Вы находитесь на странице: 1из 6

Journal of The Electrochemical Society, 147 (10) 3661-3666 (2000) 3661

S0013-4651(00)01-028-4 CCC: $7.00 © The Electrochemical Society, Inc.

A Scanning Vibrating Electrode Study of Chromated-Epoxy Primer on


Steel and Aluminum
Jie He,a Victoria Johnston Gelling,a Dennis E. Tallman,a,*,z and Gordon P. Bierwagenb,*
aDepartment of Chemistry and bDepartment of Polymers and Coatings, North Dakota State University, Fargo,
North Dakota 58105, USA

Chromated-epoxy primers are often used for corrosion control of iron and aluminum structural alloys. However, due to environ-
mental concerns and adverse health effects surrounding such use of chromates, there is an intensive effort to find suitable replace-
ments for chromate-based coatings. This effort is hampered by lack of a detailed understanding of how the chromate-based coat-
ings function. In this work, the scanning vibrating electrode technique (SVET), also known as the current density probe, was used
to measure current flowing at chromated-epoxy coated steel and aluminum substrates. The coatings were scribed to simulate a
defect through the coating to the metal substrate surface. A plain (nonchromated) epoxy coating was also studied for comparison.
The SVET was used to map the current flowing in and around the defect while the sample was immersed in either 3% NaCl (steel)
or in dilute Harrison solution (aluminum). Both steel and aluminum substrates coated with the chromated epoxy exhibited a sig-
nificant delay before the onset of corrosion within the scribe, compared with substrates coated with nonchromated epoxy. Fur-
thermore, the current density maps for steel suggest that the reduction reaction may occur at the surface of the chromated-epoxy
coating. With the nonchromated-epoxy coating, the reduction reaction was always confined to the defect area.
© 2000 The Electrochemical Society. S0013-4651(00)01-028-4. All rights reserved.

Manuscript submitted January 10, 2000; revised manuscript received June 19, 2000.

Historically, coatings designed to protect structural metals such as Products). The chromated-epoxy coating was a high solids primer
steel and aluminum from corrosion have incorporated slightly soluble coating (Mil-P-23377G, DEFT, Inc.). Cold-rolled steel (Q-Panel) or
lead and chromate-based pigments, such as zinc chromate, zinc aluminum alloy (Al 2024-T3, Q-Panel) samples were first polished
tetroxy chromate, strontium chromate, and basic lead silicochro- using 600 grit silicon carbide, washed with hexane, and air-dried.
mate.1 For example, the United States Air Force uses chromate-rich After this pretreatment, the corresponding coatings were applied by
primers for corrosion control on most of its aircraft, which typically bar coating and were allowed to dry overnight. The average coating
are made of aluminum 2024-T3 alloy.2 These primers are typically thickness for the plain epoxy was 30 6 8.5 mm, for the chromated-
used in combination with a surface pretreatment employing Alo- epoxy 21 6 4.5 mm. The coating of each sample contained a defect
dine™, a solution containing chromate and other components such as that was introduced by scribing, the area of the defect ranging from
sodium fluoborate, potassium ferricyanide, and potassium fluozir- 0.1 to 0.3 mm2. No conversion coatings and no topcoats were used
conate. Although these chromate-based systems are highly effective in this study.
for corrosion protection, their use will soon become sharply curtailed For the Al 2024-T3 samples, the immersion solution was dilute
due to new government regulations concerning both protection of the Harrison solution, containing 0.35% (NH4)2SO4 and 0.05% NaCl in
environment and protection of industrial workers.1,3 Thus, an inten- H2O. For the cold-rolled steel samples, the immersion solution was
sive effort is underway to develop suitable replacements for chro- 3% NaCl solution. These solutions were prepared from reagent
mate-based coatings.4 This effort is hampered by lack of a detailed grade salts and distilled water.
understanding of how the chromate-based coatings function.4 Instrumentation.—SVET permits the mapping of current density
In this work, the scanning vibrating electrode technique in an electrolyte immediately above a substrate surface. The instru-
(SVET),5,6 also known as the current density probe (CDP), was used ment used in this study was from Applicable Electronics (Forestdale,
to measure current flowing at chromated-epoxy coated steel and alu- MA). A microelectrode, an electrochemically etched Pt/Ir
minum substrates. The coatings were scribed to simulate a defect (80%/20%) wire coated with parylene and arced at the tip to expose
through the coating to the metal substrate surface. A plain (nonchro- the metal, is platinized to form a small ball of platinum black at the
mated) epoxy coating was also studied for comparison. The SVET tip, approximately 20 mm in diameter in this study. This electrode or
was used to map the current flowing in and around the defect while probe is vibrated in two dimensions (parallel and perpendicular to
the sample was immersed in either 3% NaCl (steel) or in dilute Har- the surface, approximately one tip diameter in each dimension) as
rison solution (aluminum). the tip is rastered above the surface of the substrate at a height of ca.
The aim of this work was twofold. First, the use of the SVET to 100 mm. Current flow in the electrolyte due (for example) to corro-
study chromated-epoxy primers does not appear to have been report- sion processes occurring on the substrate surface results in a voltage
ed previously. Thus, it was anticipated that these experiments would (IR) drop within the electrolyte. This voltage drop is sensed by the
provide additional insight into the mechanism of protection of struc- vibrating probe and resolved by two phase-sensitive detectors into a
tural metals by such primers. Second, in a companion paper we voltage vector having magnitude and direction. The voltage vector is
report the results of SVET measurements on a conducting polymer converted into a current density vector using calibration data gener-
coating, poly(3-octyl pyrrole), on steel and aluminum.7 The results ated in a separate experiment (calibration involves generation of a
described here for the chromated-epoxy primer provide a foundation known current density from a point source calibration electrode).
for discussing the similarities and differences in behavior of the con- The measured current density vectors permit mapping of both the
ducting polymer system. magnitude and direction of current flow immediately above the sub-
Experimental strate surface. This current density map can be overlaid onto an opti-
Sample preparation and experimental conditions.—The plain cal micrograph of the substrate captured by a video microscope. This
epoxy coating was a polymer of epichlorohydrin and bisphenol A capability permits correlation of anodic and cathodic current pro-
(Dow Chemical D.E.R. 331) with Ancamide 2353 curing agent (Air cesses with visual features on the substrate surface. The SVET is
capable of providing detailed spatially resolved information not
* Electrochemical Society Active Member. readily obtainable by other techniques.
z E-mail: dtallman@plains.nodak.edu
3662 Journal of The Electrochemical Society, 147 (10) 3661-3666 (2000)
S0013-4651(00)01-028-4 CCC: $7.00 © The Electrochemical Society, Inc.

Samples for SVET measurement were cut into 1 3 1 cm squares scanning probe. With Al 2024-T3, the activity was much more
and were masked by a Polyester 5 adhesive tape (3M Company) such dynamic, with anodic sites appearing (onset of pitting corrosion) and
that only a 2 3 2 mm square opening of the sample was exposed. The then disappearing (passivation) on a time scale of tens of minutes. Of
artificial defect was made using a scribing tool, the sample was importance to the following discussion is the observation that corro-
mounted in a Teflon sample cell, and ca. 5 mL of the appropriate sion onset at these bare metal samples was virtually immediate, with
immersion solution was added. Scans were initiated within 5 min of the observation of significant current flow during the first SVET
immersion and were collected every 20 min for the duration of the scan, which was initiated within 5 min of immersion.
experiment, typically 20 h. Each scan consisted of 400 data points As an additional control experiment, current density maps were
obtained on a 20 3 20 grid, with an integration time of 1 s per point. obtained for steel and Al alloy samples coated with a plain (nonchro-
A complete scan required 10 min, followed by a 10 min rest period mated) epoxy primer. A defect was introduced by scribing through
prior to the next scan. The current density maps of this paper are dis- the coating to the bare metal surface. The results of these experi-
played in two ways. In one method, the normal or z-component of the ments are exemplified by the result for steel shown in Fig. 1. Clear-
measured current density in the plane of the vibrating electrode is ly visible in the micrograph of Fig. 1 is the 2 3 2 mm scan area
plotted in 3D format over the scan area, with positive and negative defined by the Polyester 5 adhesive tape. The defect, which is paral-
current densities representing anodic and cathodic regions, respec- lel to the x axis, is visible within this scan area. From Fig. 1, it is
tively. In the other method, vectors representing current density mag- clear that the major current flow is from a single anodic site to a sin-
nitude and direction are superimposed onto an optical image of the gle cathodic site, each located within the defect. This confirms that
immersed sample. In all cases, the bottom edge of the optical micro- the coating is effectively removed by scribing, exposing the bare
graph corresponds to the x axis of the 3D plot. metal. In every such control experiment for steel and aluminum,
The measurements were taken at the open-circuit potential. At similar behavior was observed, with oxidation and reduction cur-
least six specimens of each sample type were prepared and scanned rents confined to the defect. With steel, the onset of corrosion with-
to assess reproducibility of the observed phenomena. In each case, in the defect was very rapid, typically observed in the first scan (i.e.,
representative scans are presented. The background noise in our within 5 min). With aluminum, the onset of corrosion within the
SVET instrumentation was estimated to be 3 mA/cm2, determined defect was somewhat more variable but was typically observed by
from the standard deviation of the variations in current density at the second scan (i.e., within 20 min). For the steel sample of Fig. 1,
sufficiently large distance (ca. 1 mm) from the point source calibra- the current slowly diminished toward zero over the course of approx-
tion electrode passing 60 nA of current. imately 3.5 h, at which point the leftmost two-thirds of the defect
Results and Discussion
The mapping of current density.—The electrochemical tech-
niques commonly used in corrosion studies (for example, linear
polarization, electrochemical impedance spectroscopy, or electro-
chemical noise methods) provide data that reflect an average re-
sponse over the entire sample surface. While such data is very use-
ful, no local or spatial information is obtained. A number of tech-
niques have been developed recently which permit mapping of local
current flows in the electrolyte above an immersed substrate, pro-
viding a more detailed view of corrosion processes. Typically, these
techniques map the local potential field that arises due to current
flow through the resistive electrolyte and the potential map is then
converted to a current density map by a calibration process.
The scanning reference electrode technique (SRET) as originally
implemented maps the potential by scanning two microreference
electrodes near the surface of the substrate under immersion.8 More
recent versions of the SRET use a single vibrating microelectrode,
leading to improvements in spatial resolution.9 This technique is
often referred to as SVET, although there are several variations in
how the electrode is actually vibrated.9 These techniques are limited
to measuring dc currents which restricts their application to the study
of bare metal substrates or coated metals having coating defects. 10,11
Local electrochemical impedance spectroscopy (LEIS) uses a
twin microelectrode probe to measure the ac current flowing near the
surface of a substrate that results from an ac potential perturbation
applied to the substrate.12 Thus, LEIS is applicable to both bare and
coated metals under immersion,13,14 but the requirement of a twin
microelectrode probe sacrifices some spatial resolution. Recent vari-
ations of the SVET permit local impedance measurements using the
single vibrating electrode.15,16
Control experiments.—In this work, the current density in and
around a defect in a chromated-epoxy coating applied either to cold-
rolled steel (and immersed in 3% NaCl) or to aluminum 2024-T3
(and immersed in dilute Harrison solution) was studied using the
SVET. Control experiments involving bare cold-rolled steel and bare
Al 2024-T3 in the corresponding immersion solutions were also
conducted. In every case, these bare metals exhibited one or more (in
the case of the Al alloy, typically several) regions of anodic activity, Figure 1. Current density maps for the plain epoxy primer on steel at 5 min
reflecting sites of localized corrosion, with current densities of 100- immersion in 3% NaCl: (top) 3D representation of the z component of the
200 mA/cm2. With steel, the corrosion often continued until the current density and (bottom) optical micrograph of the sample with current
buildup of corrosion products at a particular site led to a crash of the density vectors superimposed.
Journal of The Electrochemical Society, 147 (10) 3661-3666 (2000) 3663
S0013-4651(00)01-028-4 CCC: $7.00 © The Electrochemical Society, Inc.

area was covered with corrosion product (Fig. 2). The cathodic site formly cathodic), with little or no reduction current within the defect.
remained free of corrosion product throughout the experiment. With the plain epoxy control experiments, the total anodic current and
We note that Kinlen and co-workers recently conducted a similar total cathodic current flowing above the defect were approximately
SRET experiment at epoxy-coated carbon steel with a 1 mm diam equal. In these chromated-epoxy experiments, the anodic current was
pinhole drilled through the coating to the metal surface. They also always greater. We conclude that reduction in these experiments did
observed anodic and cathodic reactions occurring within the defect. indeed occur over the chromated-epoxy coating.
The onset of corrosion was immediate and the corrosion rate did not There are two reduction mechanisms that might explain these ob-
appear to diminish over the 8 h duration of the experiment. servations. In one mechanism, the reduction reaction could occur at
Chromated-epoxy primer on steel.—Two differences were the coating-electrolyte interface, with electron transfer to the under-
observed in the experiments involving chromated-epoxy primer on lying metal mediated by the chromate pigment in the epoxy coating.
steel. First, whereas large currents were observed within 5 min in the In the other mechanism, the reduction reaction could occur at the
control experiments, there was a significant delay of 40 min or metal-coating interface, the result of diffusion of the oxidant through
longer before the onset of any observable corrosion current in these the coating to the metal surface. Both mechanisms would give rise to
experiments. Second, significant reduction current on the exposed a cathodic current flow above the coating surface.
metal in the defect, as always observed in the control experiments, We favor the first mechanism for the following reasons. First, the
was not observed in these experiments. reduction current always appeared to be rather uniformly distributed
Figure 3 shows a sample after 169 min of immersion (the first across the coating surface. Reduction at the metal-coating interface
observable current for this particular sample was at 42 min immer- would likely be more localized, with transport (e.g., of oxygen) to
sion). Oxidation was clearly occurring within the defect and corrosion the metal interface occurring primarily through coating defects. Sec-
product is visible at the anodic site in the optical micrograph. How- ond, the first occurrence of observable oxidation within the defect at
ever, little if any reduction occurred within the defect. Rather, reduc- ca. 40 min is accompanied by uniform reduction over the coating
tion appeared to occur more or less uniformly across the coating sur- surface, a time we consider too short for uniform penetration of
face. Since the reduction current was spread over a relatively large water and oxidants through the coating to the metal interface.18 It is
area, the current density at any one point was rather small, similar in noted, however, that the pigment volume concentration of this high-
magnitude to the noise level. Therefore, we repeated this experiment
eight times to confirm this general behavior. In every case the surface
of the coating was uniformly depressed below zero current (i.e., uni-

Figure 2. Current density maps for the plain epoxy primer on steel at 192
min immersion in 3% NaCl: (top) 3D representation of the z component of Figure 3. Current density maps for chromated-epoxy primer on steel at
the current density and (bottom) optical micrograph of the sample (current 169 min immersion in 3% NaCl: (top) 3D representation of the z component
density vectors omitted for clarity). Corrosion product is visible over much of the current density and (bottom) optical micrograph of the sample with
of the defect area. current density vectors superimposed.
3664 Journal of The Electrochemical Society, 147 (10) 3661-3666 (2000)
S0013-4651(00)01-028-4 CCC: $7.00 © The Electrochemical Society, Inc.

ly pigmented primer (ca. 40-45%)19 is close to the critical pigment sion is displayed in the 5.5 h map (Fig. 4), with both oxidation and
volume concentration (ca. 50%), and pathways of conductivity due reduction being confined to the defect area. As with the steel sam-
to electrolyte channels may form in the partial porosity that occurs ples, this corrosion current continued to increase over the next sev-
in such coatings.20 Finally, if the mechanism involves simple diffu- eral hours of immersion, and most of the exposed metal area in the
sion through the coating and reduction at the underlying metal, then defect turned dark. Figure 5 shows the current density maps for the
similar behavior might be expected for an aluminum alloy. Howev- same sample after 17.5 h immersion, by which time the current had
er, at aluminum alloy the reduction occurs preferentially within the increased substantially but still appeared to be confined to the defect
defect (see the following section). This suggests a more metal-spe- area. Two local cathode sites, one on each side of the anode site, are
cific mechanism, such as might occur if coupling of electron trans- observed. As noted in the previous section, this behavior is in con-
fer between metal and coating were involved. trast to that on steel where cathodic currents above the coating were
If reduction is occurring at the coating-electrolyte interface, the observed. The site of corrosion is clearly visible in Fig. 5 as the re-
implications are twofold. First, the chromated-epoxy coating must maining shiny metal spot in the micrograph.
be capable of shuttling electrons from the metal to an oxidant in Long-term immersion experiments were conducted to determine
solution. That is, the coating would need to be electronically con- the time required for the corrosion current to diminish to background
ducting or semiconducting. The mechanism of conduction in this levels. For the plain epoxy control experiment, the average time re-
coating is unclear but could involve electron transfer between con- quired was approximately 16 h. For the chromated-epoxy coated
tacting chromate particles within the coating, similar to that occur- samples, the average time required was greater than 24 h. As for
ring in redox polymers such as poly(vinylferrocene).21 Experiments steel, the presence of the oxidant chromate appeared to sustain the
are underway to assess the conductivity of the chromated-epoxy corrosion current.
coating. Second, cathodic delamination of a coating is generally In a related study, the SRET was used to investigate a scratched
associated with a high local hydroxide concentration generated by chromated-epoxy primer on galvanized steel immersed in an artifi-
the reduction reaction occurring at the coating-metal interface.22,23 cial acid rain solution.25 In that work, as in this, the chromate-con-
Removal of the reduction reaction from the metal interface may be taining epoxy coating delayed the onset of corrosion within the
an important aspect of the mechanism by which chromated coatings defect compared to a chromate-free epoxy coating. Once corrosion
extend the lifetime of the complete coating system. A mechanism of ensued, the chromated-epoxy sample displayed a more rapid passi-
this type has been proposed for conducting polymers24 and, indeed, vation, requiring 2-3 days under their immersion conditions. It was
reduction reactions at poly(3-octyl pyrrole) coatings7 and at polyani- suggested that the chromated-epoxy coating released chromate ions
line coatings17 on steel have been observed.
In these experiments, we observed no tendency for the chromat-
ed-epoxy primer to rapidly passivate the steel. In fact, once corrosion
current began to flow (following the initial delay), it continued for
18 to 20 h, until corrosion products eventually covered the anodic
site and diminished the current to near background levels. In con-
trast, the current at plain epoxy coated steel diminished after only 3-
4 h (Fig. 2). This suggests that the presence of the oxidant (chro-
mate) in the epoxy primer actually sustains the current under our
immersion conditions. Very similar behavior was observed for a
polypyrrole coating (also an oxidant) on steel.7 An additional obser-
vation may be relevant here. The corrosion products formed at the
defect of the plain epoxy-coated steel appeared to be more volumi-
nous and less adherent, with considerable suspended matter forming
in the immersion solution during the course of the experiment. The
products formed at the defect of the chromated-epoxy coated steel
appeared to be much more adherent, with little or no suspended
material appearing in the immersion solution in spite of the consid-
erably greater time over which oxidation occurred.
An insulating topcoat such as polyurethane would be expected to
block the reduction reaction observed at the chromated-epoxy coat-
ing surface, forcing the reduction process to occur at the defect. In
preliminary experiments with chromated-epoxy samples prepared as
described previously but with a ca. 50 mm thick polyurethane top-
coat and a defect introduced through both coatings to the metal sur-
face, the oxidation and reduction reactions were indeed confined to
the defect. Importantly, there was still a delay of approximately 1 h
before observation of any redox activity, a delay similar to that
observed without the topcoat. Further studies of topcoated substrates
are in progress.

Chromated-epoxy primer on Al 2024-T3.—The chromated-


epoxy primer on Al 2024-T3 alloy, as on steel, significantly retard-
ed the onset of corrosion. Whereas corrosion activity was observed
within ca. 20 min for the plain epoxy control samples, the first
detectable current with the chromated-epoxy samples was typically
after 5 h of immersion. The results are exemplified by Fig. 4, which
shows the current density maps for a particular sample after 5 min
immersion and after 5.5 h immersion. The 5 min map is typical of Figure 4. Current density maps for chromated-epoxy primer on Al 2024-T3
those obtained over the first 5 h of immersion, showing no detectable alloy at (top) 5 min and (bottom) 5.5 h immersion in dilute Harrison solution
corrosion of the metal exposed in the defect. The exposed metal in (3D representation of the z component of the current density). See Fig. 5 for
the defect remained shiny during this period. The first sign of corro- the optical micrograph of this sample.
Journal of The Electrochemical Society, 147 (10) 3661-3666 (2000) 3665
S0013-4651(00)01-028-4 CCC: $7.00 © The Electrochemical Society, Inc.

into the defect site, promoting passivation of the exposed metal. initial stages of immersion and these mechanisms may play a role in
However, except at very low pH, the leaching of chromate from the initial delay to the onset of corrosion observed at both alloys of
epoxy coatings appears to be a relatively slow process.25,26 Such a this work.
process probably plays only a minor role over the time course of our The sustained oxidation current observed (after the initial delay)
experiments, which typically are 20-30 h in duration. It is also pos- at the defect on chromated-epoxy coated steel and the apparent alter-
sible that the mechanical scribing technique used in these studies ation of corrosion products noted earlier are consistent with the
embeds or otherwise deposits chromate pigment particles within the observation that, typical of chromate-inhibited iron corrosion,31
defect area, influencing behavior during early immersion stages. Fe21 is further oxidized to Fe31. For example, the reactions
Mechanisms of chromate inhibition.—The detailed mechanisms 6FeO 1 2CrO422 1 2H2O r Cr2O3 1 3Fe2O3 1 4 OH2
by which Cr61 (either CrO422 or Cr2O22 7 , depending on concentra-
tion and pH) decreases the corrosion rate of Fe and Al alloys are still 3Fe21 1 HCrO2
4 1 8H2O r Fe3Cr(OH)12 1 5H
1

unclear. Chromate-containing coatings have the reputation of pro- have been identified under near neutral pH conditions,35,36 resulting
tecting these alloys from corrosion, especially in damaged areas of in the formation of more adherent composite oxide/hydroxide films
the coating system (such as the defects introduced into the coatings with enhanced resistance to dissolution.
used in this work).27,28 Clearly the oxidizing power of CrO422 is Similarly, the formation of composite oxide/hydroxide films are
important and several redox reactions are possible, including direct believed to play an important role in the chromate inhibition of alu-
oxidation of metal substrates.29,30 In these processes, the Cr61 is minum alloys,31,32 although the composition of such films have only
reduced to Cr31 and, as a consequence, composite oxide/hydroxide recently been examined in detail.34,37 As for steel, the sustained oxi-
adherent films are formed with the general composition dation current in the presence of chromate may ultimately lead to a
x?M2O3/y?Cr2O3 or x?M(OH)3/y?Cr(OH)3, where M is Fe or Al and passive film of different composition and perhaps more resilience
x and y are variable, depending on conditions.31 The negative charge than that produced in the absence of chromate. For example, the in-
on CrO422 (or Cr2O722) facilitates migration of Cr61 to anodic sites creased time to passivation may allow for diffusion/migration of
where such passivation processes occur.32 Furthermore, chromate at Cr61 species to the anodic site and subsequent incorporation of Cr31
low concentrations is also known to function as a cathodic inhibitor (and perhaps Cr61) into the passive film.37
for both Fe33 and Al 2024-T3.34 As noted in the previous section,
low concentrations of CrO422 in the defect are expected during the Conclusions
The chromated-epoxy coating delayed the onset of corrosion of
both the steel and aluminum substrates compared to a plain epoxy
coating. Apparently the chromated-epoxy coating initially promoted
formation and/or stabilization of a passive layer within the defect
region, slowing the corrosion process. However, the aggressive
immersion conditions eventually led to a breach of this passive layer
and corrosion current began to flow. Cathodic inhibition by CrO422
may also play a role during this initial delay period. It should be noted
that the SVET detects only those reactions that involve one or more
ionic species. If a reaction such as M 1 O2 r MO2 occurred, there
would be no attendant ion flux in the electrolyte and such a reaction
would not be detected by the SVET. Thus, it is possible (though
rather unlikely in view of the control experiments) that such a corro-
sion process occurred during this apparently inactive delay period.
Once detectable corrosion commenced, the chromated-epoxy
coating actually appeared to sustain the current at both metals for a
longer period of time than observed with the plain epoxy coating.
That is, the time to passivation of both metals, as determined by the
decrease in measured anodic current within the defect, was longer
with the chromated-epoxy coating than with the plain epoxy coating.
This surprising observation may be an important aspect of the long-
term protection mechanism of chromate-based coatings, leading to
passive films of different composition and perhaps more resilience
than that produced in the absence of chromate.
The location of the reduction reaction with the chromated-epoxy
coating apparently depends on the metal substrate, occurring pre-
dominately over the coating surface with steel but being confined to
the defect with the aluminum alloy. A similar observation was made
with a coating of the inherently conducting polymer poly(3-octyl
pyrrole), where reduction current was observed only across the coat-
ing surface when the substrate was steel, but often in the defect when
the substrate was Al 2024-T3.7 Such metal specificity is difficult to
understand at this juncture, but may involve differences in the facil-
ity with which electrons are transported across the coating-metal
interface and through an intervening passive layer. Further studies
are planned to help elucidate the details of these processes.
Acknowledgments
This work was supported by the Air Force Office of Scientific
Research, grant no. F49620-96-1-0284 and F49620-97-1-0376
Figure 5. Current density maps for chromated-epoxy primer on Al 2024-T3 (AASERT).
alloy at 17.5 h immersion in dilute Harrison solution: (top) 3D representation
of the z component of the current density and (bottom) optical micrograph of North Dakota State University assisted in meeting the publication costs
the sample with current density vectors superimposed. of this article.
3666 Journal of The Electrochemical Society, 147 (10) 3661-3666 (2000)
S0013-4651(00)01-028-4 CCC: $7.00 © The Electrochemical Society, Inc.

References 17. P. J. Kinlen, V. Menon, and Y. Ding, J. Electrochem. Soc., 146, 3690 (1999).
18. H. N. Rosen and J. W. Martin, J. Coat. Tech., 63, 85 (1991).
1. C. Simpson, Organic Coatings for Corrosion Control, G. P. Bierwagen, Editor, 19. C. R. Hegedus, D. F. Pulley, S. J. Spadafora, A. T. Eng, and D. J. Hirst, J. Coat.
ACS Symposium Series 689, p. 356 (1998). Technol., 61, 31 (1989).
2. R. L. Twite, V. Balbyshev, G. P. Bierwagen, and D. E. Tallman, in Environmental- 20. R. S. Fishman, D. A. Kurtz, and G. P. Bierwagen, Prog. Org. Coat., 21, 387 (1993).
ly Acceptable Inhibitors and Coatings, S. R. Taylor and H. Issacs, Editors, PV 95- 21. W. H. Smyrl and M. Lien, Applications of Electroactive Polymers, B. Scrosati, Edi-
16, p. 202, The Electrochemical Society Proceedings Series, Pennington, NJ tor, p. 29, Chapman and Hall, New York (1993).
(1995). 22. D. A. Jones, Principles and Prevention of Corrosion, 2nd ed., p. 490, Prentice Hall,
3. EPA Federal Register, National Emission Standards for Hazardous Air Pollutants Upper Saddle River, NJ (1996).
for Source Categories: Aerospace Manufacturing and Rework Facilities, 60(170), 23. Z. W. Wicks, Jr., F. N. Jones, and S. P. Pappas, Organic Coatings: Science and
p. 45947, Sept 1, 1995. Technology, 2nd ed., pp. 131-133, Wiley-Interscience, New York (1999).
4. S. M. Cohen, Corrosion, 51, 71 (1995). 24. P. J. Kinlen, D. C. Silverman, and C. R. Jeffreys, Synth. Met., 85, 1327 (1997).
5. C. Scheffey, Rev. Sci. Instrum., 59, 787 (1988). 25. I. M. Zin, R. L. Howard, S. J. Badger, J. D. Scantlebury, and S. B. Lyon, Prog. Org.
6. H. S. Isaacs, A. J. Davenport, and A. Shipley, J. Electrochem. Soc., 138, 390 (1991). Coat., 33, 203 (1998).
7. J. He, V. Johnston Gelling, D. E. Tallman, G. P. Bierwagen, and G. G. Wallace, 26. K. C. Tsai and C. H. Tsai, in Proceedings of the 4th Asian-Pacific Conference,
J. Electrochem. Soc., 147, 3667 (1999). Materials Conservation and Corrosion Control, Vol. 1, p. 491, Tokyo, May 26-31,
8. H. S. Isaacs and B. Vyas, in Electrochemical Corrosion Testing, ASTM STP-727, 1985.
F. Mansfield and U. Bertocci, Editors, pp. 3-31, American Society for Testing and 27. H. Wienand and W. Ostertag, Modern Paint Coat., 38 (Nov 1984).
Materials, Philidelphia, PA (1981). 28. G. Adrian and A. Bittner, J. Coat. Technol., 58, 59 (1986).
9. D. A. Sargeant, Corros. Prevent. Control, 44, 91 (1997). 29. J. R. Waldrop and M. W. Kendig, J. Electrochem. Soc., 145, 1 (1998).
10. I. Sekine, M. Yuasa, K. Tanaka, T. Tsutsumi, F. Koizumi, N. Oda, H. Tanabe, and 30. Z. Szklarska-Smialowska and R. W. Staehle, J. Electrochem. Soc., 121, 1146 (1974).
M. Nagai, Shikizai, 67, 424 (1994). 31. J. Sinko, Prog. Org. Coat., In press (2000).
11. I. Sekine, Prog. Org. Coat., 31, 73 (1997). 32. M. W. Kendig, A. J. Davenport, and H. S. Issacs, Corros. Sci., 34, 41 (1993).
12. R. S. Lillard, P. J. Moran, and H. S. Isaacs, J. Electrochem. Soc., 139, 1007 (1992). 33. G. B. Hatch, Influence of Inhibitors on Differential Aeration Attack on Steel, II-
13. M. W. Wittmann and S. R. Taylor, in Advances in Corrosion Protection by Organ- Dichromate and Orthophosphate, NACE National Conference, NACE, Houston,
ic Coatings II, D. Scantlebury and M. Kendig, Editors, PV 95-13, p. 158, The Elec- TX (1964).
trochemical Society Proceedings Series, Pennington, NJ (1995). 34. J. Zhao, G. Frankel and R. McCreery, J. Electrochem. Soc., 145, 2258 (1998).
14. R. B. Leggat and S. R. Taylor, Corrosion, 55, 984 (1999). 35. B. P. Boffardi, in Reviews on Corrosion Inhibitor Science and Technology, A.
15. E. Bayet, F. Huet, M. Keddam, K. Ogle, and H. Takenouti, J. Electrochem. Soc., Raman and P. Labine, Editors, NACE, Houston, TX (1993).
144, L87 (1997). 36. I. J. Buerge and S. J. Hug, Environ. Sci. Technol., 31, 5 (1997).
16. E. Bayet, F. Huet, and H. Takenouti, Electrochim. Acta, 44, 4177 (1999). 37. L. Xia and R. L. McCreery, J. Electrochem. Soc., 145, 3083 (1998).

Вам также может понравиться