Embed Size (px)
Citation preview
Lehigh UniversityLehigh Preserve
Theses and Dissertations
1999
The effect of chromate on adhesion between epoxycoating and Al substratePo-Nien ChenLehigh University
Follow this and additional works at: http://preserve.lehigh.edu/etd
This Thesis is brought to you for free and open access by Lehigh Preserve. It has been accepted for inclusion in Theses and Dissertations by anauthorized administrator of Lehigh Preserve. For more information, please contact [email protected].
Recommended CitationChen, Po-Nien, "The effect of chromate on adhesion between epoxy coating and Al substrate" (1999). Theses and Dissertations. Paper576.
Chen, Po-Nien
The effect ofchromate on .adhesion betweenepoxy coating and·AI substrate
May 31, 1999
The Effect of Chromatei/
on Adhesionbetween Epoxy Coating and Al Substrate
byPo-Nien Chen
A ThesisPresented to the Graduate and Research Committee of
Lehigh Universityin Candidacy for the Degree of Master of Science
In
Polymer Science and Engineering
Lehigh UniversityMay, 1999
Acknowledgments
First, I would like to thank my thesis advisor, Dr. Raymond A. Pearson for his advice,
guidance, kindness, and understanding through this research. His literary support and
valuable advice were very helpful to me.
I would like to express my great appreciation to Dr. Richard D. Granata for his very
helpful discussions, assistance, and support during this research.
I would also like to thank other colleagues at Polymer Interfaces Center, Polymer
Engineering Laboratory, and Zettlemoyer Center for Surface Studies who all have
contributed in some manner, especially Dr. Roy R. Miron, Dr. Mahmoud Moussavi-Madani,
Jason Goodelle, and Jen-Chou Hsiung. Technicians Dave Ackland, Arlan Benscoter, Gene
Kozrma, and Kathy Repa were also of great help and I thank them all.
The financial support provided by AFOSR (Contract No. F49620-96-1-0479)
Multidisciplinary University Research Initiative program is greatly appreciated.
Finally, I would like to acknowledge my parents, Shyan-Hyong Chen and Lo-Yu-Chun
Chen, for their support and encouragement. I also appreciate my sisters for their support
through this research.
-111-
Table of Contents
Certificate of Approval 11
Acknowledgments iii
Table of Contents iv
List of Tables vii
L o t f FO ...IS 0 Igures VI11
Abstract : ,. 1
Ao Introduction 2
1. Background 4
1.1. Chromate conversion coating on Al substrate....... 4
Effect of aging 11
1.2. Epoxy polymers 11
1.3. Polymeric coatings with chromate (pigment) 12
Strontium chromate 12
2. Mechanisms of Adhesion 16
2.1. Mechanical interlocking 16
2.2. Diffusion theory 18
2.3. Electronic theory 18
2.4. Adsorption theory 19
Secondary force interactions 19
The role of primary interfacial bonding 20
3. De-adhesion mechanisms (processes) of polymeric coatings on metals 21
3.1. Lost of adhesion when wet 21
3.2. Cathodic delamination 22
3.3. Swelling of the polymer 22·.n.:-:-:-:·.·.::· --
-lV-
3.4. Gas blistering by corrosion 22
3.5. Osmotic blistering 23
3:6. Thermal cycling 23
3.7. Anodic undermining 23
B. Objective 25
C. Experimental 26
1. Sample Preparation 26
1.1. Compositions & Procedures 26
1). Substrate preparation 26
2). Coating preparation 28
2. Accelerated stress testing 32
2.1. Water immersion test 32
1). Water immersion test at 150 0 F 33
2). Water immersion test at room temperature 33
3. Adhesion Measurements ~ 33
3.1. Tape test 33
3.2. Microhardness indentation test 35
4. Miscellaneous testing 38
4.1. Scanning electron microscopy 38
4.2. Tensile test 38
D. Results and Discussion 41
1. SEM results 41
2. Polymers with or without chromate (SrCr04) 41
3. Comparison between tape test and indentation test 54
4. Effects of substrates 54
E. Conclusions ; ;, ;.; 71
-v-
F. Future Work 72
G. References 73
Vita 79
-Vl-
List of Tables
Table 1. X-ray photoelectron spectroscopy analysis of the chromate conversion coating
during aging 10
Table 2. Concentration of chromate ions provided in water solution by zinc and strontiUm
chromates 13
Table 3. <;omparison of the trends of adhesion curves between tape test and indentation
test in different conditions (yes: they agree well, no: they don't agree, +/-:
between yes and no) 54
Table 4. Results of the tape adhesion tests for the chromated Al panels of different
coating weights 70
-Vll-
List of Figures
Figure 1. Model of the chromate conversion coating formed on 2024 aluminum alloy
based on XPS, SEM and EDX analysis. The external layer is composed of
CrOOR, with significant levels of Fe(CN\ 3 - and smaller amounts ofCr 6+. The
bulk of the coating is made up of Cr20 3 'CrOOR, with F -and Fe(CN)6 3- anions
within the coating. Some of the fluoride may be present as (Cr, AI) OF. The
interface region with the alloy appeared to have Al20 3 as well as increased
levels of Cu. Both F - and Fe(CN)6 3 - (infelTed from Fe XPS signals) were
present at the interface 6
Figure 2. X-ray photoelectron spectroscopy depth profile through the conversion coating:
(a) maj or components; (b) minor components; (c) percentage of oxygen species.
(Specimens immersed for 90 seconds in fresh Alodine® 1200S but no significant•
difference was observed between this profile and that for a coating obtained from
an aged solution) 7
Figure 3. Sputtering profiles measured by Auger electron spectroscopy ofO, Cr, and Al in
a 1900 Athick fresh Alodine® 1200S surface on 2024 Al (immersion time: 1
minute) 8
Figure 4. COlTosion CUlTent i eoIT vs. ageing time in laboratory air. The cOlTosion CUlTent was
determined from the linear extrapolation of the cathodic arm of the polarization
curve to the cOlTosion potential 9
Figure 5. Polarization curves in .0.1 M NaCI for the chromate convyrsion coating as a
-Vlll-
function of ageing time in laboratory air. Samples were conditioned for 20 minutes
at E eorr prior to the experiment. Cathodic and anodic polarization curves were
recorded at a scan rate of 1 mV s -1 at 21 0 C: (A) 1 hour; (B) 44 hours; (C) 165 hours
............................................................................................................................... 10
Figure 6. A typical synthesis of Epoxy resin 13
Figure 7. Passivation of iron by chromate pigments 15
Figure 8. Procedure of the chromate conversion treatment 29
Figure 9.S~tti~_~Jor water immersion test 34- ~)~ v
Figure 10. Classification of adhesion tape test results (ASTM D 3359) 36
Figure 11. Schematic of the three types of debonding on indentation of the coating 37
Figure 12. Some micrographs illustrating debonding by Vickers indentation of epoxy
coating 39
Figure 13. (a) Schematic of indentation damage; (b) schematic of the annular plate model
......................................................................................................................... 39
Figure 14. SEM image of 1000 substrate (Magnification 20000 X, 20 KV) 42
Figure 15. SEM image ofLCCC substrate (Magnification 20000 X, 20 KV) : 43
Figure 16. SEM image ofLCCC substrate (Magnification 60000 X, 20 KV) 44
Figure 17. SEM image of MCCC substrate (Magnification 20000 X, 20 KV) 45
Figure 18. SEM image ofMCCC substrate (Magnification 60000 X, 20 KV) 46
Figure 19. SEM image ofHCCC substrate (Magnification 20000 X, 20 KV) 47
Figure 20. SEM image ofHCCC substrate (Magnification 60000 X, 20 KV) 48
Figure 21. Dry adhesion (Ge) of coating 828 and 828 (~rCr04) 'on different substrates.. 49
-IX-
Figure 22. Comparison of interfacial fracture energy (Gc) between coating 828 and coating
828 (SrCr04) on Sanchem® CC 1000 cleaned substrates versus exposure time
from indentation test (room temperature) : : 50
Figure 23. Comparison ofretained percentage between coating 828 and coating 828 (SrCr04)
on LCCC cleaned substrates versus exposure time from tape test (room temper-
ature) 51
Figure 24. Comparison ofretained percentage between coating 828 and coating 828 (SrCr04)
on MCCC substrates versus exposure time from tape test (150 0 F) 52
Figure 25. Comparison of interfacial fracture energy (GJ between coating 828 and coating
828 (SrCr04) on HCCC substrates versus exposure time from indentation test
(150 0 F) 53
Figure 26. Retained percentage compared to indentation test records for coating 828 on 1000
substrates versus exposure time (room temperature) 55
Figure 27. Retained percentage compared to indentation test records for coating 828 on
LCCC substrates versus exposure time (150 0 F) 56
Figure 28. Interfacial fracture energy (Gc) of coating 828 on different substrates versus
exposure time from indentation test at ro_om temperature 57
Figure 29. Interfacial fracture energy (Gc) of coating 828 on different substrates versus
exposure time from indentation test at room temperature (Fig. 28, expanded
scale) 58
Figure 30. Retained percentage of coating 828 on different substrates versus exposure time.
from tape test at roQm temperature 59
-x-
Figure 31. Retained percentage of coating 828 on different substrates versus exposure time
from tape test at room temperature (Fig. 30, expanded scale) 60
Figure 32. Interfacial fracture energy (Gc) of coating 828 (SrCr04) on different substrates
versus exposure time from indentation test at room temperature 61
Figure 33. Retained percentage of coating 828 (SrCr04) on different substrates versus
exposure time from tape test at room temperature 62
Figure 34. Interfacial fracture energy (Gc) of coating 828 on different substrates versus
exposure time from indentation test at 150 0 F 63
Figure 35. Retained percentage of coating 828 on different substrates versus exposure time
from tape test at 150 0 F 65
Figure 36. Retained percentage ofcoating 828 on different substrates versus exposure time
from tape test at 150 0 F (Fig. 35, expanded scale) 66
Figure 37. Interfacial fracture energy (Gc) of coating 828 (SrCr04) on different substrates
versus exposure time from indentation test at 150 0 F 67
Figure 38. Retained percentage of coating 828 (SrCr04) on different substrates versus
exposure time from tape test at 150 0 F 68
Figure 39. Retained percentage of coating 828 (SrCr04) on different substrates versusIv
exposure time from tape test at 150 0 F (Fig. 38, expanded scale) 69
-XI-
ABSTRACT
Adhesion tests such as the microindentation test and the tape test were evaluated for
measuring the adhesion properties of polymer coatings on chromate converted 2024 T3 Al
substrates, which are used for protecting aircraft and aerospace components.
Three different types ofchromate conversion coatings with lower, medium (normal), and
high chromate concentrations on aluminum substrates were studied. These substrates were
also coated with epoxies to further passivate the surfaces. Some epoxy coatings contained
strontium chromate. After subjecting the substrates to water immersion tests at room
temperature or higher temperature, all coatings with strontium chromate have much better
adhesion properties when compared to those without. In most cases, the chromate conversion
coatings with medium chromate weight were better than those with low or high chromate
weight and those without chromate pretreatment.
It was also observed that the adhesion increased after a certain period ofexposure in most
cases. Also, the roughness or surface morphology was different in each kind of substrate:
There was no direct relationship found in the adhesion data. It indicated that mechanical
interlocking was not the main factor affecting the adhesion strength in our experiments.
-1-
A. Introduction:
A conversion coating is one which results when the physical, chemical nature or
properties ofa metal surface are altered. When the surface is reacted with a suitable chemical
solution, a partial or complete chemical film occurs at the metal-solution interface. The thin,
gel-type film which dries to amorphous coating includes a portion of the base metal as one
ofthe components ofthe film. The surface ofthe metal is thereby converted from its initially
rather active condition to an inert film which is commonly known as a conversion coat-
ing 1,2. On aluminum, chromate conversion coatings usually have excellent paint bonding
qualities and also often exhibit good, unpainted corrosion resistance.
The simplest type of bath used to produce chromate conversion coatings contains
hexavalent chromium and a source of fluoride. It is also acidic, either by the addition of the
ingredients themselves or by the addition ofa suitable mineral acid. The coating appearance
can be varied, according to the weight obtained, from a light iridescence to a golden tan, with
coating weights offrom 5to 80 mg / ft 2 being readily obtainable. Although their morphology
and chemical composition depend on the application techniques, these layers often consist
of aluminum hydroxide-oxide containing Cr 3+ and Cr 6+ species. The Cr species are largely
located in the outer layers of the film 3.
For over 40 years chromated pigments and chromate conversion coatings have been
widely accepted as the key ingredient to corrosion protection for aluminum alloys. It is still
unknown the exact process or mechanism that takes place when Al surfaces are treated with
chromates but it is known that there is passivation of the galvanic cells and enhanced
adhesion (in the case of chromate conversion coatings and anodic deposition treatments)"'~-.'.""""""_~'l""\~~~I">':"f"1...";'~.~.~ . ~ .
<>.
-2-
between the metal substrate and primer system. The dilemma is that chromates are
considered hazardous materials and pose problems for paint and disposal waste products. For
these reasons, federal regulations are mandating the reduction of the use of chromates in
coating systems. Therefore, much work has been performed to' examine the effects of
chromates on the Al substrates or within the polymeric top coatings to determine the
mechanism for passivation. Once the passivation mechanism is understood then promising
alternatives may be found.
The adhesion between the coating and its substrate depends critically on the mechanical
properties, chemical structure, and physical characteristics ofa coating. The adhesion is also
dependent on the chemistry and physics of the interface, and on stresses in the film and
substrate 4. Poor adhesion of polymeric coating at an interface will lead to high
concentrations of mobile water molecules and as a result will cause a loss in its passivation
functi~n, promoting corrosion and coatings failures. To avoid moisture and ionic
contamination-induced failures of adhesion on aerospace equipment, it is required to have
good adhesion of polymer to an underlying surface, preventing the formation of water
films at the interface.
Reliable methods of measuring the adhesive bond strength between passivation layers
and the substrates are needed to determine the performance ofpolymeric coatings. Adhesion
bond strength involves either cohesive failure which is dependent on film properties or
interfacial failure which is the failure at the interface between the film and substrate. A great
deal of work has been put into the development of reliable techniques for determining the
~ . .. n""~~"",~~.~,,,, ...m.• ~gh~§j~,Jl!mij.g~h,,Qfgoatings by manyresearchersj;i;~,s1iohJ1OO~~peel test, blister
-3-
test, scratch test, lap-shear test, and microindentation test.
1. Background:
1.1. Chromate conversion coating on Al substrate:
Not only are there a wide variety of commercial chromate formulations in use, but the
conversion coatings formed also differ in composition and experts disagree about the detailed
chemistry. There is, however, general agreement that in amorphous chromate treatments the
following reactions are involved 7 :
(1) Acid attack upon the metal
2AI +6H+ + 2 Al 3+ + 3 H2
(2) Reduction of hexavalent chromium to trivalent chromium
3H2 +2Cr2 0 72
- + 2Cr(OH)3 +2Cr042
-
(3) Formation of hydrated aluminum oxide
2 Al 3+ + 4 H2° -~ A12 0 3 ' H2°+ 6 H+
(4) Formation of aluminum chromate
2 Al 3+ + 3 Cr 0 42- -~ Al2 (Cr 04)3
(5) Formation of complex chromic chromates
2 Cr (OH)3 + Cr 0 42- + Cr (OH)3' Cr (OH) Cr 0 4 + 2 OH-
For many years, the amorphous nature ofthis protective conversion coating prevented its
characterization by traditional methods of analysis, such as x-ray diffraction. In the last 20
years, however, surface sensitive techniques such as x-ray photoelectron spectroscopy
-~(XPSr8"H~=Alfger electron spectroscopy (AES) 11, 13, Fourier transform infrared (FTIR) .
-4-
spectroscopy 14, secondary ion mass spectrometry (SIMS) 15, and x-ray absorption
spectroscopy (XAS) 14,16, including extended x-ray absorption fine structure (EXAFS) and
x-ray absorption near-edge structure (XANES), have been used to study chromate conversion
coatings. The information provided by these techniques, in combination with electron
microscopic studies such as those by Brown et al. 17-
19, have provided a greater under
standing of the coating process, particularly on pure aluminum.
As noted by Hagans and Haas 13, very little attention has been given to the deposition of
chromate conversion coatings under typical industrial conditions, i.e. using Al alloys as
opposed to pure aluminum, and using chemical pretreatment. Recently, several studies have
been focused on the chromate conversion-coated Al 2024-T3 alloy. Some of the coatings
have been processed in a fashion similar to our own procedure 14, zo. The structure is
schematically illustrated in Fig. 1 and the analysis of the depth profiles are shown in Figs.
2&3.
The depth profile can be divided into three regions: the top surface (region 1), the bulk
of the conversion coating (region 2) and the interface region with the alloy (region 3). In
region 1, it is generally the Cr 3+ oxides such as CrZ0 3or CrOOH (Cr(OH)3) on this layer.
There were only slight changes in the composition ofthe conversion coating throughout most
of region 2. The relative intensity of Cr and 0 remained approximately constant up to the
interface with region 3, and is mainly composed of chromium with oxygen with a small
amount of Fe cyanide, and Al usually was not present. Other elements Cu, Mg and C
occurred at detectable levels thrQughout the coating. In region 3, the concentration of F, Cr,
and 0 declined sharply·where~aU(»)k"eempill1613..tB,..auch as Cu, AI, and-Mg displayed_an__nn
.",;..",.. ~
-5-
O "d Fa(CN)3- Cr6+Hydrated Chromium XI a,~ -6
-----------~ ..............'~-----~------------------- -.... ~~------
-
, (Cr, AI)OF and AI203
~~u ~~7JL@r-rrr-r-t_-r---........-l2024 Aluminium Alloy
Fig. 1. Model ofthe chromate conversion coating formed on 2024 aluminum alloy based onXPS, SEM and EDX analysis. The external layer is composed of CrOOH, with significantlevels of Fe (CN)6 3
- and smaller amounts of Cr 6+. The bulk of the coating is made up ofCr20 3 ·CrOOH, with F-and Fe(CN)6 3
- anions within the coating. Some of the fluoride maybe present as (Cr, AI) OF. The interface region with the alloy appeared to have Al20 3 as wellas increased levels of Cu. Both F -and Fe(CN)6 3 - (inferred from Fe XPS signals) werepresent at the interface 20.
-6-
Chromate Depth Profile
60
Cu
~-JCO
40
~++
~YdrOXYIS
andAluminium
+ Oxide
20
: Oxide
II
I ~
I /1: /+
+",,1 ....---+'+-+
Adsorbed H20
o
20
I(a) I I60 I I Al
1 ~:-_lC-""'" 2 III x-x
x I '" ,I
40 ~ i x
+ I
3.0 I?fi I
.Sd \ IE2.0 -\ I.8 +"" Fe-( +-+~
I +-+1.0 I
1
Sputtering Time (mins)
Fig. 2. X-ray photoelectron spectroscopy depth profile through the conversion coating: (a)major components; (b) minor components; (c) percentage of oxygen species. (Specimensimmersed for 90 seconds in fresh Alodine ® 1200S but no significant difference wasobserved between this profile and that for a coating obtained from an aged solution) 20.
-7-
oDepth (A)
o 400 800 1200 1600 2000 2400 2800
ESCA
6
oElectro 1900 Ayield t :
""'----t:-~~~~-__""__ISurface layer Interface re ion O-H
5
•••••••••••••••• eO •••••• ,. •••••• .0 •• :- •• 0' ••••• o •••••••• ~ ••••••••••••••••••• : •••••••
, "
6
......................: : .: _.. ~ .. . . .: : .•~..-.........,: : Al
4
Cr
2
o
o
"0 ..",4
~dUC.u 3o~..d-u
0:::
>....0;;jcu
.5
Sputter time (min)
Fig. 3. Sputtering profiles measured by Auger electron spectroscopy of 0, Cr, and Al in a1900 Athick fresh Alodine® 1200S surface on 2024 Al (immersion time: 1 minute) 1-1,
-8-
200150100
Time (hrs)
50O'----=----L..----L....----~---~
a
1
4..---------------------..,
3
,N
~::t
t:: 20(.).-
Fig. 4. Corrosion current i carr vs. ageing time in laboratory air. The c.orrosion current \vasdetermined from the linear extrapolation of the cathodic arm of the polarization curve tothe corrosion potential 20•.
-9-
-150r-------------------,
10" 10° 10'
I(~A/cri12 )10-2
-900 '---J'--'-L...L.L.LLJu...-..J-l-I..J..L.I.Ll..I.--l.....J..J...l.J..JJ..I;L..-..-I--'-L~~.....J..J...u..LU."___l.__L..L..~
10-3
-750
-600
-450
-300
->E-w
Fig. 5. Polarization curves in 0.1 M NaCI for the chromate conversion coating as a functionofageing time in laboratory air. Samples were conditioned for 20 minutes at Ecorr prior to theexperiment. Cathodic and anodic polarization curves were recorded at a scan rate of 1mVs -I at 21°C: (A) 1 hour; (B) 44 hours; (C) 165 hours 20.
Cr 0 Fe C N Al F eu
Fresh 8.7 35.2 1.8 35.1 12.8 0.9 5.4 0.024h 8.2 37.6 1.0 36.3 11.7 0 4.0 1.2312 h 6.6 31.9 1.2 39.7 10.1 6.5 2.2 1.6
Table 1. X-ray photoelectron spectroscopy analysis of the chromate conyersion coatingduring aging 20.
,
-10-
Increase.
Effects ofAging
The effects of aging on structure and properties of chromate conversion coatings is the
likely reason why so many discrepancies existed in the literature. For example, Hughes et
al. 20 analyzed the chromate conversion-coated Al 2024-T3 alloy, and found that during
ageing experiments, potentiodynamic measurements indicated that the corrosion current
(i corr) decreased from ~ 0.4 to ~ 0.04 IlA cm-2 during the first 40 hours after preparation but
thereafter slowly increased (Fig. 4). After 165 hours of aging in air (Curve C, in Fig. 5), the
chromate coating had deteriorated to the extent that its performance was approaching that
of the uncoated sample. These experiments indicated that the corrosion protection afforded
by the conversion coating decreased with aging time in air. No significant changes were
observed in the chemistry ofthe coating by XPS after 24 hours ofaging but more significant
changes were observed for 312 hours of aging, which included decreases in the atomic
percentages ofCr, 0, N, and F and an increase in Cu (Table 1). And as the coating aged, a
network of micro-cracks developed across the surface.
1.2. Epoxy Polymers:
Epoxies are one class of polymers that are widely used in the coating industry. Most
epoxies exhibit shrinkage during cure, good adhesion to most surfaces, excellent
environmental stability, and are easy to process. The excellent adhesion characteristics of
epoxies are due to their highly polar and active surface nature which can provide both
chemical and mechanical bonds. Fillers such as carbon black and silica are often added to
epoxies to reduce curing shrinkage, increase the strel1g!hs~Hh~~1?~"y, and lower th~ tll,~ItneL...,,=~..
-11-
expansivity. Epoxies are generally stable up to 150°C, above this temperature electrical and
mechanical properties can degrade. A drawback ofepoxy is its hydrolytic instability. Water
can act as plasticizer 21, reducing the glass transition temperature 22, changing mechanical
properties 23, and changing the dielectric properties of epoxies 24-26, Water can also partially
hydrolyze epoxy networks 27, causing dimensional changes and increased stresses 28.
The synthesis of epoxies involves a step polymerization reaction of 2, 2' - bis (4 -
hydroxyphenyl) propane of bisphenol A and epichorohydrin to form epoxy prepolymer,
called the diglycidyl ether of bisphenol A (DGEBA). The low molecular weight liquid
prepolymer is then cured, or cross-linked, at room temperature or at higher temperature
about 50 - 130 ° C by the addition ofdianhy-dride or polyamines 29. The crosslinking occurs
either through the epoxy end groups or the hydroxyl groups. Figure 6 shows the typical
synthesis of epoxy resin.
1.3. Polymeric coatings with chromate (pigment):
According to accepted standards (Table 1, Ref. 30), the word pigment means a substance
consisting of small particles that is practically insoluble in the applied medium and is used
for its coloring, protective, or magnetic properties 30. Common chromate pigments include
the yellow and orange lead chromate and chromate-molybdate derivatives, the less toxic
yellow barium chromate, and the corrosion-inhibiting pigments, zinc chromate and strontium
chromate. Our interest is in strontium chromate which is discussed in detail below.
Strontium Chromate
Strontium chromate pigment was first used commercially (toward the end of tht2
nineteenth century) as a coloring matter in artists' paints under the name citron yellow. Such
-12-
Fig. 6. Typical synthesis of a bisphenol A resin 29.
Pigment
Zinc YellowStrontium chromateBasic zinc chromate
Lead Chromate (reference)
CrOJ / H,O (g/liter)
1.10
0.500.02
0.00005
Table 2. Concentration of chromate ions provided in water solution by zinc and strontiumchromates 32.
-13-
use in artists' paints persisted until 1936 when it was replaced by organic pigments. At about
this time some strontium chromate was also being used for corrosion resistance on alumi
num, magnesium, and their alloys in accordance with a British patent 31. Emmet Lalor 32,
described that the major reason for the use of chromate pigments is because of its ability to
release chromate ions gradually (due to its slight water solubility, Table 2). Later it was used .
in chemical resistant coatings because of its low reactivity in epoxy-polyamide vehicle (in
paints).
The mechanism of passivation of iron-rich surfaces is well understood. As moisture
diffuses through the paint coat, Cr03 is extracted from the strontium chromate and chromate
ions are produced. These chromate ions oxidize the ferrous ions that were released from the
metal to produce a protective film ofhydrated ferric oxide. Strontium chromate is about half
as soluble as zinc yellow and thus offers a slightly lower level of effective inhibition for a
longer period. Therefore strontium chromate is better for those primer coatings requiring a
steady release of chromate ions over a long period.
In summary, the anticorrosive action of chromate pigments is based both on chemical
and electrochemical reactions 33. 34. Electrochemical passivation and film formation reac
tions are illustrated in Fig. 7. Passivation is based on electrochemical processes in the
cathodic region. A protective film is formed by reaction of chromate ions with metal ions at
the surface of the substrate to form metal oxide hydrates and seal the defects 35.
The anticorrosive properties of this class of pigments depend on 30:
1) The content of water-soluble chromate ions.
2) The ratio of water-soluble chromate ions to water-soluble corrosion-promoting ions
-14-
---- CathodeAnode
--- -........ ,,~
Fe - Fe 2+ + 2e- Fe(QH)2
II
'--e
- ---- - ... .... ...
0, ~O(r 7
-0" ~O' 3e..: (r':ll+_ Chromate -- Cr6+ --- ;1
0, ~OCr0.... ~O
'-"-"-'r,-<"
Fe{OH)2' 2CrOOH _-----J
Fig. 7. Passivation of iron by chromate pigments 30.
-15-
(chloride and sulfate ions).
3) The active pigment surface in the coating (i.e., particle size distribution and dispers
ibility).
2. Mechanisms of Adhesion 36:
The mechanisms of adhesion are still not fully understood and test methods commonly
employed to measure the strengths of adhesive joints are not well suited to theoretical
analysis. They introduce geometrical factors and loading factors which are difficult to
analyze and the measured joint strength includes indeterminate contributions from
rheological energy losses in the adhesive and substrate. Thus, although the intrinsic adhesion
forces acting across the adhesive/substrate interface may affectjoint strength they are usually
completely obscured by other contributions, and information concerning the magnitude of
such forces may only be indirectly obtained. This inability to measure the interfacial
interactions has been the main obstacle to the development of a comprehensive theory of
adhesion. The four main mechanisms of adhesion which have been proposed are: (l)
mechanical interlocking, (2) diffusion theory, (3) electronic theory, (4) adsorption theory.
2.1; Mechanical interlocking:
This theory proposes that mechanical keying, or interlocking, of the adhesive into the
irregularities of the substrate surface is the major source of intrinsic adhesion. Borroff and
Wake 37 convincingly demonstrated that the most important single feature in the adhesion of
a simple rubber to an uncoated fabric is the penetration of the protruding fiber ends of the
spun yarn into the rubber. The degree ofpenetration necessary is such that, upon the rubber-
.-_.- ."-_.'-""r.0"""--'~"'-'~".'!'n.~._....... _ ..._
-16-
textile structure being stressed, the length of fiber embedded is sufficient for the total
shearing force on the fiber/rubber interface to exceed the breaking strength of the fiber.
Another, but 'somewhat more contentious example is the metal plating of polymers
where a chemical pretreatment ofthe polymeric substrate is employed prior to plating. Some
workers 38. 39 have argued that the adhesion of metal plating to polymeric substrates is a
function of the surface topography. Others 40.41 have emphasized the role that increased
oxidation of the polymer surface, commonly induced by the pretreatments employed prior
to plating, increases the number ofsurface force interactions. A balanced view emerges from
the quantitative experiments of Perrins and Pettett 42, who separated the contributions form
mechanical interlocking and surface forces of electroplated copper to polypropylene.
Following the theories of Andrews and Kinloch 43 and Gentand Schultz44, Wake 45 has
suggested that the effects of mechanical interlocking and surface free components could be
multiplied to give a result for the measured joint strength:
Joint strength = (constant) x (mechanical component) x (surface free components)
This equation reveals that the substrate should possess, simultaneously, the topography
and surface chemistry necessary to produce the highest extent ofmechanical interlocking and
surface force contribution. Thus, this interaction results in the highest joint strengths.
Works by Packham and co-workers 46 provides further evidence for the importance of
substrate surface topography when considering the strength of certain interfaces. In their
studies on the adhesion of polyethylene to metallic substrates they found that high peel
strengths were obtained when a very rough, fibrous-type, oxide surface was formed on the. ~ )
-17-
substrate.
2.2. Diffusion theory:
Voyutskii 47 advocated the diffusion theory of adhesion which states that the intrinsic
adhesion of high polymers to themselves (autohesion), and to each other, is due to mutual
diffusion of polymer molecules across the interface. This requires that the macromolecules
or chain segments of the polymers (adhesive and substrate) possess sufficient mobility and
are mutually soluble.
An area where interdiffusion is ofcardinal importance is the solvent-welding ofplastics.
This is a technique in which the adhesion of plastic components is promoted by the
temporary presence ofa solvent in the absence ofan extraneous adhesive. A requirement of
the sol,vent is that it strongly plasticises the surface of the polymers, which results in a large
increase in free volume and hence in the chain mobility of the polymer in the interfacial
region, increasing the rate arid extent of inter diffusion of the polymer chains.
2.3. Electronic theory:
If the adhesive and substrate have different electronic band structures there is likely to
be some electron transfer on contact to balance Fermi levels which will result in the
formation of a double layer of electrical charge at the interface. The electronic theory of
adhesion is due primarily to Deryaguin and coworkers 48 and they have suggested that the
electrostatic forces arising from such contact or junction potentials may contribute
significantly to the intrinsic adhesion.
Deryaguin's theory essentially treats the adhesive/substrate system as a capacitor'which
is charged due to the contact of the two different materials. Separation of the parts of the
-18-
capacitor, as during interface rupture, leads to a separation of charge and to a potential
difference which increases until a discharge occurs. Adhesion is presumed to be due to the
existence of these attractive forces across the electrical double layer.
2.4. Adsorption theory
The adsorption theory of adhesion is the most generally accepted theory and has been
discussed in depth. This theory proposes that, provided sufficiently intimate intermolecular
contact is achieved at the interface, the materials will adhere because of the surface forces
acting between the atoms in the two surfaces; the most common such forces are van der
Waals forces and are referred to as secondary bonds. In addition, chemisorption may well
occur and thus ionic, covalent and metallic bonds may operate across the interface; these
types of bonds are referred to as primary bonds.
Secondary Force Interactions
The thermodynamic work of adhesion required to separate a unit area of two phases
forming an interface, WA , may be related to the surface free energies by the Dupre' equation.
In the absence of chemisorption and interdiffusion the reversible work of adhesion, WA , in
an inert medium may be expressed by:
where Y a corresponds to the specific surface free energy of sub.strate A, where Y b
corresponds to the specific surface free energy of substrate B, and Y ab corresponds to the
interfacial specific surface free energy. This equation only applies strictly to a solid/liquid
-19-
By adopting a continuum fracture mechanics approach, the works of Andrews and
Kinloch 49 and Gent and Kinloch 50 defined a geometry independent measure of joint
strength, the adhesive failure energy P. Model joints were prepared between a crosslinked
amorphous (a-styrene-butadiene-rubber) and rigid polymeric substrates and the adhesive
failure energy P, was determined for a wide range of temperatures and rate of crack
propagation. For any given rubber-substrate joint the results yielded a single master curve
when normalized to a reference temperature by means of the Williams-Landel-Ferry rate
temperature-equivalence for simple viscoelastic materials.
Further experimental and theoretical considerations demonstrated that the adhesive
failure energy, P, for a crosslinked rubbery adhesive/rigid plastic substrate interface could
be divided into two major components:
(a) The energy required to propagate a crack through unit area of interface in the absence
of viscoelastic energy losses, i.e., an intrinsic adhesive failure energy, Po.
(b) The energy dissipated viscoelastically within the rubbery adhesive at the propagating
crack, again referred to unit area of interface.
The Role ofPrimary Interfacial Bonding
Although it is evident that intrinsic adhesion arising from secondary bonding forces alone
may result in adequate and high joint strengths, many adhesion scientists believe that the
presence of additional primary bonds may often increase the measured joint strength and is
certainly a necessary requirement for securing environmentally stable interfaces. The use of
sophisticated, surface-specific, analytical techniques such as laser Raman spectroscopy, .
X-ray photoelectron spectroscopy, and. secondary-ion mass spectroscopy have produced
-20-
definitive evidence that primary interfacial bonding may occur in certain circumstances and
may make a significant, indeed often vital, contribution to the intrinsic adhesion.
Until recently there has been little known of the mechanism by which the intrinsic
adhesion of the interface is enhanced, especially with respect to environmental attack. The
generally, but not universally, accepted mechanism by which the durability of the joint is
increased is the formation of strong, primary interfacial bonds. For example, the
polysiloxane/substrate interface, this arises from the formation ofSi-O-substrate bonds (the
general structure of silane is X3Si(CHz)nY, where n = 0 and 3, X is a hydrolyzable group on
silicon and Y is an organo-functional group selected for compatibility with 'it given resin).
For the adhesive/silane interface this arises from the reaction of the Y-group on the silane
with reactive groups in the adhesive. Alternatively, other features such as wetting and the
possible formation of a boundary layer in the adhesive differing in chemic~l and physical
properties to that of the bulk resin have been suggested to be of importance.
3. De-adhesion Mechanisms (Processes) of Polymeric Coatings on Metals 51:
The role of the paint is to serve primarily as a partial barrier to environmental
constituents such as water, oxygen, sulfur dioxide, and ions and secondarily as a reservoir
for corrosion inhibitors. Some formulations contain very high concentrations ofmetallic zinc
or metallic aluminum such that the coating provides galvanic protection as well as barrier
protection.
3.1. Loss of Adhesion When Wet:
Many coatings, particularly those applied to a roughened surface, show excellent tensile--~-. ·.-·ce.:.·..:.- .... ,::;:'~,.__ ._""'-__... __.. __-" ....._
-21-
adhesion to metal but lose this adhesion after exposure to pure water at room or elevated
temperatures. A thin film of water at the interface is apparently responsible for the loss of
adhesion. Ifthe coating is allowed to dry without destructively testing the adhesion, the dried
coating often exhibits the original tensile adhesion. The phenomenon is reversible: the
adhesion is poor when the coating is wet and is satisfactory when it is dry.
3.2. Cathodic Delamination:
Most organic coatings on most metal surfaces lose their adhesion when alkali is
generated at a defect in the coating or at weak spots in the coating. Alkali can be generated
by the cathodic half of the corrosion reaction or by driving the cathodi~ reaction by means
of an applied potential.
The alkali is generated by the cathodic reaction,
H2°+ 1/2 O2 + 2 e - = 2 OH-
which occurs at a defect in the coating or through an electrolytic pathway at weak spots in
the coating.
3.3. Swelling of the Polymer:
Some polymer formulations have the property of swelling, i.e., increase in dimensions
when exposed to certain environments. An example of this effect is the swelling of some
epoxy coatings when exposed to strong sulfuric acid solutions at elevated temperatures.
3.4. Gas Blistering by Corrosion:
This phenomenon has been observed in a velY few cases. The effect was attributed to
gas blistering rather than swelling of the polymer because the blister contained a large
quantity of hydrogen as judged by extraction of the gas in the blister with a hypodermic
-22-
needle followed by gas chromatographic analysis. The blistering must occur as a
consequence of rapid penetration ofthe coating by hydrogen ions and slow diffusion ofthe
hy-tfogen gas out through the coating. The blistering requires that the coating possess a
degree ofductility since a brittle coating would be expected to fracture rather than to deform.
3.5. Osmotic Blistering:
Osmotic pressures are very powerful and are a driving force for blistering. They are
especially destructive under conditions where a soluble salt impurity is present beneath the
coating and the coated metal is exposed to water with a low ionic content. The driving force
is the attempt by the system to establish two liquids, one under the coating and the other
external to the coating, with the same thermodynamic activity. The direction of water flow
through the coating is inwards since dilution ofthe concentrated solution at the interface is
the mechanism by which the two liquids strive for equal thermodynamic activity.
3.6. Thermal Cycling:
Coatings that are brittle and have different coefficients of expansion than the substrate
metal are very susceptible to disbonding upon thermal cycling. This disbonding may occur
locally in small areas or it may occur in the most drastic cases over very large areas. It is
believed that one of the functions of the rough surface generated by abrasive blasting is to
provide many anchor points that reduce the likelihood oflarge area disbonding upon thermal
cycling.
3.7. Anodic Undermining:
Anodic undermining represents one class of con-osion reactions underneath an organic
coating in which the major separation process is the anodiccOlTosion reaction under the
-23-
coating. An example is the very slight delamination that occurs when a thin copper layer is
overcoated with an organic coating such as a photoresist and the system is made anodic. The
rate of disbonding is a function of the applied potential and hence the rate of dissolution of
the copper beneath the coating. Anodic delamination occurs very slowly relative to cathodic
delamination at equal potential differences from the corrosion potential.
-24-
B. Objective:
It has been realized for years that proper chemical pretreatment of the aluminum prior to
bonding is essential for developing bond strengths necessary for high performance aircraft
applications. The amount of chromate on Al substrate has been shown to be an important
parameter that influences the corrosion inhibition and adhesion properties of polymeric
coatings on Ai. However, quantitative correlation between the amount of chromate and
the surface morphology with the adhesion performance were not illustrated clearly.
The objective of this research is to increase our understanding between chromate levels
and adhesion strength. Two types of adhesion tests, tape test and indentation test, were
chosen to characterize the adhesion properties ofcommercial polyamide and epoxy (with or
without chromate blended) on chromate conversion pretreated aluminum substrates and on
aluminum substrates cleaned with only general detergent and deoxidizer. By controlling the
experimental factors (immersion time) , three different kinds ofconverted Al substrates with
different ranges of chromate weights were evaluated by water inunersion tests at room
temperature or elevated temperatures. The surface morphology were also examined by
electron microscopy.
,..-25-
C. Experimental:
1. Sample Preparation:
1.1. Compositions & Procedures:
1). Substrate Preparation
Panels of size 3" x 3" were cut off from 2024-T3 sheet from the same lot number were
mounted in immersion racks supplied by the National Rack Co., Paterson, NJ. These racks
are covered with a plastisol to protect the steel rack. Steel springs are used for the rack
fingers - three fingers per panel. Each rack was capable of holding eight panels.
Tanks manufactured from 316 stainless steel and equipped with L-shaped immersion
heaters, controlled by means of a teflon coated hydrostatic immersion probe, contained in
a stainless steel well, supplied by the Technic Corp., Pawtucket, RI, along with a variable
speed motor and 14" shaft coated with teflon, sfabilized by means of a stainless steel guide
were used on the tanks containing Sanchem® CC 500, Sanchem® CC 1000 and Alodine®
1200S. The Sanchem®tanks each contained approximately 2.5 gallons ofsolution prepared
according to the manufacturer's recommendation. The Alodine® 1200S contained 3 gallons
ofsolution also prepared according to the manufacturer's recommendation. Next to each of
the stainless steel tanks were 3 gallon capacity polypropylene tanks each of which was
equipped with an overflow weir and fed by 1/4" diameter tube connected to the city water
supply. PVC 40 gauge rigid tubing was used for the overflow connections, which drained
into the waste system of the university.
The first tank which contained Sanch-em® CC 500, a non-silicated, caustic-free mild
alkaline cleaner for aluminum, was heated to 65.5 DC '± ~ DC withstining. Th~ plastisol
-26-
coated racks, holding 8 pieces of2024 T3 aluminum panels were immersed in this solution
for 3 minutes, removed, allowed to drain briefly, and then placed in the rinse tank for 5
minutes. The rate ofwater flow was sufficient to maintain good circulation and remove carry
over cleaning solution. After 5 minutes, the rack was removed, allowed to drain for a shoD:
period oftime, approximately 10 seconds, and then placed in the tank containing Sanchem®
CC 1000, a chrome-free surface conditioning deoxidizer and smut removal agent. The
temperature ofthis tank was maintained at 43.9 DC ± 2 DC with stirring. The panels were kept
in the tank for 5 minutes, removed, allowed to drain for a few seconds and then placed in the
rinse tank. The panels remained in the rinse tank for 5 minutes before removal, drained
briefly and then immersed in the Alodine® 1200S, a chromate conversion solution. The
temperature ofthis solution was 30 ± 1 DC with stilTing. The duration ofexposure in this tank
was dependent on the weight of conversion coating needed on the panel.
In our particular system, it was found that for the first two minutes the amount ofcoating
deposited was a direct function of the exposure time (the immersion time in the Alodine®
1200S soluti~n). In order to obtain a standard coating weight of 40-60 mg / ft 2, immersion
time was 55-70 seconds. After removal, the panels were rinsed for 5 minutes in the wash
tank, removed, allowed to drain briefly and sprayed with deionized water before being
allowed to dry on the rack at room temperature for 2 hours. The substrates which were
pretreated with a chromate conversion coating of chromate weight in the lower range 20 -
30 mg / ft 2 are denoted as LCCC, those in the middle range 40 - 60 mg / ft 2 are denoted as
MCCC, and those in the higher range 80 - 100 mg / ft 2 are denoted as HCCC. For those
_~~~,§l!.Q~Jrate~leaned,.hy~SMGhem®"CC 500 and Sanchem®CC 1000 solutions, removed and.. _.' _,,,..~> .• --''7 . - --...~"•.~~~------ ----- , -
-2-7-
without further immersion in the Alodine® 1200S solution are denoted as 1000.
The supplier of the Sanchem®materials, Sanchem, Inc., Chicago, IL, list quality control
procedures for their materials. The same is true for Alodine® 1200S, supplied by the Henkel
Corporation (Parker Amchem) ofMadison Heights, MI. The chromate conversion procedure
is illustrated in Fig. 8.
2) Coating Preparation
The following is a step by step procedure for coating 2024 T3 aluminum panels, which
mayor may not have been treated with a Alodine® 1200S chromate conversion solution
supplied by Parker Amchem. After treatment, these panels are typically stored at RT in a
open box with a lid to protect them from dust for a period of48 hours to several weeks. An
arbitrary period ofone month has been established as the maximum time allowed before the
panels are coated. The panels are handled by holding the edges using plastic gloves, and are
treated in the following ways prior to coating:
A 3" X 3" panel is sprayed with 95 % ethanol both with the roll marks as well as across
the roll marks and gently wiped using a extra low-lint free Kimwipes® EX-L wipers. All
alcohol is removed in this manner and the panel is then sprayed again with the ethyl alcohol.
The panel is not wiped again. Panels are hung on a suspended wire by means of a clip
attached at the comer of the panel. These panels are allowed to dry at RT for a minimum
of 30 minutes.
(1). Using the above procedure, panels were prepared using Epon 828 (epoxy resin) and
Epon 3141 (a polyamide curing agent supplied by the Shell Chemical Co.). A sm~ll amount
of BYK 306 (mmllJ(actllrs:QQyI!Xfu,£!legl~~L~~s_.~~~~~ _t~ the Epon 828 prior to the• -# ~-,---_•• ".-.'.- ~. ~ .. ·-c~.t{;.,:-Al"=~."r.o::.t.·~"~~).
-28-
Sanchen1® 5003 n1in
65.5 0C
Non-silicated,caustic free,mild alkalinecleaner
RinseRT talS water
5 min
1
Rinse Sanchem® CC 1000Drained RT tap water Drained .. 5 min.. r
43.9 OC5 - 15 sec 5 min 5 - 15 sec
Chrome ftee,surface conditioningdeoxideI' andsmut remover Drained
5 - 15 sec
Alodine ® 1200S RinseDrained up to 5 min Drained RT tap water 1..-... ...
5 - 15 sec 30.0 oC 5 - 15 sec 5 min
Chromate treatment
Deionizedwater spray • Air dry 2 hour
....... Minimum .. Polymer coat40 hours aging and!or test
Fig. 8. Procedure of the chromate conversion treatment.
-29-
'\
addition of the Epon 3141. The panels were prepared in the followi~1g manner:
a. 50g of Epon 828, which had been allowed to stand overnight at room temperature in a
covered plastic beaker, was weighed into a LDPE plastic sandwich bag. The bag was
supported on the balance by means of a section of circular cardboard pipe. After weighing
the Epon 828, a silane, BYK 306 was added. Note: The BYK 306 was dispensed from a
plastic hypodermic syringe in the amount of0.28 ml. The bag was sealed and kneaded gently
until all the BYK had been dispersed in the resin. The sealed bag was placed in a bath at
43 DC and allowed to stand for approximately 30 minutes, or until no bubbles were visible
in the resin and then kneaded again. Another bag was placed in the holder and 37.23 g of
Epon 3141, was weighed into the bag. The bag was zipped sealed and also placed in the
water bath mentioned above for the same period of time. The polyamide was then poured
into the bag containing the Epon 828 resin and the two resins mixed by kneading them
together.
b. The bag was placed in a warm water bath for 15 minutes and the resin poured into a
cylindrical plastic cartridge equipped with a nozzle and a plastic tube with clamp. The
cylinder was degassed for 15 minutes. After standing for 15 minutes, the resin was ready to
apply to the panel by allowing the resin to pour onto the panel, while keeping the end ofthe
tube below the surface of the resin. Approximately 15 g of resin was used for each panel.
c. The spin coater was set as follows (results in coating thickness of approximately 2.0
mils):
Acceleration at 1/2 maximum settingRPM 2500Spinning time 30 seconds
d. The resin was applied to the panel starting in the center and poring an amount the size
ofa silver dollar to cover approximately three quarters ofthe panel. After coating, the panels
were placed on a stainless steel plate 12 x 12 inches, which was then placed in an circulating
air oven and cured at 100 DC for 2 1/2 hours.
Those panels prepared by this procedure were denoted as 828.
(2). Wayne Pigments Co. supplied both Shell Epon 828 epoxy resin and Epon 828 epoxy
resin containing 10 % SrCr04 which had been mixed into resin. In order to keep the pigment
from settling, a SS rod approximately 12 cm by 2.4 cm was placed in each one liter contain
er. The container was sealed with friction tape and placed on rollers turning at 1/2 rpm.
The resins containing the chromate pigment appeared to have minute bubbles of air
trapped in the mixture. These bubbles would appear after coating the panel and when cured
produced a panel that had a marred surface. It was also detennined that the amount of13YK
306 had to be increased to prevent the coating from cissing and cratering. Special precautions
must be taken when preparing these resin mixtures. It is essential that the resin contains no
air when placed on the panel. Even minute bubbles in the mixture will rise to the surface of
the coating after it is on the panel and will not dissipate when placed in the oven. Many
difficulties were encountered in trying to prepare panels that were bubble free.
a. The following procedure was found to produce Epon 828 with SrCr04 (Epon 828
coating containing 5 % SrCr04
by weight) panels that were satisfactory for adhesion
experiments. 8.7 g ofEpon 828 was weighed into a plastic bag. Then 0.47 ml ofBYK 306
was added to the resin and mixed. The mixture was placed in a 43 DC water bath and allowed
to stand for 20 minutes. The bag was thenkneaded until the BYK 306 was thoroughly.mix.ed
-31-
into the resin.
b. The bag was replaced on the balance and 41.3 g of Epon 828 containing 10 % SrCr04
was weighed into the Epon 828 containing the BYK 306. Note: This mixture had been
allowed to stand in the warm water bath for at least 1 h. The two resins were kneaded in
order to mix them and the mixture was placed in the 43°C water bath. The mixture was
allowed to remain in the water bath for 20 minutes. During this time the ~pon 3141 was
weighed, allowing approximately 10 grams for hold-up(excess) in the bag and this was also
placed in the 43°C water bath for the same period oftime. Both bags were degassed in the
vacuum desiccator for 20 minutes.
c. The Epon 3141 was mixed with the Epon 828 containing the SrCr04 and the mixture
degassed for 15 minutes. The mixture was then poured into the plastic cartridge described
previously and once again degassed for 15 minutes. At this point all air bubbles should be
on the surface of the resin. The objective is to dispense the resin onto the panel from the
bottom ofthe cartridge, without introducing additional air. This can be done by keeping the
plastic tube below the surface of the resin being dispensed on the panel until sufficient resin
has been placed on the panel to ensure complete coverage. The clamp is then closed and the
dispenser removed from the area of the panel.
d. The panels were spin coated at 3000 rpm for 30 seconds, with the acceleration set at 3/
4. The panels were cured at 1000e for 2 1/2 hours This method produced satisfactOlY panels.
Those panels prepared by this procedure were denoted as 828(SrCrOJ or 828 Cr.
2. Accelerated Stress Testing:
-2:1. WaterImmersiurrTest-·---
-32-
1). Water immersion test at 150 ° F (65.6 ° C), then equilibrate to room temperature for
24 hours (still immersed in water).
a. By using apparatus shown in Fig. 9, the coated panels were partially exposed to
distilled water, then the fixture and samples were placed in an oven kept heated at constant
temperature of 150 ° F (65.6 °C).
b. Panels were kept in the oven for desired exposure time. After the exposure time, the
whole setups were taken out and placed in the general laboratory environment to let them
cool down to room temperature for 24 hours. Then the panels were wiped dry and
immediately subjected to adhesion tests (within one minute).
2). Water immersion test at room temperature.
a. By using the same apparatus shown in Fig. 9, the coated panels were partially
exposed to distilled water, then the fixture and samples were kept in the general
laboratory environment for desired exposure time. After the exposure time, the panels
were wiped dry and immediately subjected to adhesion tests (within one minute).
3. Adhesion Measurements:
3.1. Tape Test:
1). Tape test, Cross-cut tape test (ASTM D 3359, Test method B)
a. By using sharp razor blades, make cross pattem of cuts on polymeric coatings. For
coatings having a illy-film thickness between 2.0 mils (50 /-un) and 5 mils (125 !lm), space
the cuts 2 mm apart and make six parallel cuts in each direction.
b. Make all cuts about 3/4 in. (20 mm) long. Cut through the film to the substrate in one
steady motion usingjust sufficient pressure olithe-cultiiigtoolToliave tl1.e CTlttihg' edgeI'eich~~:
-33-
DistilledWater
Coated Panel
Fig. 9. Setup for the water immersion test.
-34-
O-Ring Seal
Spring LoadedClamp
Non-conductiveBase
the substrate. After making the required cuts, brush the film lightly with a soft brush or tissue
to remove any detached flakes or ribbons of coatings.
c. Dispense a tape at a steady (that is, not jerked) rate and cut a piece about 3 in. (75 mm)
long. Place the center of the tape over the grids and in the area ofthe grid smooth into place
by a finger. To ensure good contact with the film, rub the tape firmly with the eraser on the
end ofapencil. The color under the tape is a useful indication ofwhen good contact has been
made.
d. Within 90 ± 30 seconds of application, remove the tape by seizing the free end and
rapidly pull (not jerked) back upon itself at as close to an angle of 180 0 as possible. Inspect
the grid area for removal ofcoating from the substrate using the illuminated magnifier. Rate
the adhesion in accordance with the scale illustrated in Fig. 10.
3.2. Microhardness Indentation Test:
A microindentation technique was developed by 1. E. Ritter, et ai. 52- 54 to measure the
adhesive shear strength of thin polymer coatings on glass substrates. Indentation-induced
debonding ofthe coating was observed to occur under three different conditions: Type I was
with the deformations underneath the indenter being essentially elastic; Type II was with the
deformations underneath the indenter being plastic; and Type III was after the indenter
penetrated the substrate (Fig. 11).
In Type II, when the load increases beyond the critical load which initiates an interface
crack, further loading ofthe indenter causes this interface crack to grow in a stable fashion.
An analytical model was developed for this state 55, and the interfacial fracture energy
between this th.in coating and rigid substrate' can be~alculated by!l~e,f2!!9~iB[ equation:
-35-
Surface of cross-cut area fromCliI'sslficalion which rla~ing has occurred.
(Example for !>ix paralled cutsi
58 None.,
48 II38 II28 II18 •OS Greater than 65'10
58 The edges of the cuts are completely smooth: none of the squares of the lattice is detached.48 Small flakes of the coating are detached at intersections; less than 5 % of the area is affected.38 Small flakes of the coating are detached along edges and at intersections of cuts. The area
affected is 5 to 15 % of the lattice.28 The coating has flaked along the edges and on parts of the squares. The area affected is 15 to
35 % of the lattice.1B The coating has flaked along the edges of cuts in large ribbons and whole squares have
detached. The area affected is 35 to 65 % of the lattice.OB Flaking and detachment worse than Grade 1.
Fig. 10. Classification of adhesion tape test results (ASTM D 3359).
....-36-
TYPE I (Elastic deformation under the indenter)
P
t
~o..----Oebond~ Coating
. / / / / / / / / / / / / / / / / /Sub~traie/ /
TYPE'II (Plastic deformation under the indenter)
P
t
Debond __ S1~6 Coating
TYPE 111 (Indenter penetrates substrate)
P
t_______~s~ ..r-Z _
Cebond---..~./l Coating
/ / / / /' / // /' / /\:::77/ / / AUb~trate/ /
Fig. 11. Schematic of the three types of debondirtg on indentation of the coatina 52. .•.,.. _.__~~." ." ••:-_,_, ....T"..,~ •..'~.":r'-':Y~~n-r:.~l'..:a.:r~h-:.-o.:rT::r.r!:I':1;VJ...V,l-..;l 'i...;:~~.,.L'~U--.':., !.l..~.1"...~ •.- ...... -'.~:.'._; ' ••-".''''',. '~.-:..~:"-' "", -- _. " .....,-Tn~'......,.·c~-rr;'y.'r:T":TT1~I."':7,':_r-~~-s'-~--;.r-·,·_"·
-37-.
." '._ .._.•__ ..._,.",...,...,. .... ,......•. ,.>~~.c,,:}_ ....~ • .••• >
where the Ge is the interfacial fracture energy, He is the hardness of the coating, h is the
thickness ofthe coating, Ve is the Poisson's ratio ofthe coating, Ee is the elastic modulus of
the coating, c is the crack radius, and the P is the load of indentor. It is assumed that the
coating has a constant hardness, hence b = (P / 2 H) ~\ where b is the halfofaverage diagonal
length of the indentation. And c is the radius of the furthest ring and can be illustrated in
Figures 12 & 13.
A LECO model M - 400 FT hardness tester was used to measure the adhesion of the
coatings to substrates. The specimens were indented with apyramid-shaped Vicker's indenter
at load of500 grams and dwell times o£1 0seconds. The amount ofthe force applied, the size
of the indentation, and the size of the debonded area (radius of the furthest ring) were
recorded for each measurement. Five indentations were made on each specimen and the
values were averaged.
4. Miscellaneous Characterizations:
4.1. Scanning Electron Microscopy:
The specimens were cut from the bare Al substrate, Al substrate cleaned by Sanchem®
CC 1000, and by Alodine® 1200S with three different immersion time (LCCC, MCCC,
HCCC). Then they were mounted on SEM stubs using a conductive cement. SEM was
perfonned by JEOL - 6300F instrument at 20 KV and 5 KV .
4.2. Tensile Behavior:
The elastic mQdulus, Ee, ofthe 828 and 828 (SrCr04) coatings· were measured at Instron .
-38-
Fig. 12. Some micrographs illustrating debonding by Vickers indentation of epoxy coating 55.
(b)
-39-
4206 by following ASTM D638 (M) with specimen specification of type I and at testing
speed of 5 mm/min. Five dumbbell samples were made from each kind of polymers. The
tensile values reported were an average of the five specimens tested.
-40-
D. Results and Discussion:
1. SEM results:
The image of the 1000 substrate is the general type of Al surface morphology after
detergent and deoxidizer cleaning (Fig. 14), that has been seen by other workers 56.57. SEM
images ofLCCC, MCCC, and HCCC substrates (Fig. 15-20) reveal that there is an increase
of the number (density) of mud-cracks. Some works have suggested that it is this kind of
surface morphology that increases the adhesion property between coating and substrate.
From the results of our dry adhesion tests on all kinds of substrates (Fig. 21), there is no
direct relationship between the number ofmud-cracks and the interfacial fracture energy. If
from the view ofmechanical interlocking mechanism, the surface ofHCCC substrate has the
roughest appearance and highest density ofmud-cracks, but the adhesion property ofHCCC
was not superior to LCCC and MCCC, and in some cases it decreases in the early stage
of exposures when compared to LCCC and MCCC substrates.
Therefore, we suggest that mechanical interlocking is not the main factor affecting the
adhesion strength which was also observed by Scantlebury et ai. 58.
2. Polymers with or without chromate (SrCr04):
The effect of strontium chromate mixed within the polymeric top coating can be clearly
seen from the results of tape test and indentation test at room temperature or 150 0 F. In all
conditions, the retained percentage or interfacial fracture energy for the polymers having the
strontium chromate are higher than those without (Fig. 22-25). The reinforcement property
(combined with inhibition to corrosion) was also observed in other works 59. 60 when
----,. "..~pigments were added.
-41-
..---~-:---._-- -- .. ~.~-Fig ... l-4.-SRM-imageof 1000 substrate (Magnification 20000 X, 20 KV).
-42-
INTENTIONAL SECOND EXPOSURE
Fig. 14. SEM image of WOO substrate (MagniflC<lli-on...2.0.cu)G...X<.-2O-KV).
-42-
INTENTIONAL SECOND EXPOSURE
Fig. 15. SEM image ofLCCC substrate (MagI1!llcation 20000 X. 20 KVl.
,-.-~--'-
Fig: 16. SEM image ofLCCC substrate (Magni~cation60000 X, 20 KV).
-44-
INTENTIONAL SECOND EXPOSURE_-----------
Fig, th, SE\! image ofLCCC substrate (\lagnificmion 6UOOO~. 2U K\'l.
-4-+-
Fig. 17. SEM image ofMCCC substrate (Magnification 20000 X, 20 KV) ..
-45-
INTENTIONAL SECOND EXPOSURE
--+~ -
Fig. 18. SEM image ofMCCC substrate (Magnification 60000 X, 20 KV).
-46-
INTENTIONAL SECOND EXPOSURE
--J.(J-
Fig. 19. SEM image ofHCCC substrate (Magnification 20000 X, 20 KV).
-47-
COND EXP05UKI:
Fig. 19. SEM
Fig. 20. SEM image ofHCCC substrate (Magnification 60000 X, 20 KVY.
-48-
INTENTIONAL SECOND E.XPOSURE_-------------
Fig. 20. SE\I image of HCCC substrate (lvlagnification 6Cll)OO X. 20 KV).
-48-
35
34
-C\I<
~ 33cSC)-l:.2 32IIIQ)..l:1J«
31
30
Dry Adhesion on different substrates828& 828 (SrCr04) coatings
~ .'
/ I~/ ~
/ Ri"
/~
I;~~~
/ .--- ~/ ~.v 'I-
-1 ( 00 LCCC MCCC HCCCSubstrates
1-111-- 828 --s- 828 (srCro4ll
Fig. 21. Dry adhesion (Gc) of coating 828 and 828 (SrCr04) on different substrates.
-49-
Gc v.s. Exposure TimeWater Inrnersion Test (RT)
33 .,----------------------------,
----_.__._- -_..- ~. - - .. - .- _._-----_ .. _-
23· -----.~-_____r----~----.---~--.---------_l
f 29 ;--.-----.::E-.,cS~ 27 .--...--~ ..---------.-.-~-~--.-- .... - .-- ..._.--_....-c.2(/)Q)
.c: 25~
35302515 20Exposure Time (days)
105
21 -f-------;-------,-----c----..,...--------..,....-----l
a
r-... - _.. ::_ ._./.:" -c- - ••••-.----.:-- ,•..•_ •. !
I-.- 828 on -1000 -m- 828Cr on -1000 II . ' .. :
Fig. 22. Comparison ofinterfacial fracture energy (Gc) between coating 828 and coating 828(SrCrO4) on Sanchem ® CC 1000 cleaned substrates versus exposure time from indentationtest (room temperature).
-50-
Retained %v.s. Exposure TimeWater Inmersion Test (RT)
100 ~---m-~==m==--m----1B------I!k:::::----------ii-=::::::m---,
80 ---.---~---------.--.-.
70 -~----- ..---=-~----.-.- - ..-- ._-. ----------.-.--
90 ---- -
~ 60 -/----------.------ --.--------.-- --..------..----
30 -------
50 -------------+-----
20 -.-.-----.-----~-- -- -
c:o'(ijQl
:§ 40«
4035
._-- .__ .__ .
10 15 20 25 30
Exposure Time (days)
l. 828.0n Leee -m.=- 8~~~ron_:~~e]
5
0--\--- ,...--__---;-__--,-__---:......_-----;----,.---
o
10 -------------
Fig. 23. Comparison of retained percentage between coating 828 and coating 828 (SrCr04)
on LCCC substrates versus exposure time from tape test (room temperature),
-51-
Retained %v.s. Exposure TimeWater Immersion Test (150 F)
100 m--:----£!Io=---=--------------------------,
90
80 --------'1- _
70 -------\--~--------------- 1
-----------------------
-- ----- ---- - -- ~ - - ---------
20 -1---1e-------\-----------
30
c.2 50 - -- ------\C/)Q)
:§ 40«
~ 60 --- - ---I-
20018016060 80 100 120 140
Exposure Time (hours)
4020
o _L,#-~-....,.__4.-_;___+___:_-=:====;====;:====::;===T___~o
Fig. 24. Comparison of retained percentage between coating 828 and coating 828 (SrCr04)
on MCCC substrates versus exposure time from tape test (150 0 F).
-52-
Gc v.s. Exposure TimeWater Inmersion Test (150 F)
34 ..r----------------------------,
32 - ---.-------.-.-
---------------
---- _. ------------.---. ----- -m=-~---~..-=-=-----==----=-==-~~30 --------N<E--, 28 ++-__,,----0'
C)-c.2 26IIIQ).c"'C« 24
22 -----
!!=====-=~----_.._- ._-- -
~~---------
504540353025201510520+,-:--~:----.,..---_____;--____;_--_r_---r---__:--__r--__,__--
oExposure Time (hours)
-.--.....--;~---;--..-.- --------.- ~-;T;·-c,---~
~828(JnHGGG--fiI-.828Gron HCGC !
Fig. 25. Comparison ofinterfacial fracture energy (GJ between coating 828 and coating 828(SrCr04) on HCCC su.!Jstrates versus exposure time from indentation test (150 0 F).
-53-
3. Comparison between tape test and indentation test:
The tape test is not a precise way ofmeasuring adhesion property of the coatings, it can
be severely influenced by the techniques of the applier, variation of cutting instruments,
visual bias, and other effects. Besides, its limitation to distinguish between higher levels of
adhesion may be a problem with those having strong adhesion properties. But with careful
control of every factor, it is possible to give a semi-quantitative determination to establish
whether the adhesion ofa coating to a substrate is at a generally adequate level. Thiscan be
seen in our work in Figs. 26 & 27 and Table 3.
Room Temperature 150 0 FSubstrate \ Coating 828 828 (SrCr04) 828 828 (SrCr041
1000 yes yes yes yesLCCC +/- no yes + /-
MCCC +/- no yes yesHCCC yes no yes no
Table 3. Comparison ofthe trends of adhesion curves between tape test and indentation testin different conditions (yes: they agree well, no: they don't agree, +/- : between yes and no).
4. Effects of substrates:
Summary ofindentation test results after water immersion test (room temperature) for 828
coatings on different substrates are shown in Figs. 28 & 29; and the tape tests are shown in
Figs. 30 and 31. Summary of indentation test results after water immersion test (room
temperature) for 828 (SrCr04) coatings on different substrates is shown in Fig. 32; and the
tape tests is shown in Fig. 33. Summary ofindentation test results after water immersion test
(150 0 F) for 828 coatings on different substrates is shown in Fig. 34; and the tape tests are
-54-
Gc v.s. Retained 0/0
Water Irrrrersion Test (RT)
201816
- - ,--- - - r-- -~ - •
16 18 20
14
14
12
1210
10
8
8
6
6
4
4
-- I --
2
,..- ~ -- -.- - - ...... - - I - - -I -- - - I - -
2
10 '
o 'o
31
30
29 -C'l<E 28 --..,0 27~s:: 26 -.21/1Q) 25 _J:"C« 24
23 ~22 j
II
21 L _
0
100
90 -
80 c
70:- ,
~ I0 60 I-l: l0'iiiQ)
.c"C<
Exposure Time (days)
Fig. 26. Retained percentage compared to indentation test records for coating 828 on 1000substrates versus exposure time (room temperature).
-55-
Gc v.s. Retained %Water Inmersion Test (150 F)
32 ;
31 -
C'\l 30l<E I- 29 _:..,
I
cJ 28 i(.!) :-l: 27.S!UI 26(1).l:"C 25<t -,,
24!
23 Il
I
22 .~. - ........-- -- - -- -,-,
0 10 20 30 40 50 60 70
100
90
80
~
,
~ 60;1
l:.S! 50 JUI(1).l: 40 ,"C<t ,
30 ~I
20 ~I
10 ~II
o -i- T - --- ~-- ~ .-- -
0 10 20 30 40 50 60 70
Exposure Time (hours)
Fig. 27. Retained percentage compared to indentation test records for coating 828 on LCCCsubstrates versus exposure time (150 0 F).
-56-
r
Adhesion after RT Immersion Test828 coatings on different substrates
33 ,-- --,
31 -+\\-----------------------------1
N 29<E-.,c.SQ. 27 --------
c:.2UlCllJ:"C 25<l:
252010 15Exposure Time (days)
5
21 -I-------,--------r--------;--------,----------'
o
Fig. 28. Interfacial fracture energy (Gc) of coating 828 on different substrates versusexposure time from indentation test at room temperature.
-57-
Adhesion after RT Immersion Test828 coatings on different substrates
33 ,-- ----,
--=-..c~------~.----.--------
23 -t-\11--t--f--------Y------------------------j
N<.§..,e5~c,211lQ).c~ 25 -H-+I--\-----.l:-f--------'l~'----------~----==-_==__---------"'~
5.04.54.03.53.02.52.01.51.00.5
21 +---,.--------;-----r----,---...,---,.--------;------,---......,..------4
0.0
Exposure Time (days)
I~ -1000 -m- Leee -e-- rvcee -.- Heee I
Fig. 29. Interfacial fracture energy (Gc) of coating 828 on different substrates versusexposure time from indentation test at room temperature (Fig. 28, expanded scale).
-58-
Retained % v.s. Exposure TimeWater Immersion Test (RT)
201816141210864
+-+-II---+-'j----------------------_·----~-----_· ------ --
-lHH--\-------+-----'--------->,----==""'---=::----'.----~------~--
100
90
80 -
70
~ 60~
c.2 50IIIC1l.c'C 40<t
30
20 -
10
00 2
Exposure Time (days)
Fig. 30. Retained percentage of coating 828 on different substrates versus e~posure timefrom tape test at room temperature.
-59-
Retained % v.s. Exposure TimeWater Il111lE!rsion Test (RT)
100
90
80
70
~ 600-~
.2 50IIICIl.c
40"tI«30 -
20
10
00.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0
Exposure Time (days)
[-+- -1000 -m- Leee -A- MCee -G- HCee-j
Fig. 31. Retained percentage of ~oating 828 on different substrates versus exposure timefrom tape test at room temperature (Fig. 30, expanded scale).
-60-
Adhesion after RT Immersion Test828 (SrCr04) coatings on different substrates
35 ,-------------------------------,
33 dr---",,--------------------------I
f 31 -H-\--\---~------------E-..o~ 29 -t-t--~-\----~--~""""'------__=~~---------Is:.2IIIQ)J:"'C 27 -I-"\------a::::::---....=::::::a~----------_____:=___-----I«
25 -l--~--__F----~-----_/'_--------~----
45403515 20 25 30Exposure Time (days)
10523 -I----,.....----,------,-----..,....----,-----,-----,------,------J
o
Fig. 32. Interfacial fracture energy (Gc) of coating 828 (SrCr04) on different substratesversus exposure time from indentation test at room temperature.
-61-
Retained %v.s. Exposure TimeWater Inmersion Test (RT)
90 _
80 -/-t-------------\---J'--\--I---\-----r-------
70 -/-t-------------x---------------j
~ 60 -/-t--------------------------~--lo-c.2 .r 50 -l--.lr.,------------------------~--""------------"----"II)Q).c~ 40 __I ""_·· ~ " --------
30 -1--1--------------------------_1
20 -l~.__-------------------------_I
605010 20 30 40
Exposure Time (days)
o -!----.>----.O"-"--.......---;---.....,....-~>----__,_----_,__---_l
o
10 -1_-\- --.- 1
1-.-:- -1 000 --0-- Leee -.-Meee ---*- Heee I
Fig. 33. Retained percentage ofcoating 828 (SrCr04) on different substrates versus exposuretime from tape test at room temperature.
-62-
Adhesion after 150 F Immersion Test828 coatings on different substrates
32 -k--------------------------------,
706030 40 50Exposure Time (hours)
2010
20 -l-----,...-------,-----,-----..,-------,-----,------..,..-----J
o
22 -\H--------~------------------~I
N 28 -\I/-+---I_--'l;:--- ------'r-- ~--
<E-..,cj
~c.S!fI)C1l.c'C«
Fig. 34. Interfacial fracture energy (Gc) of coating 828 on different substrates versusexposure time from indentation test at 150 0 F.
-63-
shown in Figs. 35 & 36. Summary of indentation test records after water immersion test
(150 0 F) for 828 (SrCr04) coatings on different substrates is shown in Fig. 37; and the tape
tests are shown in Figs. 38 & 39.
In room temperature immersion tests, the 1000 and HCCC substrates dropped their
adhesion properties significantly in the early stage when compared to the LCCC or MCCC
substrates. The only exception was the 828 (SrCr04) on HCCC substrates when tested by
tape test: 828 (SrCr04) on 1000 substrates failed to remain attached after approximately 4
days of immersion, yet 828 (SrCr04) on HCCC substrates maintained 70 % ofretention for
27 days of immersion which was only a little lower than LCCC substrates. When one
compares the retained percentage graph (Fig. 33), the indentation data (Fig. 32) and Table
3, it can be seen that the 828 (SrCr04) coatings had many discrepancies and these were also
observed for the same coatings in the higher temperature immersion tests. The tape test,
unlike the indentation test, involves the competition between two adhesion properties: the
adhesion between the tape and coating surface; and the adhesion between coating and
substrate. When engaged in any kind ofexposure tests or blended with other materials, these
properties will definitely change. If surface properties such as wetting ability, surface
morphology, chemical function groups' distribution, and any other mechanical properties
changes, the results oftape test may not be sufficient to reveal the real changes of adhesion
properties of the coatings/substrate interface.
The similar trends for the 1000 and HCCC substrates to fail in the early stages were also
observed in the 150 0 F immersion tests. In most cases, the adhesion ofpolymeric coatings
on MCCC substrates (which were the recommended condition by manufacturer) had the best
-64-
il~
...._---~-~~------ ---------------. -
~I,_._._-~~._.~-
I'I 1\I I \I , I
\ , I
I , I
I
" ]1
,\I ,\ ... ...1:>
I ,~-
I , ---I Q \ --\ - --_L_ -I -----------~----_!:r~-~...... _---- -----
I I I ",I I 1_J\ " -~._--~----_. __ . __ ._-_._---- - _.. -
I-,.
\ "I , l.t. \ " ----'I M
P-J/\ ~ ~I,
"- II
\ I
~v1 -><' ~I ..J --- , , - , , -
100
90
80
70
- 60~-I:.2 50l/lQ)J:"C 40«
30
20
10
oo
Retained % v.s. Exposure TimeWater Irrmersion Test (150 F)
10 20 30 40 50Exposure Time (hours)
1- -4- - -1000 -0- LCCC A MCCC :oK
60
HCCCIJ
70
Fig. 35. Retained percentage of coating 828 on different substrates versus exposure timefrom tape test at 150 0 F.
-65-
Retained % v.s. Exposure TimeWater Immersion Test (150 F)
14
,\
\
\
~~-----~
\
\ ,
,\
\
\
\
,,
•. ,, ,I \ ,. ,
I \.,
I
4 6 8 10 12Exposure Time (hours)
I-;----&;::-------+---------.'---------',-------~--
,,-/-1~------__1r_~_H__\_-Q.;;::_______\___-------~-~---~
,
,,+h----\-----\---------/r--~\:~~,.
I
100
90 - I
\
80 - ~
70
~ 60\0-l: ,
.2 50 IUl ,Q)J:
" 40«30 -
\20
,
10 - -l
0-
0 2
1-· ....·-1000 -a- Leee -.-lVCee --r- HCee I
Fig. 36: Retained percentage of coating 828 on different substrates versus exposure timefrom tape test at 150 0 F (Fig. 35, expanded scale).
-66-
Adhesion (Ge) after 150 F Immersion Test828 (SrCr04) coatings on different substrates
38 -,----------------------------,
36 -j--------------------------~___j
20018040 60 80 100 120 140 160
Exposure Time (hours)20
24 +,--...,----r------,---..,.....------,---,--...,..-------;-----;------j
o
26 -j---------~---------------". --__j
34--N<E- 32..,<i
C>-c.2 30- -1IlQ).c'0<C 28
I--+- -1000 -- LeeC -A- rvcee -0- HCCC I
Fig. 37. Interfacial fracture energy (Gc) ofcoating 828 (SrCrO4) on differel\t substrates versusexposure time from indentation test at 150 0 F.
-67-
Retained %v.s. Exposure TimeWater Inmersion Test (150 F)
20018016060 80 100 120 140Exposure Time (hours)
4020
tr-~I_---__\---~---~~~---=""__=:\-~~--~~------
I--\l--~--\-----/--~--~\----~------ ------ ------
100
90 -
80 -
70 -
~60 --c
0 50 -'iiiQlJ:"0 40«
30
20 -
10 -
0
0
Fig. 38. Retained percentage ofcoating 828 (SrCr04) on different substrates versus exposuretime from tape test at 150 0 F.
-68-
100
90 -
80 -
70
- 60 -~0-c.S! 50 -11ICIlJ:'0 40<t
30
20 -
10 -
0
0 5 10
Retained %v.s. Exposure TimeWater Immersion Test (150 F)
15 20 25 30 35 40Exposure Time (hours)
1-<>-- -1000 -m- Leee --.- rvcee --*- HCee I
45 50
Fig. 39. Retained percentage ofcoating 828 (SrCr04) on different substrates versus exposuretime from tape test at 150 0 F (Fig. 38, expanded scale).
-69-
overall properties and the 1000 or HCCC substrates were poorest. And all of them seem to
increase their adhesion after certain period of exposure which were also observed by other
works 58,61. Arslanov et ai. 62 used peel test and tape test to measure the adhesion properties
of epoxy coatings on aluminum foils (AI, 99 %) and panels ofaluminum alloy. They found
that after an initial decrease the adhesive strength increased with the time of exposure time.
They suggested it was the hydration ofaluminum oxide adjacent to the adhesive joint which
enabled additional hydrogen bonding between the organic and its support. This may have
also happened to our 1000 substrates, but for the other chromate substrates, other chemical
analysis should be applied to determine the formation aluminum hydroxide or other metal
hydroxide.
Kovaleski et ai.63, used the tape tests to discriminate the adhesion properties of similar
coatings and substrates. Their results are shown in Table 4. It can be seen that substrates with
heavy weight of chromate possessed the worst adhesion.
Light Standard HeavyDry Tape
Primed SA SA SA SA SA SA 3A 3A 3APrimed + Topcoat 4A 4A 4A 4A 4A 4A 3A 3A 3AWet Tape (24h)
Primed SA SA SA SA SA SA 4A 4A 4APrimed + Topcoat 4A 4A 4A 4A 4A 4A 4A 4A 4AWet Tape (7 Days)
Primed SA SA SA SA SA SA 3A 3A 3APrimed + Topcoat 4A 4A 4A 4A 4A 4A 4A 4A 4A
Table 4. Results of the tape adhesion tests for the chromated Al panels of different coatingweights 63.
-70-
E. Conclusions:
1. An indentation-induced debonding test can be used to determine the adhesive strength
ofpolymer coating/substrate. The adhesion measurements are reproducible and simple. The
test appears acceptable in establishing trends in the adherence of a given coating/substrate
system and in comparing adherence between different coating/substrate combination.
2. From the SEM images and the dry indentation test data, we suggest that mechanical
interlocking in not the main factor affecting the adhesion strength in our experiments.
3. The strontium chromate mixed within the polymeric top coating certainly improved
adhesion properties which were seen from the results of tape and indentation tests in room
temperature or 150 0 F immersion.
4. The tape test was good as a semi-quantitative determination of coating adhesion
adequate level. However, it was unable to distinguish between coatings at higher levels of
adhesion.
5. In most cases, the adhesion ofpolymeric coatings on MCCC (normal) substrates had
the best overall properties and the 1000 or HCCC substrates were the poorest. All of these
samples seem to increase their adhesion after acertain period ofexposure following an initial
decrease.
-71-
F. Future Work:
Further work is needed to characterize the adhesion ofother types ofpolymeric coatings:
high permeability or mixture of high and low permeability polymeric coatings, or
commercial coatings with high percentage of strontium chromate which is usually used as
top coatings. Since the coatings with higher percentage of strontium chromate have strong
yellow color, it is very difficult to determine the size of Newton Ring by microindentation
test. Other types ofadhesion measurements such as peel test or double cantilever beams test
should be considered as an alternative method.
The main emphasis of this work was on the glassy polymers. When using the high
permeability soft coatings, the adhesion measurement such as the blister test should be
considered, since it is impossible to be done by microhardness indentation or other types of
mechanical tests due to the softness. Other cleaning and treatment procedures such as
phosphoric acid anodizing may be useful as a comparison. Further more, the thickness effect
on adhesion should be considered. Finally, micrographs by TEM and surface roughness
determination by ATM would be very useful for surface profile and morphology.
-72-
G. References:
1. D. J. George, C. J. Walton, and W. G. Zelley, "Chemical Treatment and Finishing",
Chapter 17, Volume III, Aluminum Fabrication & Finishing, John Wiley & Sons, New
York, 1973, pp. 609.
2. N. J. Newhard, "Conversion Coatings for Aluminum", Met. Finish., 70, 1972, pp. 49.
3. H. J. W. Lenderink and J. H. W. de Wit, "The Effect of Pretreatments on the Corrosion
Behavior of Aluminium", Modifications of Passive Film, European Symposium on
Modifications ofPassive Films, London, 1994, pp. 214-219.
4. K. S. Kim, and J. Kim, "The Mechanic of Peel Test", 1. Eng. Mater. Techno/., ASME
Trans., 110, 1988, pp. 226-232.
5. K. L. Mittal, "Adhesion Measurement of Thin Films", Electrocomponent Science and
Technology, 3, 1976, pp. 21-42.
6. K. L. Mittal, "Techniques to Measure Adhesion", 1. Adhesion Sci. Techno/., 1, 1987, pp.
247- 259..
7. G. P. A. Turner, Chemical Treatment of Substrates, Chapter 17, Introduction to Paint
Chemistry and Principles ofPaint Technology, 3rd ed., Chapman and Hall, New York,
1988, pp.233.
8. J. A. Treverton and N. C. Davies, Surf Interface Anal., 3, 1981, pp. 194.
9. J. A. Treverton and N. C. Davies, Met. Technol,A, 1977, pp. 480.
10. H. A. Katzman, G. M. Malouf,R. BauerandG. W. Stupian, App/. Surf Sci., 2,1979, pp.
416.
11. Z. Yu, H. Ni, G. Zhang and Y. Wang, Appl. Surf Sci., 62, 1992, pp. 217.
12. M. Koudelkova, J. Augustynski and H. Berthou, 1. Electrochem. Soc., 124, 1977, pp.
1165.
-73-
13. P. 1. Hagans and C. M. Haas, Surf Interface Ana!., 21, 1994, pp.65.
14. F. W. Lytle, G. 1. Bibbins, K. Y. Blohowiak, R. B. Greegor, R. E. Smith and G. D. Tuss,
Carras. SeL, 37 (3), 1995, pp. 349 .
15. M. F. Abd Rabbo, J. A. Richardson and G. C. Wood, Carras. SeL, 18, 1978, pp. 117.
16. J, K. Hawkins, H. S. Isaacs, S. M, Heald, J. Tranquada, G. E. Thompson and G. C.
Wood, Carras. Sci., 27, 1987, pp. 391.
17. G. M. Brown, K, Shimizu, K. Kobayashi, G. E. Thompson, and G. C. Wood, Carras.
Sci., 34, 1993, pp. 1045.
18. G. M. Brown, K, Shimizu, K. Kobayashi, G. E. Thompson, and G. C. Wood, Carras.
Sci., 33, 1992, pp. 1371.
19. G. M. Brown, K, Shimizu, K. Kobayashi, G. E. Thompson, and G. C. Wood, Carras.
Sci., 35, 1993, pp. 253.
20. A. E. Hughes, R. J. Taylor, and B. R. W. Hinton, "Chromate Conversion Coatings on
2024 Al Alloy", Surf Interface Anal., 25,1997, pp. 223-234.
21. T. S. Ellis and F. E. Karasz, "Interaction of epoxy resins with water: the depression of
glass transition temperature", Polymer, 25, 1984, pp. 664-669.
22.1. Banks and B. Ellis, Polymer Bull., 1, 1979, pp. 377-382.
23.1. T. Drzal, Advances in Polymer Science, ed. by K. Dusek, 75, 1986, pp. 1-50.
24. P. D. Aldrich, S. W. Thurow, M. J. McKennon, and M. E. Lyssy, "Dielectric
relaxation due to absorbed water in various thermosets", Polymer, 28, 1987, pp. 2289-
2296.
25.1. D. Maxwell and R. A. Pethrick, "Dielectric Studies of Water in Epoxy Resins", J.
App!. Polymer Sci., 28, 1983, pp. 2363-2379.
26. G. E. 'Johnson, H. E. Bair, S. Matsuoka, E. W. Anderson, and J. E. Scott, "Water. ...
-74-
Sorption and Its Effect on a Polymer's Dielectric Behavior", Polymeric Materials for
Corrosion Control, ed. by R. A. Dickie and F. 1. Floyd, American Chemical Society,
Washington, D.C., 1986, pp. 451-468.
27. P. Bonniau and A. R. Bunsell, "A Comparative Study of Water Absorption Theories
Applied to Glassy Epoxy Composites", 1. Camp. Mater., 15, 1981, pp. 272-293.
28. D. A. Blackadder and 1. S. Keniry, "Difficulties Associated with the Measurement of
the Diffusion Coefficient of Solvating Liquid or Vapor in Semicrystalline Polymer",
1. Appl. Polymer Sci., 18, 1974, pp. 699-708.
29. G. Odians, Principles of Polymerization, 3rd ed., John Wiley & Sons, New York,
1991, pp. 134-136.
30. Heinrich Heine and Hang G. Volz, Introduction, Chapter 1, Industrial Inorganic
Pigments, ed. by Gunter Buxbaum, 1st ed., Weinheim, New York, 1993.
31. H. Sutton and 1. F. LeBrocq, British Patent 370,949, April, 11, 1932.
32. Emmet Lalor, "Zinc and Strontium Chromates", Chapter G-c, Pigment Handbook, ed.
by Temple C. Patton, 1st ed., John Wiley & Sons, New York, 1973.
33. J. 1. Rosenfeld, Lakokras. Mater. Ikh. Primen., 1961, pp. 50.
34. G. H. Cartledge, Corrosion (Houston), 18, 1962, pp. 316.
35. M. Piens,1. Coatings Techno!., 51 (655), 1979, pp. 66.
36. A. J. Kinloch, "Review: The Science of Adhesion, Part I", 1. Mater. Sci., 15, 1980,
pp. 2141- 2166.
37. E. M. Borroffand W. C. Wake, Trans. Institute of the Rubber Industry, 25, 194, pp.
190.
38. M. Matsunaga, Y. Haguida, and K. Ito, Metal Finish, 66 (11), 1968, pp. 80.
39. A. Rantell, Trans. Inst. Metal Finish, 47, 1969, pp. 197.
-75-
40.1. A. Abu-Isa, J Appl. Polymer Sci., 15, 1971, pp. 2865.
41. R. Roberts, F. W. Ryan, H. Schonhom, G. M. Sesssier and 1. E. West, J Appl.
Polymer Sci., 20, 1976, pp. 255.
42.1. E. Pemns and K. Pettett, Plastics and Polymers, 39, 1971, pp. 391.
43. E. H. Andrews and A. 1. Kinloch, Proc. Roy. Soc., A332, 1973, pp. 401.
44. A. N. Gent and 1. Schultz, J Adhesion, 3, 1972, pp. 281.
45. W. C. Wake, "Adhesion and the Formulation of Adhesives", Applied Science
Publishers, London, 1976, pp. 69.
46.1. R. Evans and D. E. Packham, "Adhesion - 1", ed. by K. W. Allen, Applied Science
Publishers, London, 1977, pp. 297.
47. S. S. Voyutskii, "Autohesion, and Adhesion of High Polymers", John Wiley and Sons
(Interscience), New York, 1963.
48. B. V. Deryaguin and V. P. Smilga, "Adhesion, Fundamentals and Practice", McLaren
and Son, London, 1969, pp. 152.
49. E. H. Andrews and A. 1. Kinloch, J Polymer Sci. Polymer Phys., 11, 1973, pp.269.
50. A. N. Gent and A. 1. Kinloch, J Polymer Sci., A2, 1971, pp. 659.
51. Henry Leidheiser, Jr., "Mechanisms of De-Adhesion of Organic Coatings from Metal
Surfaces", Polymeric Materials for Corrosion Control, ed. by R. A. Dickie and F. 1.
Floyd, American Chemical Society, Washington, D.C., 1986, pp. 124-135.
52.1. E. Ritter, T. 1. Lardner, 1. Rosenfeld, and M. R. Lin, "Measurement of Adhesion of
Thin Polymer Coatings by Indentation", J Appl. Phys., 66, 1989, pp. 3626-3634.
53. 1. E. Ritter and 1. Rosenfeld, "Use of the Indentation Technique for Studying
Delamination of Polymeric Coatings", J Adhesion Sci. Technol., 4, 1990, pp. 551
571.
-76-
54. L. Rosenfeld, 1. E. Ritter, and T. 1. Lardner, Interfaces in Composites, Materials
Research Society symposi1.¥TI proceedings, v. 170, Pittsburgh, PA., 1990, pp. 11-16.
55. L. Rosenfeld, 1. E. Ritter, T. 1. Lardner, and M. R. Lin, "Use of the Microindentation
Technique for Determining Interfacial Fracture Energy", J Appl. Phys., 67, 1990, pp.
3291-3296.
56. T. Abe, K. Alzawa, T. Uchiyama, and E. 1. Soyama, "Adhesion to aluminum heat
treated in hot water.", J lap. Inst. Light Metals, 24, 1974, pp. 489.
57. J, A. Treverton and M. P. Amor, "The Effects of Subsurface Topography on
Secondary Electron Images from Chromate Pretreated Aluminum Surfaces", J
Microscopy, 140 (3), 1985, pp. 383-393.
58. Xiaohong Jin and 1. D. Scantlebury, "An Investigation ofPaint Adhesion to Chromate
Conversion Coating on Aluminum Alloys", Cailiao Baohu, 25 (4),1992, pp. 4-13.
59. R. Twite, S. Balbyshev and G. Bierwagen, "Electrochemical Studies of the Effects of
Chromates (Cr 6+) in Aircraft Coatings", Proceedings ofSymposium on Environmen-
tally Acceptable Inhibitors and Coatings, Electrochemical Soc., Pennington, N. 1.,
1995, pp. 203- 217.
60. A. T. Evans, 1. D. Scantlebury, and L. M. Callow, "The Adhesion and Corrosion of
Chromate Containing Coatings on Aluminum", Proceedings of Symposium on Ad-
vances in Corrosion Protection by Organic Coatings II, Electrochemical Soc.,
Pennington, N. 1., 1995, pp. 267-273.
61. X. H. Jin, 1. D. Scantlebury, and G. E. Thompson, "Laquer Adhesion Mechanically
Treated and Conversion Coated Aluminum surfaces", Trans. Inst. Metal Finish., 68
(2), 1990, pp. 65-68.
62. V. V. Arslanov and W. Funke, "The Effect of Water on The Adhesion of Organic
-77-
Coatings on Aluminum", Progress in Organic Coatings, 15, 1988, pp. 355-363.
63. K. 1. Kovaleski and D. Hirst , "Evaluation of Chromate Pretreated 2024 T3
Aluminum Panels", unpublished work, 1994.
-78-
VITA
Po-Nien Chen was born to Shyan-Hyong Chen and Lo-Yu-Chun Chen on February 4,
1971 in Taipei, Taiwan, R. O. C. He received his B. E. in Materials Science from Feng
Chia University (Taichung, Taiwan, R. O. C.) in June 1993.
He worked for Department of Materials Science from 09/90 to 06/91 as a research
assistant, worked for Chung Hwa Chemical Industrial Works, Ltd. from 07/92 to 09/93
and from 09/95 to 08/96 as a research assistant, and worked for the Army of Republic of
China as a secretary. Joined the Carbon Fiber Laboratory in Feng Chia University from
02/92 to 01/93. He received the Certificate of Merit for devoting to organizing and devel
oping the Human Philosophy Club at 06/93, the Scholarship of Chung Hwa Chemical
Industrial Works, Ltd. for research works on summer vacation at 09/93, and Research
Assistantship funded by AFOSR Multidisciplinary University Research Initiative pro
gram from 03/97 to 12/98.
He joined the Materials Science and Engineering department of Lehigh University in
08/96 and joined Polymer Science and Engineering program at 01/97. In 03/97, he began
to do research under the direction ofDr. R. D. Granata.
-79-
ENDOFTITLE
