NCER Assistance
Agreement Final Report Executive Summary
Date of Final Report: January 31, 2009
EPA Agreement Number: X-83254101-1
Center: Center for Environmental and Energy Research (CEER)
Project Title: Magnesium
Rich Coatings for Corrosion Control of Reactive Metal Alloys
Investigator(s): Rebecca
DeRosa
Institution(s) of PI(s): Alfred University
Research Category: Congressionally Mandated Center
Project Period: September
1, 2006 – October 30, 2008
Description and Objective
of Project:
The main objective of this work is to understand specifically the role of magnesium particles as sacrificial anodes in magnesium rich primers (MRPs) of different formulations (organic and inorganic) on 2024-T3 and 7075-T6 in different corrosive environments. The corrosion protection of MRPs received from North Dakota State University (NDSU) were evaluated using: open circuit potential (OCP), potentiodynamic polarization analysis, and electrochemical impedance spectroscopy (EIS). The changes in electrochemistry are correlated to the chemical and physical changes of the MRPs using: thickness measurements, X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM), energy dispersive spectroscopy (EDS), and wavelength dispersive spectroscopy (WDS).
1.
MRPs Tested
Three
different primer formulations on two different aluminum panel types were
investigated. The first
formulation is an epoxy polyamide magnesium rich primer (CPM1040) formulated at
45% pigment volume concentration (PVC) on Al-2024-T3 and Al-7075-T6
panels. CPM1040 is an in-house MRP
formulated by NDSU. The formulation
consisted of Mg particles with an average diameter of 30–40 μm. The coating was spray applied to an
optimal thickness of 150 ± 40 mm.
The
second formulation is a commercial epoxy polyamide magnesium rich primer (Akzo)
formulated at 45 % PVC on Al-2024-T3 and Al-7075-T6 panels. The primer also contains additional
additives to aid in the dispersion of Mg particles. The coating was spray applied to a nominal thickness of 47 ± 12 mm. The magnesium rich primer was prepared using Mg
particles, with an average diameter of 30–40 μm, manufactured by
Ecka Granules, Salzburg, Austria.
The
third formulation is a silane-based inorganic magnesium rich primer with
plasticizer (SMT30) formulated at 30 % PVC on Al-2024-T3. The coating was spray applied to a
nominal thickness of 112 ± 6 mm. r of
30–40 μm, manufactured by Ecka Granules, Salzburg, Austria.
2.
Exposure Conditions
Samples were
analyzed during and after either A) constant immersion in dilute HarrisonÕs
solution (DHS-CI) or B) 90 ± 6.5 % relative humidity (90 % RH). DHS-CI consisted of 0.35 wt% (NH4)2SO4
and 0.05 wt% NaCl in distilled water (pH =5.4). The 90% RH exposure consisted of a saturated salt solution
containing equal amounts of (NH4)2SO4 and NaCl
in distilled water (pH =3.9). For 90 % RH exposure; a set of samples was
exposed without a defect and another set was scribed. For the scribed samples an X was scribed though each primer
with a razor blade prior to exposure.
The data from each exposure type were compared to as-received samples.
Each
bare aluminum panel type and respective primers was exposed to both the DHS-CI
and 90% RH exposures and samples were evaluated using electrochemical
techniques, SEM, XPS, EDS, and WDS during exposure times ranging from the
initial start of exposure to over 6 months.
Summary of Findings:
1. Open Circuit Potential (OCP)
Each MRP system provided initial cathodic protection to the aluminum
substrates, with OCP values ranging (-1.0 V to -1.5 V) versus. saturated calomel
electrode (SCE). When the OCP for
an MRP reaches the OCP of bare aluminum (baseline), ~ -0.5 to -0.6 V versus
SCE, the magnesium particles in the MRP are no longer providing cathodic
protection. A compiled list of
when each primer formulation has reached the baseline according to exposure
type is provided in Table 1. With
respect to OCP measurements, it is apparent that there is a distinct difference
in the behavior of the coating with respect to the exposure method. OCP curves for the 90%RH samples
indicate that the samples have some passivating/barrier layer protection
whereas the DHS-CI samples show no indication of a barrier layer.
Table 1. Time to Reach Baseline
(Days) for Eeach Primer Formulation and Exposure Type
|
Formulation |
Time
to reach baseline (days) DHS-CI |
Time
to reach baseline (days) scribed 90 % RH |
Time
to reach baseline (days) non-scribed 90 % RH |
|
CPM1040
2024 |
25 |
1 |
45 |
|
CPM1040
7075 |
37 |
1 |
85 |
|
Akzo
2024 |
10 |
7 |
85+ |
|
Akzo
7075 |
18 |
100 |
85+ |
|
SMT30
2024 |
8 |
6 |
85 |
2. Potentiodynamic Polarization Curves
Potentiodynamic scans provide information about 1) the rate of
corrosion, i.e. reaction of the Mg particles through current density, and 2)
cathodic protection, through corrosion potential. A high current density with a low potential indicates that
Mg particles are abundant and available to provide sacrificial protection. Therefore, samples that show high
current density and low potential in theory should provide substantial cathodic
protection. For MRPs on 2024-T3, CPM1040
provided the best cathodic protection (lowest potential), followed by Akzo and
lastly SMT30. CPM1040 has the
highest current density, therefore it is reacting the fastest with DHS-CI due
to the high amount of available Mg on the primer surface. The abundance of Mg particles was later
confirmed by XPS, SEM, EDS and WDS.
Akzo and SMT30 have lower current densities than CPM1040, indicating not
as much available Mg at the primer surface, confirmed as well by XPS, SEM, EDS
and WDS.
For MRPs on 7075-T6, Akzo provides the
best initial cathodic protection, followed by CPM1040. The difference in potential is not
much, however CPM1040 displays a much higher (one order of magnitude) current
density, indicating more available Mg at the primer surface compared to Akzo
confirmed by XPS, SEM, EDS and WDS.
3. Thickness Measurements
DHS-CI
exposure resulted in a decrease in primer thickness with increasing exposure
time for all MRP formulations due to Mg particles dissolving out of the coating.
The loss of Mg particles was
confirmed by SEM, EDS and WDS. 90
% RH exposure showed an increase in thickness with increasing exposure time for
all MRP formulations due to the lack of Mg dissolution and the build up of
corrosion products surrounding the Mg particles confirmed by SEM and EDS.
4. X-ray Photoelectron Spectroscopy (XPS)
XPS
was used to evaluate the as-received primers and corrosion products present
after exposure. CPM1040 on 2024-T3
and 7075-T3 as-received displayed 60 % atomic concentration of MgO available at
the sample surface. Akzo on
2024-T3 and 7075-T6 as-received displayed only ~ 5 % atomic concentration of
available MgO at the sample surface.
SMT30 on 2024-T3 as-received displayed 10 % atomic concentration of MgO
available at the sample surface.
DHS-CI caused the formation of Mg(OH)2 for all MRP
formulations. However DHS-CI did
not cause the formation of MgCO3 for any MRP formulation, due to the
lack of available CO2 during exposure. After 24 hours of DHS-CI the atomic concentration of MgO
dropped below 10 % at the sample surface for CPM1040, below 2 % for Akzo and
SMT30, indicating Mg dissolution.
This is confirmed by SEM, EDS and WDS.
90
% RH exposure allowed for the formation of MgCO3 due to the
availability of CO2.
The formation of MgCO3 varied for each MRP. CPM1040 on 2024-T3 and 7075-T6
displayed MgCO3 formation after the first hour of exposure. Akzo on 2024-T3 and 7075-T6 displayed
MgCO3 formation after 600 hours of 90 % RH exposure. SMT30 on 2024-T3 displayed MgCO3
formation after 48 hours of 90 % RH exposure. The formation of carbonates during 90 % RH exposure is
important due to their passivating properties providing an additional form of
corrosion protection. Of
importance is that samples under DHS-CI exposure do not show the formation of
carbonates due to the lack of CO2. Therefore, the environment used for accelerated testing is
an integral part of determining a more accurate depiction of how the coating
will behave in the field.
5. Scanning Electron Microscopy (SEM)
SEM was used to determine two specific characteristics associated with
the primers. First, we were able
to look at particle connectivity in the as-received samples. CPM1040 displayed the best particle
connectivity and distribution. Akzo
did not show as high degree of particle connectivity compared to CPM1040. SMT30 displayed the lowest amount of
particle connectivity and also had several voids throughout the primer.
Second, we were able to monitor the
dissolution of Mg particles from the coatings with respect to exposure
condition. DHS-CI exposure caused
the Mg particles to dissolve out of the MRPs, leaving behind little to no
corrosion products on the sample surface, a depleted Mg layer, and a porous
structure that facilitated faster diffusion of corrosive species to the
aluminum substrate. Alternatively,
magnesium particles did not dissolve out of the MRPs when using our 90 % RH
exposure. Corrosion products from
magnesium oxidation were able to build up on the surface as indicated from XPS
analysis. In this case, 90 % RH
can be considered a more realistic exposure condition than DHS-CI or any other
constant immersion technique. It
is more likely that aircraft will encounter high humidity with the presence of
salt, as in marine environments, rather than pooled water or immersion as
emulated by the DHS-CI conditions.
6. Energy
Dispersive Spectroscopy (EDS) & Wavelength Dispersive Spectroscopy (WDS)
EDS and WDS results further reinforce
the variation in the MRPs response to different exposure types. Samples exposed to DHS-CI displayed a
thick oxide layer surrounding the Mg particles that remained in the
primer. By comparing EDS and WDS
maps of samples as-received, 24 hours and 480 hours DHS-CI exposure, Mg
particles are seen to develop a thick oxide layer, shrink in size and
eventually dissolve into solution.
90
% RH did not cause the Mg particles to dissolve out of the primer, confirmed by
EDS and WDS. During 90 % RH
exposure the initial oxidation of the Mg particles at the sample surface is
seen. Increased exposure time
causes the oxide layer to grow surrounding the Mg particles and penetrate
deeper into the primer eventually creating an oxide layer that surrounds a
majority of the particles in the primer.
MRP SMT30 was the only primer to show an Mg based oxide layer that
re-deposited on the sample surface after 480 hours exposure. The formation of carbonates at the
sample surface was not detected by EDS and WDS.
Even
though OCP results may indicate that an MRP has reached the Al baseline, Mg
particles may still remain in the primer according to XPS, SEM, EDS, and
WDS. This result is further
reinforced by CPM1040 samples that were scribed after 900+ hours of DHS-CI. Upon scribing, the OCP of the primer
dropped from the baseline to below -1.0 V. This proves that Mg particles still remain in the primer,
however their activity is not detected by OCP measurements. Unless a mechanical defect exposes the
Mg particles, they are essentially inactive.
Conclusions:
All
of the MRPs provided initial cathodic protection. Due to the complexity of these systems, it is not possible
to say one MRP provided the best overall protection regardless of exposure
environment. However, it is
apparent that the environment to which MRPs were exposed played a distinct role
on the primersÕ responses.
Electrochemical
analysis using OCP indicated rapid deterioration of the cathodic protection of
the DHS-CI samples with no passivating layer forming to provide extra
protection. In contrast, OCP of
90% RH samples indicated the formation of a passivating layer, which was later
confirmed by XPS to be a carbonate species.
XPS,
EDS and WDS were used to monitor the particle connectivity and dissolution with
respect to exposure condition. If
we look at the amount of Mg remaining in the primer systems after 480 hours of
exposure to DHS-CI the MRPs are ranked as follows: 1) CPM1040 on 7075-T6, 2)
CPM1040 on 2024-T3, 3) SMT30 on 2024-T3, 4) Akzo on 7075-T6, 5) Akzo on
2024-T3. This ranking almost
directly correlates to OCP results with the exception of SMT30 on 2024-T3.
The
primers exposed to 90 % RH did not show Mg dissolution. However, all three formulations showed
MgCO3 formation when exposed to 90 % RH. The time required for the formation of MgCO3
varied with respect to the type of primer. Carbonates formed the fastest on CPM1040, followed by SMT30
and lastly Akzo. Carbonate
formation due to 90 % RH may be the reason why the OCP trends are essentially
opposite to DHS-CI exposure. The
only formulation to show an Mg based layer that re-deposited on the primer
surface was SMT30 on 2024-T3.
Publications/Presentations:
R.L. DeRosa, I. Szabo, D.
Battocchi, and G.P. Bierwagen, ÒAssessing the Role of Magnesium in Magnesium
Rich CoatingsÓ, proceedings of the FSCT, 2007 FutureCoat, Toronto, Ont., Oct
3-5 2007.
I. Szabo and R.L. DeRosa,
ÒSurface Behavior of Magnesium Rich CoatingsÓ, poster presented at FSCT, 2007
FutureCoat, Toronto, Ont., October 4, 2007.
Supplemental Key Words: corrosion, metal rich coating, magnesium rich primer,
aluminum alloy, magnesium alloy
Relevant Web Sites:
http://www.estcp.org/Technology/WP-0731-FS.cfm
http://www.ndsu.nodak.edu/research/article.php?article_number=75
http://ceer.alfred.edu