close

Вход

Забыли?

вход по аккаунту

?

The use of high-resolution XPS and ToF-SIMS to investigate segregation phenomena of minor components of a model coil coating formulation

код для вставкиСкачать
SURFACE AND INTERFACE ANALYSIS
Surf. Interface Anal. 26, 444È454 (1998)
The Use of High-resolution XPS and ToF-SIMS to
Investigate Segregation Phenomena of Minor
Components of a Model Coil Coating Formulation
S. R. Leadley,1,¤ J. F. Watts,1,* C. J. BlomÐeld2 and C. Lowe3
1 School of Mechanical and Materials Engineering, University of Surrey, Guildford, Surrey, GU2 5XH, UK
2 Kratos Analytical, Tra†ord Wharf Road, Manchester, M17 1GP, UK
3 Becker Industrial Coatings, Goodlass Road, Speke, Liverpool, L24 9HJ, UK
The segregation of minor componentsÈa melamine–formaldehyde resin and a poly(acrylic) Ñow control agentÈin
a model epoxy resin coil coating applied to hot-dipped galvanized steel has been investigated by high-resolution
XPS and time-of-Ñight SIMS (ToF-SIMS). It is shown that by using the highest resolution monochromated XPS
currently available, the C 1s spectrum can be peak Ðtted to account for all eleven carbon functionalities present in
the three components of the organic coating. A ToF-SIMS analysis of the melamine–formaldehyde resin has been
undertaken and a comprehensive ion fragmentation scheme for the positive ion spectrum of this molecule in the
range 0–500 Da is proposed. It is shown, by both surface analytical techniques, that when the Ñow control agent is
excluded from the formulation the surface of the paint Ðlm is enriched in the melamine–formaldehyde component.
On addition of the Ñow control agent such segregation is only identiÐed by XPS ; the ToF-SIMS spectrum resembles that of the Ñow control agent. This is taken to be an indication of monolayer segregation of this component at
the paint/air interface. Such segregation phenomena are shown to be insensitive to substrate surface pretreatment.
( 1998 John Wiley & Sons, Ltd.
KEYWORDS : XPS ; ToF-SIMS ; surface segregation ; coil coatings ; adhesion
INTRODUCTION
Coatings have been used for centuries, primarily for two
purposes : to improve appearances and to protect substrates. Since the 1930s, processes have been developed
that both pretreat and coat continuous coil stock.1 The
strip, received in the form of a coil, is uncoiled,
degreased, pretreated, coated with paint, cured and
recoiled. Both sides of the coil can be coated in a continuous production run that may involve the application of both a primer and a topcoat. The pretreatment
and the coil coating applied will be determined by the
nature of the metal substrate and the end use of the
coated coil stock. The current European market for coil
stock is some 8 ] 108 m2, of which D80% is coated
steel and the remainder is coated aluminium.
The coating must be capable of being applied and
cured at line speeds in excess of 50 m min~1. It must
also be able to withstand quite violent forming processes involving large amounts of mechanical deformation without cracking, and more importantly without
losing adhesion. In addition, the majority of coil-coated
* Correspondence to : J. F. Watts, School of Mechanical and
Materials Engineering, University of Surrey, Guildford, Surrey GU2
5XH, UK.
¤ Present address : Dow Corning Limited, Barry, South Glamorgan, CF63 2YL, UK.
CCC 0142È2421/98/060444È011 $17.50
( 1998 John Wiley & Sons, Ltd.
products are used in harsh environments and are
expected to resist corrosion and photodegradation for
25 years or more. Coil coatings are used extensively for
a wide variety of di†erent purposes and they are considered as one of the most technically advanced paint
systems.2 The durability of coil-coated material is
reliant on good adhesion throughout its lifetime.
However adhesion is perhaps one of the least understood phenomena associated with coil coatings. Historically the approach to achieve good performance has
essentially consisted of an empirical trial and error.
Conventional thermally cured coatings consist of
complex mixtures of organic resins and inorganic pigments dissolved or dispersed in non-reactive volatile
liquids. Such coatings usually require the application of
heat to ensure a rapid rate of cure. During heating the
volatile solvents evaporate, which leaves a coating that
consists of inorganic pigments in a cross-linked polymer
network. When a topcoat is applied to the primer it is
essential that a robust bond be formed between the two
coatings. Detailed knowledge of the surface chemistry of
primers is important so that a greater understanding of
the interaction between primers and topcoats is gained.
This is of particular importance as the concept of
surface segregation of components within coatings and
adhesives is now recognized.2,3
This study is part of a research programme studying
segregation phenomena in coatings and adhesives, and
the e†ect on adhesion.4 This programme aims to
develop an underpinning theory of the manner in which
Received 26 August 1997
Accepted 17 February 1998
SEGREGATION PHENOMENA IN COATINGS USING XPS AND ToF-SIMS
445
Table 1. Formulations of thermally cured primers used in this investigation
Formulation A
Wt.%
Formulation B
Wt.%
7-Type epoxy resin
Hexamethoxymethyl melamine
Non-reactive solvents
27.0
4.0
69.0
7-Type epoxy resin
Hexamethoxymethyl melamine
Acrylic flow control agent
Non-reactive solvents
26.8
4.0
0.2
69.0
the individual components behave in fully formulated
systems. This behaviour will then be related to the performance of the coatings and adhesives under accelerated test methods. This study investigates a model
thermally cured primer for use in coil coating. In order
to reduce the complexity of the fully formulated model
primer, this initial stage of the programme is concerned
with an unpigmented formulation. Both high-resolution
x-ray photoelectron spectroscopy (XPS) and time-ofÑight secondary ion mass spectrometry (ToF-SIMS)
have been used to investigate the surface chemistry of
this model coil coating.
EXPERIMENTAL
Sample preparation
The unpigmented thermally cured primers under investigation were prepared from commercially available
products to achieve model primers suitable for use in
his work. For reasons of commercial conÐdentiality full
details of these products are not available and for this
reason generic descriptions are employed. The formulations were prepared as described in Table 1, where for-
Figure 1. The chemical structures of the components used in the primer formulations A and B.
( 1998 John Wiley & Sons, Ltd.
Surf. Interface Anal. 26, 444È454 (1998)
S. R. LEADLEY ET AL .
446
Table 2. The elemental concentrations detected by XPS
Sample
HDGS ½ formulation
HDGS ½ formulation
HDGS ½ formulation
HDGS ½ formulation
A
A
B
B
(uncured)
(cured)
(uncured)
(cured)
HDGS ½ pretreatment ½ formulation
HDGS ½ pretreatment ½ formulation
HDGS ½ pretreatment ½ formulation
HDGS ½ pretreatment ½ formulation
A
A
B
B
mulations A and B are essentially the same but with the
addition of an acrylic Ñow control agent at a level of 0.2
wt.% to formulation B. The hexamethoxymethyl melamine component is produced by the reaction between
melamine and formaldehyde in the presence of a base to
yield a substituted methanol
wNH ] H CxO H wNHwCH wOH
2
2
2
Reaction of the hydroxy group with an alcohol yields
the Ðnal reaction product, with water as the condensation product
wNHwCH wOH ] CH OH
2
3
H wNHwCH wOCH ] H O
2
3
2
The product is therefore usually referred to as a
melamineÈformaldehyde resin, and for this reason we
have adopted this description throughout the paper.
The components of Table 1 have the chemical structures shown in Fig. 1. Two substrates were used in this
investigation : hot-dipped galvanized steel (HDGS) in
the as-received condition and HDGS pre-treated with a
(uncured)
(cured)
(uncured)
(cured)
%C
%O
%N
% Cl
75.9
75.3
73.5
73.5
21.1
21.5
24.6
24.4
2.9
3
1.8
1.9
0.1
0.3
0.3
0.2
76.3
77.2
75.2
74.2
20.3
20.2
23.4
24
3.2
2.4
1.4
1.8
0.3
0.2
0.1
–
commercial chromating protocol (Bonder 1303 alkaline
cleaning, followed by Parcolene 62 chromate rinse).
Formulations A and B were applied to the HDGS
substrates using a wire-wound bar coater to deposit a 6
lm wet Ðlm. All the coated samples were then placed in
an oven at 80 ¡C for 30 min so that the non-reactive
solvents were removed. This stage is clearly unrepresentative of a commercial high-speed cure line but was
undertaken in order to produce a coherent, dry, but
unreacted Ðlm. Cure does not occur during this period
because the trans-etheriÐcation reaction between the
alkoxy groups on the melamine resin and the hydroxy
groups in the epoxy resin requires a minimum temperature of 150 ¡C for the initiation. The coatings were
cured by heating from room temperature to a peak
metal temperature of 230 ¡C for 40 s. Prior to surface
analysis, 10 mm diameter discs were punched from the
coated panels.
Surface analysis
All the XPS analysis was performed on a Kratos Axis
Ultra electron spectrometer. The instrument is equipped
Figure 2. The high resolution C 1s spectrum of samples coated with formulation A, uncured.
Surf. Interface Anal. 26, 444È454 (1998)
( 1998 John Wiley & Sons, Ltd.
SEGREGATION PHENOMENA IN COATINGS USING XPS AND ToF-SIMS
Table 3. Eight-peak Ðtting data from the C 1s
spectra of samples coated with formulation A
Peak
C wC/C wH
1
1
(aromatic)
C wC/C wH
2
2
(aliphatic)
C wO
3
(aromatic)
C wO
4
(aliphatic)
C wO
5
(melamine)
C wC
6} z 6
O
NwC wO
7
NwC xN
8
Binding energy
shift
(eV)
FWHM
(eV)
% Area
É0.4
0.85
43.4
–
1.00
18.6
1.1
1.00
9.4
1.5
0.90
11.9
1.7
1.00
2.7
2.1
1.00
9.6
447
magniÐcation of ] 500. This leads to an ion dose well
below the static limit of 1013 ions cm~2 per analysis.
Spectra were acquired using a 20 keV Ga` primary ion
beam with pulsed charge compensation and 4 keV
sample bias. The instrument has been modiÐed by
replacing the original time-to-digital converter (TDC)
and DEC PDP 11/73 computer with a custom-built
TDC unit (Kore Technology, Cambridge, UK) and a
personal computer. Data acquisition and analysis were
performed using Kore Technology software that operates under GRAMS/32 (Galactic Industries Corporation, Salem NH, USA).
RESULTS
X-ray photoelectron spectroscopy analysis
2.8
3.4
1.00
1.00
2.9
1.3
with a unique spherical mirror analyser (a 165 mm
mean radius HSA) an integral automatic charge neutralizer and a magnetic lens. A monochromatic Al Ka
x-ray source was used at a nominal power of 450 W to
record spectra at normal emission. All of the samples
under investigation required charge compensation.
The ToF-SIMS spectra were acquired using a VG
ScientiÐc IXL23 instrument. This instrument is
equipped with a Poschenrieder analyser and a
MIG300PB pulsed liquid metal ion source. The pulsing
conditions were 5 kHz at a pulse width of 20 ns ; the
spectra were acquired whilst rastering at TV rate at a
All of the coated samples contained carbon, oxygen and
nitrogen at the surface. The concentrations of these elements are shown in Table 2. Analysis by XPS also
showed that trace amounts of chlorine were present in
most of the coatings. The most likely source of this
chlorine contamination is epichlorohydrin, which is a
precursor used in the synthesis of epoxy resins and consequently is a common contaminant.
The peak-Ðtting discussed in this paper has been
assessed by the plot of the residuals and visual assessment of the quality of Ðt. In this manner a selfconsistent protocol for peak Ðtting of the C 1s spectra
considered in this work was eventually developed. The
full width at half-maximum (FWHM) of the individual
components of the C 1s spectra were constrained within
the range 0.9È1.0 eV, which enabled all spectra to be
Ðtted satisfactorily in the manner described below. At
Figure 3. The high resolution C 1s spectrum of samples coated with formulation B, uncured.
( 1998 John Wiley & Sons, Ltd.
Surf. Interface Anal. 26, 444È454 (1998)
S. R. LEADLEY ET AL .
448
Table 4. Eleven-peak Ðtting data from the C 1s
spectra of samples coated with formulation
B
Peak
C wC/C wH É 0.4
1
1
(aromatic)
C wC/C wH
2
2
(aliphatic)
C wO
3
(aromatic)
C wO
4
(aliphatic)
C wO
5
(melamine)
C wC
6} z 6
O
NwC wO
7
NwC xN
8
C wCxO
9
C wO
10
OxC wO
11
Binding energy
shift
(eV)
FWHM
(eV)
% Area
É0.4
0.90
38.4
–
0.96
19.4
1.1
0.96
7.7
1.5
0.96
13.5
1.7
0.96
0.8
2.1
0.96
3.7
2.8
3.4
0.6
1.5
4.1
0.96
0.96
0.62
0.96
0.89
0.8
0.4
5.0
5.6
4.8
this level of resolution it is sometimes possible to
observe vibrational e†ects in the C 1s spectra of simple
polymers such as the poly(oleÐnes). In the case of
complex formulations of several polymeric constituents,
as observed in this work, it is unlikely that such e†ect
will be resolvable in the spectra, even when recorded at
such high resolution. For this reason no attempt has
been made to take any of these unknown e†ects into
account in the peak Ðtting of the C 1s spectra. The C 1s
spectrum of uncured formulation A coated on HDGS
with no pretreatment is shown in Fig. 2. This spectrum
has been Ðtted with eight peaks using the following protocol. The epoxy resin used in both primer formulations
has Ðve di†erent carbon environments, as shown in Fig.
1. Therefore, Ðve peaks associated with the epoxy resin
were Ðtted to the C 1s spectra of the coated samples, i.e.
aromatic hydrocarbon (C wC/C wH), aliphatic hydro1
1ether carbon (C wO),
carbon (C wC/C wH), aromatic
2
2
3
aliphatic ether carbon (C wO) and carbon associated
4
with the epoxy end groups. The melamineÈ
formaldehyde resin has three di†erent carbon environments, as shown in Fig. 1. Therefore, three peaks
associated with the melamineÈformaldehyde were also
Ðtted to the C 1s spectra of the coated samples, i.e. the
alkoxy end groups (C wO), the N-substituted formal5
dehyde carbon (NwC wO) and the carbon in the mela7
mine ring structure (NxC wN). The binding energy
8
shifts, FWHMs and the percentage areas of the eight
peaks Ðtted to the C 1s spectrum in Fig. 2 are shown
in Table 3. This eight-peak Ðtting strategy was used
successfully to Ðt the C 1s spectra for all of the samples
coated with formulation A.
The C 1s spectrum of uncured formulation B coated
on HDGS with no pretreatment is shown in Fig. 3. A
comparison with Fig. 2 shows that the C 1s spectra
acquired from samples coated with formulation A were
di†erent from the C 1s spectra acquired from samples
coated with formulation B. Initially, the eight peaks
Ðtted to the C 1s spectra of samples coated with formulation A were Ðtted to the C 1s spectrum in Fig. 3. The
di†erence between the two primer formulations is the
addition of a small quantity of a poly(acrylic) Ñow
agent, therefore three peaks associated with
poly(acrylics) were also Ðtted to the C 1s spectra of
samples coated with formulation B, i.e. the b-shiftinduced carbon environment (C wCxO), an ester
component (C wOwCxO) and9 a carboxyl carbon
10
component (OwC
xO). The binding energy shifts,
FWHMs and the 11
percentage areas of the eight peaks
Ðtted to the C 1s spectrum in Fig. 3 are shown in Table
4. Using the approach to peak Ðtting described above,
where singlets from each component of the formulation
are added to the curve Ðtting envelope, it became clear
that both the C wO and the C wOwCxO com4
ponents had binding
energy shifts10of 1.5 eV and the
same FWHM. Therefore, one peak could be Ðtted to
replace the C wO and C wOwCxO components.
However, the 4C wO and C10 wOwCxO components
4
are representative
of the10 epoxy resin and the
poly(acrylic) Ñow agent, respectively. In order that a
comparison could be made between peaks representative of the three components used in formulation B, the
eleven-peak Ðtting strategy was maintained. This
eleven-peak Ðtting strategy was used successfully to
analyse the C 1s spectra of all the samples coated with
formulation B.
Time-of-Ñight SIMS analysis
The positive ion ToF-SIMS spectrum of the melamineÈ
formaldehyde resin in the mass range m/z 100È400 is
shown in Fig. 4. This spectrum clearly shows a distinct
Figure 4. The positive ion ToF-SIMS spectrum of the melamine–formaldehyde resin in the mass range m /z 100–400.
Surf. Interface Anal. 26, 444È454 (1998)
( 1998 John Wiley & Sons, Ltd.
SEGREGATION PHENOMENA IN COATINGS USING XPS AND ToF-SIMS
449
Figure 5. The proposed positive ion fragmentation pattern of melamine–formaldehyde resin.
series of ions between m/z 163 and 359. It is interesting
to note that all of the ions in this series have an odd
mass number. As melamine contains six nitrogen atoms,
the interpretation of the structures of these ions is
subject to the Nitrogen Rule.5 Therefore, as each of the
ions has an odd mass number, each of the ions is likely
to contain an even number of nitrogen atoms. As each
of these ions has a mass greater than m/z 120, it is reasonable to assume that the ions will contain the ring
( 1998 John Wiley & Sons, Ltd.
structure of the melamine unit. This series of ions are all
separated by intervals of 14, 16 or 30 mass units, which
indicates the stepwise fragmentation of the methylated
formaldehyde side groups. These observations have lead
to the proposed fragmentation process shown in Fig. 5.
Thus the ions diagnostic of the melamineÈformaldehyde
resin in the mass range m/z 100È300 have the structures
proposed in Fig. 6.
The positive ion ToF-SIMS spectra acquired from
Surf. Interface Anal. 26, 444È454 (1998)
S. R. LEADLEY ET AL .
450
Figure 6. The proposed structures of the ions diagnostic of melamine–formaldehyde resin in the mass range m /z 100–300.
the samples coated with formulation A were qualitatively very similar. A positive ion ToF-SIMS spectrum
in the mass range m/z 100È300 representative of the
samples coated with formulation A is shown in Fig. 7.
The majority of the ions present in this spectrum are
diagnostic of the melamineÈformaldehyde resin, i.e. the
ions at m/z 163, 177, 207, 237/239, 253 and 283. Previous ToF-SIMS investigations of epoxy resins have
shown that the diagnostic positive ions in this mass
range are found at m/z 135 and 191,6 having the strucSurf. Interface Anal. 26, 444È454 (1998)
tures shown below. Thus, it is reasonable to assume
that the ions observed at m/z 135 and 191 in the positive ion ToF-SIMS spectra of the samples coated with
formulation A can be attributed to the epoxy resin. It is
interesting to note that the ions diagnostic of the
melamineÈformaldehyde resin are more dominant than
the ions diagnostic of the epoxy resin. Although negative ion ToF-SIMS spectra were acquired from the
sample coated with formulation A, little diagnostic
information was gained from these spectra. Therefore,
( 1998 John Wiley & Sons, Ltd.
SEGREGATION PHENOMENA IN COATINGS USING XPS AND ToF-SIMS
451
Figure 7. The positive ion ToF-SIMS spectrum of the samples coated with formulation A in the mass range m /z 100–300.
for the purpose of this study these negative ion ToFSIMS spectra will not be discussed further.
The positive ion ToF-SIMS spectra acquired from
the samples coated with formulation B were completely
di†erent from the spectrum shown in Fig. 7. These
spectra consisted of a large number of dominant ions,
which were so numerous that accurate identiÐcation
would be extremely difficult. It was also interesting to
note that these spectra did not show any evidence of the
ions proposed to be diagnostic of either the epoxy or
melamineÈformaldehyde resins. The negative ion ToFSIMS spectra acquired from the samples coated with
formulation B were all similar. A negative ion ToFSIMS spectrum in the mass range m/z 100È300 representative of the samples coated with formulation B is
shown in Fig. 8. Also shown in Fig. 8 is the negative ion
Figure 8. The negative ion ToF-SIMS spectrum in the mass range m /z 50–200 of (a) the samples coated with formulation B ; (b) the
poly(acrylic) flow agent.
( 1998 John Wiley & Sons, Ltd.
Surf. Interface Anal. 26, 444È454 (1998)
S. R. LEADLEY ET AL .
452
ToF-SIMS spectrum of the acrylic Ñow agent used in
formulation B, which is virtually identical to the spectra
of samples coated with formulation B. These spectra, as
well as providing a Ðngerprint match between the spectrum of the Ñow control agent and the outer surface,
also yields many of the characteristic fragments
expected from the negative SIMS spectra of
poly(acrylates), as seen in the numerous published
studies of these materials and unpublished work from
this laboratory. The most intense ions in the spectra of
Fig. 8 are at m/z \ 67, 71 and 73, which are assigned to
C H ~, C H COO~ and C H COO~.
5 7
2 3
2 5
DISCUSSION
If, as expected, all of the volatile solvents are removed
by heating the coated samples at 80 ¡C, formulation A
should contain D 87% epoxy resin and 13%
melamineÈformaldehyde resin. Therefore, the theoretical
elemental atomic concentrations of carbon, oxygen and
nitrogen in a homogenous coating of formulation A
should be : C \ 85.0%, O \ 14.6%, N \ 0.4%.
The data in Table 2 show that all of the samples
coated with formulation A had surface nitrogen concentrations greater than would be expected from the theoretical ratio above. Table 5 shows that the peaks Ðtted
to the C 1s spectra of samples coated with formulation
A did not have the areas predicted for a homogenous
coating. In particular, the peaks associated with
melamineÈformaldehyde (C wO, OwC wN, and
NwC xN) had areas greater5than expected.7 It was also
8 that the ions diagnostic of the melamineÈ
observed
formaldehyde resin were the most dominant species in
the positive ion ToF-SIMS spectra acquired from the
samples coated with formulation A. This suggests that
the samples coated with formulation A were surface
enriched with the melamineÈformaldehyde resin. Evidence for the epoxy resin was present in both the XPS
and ToF-SIMS results, which indicates that the
melamineÈformaldehyde resin was not present at the
paint/air interface as a continuous overlayer. This segregation of melamineÈformaldehyde resin to the surface of
similar primer formulations has also been observed previously.7,8
The samples coated with formulation B had surface
nitrogen concentrations lower than samples coated with
formulation A. Nevertheless, the nitrogen concentrations of the samples coated with formulation B were
greater than would be expected for a homogenous
coating. This indicates that the melamineÈformaldehyde
resin is still segregating towards the air interface in formulation B. However, it was also necessary to Ðt additional peaks to the C 1s spectra of samples coated with
formulation B. These peaks have been associated with
the poly(acrylic) Ñow agent that was present in formulation B. The ToF-SIMS investigation showed that there
was no evidence of either the epoxy or melamineÈ
formaldehyde resin at the surfaces of the samples coated
with formulation B. It was also noted that the ToFSIMS spectra of the samples coated with formulation B
were virtually identical to the spectrum of the
poly(acrylic) Ñow control agent.
Both the XPS and ToF-SIMS analyses indicate that
poly(acrylic) Ñow control agent segregates to the
surface. The absence of any evidence for the epoxy or
melamineÈformaldehyde resins in the ToF-SIMS
spectra indicates that the Ñow agent forms a thin overlayer. Due to the di†erences in sampling depth between
XPS and ToF-SIMS, it is proposed that the Ñow agent
overlayer has a thickness of the order of 2 nm. The Ñow
agent is added to the formulation of these primers to
modify the poor wetting properties of both the epoxy
and melamineÈformaldehyde resins. This results in
improved and more rapid wetting of the surface of the
substrate by the paint Ðlm. It must be recalled that coil
coating is a very high speed process, and thus efficient
spreading in a short time window is of paramount
importance. This behaviour is thought to occur as a
result of the tendency of the polymer solution to minimize its surface free energy by the surface segregation of
the component with the lowest surface energy, i.e. Gibbsian segregation. The analytical evidence presented
above indicates that this is indeed the case ; the
poly(acrylic) Ñow control agent segregates to the surface
of the paint Ðlm in order to minimize the surface free
energy of the paint Ðlm.
Both the XPS elemental ratios in Table 2 and the
peak-Ðtting data in Tables 5 and 6 show that no
changes occur as a result of curing. The epoxy/
melamineÈformaldehyde curing mechanism is based on
trans-etheriÐcation, which results in the cured coating
having the cross-linked structure shown in Fig. 9.
Therefore, the cross-linked coating would not have any
chemical environments di†erent from the individual
components of formulation A, which would provide
characteristic XPS spectral features. The only di†erence
that might be expected would be a reduction in the size
of the C wO peak. Evidence for cross-linking is seen in
5
the ToF-SIMS
spectra of Fig. 4 (uncured) and Fig. 7
(cured), which exhibit the reduction in intensity that
would be expected as a result of cross-linking by the
trans-etheriÐcation reaction described in Fig. 9.
Table 5 Percentage areas of peaks Ðtted to the C 1s spectra of samples coated with formulation A
C wC/C wH
1
1
C wC/C wH
2
2
C wO
3
C wO
4
C wO
5
Sample
(aromatic)
(aliphatic)
(aromatic)
(aliphatic)
(melamine)
Theoretical
HDGS (uncured)
HDGS (cured)
HDGS ½ pretreatment (uncured)
HDGS ½ pretreatment (cured)
54.1
43.4
46.7
46.6
45.7
16.2
18.6
17.8
17.5
19.4
10.8
9.4
9.4
9.4
9.2
15.7
11.9
16.5
16.4
16.1
0.5
2.7
2.2
2.2
2.0
Surf. Interface Anal. 26, 444È454 (1998)
C wC
6} z 6
O
NwC wO
7
NwC xN
8
2.0
9.6
4.0
4.4
4.3
0.5
2.9
2.2
2.2
2.0
0.2
1.3
1.0
1.0
0.9
( 1998 John Wiley & Sons, Ltd.
( 1998 John Wiley & Sons, Ltd.
(aliphatic)
(aromatic)
38.4
37.1
36.8
38.9
Sample
HDGS (uncured)
HDGS (cured)
HDGS ½ pretreatment (uncured)
HDGS ½ pretreatment (cured)
19.4
21.3
19.6
18.9
C wC/C wH
2
2
C wC/C wH
1
1
7.7
7.5
7.4
7.8
(aromatic)
C wO
3
13.5
13.1
13.0
13.7
(aliphatic)
C wO
4
0.8
0.8
0.8
1.0
(melamine)
C wO
5
Table 6. Percentage areas of peaks Ðtted to the C 1s spectra of samples coated with formulation B
3.7
3.5
3.5
3.7
C wC
6} z 6
O
0.8
0.8
0.8
1.0
NwC wO
7
0.4
0.4
0.4
0.5
NwC xN
8
5.0
5.4
5.9
5.2
C wCxO
9
5.6
5.3
6.6
4.7
C wO
10
4.8
4.8
5.2
4.6
OxC wO
11
SEGREGATION PHENOMENA IN COATINGS USING XPS AND ToF-SIMS
453
Surf. Interface Anal. 26, 444È454 (1998)
S. R. LEADLEY ET AL .
454
Figure 9. The proposed structure of the model coil coating primer after curing.
It is also interesting to note that the chromate
pretreatment of the HDGS substrate does not a†ect the
surface chemistry of the samples coated with the primer
formulations. This is of particular interest because it
gives an indication as to the driving force behind the
surface segregation phenomena observed. As no di†erences were observed between the two substrates used, it
can be assumed that the surface chemistry of the substrates did not e†ect the chemistry at the air interface,
therefore it is clear that the surface enrichment of the
melamineÈformaldehyde resin is controlled by a process
within the paint Ðlm that involves the minimization of
surface free energy.
CONCLUSIONS
High-resolution XPS and ToF-SIMS have been used to
investigate the surface chemistry of model thermally
cured primers that consist of complex mixtures of
organic components. The high-resolution of the XPS
spectrometer used enabled the identiÐcation of the individual components within the C 1s spectra. The C 1s
spectra were then used to comment on the relative
amounts of each component present at the surface. Both
XPS and ToF-SIMS have shown that the surfaces of
these primers are enriched with the melamineÈ
formaldehyde resin. The addition of a small quantity of
a polyester Ñow agent to the primer formulation caused
a change in the surface chemistry of the coated samples.
It was found that the Ñow agent surface segregates,
which is probably due to surface energy requirements.
The surface chemistry of these thermally cured primers
was not a†ected by curing or surface pretreatment of
the HDGS substrate. Further work is in hand to ascertain whether the segregation seen here also occurs in the
much shorter time schedule of the industrial process.
Acknowledgement
The authors wish to thank the EPSRC for funding this work.
REFERENCES
1. A. Brandau, Introduction to Coatings Technology , Federation
Series on Coatings Technology, Published by Federation of
Societies for Coatings Technology, Philadelphia, PA, USA
(1990).
2. A. M. Taylor, C. H. McLean, M. Charlton and J. F. Watts, Surf .
Interface Anal . 23, 342 (1995).
3. S. R. Leadley, J. F. Watts, A. Rodriguez and C. Lowe, Int J
Adhes . Adhes ., in press.
4. The Interphase Chemistry of Multi -Component Polymer -to Metal Adhesion . EPSRC Research Grant GR/L40090. (1998).
Surf. Interface Anal. 26, 444È454 (1998)
5. F. W. McLafferty and F. Turecek, Interpretation of Mass
Spectra . University Science Books, Mill Valley, CA (1993).
6. The Static SIMS Library . Surface Spectra, Manchester, UK
(1997).
7. T. Hirayama and M. W. Urban, Prog . Org . Coat . 20, 81 (1992).
8. T. Hamada, H. Kanai, T. Koike and M. Fuda, Prog . Org . Coat .
30, 276 (1997).
( 1998 John Wiley & Sons, Ltd.
Документ
Категория
Без категории
Просмотров
11
Размер файла
273 Кб
Теги
resolution, mode, coil, simsn, high, components, tof, minor, phenomena, segregation, coatings, formulation, investigate, xps, use
1/--страниц
Пожаловаться на содержимое документа