Prohesion
(cyclic salt-fog): ASTM G85
Duration: 216 hours (9 days)
Test cycle: 1-hour wet/1-hour dry
Wet cycle: A Harrison Mixture of 0.35% ammonium sulfate and 0.05%
sodium chloride was used.
Fog was introduced at ambient temperature.
Dry cycle: Air was preheated to 35°C (95°F) and then purged to the
test chamber
The volatile and pigment
content of all paint material was determined using standard American
Society for Testing and Materials (ASTM) methods. The elemental contents
of the pigments were determined by a combined scanning electron microscopy
/energy-dispersive x-ray spectrometry technique (SEM/EDS). The metallic
zinc, as well as total zinc present in the isolated primer pigments,
was quantified by titration methods. The particle size distribution
of the pigments was obtained with a particle-size analyzer that measured
the reduction in light intensity after the light passed through a
particle-suspended solution. Other physical properties, such as adhesion
strength of the coating systems and hardness of the topcoats, were
also measured.
A cyclic laboratory test
method, the Freeze/UV-condensation/ProhesionÔ test, was employed to
evaluate the performance of the three MC urethane coating systems.
All tests were conducted for a total duration of 4,000 hours (eight
500-hour test cycles). The panels were examined for surface failures
(e.g., blistering and rusting) and were measured for scribe creepage
at 500-hour intervals.
Results and Discussion
The chemical composition of the three MC urethanes varied in terms
of formulation. The difference in type of binder (isocyanate) is indicated
by the Fourier Transform Infrared (FTIR) analysis of the MC urethanes.
Table 3 shows the relative amount of aromatics to aliphatics of isocyanate
that can be estimated from the ratio of peak intensity at a wave number
of 3100 cm-1 to that of 3000 to 2800 cm-1. The primer and intermediate
coats of System A are most aromatic in nature, whereas those of System
C are generally more aliphatic. However, all topcoats are virtually
aliphatic in character and provide high UV resistance and hardness.
Table 4 shows the chemical
composition of the zinc-rich primers in terms of volatile, solid,
pigment, and binder content. This composition varied slightly among
the three systems. System A was found to contain the highest amount
of binder (presumably all isocyanate) and the least amount of pigment,
System C had the lowest amount of binder and the largest amount of
volatiles. Table 5 shows the metallic zinc and total zinc content
of the three primers. Obviously, zinc is the primary ingredient in
all the zinc-rich primers, but Primer C contained the highest amount
of it. Other elements present included magnesium, aluminum, silicon,
iron, and oxygen.
The particle-size distribution
of pigments A, B, and C is shown in figure 1. It is arranged in increasing
order (Pigment B < Pigment A < Pigment C), and the peaks occur
at 5 micrometers (mm), 3 mm, and 10 mm, respectively. This distribution
indicates that different grades of zinc particles were used in the
three different primers. The smaller the particles, the tighter the
packing of the coating film became. Proper gradation of pigment particles
resulted in a less porous film, thereby decreasing permeability and
enhancing barrier properties. In addition, tight packing of zinc particles
increased conductivity, which in turn increased their cathodic protection
to steel.
The pencil hardness (2H)
of the topcoats and the adhesion strengths (10.5 megapascals [MPa]
or 1,500 pounds per square inch) stayed fairly constant throughout
the laboratory test. After performing the pull-off adhesion test,
the coating failure modes were cohesive failure of either intermediate
coats or topcoats. There was no adhesive failure between primer and
substrate, thus demonstrating that the adhesion strength between all
primers and steel was higher than 10.5 MPa. This value is large enough
to reduce underfilm corrosion.
Throughout the 4,000-hour
laboratory test period, no surface failures were observed on any of
the MC urethane test panels. This proves that MC urethanes are excellent
barrier coatings. However, the three systems developed rust creepage
at the scribe approximately 1,500 hours after the test started. By
the end of the test, all systems developed blistering and formed a
continuous bulk surface along the scribe. Figure 2 shows the creepage
growth for the blast-cleaned SSPC SP-10 surface as a function of test
time. (Creepage values are an average of the five test panels.) It
appears that all systems exhibited slightly different amounts of creepage
at each time interval, but the rates of formation did not differ much.
The amount of creepage on the chloride-free SP-10 surface increased
in the order of B<C<A, but the difference was very small. Creepage
on the chloride-doped SP-10 surface, on the other hand, increased
in the order of B<A<C, as shown in figure 3. Creepage not only
developed earlier in the test on this kind of surface (~1,000 hours),
but also the amount of creepage was higher than that on the clean
SP-10 surfaces.
Table 6 summarizes the
overall failures that occurred by the end of the 4,000 hours. These
results suggest that System B outperformed the other two systems at
the scribe. It is speculated that the differences in scribe creepage
might have been due to the difference in binder/pigment ratios. These
ratios were calculated as 0.20, 0.15, and 0.11 (table 4) for products
A, B, and C, respectively. The zinc-rich primer in System C had the
lowest ratio, meaning that there was insufficient binder to fill the
voids between the system's large pigment particles. As a result, System
C exhibited the poorest performance at the scribe among the three
chloride-doped systems. System C's failures included large rust-filled
blisters (2D, 4D) and a 5.1-millimeter creepage that was twice as
large as the one generated on the chloride-free SP-10 surface (2.5
mm). The effect of chloride on the performance of systems A and B
was less severe than that on System C.
It is clear that the chloride-doped
test panels developed more severe blistering and a larger scribe creepage
than the chloride-free panels. Chloride contamination, even at the
20-mg/cm2 level, can have detrimental effects on coating performance.
It has been reported that epoxy zinc-rich (organic zinc-rich) coatings
can tolerate up to 30 mg/cm2 of chloride contamination without having
any significant effect on performance. However, this work suggests
that a 20-mg/cm2 level of chloride contamination is large enough to
reduce the corrosion resistance of MC urethanes at the scribe on new
steel even though primers contain more than 80 weight-percent of zinc.
Table 3 — FTIR Aromatics/Aliphatics
Peak Ratio of MC Urethanes
System |
Primer |
Intermediate Coat Topcoat |
A |
0.036 |
0.042 |
0 |
B |
0.020 |
0.008 |
0 |
C |
0.015 |
0.015 |
0.006 |
Table 4 — Chemical Composition
of Zinc-Rich MC Urethane Primers
Content |
A |
B |
C |
Weight
Percent |
Volatiles1 |
10.8 |
9.3 |
13.4 |
Solid2 |
89.2 |
90.7 |
86.6 |
Pigment3 |
73.7 |
78.8 |
77.8 |
Binder4 |
15.5 |
11.9 |
8.8 |
________________________________________
1 ASTM D2369.
2 100 wt% - wt% Volatiles.
3 ASTM D2371.
4 100 wt% - wt% Volatiles - wt% Pigment.
Table 5 — Zinc Content
of MC Urethane Primers by Titration
Determination |
A |
B |
C |
Weight
Percent |
Metallic zinc1 |
72.8 |
75.9 |
80.1 |
Total zinc1 |
78.8 |
80.4 |
86.8 |
Total zinc2 |
80 |
83 |
89 |
_________________________________________
1 ASTM D521.
2 Value on product data sheet.
Table 6 — Scribe Failure
Results of MC Urethane Systems on SP-10 and Chloride-Doped SP-10 Surfaces
After a 4,000-Hour Test
________________________________________ 1 ASTM D714.
2 Chloride-doped surface
After 24 months of outdoor
exposure to a marine environment, none of the three coating systems
exhibited failures other than rust at the scribe. Although no data
is available on field failure as of yet, a good correlation is expected
to exist between laboratory and field test results.
Summary and Conclusions
The three MC urethane systems tested showed strong mechanical properties,
such as adhesion strength of the whole coating system and hardness
of the topcoats. The chemical composition of the three coating systems
varied in terms of type and amount of binders and pigments.
In general, no surface
failures were found on any of the test panels after the 4,000-hour
accelerated laboratory test. This demonstrates that MC urethane systems
are excellent barrier coatings for steel. However, the three systems
applied on SSPC SP-10 steel surfaces developed different degrees
of scribe creepage at the end of the cyclic laboratory test. Scribe
creepage measurements are quite sensitive techniques used to differentiate
coating performance prior to surface failure. As expected, the coating
systems that were applied to the 20-mg/cm2 chloride-doped SSPC SP-10
steel surfaces developed more creepage than those on the chloride-free
steel surfaces. The failure at the scribe suggests that the high degree
of cathodic protection that zinc-rich MC urethane systems provide
is not as high as that of inorganic zinc silicate coatings. After
two years of outdoor exposure, all MC urethane coating systems remained
in excellent condition and did not exhibit any failure.
The performance of the
MC urethane coating systems on steel is likely to be affected by different
chemical compositions, including the aromatic/aliphatic content ratio
of the MC urethane binders, binder/ pigment content ratio, and particle-size
distribution of pigment. Insufficient binder content and overly large
pigment particles would result in early coating failure. It is believed
that zinc grade and particle distribution play a more important role
in primer performance than zinc content. Material testing and product
quality control practices are necessary prior to coating application.
References
- Personal communication between
the authors and the Oregon Department of Transportation.
- J. Schwindt. "Moisture-Cured, Polyurethane-Based,
Surface Tolerant Coatings: An Economical Alternative for Corrosion
Control," Material Performance, p. 25, December 1996.
- S-L. Chong, M. Jacoby, J. Boone,
and H. Lum. Comparison of Laboratory Testing Methods for Bridge
Coatings, FHWA Publication No. FHWA-RD-94-112, p. 67, June 1995.
- S-L. Chong. "A Comparison of Accelerated
Tests for Steel Bridge Coatings in Marine Environment," JPCL,
14, p. 20, March 1997.
- C. Hare. "Protective Coatings —
Fundamentals of Chemistry and Composition," SSPC 94-17, p. 514
1994.
- B. Appleman, S. Boocock, R. Weaver,
and G. Soltz. Effect of Surface Contaminants on Coating Life,
FHWA Publication No. FHWA-RD-91-011, p. 242, November 1991.
Shuang-Ling Chong
is a research chemist in the Special Projects and Engineering Division,
Office of Engineering R&D, Federal Highway Administration. Her 29
years of research experience covers photolysis, ion-molecule reactions,
fractionation and characterization of organic materials in fossil
fuels and petroleum, and identification of toxic organics and metals
in coal-combustion residues. Dr. Chong has been conducting staff research
since 1989 in paint testing, coating failure analysis, and evaluation
of low volatile-organic-compound coating systems for steel bridges.
Yuan Yao is a chemist
at Soil and Land Use Technology Inc., an on-site contractor at the
Federal Highway Administration. She earned an M.S. degree in chemistry
and her past experience includes waste water analysis.