correlation of interfacial
electrode potential and corrosion resistance of plasma polymer coated galvanised steel. Part 1:
Ultra-thin plasma polymer films of varying thickness, Corrosion Science 52 (2010) 369–377.
T. Titz, F. Hörzenberger, K. Van den Bergh and G. Grundmeier, Correlation of interfacial
electrode potential and corrosion resistance of plasma polymer coated galvanised steel. Part 2:
Influence of forming induced defects, Corrosion Science 52 (2010) 378–386.
Conference presentations:
T. Titz, K. Wapner, and G. Grundmeier, Structure and properties of ultra-thin SiO2 plasma
polymer films at polymer/metal interfaces, 11th European Conference on Applications of
Surface and Interface Analysis, Vienna, Austria, (2005).
10 Publications related to this work 104
T. Titz, K. Wapner, and G. Grundmeier, Structure and properties of ultra-thin SiO2 plasma
polymer films at polymer/metal interfaces, American Vacuum Society 53rd International
Symposium, San Francisco, USA, (2006).
T. Titz and G. Grundmeier, Correlation of structure and corrosion resistance of ultra-thin SiO2
plasma polymer films at polymer/metal interfaces, 13. Bundesdeutsche Fachtagung
Plasmatechnologie, Bochum, Germany, (2007).
T. Titz, M. Giza and G. Grundmeier, Modification of Passive Films on Metals in Vacuum and
Atmospheric Pressure Plasmas, 16th International Colloquium on Plasma Processes, Toulouse,
France, (2007).
Poster presentations:
T. Titz, K. Wapner and G. Grundmeier, Struktur und Eigenschaften ultradünner SiO2-
Plasmapolymer Schichten an Polymer/Metall-Grenzflächen, 12. Bundesdeutsche Fachtagung
Plasmatechnologie, Braunschweig, Germany, (2005).
M. Giza, T. Titz and G. Grundmeier, Modification of passive layers on ZnMg-alloys by
means of oxidising and reducing plasmas, Deutsche Physikalische Gesellschaft:
Plasmaphysik, Augsburg, (2006).
T. Titz and G. Grundmeier, Correlation of structure and corrosion resistance of ultra-thin SiO2
plasma polymer films at polymer/metal interfaces, 12th European Conference on Applications
of Surface and Interface Analysis, Brussels, Belgium, (2007).
