literature

ISSN:1369 7021 © Elsevier Ltd 2008OCTOBER 2008 | VOLUME 11 | NUMBER 1014

Corrosion-resistant metallic coatings

The epidermis–dermis–subcutaneous system is able both to resist

and repair damage and also regrow over a defect that exposes

underlying tissue. These are admirable qualities in biological

systems because these processes are repetitive, autonomous,

and can be triggered by damage without external intervention1.

In corrosion of metallic materials, dip, spray, chemically or

electrodeposited organic-based, metallic, and ceramic coatings are

often used to protect a substrate from corrosive damage2.

Natural passive films, artificially grown oxide layers (i.e. anodized

and sealed), and conversion coatings3–5 can also protect underlying

metals and alloys. Successful native passive films can form a protective

layer over many metals if the ingredients for passivation (i.e. alloying

elements) are present in high enough concentration in the underlying

alloy. The oxide quickly and automatically reforms to repair defects

in many environments. Conversion coatings and anodized layers rely

on the formation of a chemically engineered layer based both on

We describe recent computational and experimental studies on the corrosion properties of metallic coatings that can be tailored (tuned) to deliver up to three corrosion-inhibiting functions to an underlying substrate. Attributes are tuned by a selection of alloy compositions and nanostructures, ideally in alloy systems that offer flexibility of choice to optimize the corrosion-resisting properties. An amorphous Al-based coating is tuned for corrosion protection by on-demand release of ionic inhibitors to protect defects in the coating, by formation of an optimized barrier to local corrosion in Cl– containing environments, as well as by sacrificial cathodic prevention. Further progress in this field could lead to the design of the next generation of adaptive or tunable coatings that inhibit corrosion of underlying substrates.

F. Presuel-Moreno1, M. A. Jakab2, N. Tailleart3, M. Goldman4, and J. R. Scully3*

1 Florida Atlantic University, Boca Raton, FL 33431, USA,†

2 Southwest Research Institute, San Antonio, TX 78238-5166, USA†

3 Department of Materials Science and Engineering, University of Virginia, Charlottesville, VA 22904–2442, USA

4Constellation Power Company, Calvert Cliffs, MD 20657, USA†

*E-mail : [email protected]

†Formerly at MSE, University of Virginia, VA, 22904-4246, USA

MT1110p14_23.indd 14MT1110p14_23.indd 14 18/09/2008 15:32:1418/09/2008 15:32:14

Corrosion-resistant metallic coatings REVIEW

OCTOBER 2008 | VOLUME 11 | NUMBER 10 15

the oxidation of elements in the metal as well as on incorporation of

selected species in a pretreatment bath to reinforce the natural oxide.

Controlled additions of specific molecules to the conversion coating

bath are required and these can seldom be applied in the field.

Corrosion protection by passive films and coatings Passive films, conversion coatings, and metallic and organic coatings

confer corrosion protection via a variety of mechanisms, including

formation of barriers to the penetration of corrodants; high ionic

resistivity in surface layers to minimize electrochemical reactions

under the coating at the metal–coating interface; active corrosion

inhibition where an inhibitor is stored, released, and delivered to a

defect; and sacrificial cathodic protection. Chromate systems possess

this active corrosion inhibition property6 but are being phased out due

to their high toxicity and carcinogenicity7. Oxides may also exhibit

ion-selective permeability, as well as a desirable potential and/or pH

of zero charge to limit detrimental anion (Cl–) adsorption and ingress8.

Enhanced adhesion, control of the liquid chemistry at the metal–

coating interface and superhydrophobic properties9,10 may also operate

in many exciting new systems.

Active corrosion inhibition Traditionally, metallic coatings serve only one or two functions.

For example, Zn has excellent corrosion resistance and functions

as a sacrificial anode11. Zn galvanizing provides sacrificial cathodic

protection and acts as a barrier12 but does not usually supply inhibitor

ions. The release of Zn ions during the sacrificial protection of

galvanized steels11,13 only provides a small additional benefit compared

to galvanic protection provided by the potential driving force. Metallic

coatings used to protect Al alloys14,15 consist of a thin layer of nearly

pure Al mechanically bonded to standard precipitation age-hardened

Al alloys. The coating sacrificially cathodically protects the Al alloy

substrate (a beneficial form of galvanic corrosion) but does not provide

active corrosion inhibition. Galvanic throwing power is limited because

the roll-bonded cladding16 has a limited electrochemical driving force

for protection. The open circuit potential (OCP) of AlcladTM alloys

is only ~80–100 mV below that of the structural substrate, such as

AA 2024-T317, and their pitting potentials (Epit) are actually below

that of underlying structural Al alloys. The consequential virtual

non-polarizability enables cathodic protection of the aerospace alloy

but also ensures a high self-corrosion rate and considerable anodic

inefficiency.

Long-range corrosion protection of coating defects In the case of a coating containing active corrosion inhibitors or

passivators that are stored, released, and transported, defects such

as a scratch in the coating may be protected over long distances

(i.e. possess a chemical throwing power) due to concentration

(e.g. chemical potential) gradient-driven random transport from

inhibitor-rich regions to unprotected sites. Sacrificial cathodic

protection (i.e. galvanic throwing power) may also operate over long

distances. The distance over which protection can be afforded is

a function of the electrochemical properties of both the sacrificial

anode and the unprotected site as well as the geometry and ionic

conductivity of the ionic phase (often a thin electrolyte exposed to the

atmosphere) over the coating. The electric field in the ionic solution

established by the potential driving force difference between the

anodic and cathodic half cell potentials facilitates long-range cathodic

polarization. Such long-range functionality is superior to that of

biological systems where an internal circulatory system is required to

repair damaged sites1,18. Similarly, self-healing of mechanical damage

in man-made materials requires locally distributed microcapsules,

or encased composites containing gels or monomeric polymers and

catalysts spaced at a short distance from defects1,19–22. The monomer

cannot be easily supplied over long distances to repair damage sites.

One vision for new coatings with multiple, tunable functions Unfortunately, existing corrosion protecting metallic coatings cannot

provide active corrosion inhibitors to protect defects via transport of

the inhibitor through the liquid corrosive phase. Moreover, corrosion

properties are rarely tunable. For instance, the behavior of metallic

claddings is often fixed by rather inflexible limits on compositions and

microstructures. Therefore, there are few user-adjustable parameters

available to tune cathodic protection to mitigate certain localized

corrosion processes triggered above certain potential thresholds

(e.g. pitting, stress corrosion, exfoliation at high potentials, or H

embrittlement at very negative potentials). In addition, cathodic

protection, active chemical inhibition, as well as the presence of a local

corrosion barrier, are desired simultaneously. The addition of inhibitors

is a particular challenge for metallic materials. While sacrificial and

barrier attributes can be adjusted somewhat by changing composition

or structure, inhibitors can rarely be added to a solid. Moreover,

inhibitors, stored in the solid state, must be released as a liquid-soluble

molecular or ionic species to enable transport through the liquid

corrosive phase to a defect. In contrast, sparingly soluble inhibiting

pigments can readily be added to organic coatings and released by

chemical dissolution. An added challenge for both organic and metallic

coatings is the desire to trigger inhibitor release on-demand so that the

stored inhibitor is used only when it is needed.

We will now discuss some recent computational and experimental

studies where several corrosion-protecting functions are achieved

simultaneously in a single metallic coating. The desired properties

include active corrosion inhibitor supplied on demand to enable

corrosion inhibition and/or autonomous repair. Sacrificial anode-based

cathodic protection and barrier properties can also be tailored to

MT1110p14_23.indd 15MT1110p14_23.indd 15 18/09/2008 15:32:5018/09/2008 15:32:50

REVIEW Corrosion-resistant metallic coatings

OCTOBER 2008 | VOLUME 11 | NUMBER 1016

protect various substrates in a range of environments. Moreover, the

electric field created by the galvanic couple between the coating and

the substrate can be manipulated to augment transport of inhibitors

over large distances. To demonstrate this possibility we discuss a

specific nanoengineered, spray-applied, amorphous metallic alloy

(Fig. 1) that illustrates some of the mitigation strategies and functions

desired and achievable using a nanoengineered metallic coating such

as a corrosion-inhibiting system. The challenges and issues involved in

achieving these functions are highlighted.

Materials and fabrication of tunable amorphous metallic coatings The challenges of tailoring the properties of metallic coatings compared

to polymer-based coatings are numerous. In the case of polymers,

a micrometer- or nanometer-scale composite can be created by

mixing in the liquid state, and thus with great flexibility, a monomer,

resin, filler, solvent, etc., and a phase encapsulating the inhibitor

(e.g. strontium chromate pigment) in a slurry that is then reacted in

various ways and solidified in a rigid state below the polymer glass

transition temperature, Tg. In contrast, metallic coatings are often

applied at high temperatures and involve delivering a liquid metal via

either a hot dip or spray process. Such heating may adversely alter

the microstructure of the underlying alloy. Physical and chemical

vapor deposition, pulsed thermal spray (PTS), cold spray (CS), and

mechanical cladding are alternative methods to avoid excessive

temperatures. Numerous nonequilibrium alloying techniques may also

be employed to add beneficial alloying elements in significant enough

concentrations to have a major impact on properties. Corrosion barrier

properties can be improved by several routes. Minimizing structural

defects such as intermetallic and metalloid phases and optimization

of certain compositions in solid solution both usually significantly

improve corrosion ‘barrier’ properties. Therefore, methods to produce

supersaturated solid solutions which lack intermetallic and metalloid

phases are highly desired. By using metallic glasses, metallic elements

may be added or mixed in a soluble liquid solution with improved,

although not unlimited, flexibility, and then solidified and trapped

below Tg to achieve chemically homogeneous solid solutions. Moreover,

active corrosion inhibitors might be supplied if some of the added

alloying elements can function in this capacity. Amongst the possible

replacements for Cr(VI)23–27, transition metal (TM) and rare-earth (RE)

metal ions (Co2+, MoO4 2– and Ce3+) are possible choices to inhibit

localized corrosion of Al alloys23–25 and steels26–27. These elements

also can be alloyed to form Al-based metallic glasses. RE metal ions

may serve as inhibitors if they can be incorporated and released by

a chemical or electrochemical process. Alloying a crystalline metal

with these elements is traditionally difficult because of their limited

Fig. 1 Schematic of a structural high-strength precipitation age-hardened alloy (in this case AA 2024-T351, a Al–Cu–Mg alloy) protected against corrosion by a

system comprising a Al–Co–Ce metallic coating, an organic primer, and an organic topcoat. The corrosive environment is the layer of aqueous electrolyte on top

of the coating. The structural alloy is a polycrystalline alloy with Cu-rich coarse constituent particles labeled ‘IMC’ (for intermetallic compounds), grain boundaries

and fine nanoscale, precipitates. Several macroscopic defects are shown. They include a coating with a scratch through to this substrate alloy, and a scratch through

the organic coating. Ideally the metallic coating (a) functions as a barrier to corrosion, (b) delivers corrosion inhibitors indicated as released Ce3+ in the case of the

Al–Co–Ce coating, and (c) serves as a sacrificial anode to deliver cathodic protection (indicated by electrons, e–). Other enhancements include a graded coating

composition and mixed metallic phases potentially optimized to enhance bonding and ductility on the inner layer to serve as a barrier to corrosion on the outer

layers as well as to place the interface in compression to suppress fatigue of the substrate. Inhibitors released from the coating system can inhibit both anodic pits

(local corrosion attack) as well as strong cathodic reactions at Cu-rich IMC. Inhibitors and cathodic protection (electrons, e–) can inhibit pitting as well as more

advanced forms of attack such as intergranular corrosion. (Courtesy of King A. University of Virginia)

MT1110p14_23.indd 16MT1110p14_23.indd 16 18/09/2008 15:32:5118/09/2008 15:32:51

Corrosion-resistant metallic coatings REVIEW

OCTOBER 2008 | VOLUME 11 | NUMBER 10 17

solubility in the solid state28. This problem can be overcome by using

a glassy alloy that dissolves these alloying elements in solid solution.

Indeed, certain Al–TM and Al–refractory metal alloys29,30 can exhibit

enhanced local corrosion barrier24 and sacrificial anode-based cathodic

protection qualities31,32. The Al–RE–TM alloy series is of particular

interest because of its good glass-forming abilities33,34, high strength,

low density, and superior corrosion properties35–37.

In fact, a new amorphous Al–Co–Ce alloy system was recently

synthesized38–40 as an environmentally compliant metal coating with

multiple corrosion protection functionalities41–43. The notion of a

multifunctional metallic coating need not be restricted to Al–TM–RE

and there are likely to be many alloys that can be designed following

these general strategies. The goal was to simultaneously provide a

corrosion barrier, sacrificial anode-based cathodic protection42,44, or

prevention44 as well as active corrosion protection (e.g. contain and

deliver a supply of inhibiting cations that can be released to suppress

corrosion) to minimize corrosion at coating defects that expose the

underlying structural alloy45–47. Fig. 1 highlights this overall strategy.

Cathodic protection abilities may also be optimized by lowering the

OCP of the coating relative to the underlying substrate. For instance,

the OCP of the Al–Co–Ce alloy can be lowered as much as 750 mV

below that of AA 2024-T3 depending on the alloy composition and

solution pH41,48–50. Amorphous Al alloys also have the ability to store

beneficial alloying elements in solid solution. For instance, the selected

transition and RE metals (e.g. Ni, Fe, Co, Ce, Y, Gd) added as alloying

elements improve amorphicity in concentrations up to about 15 at.%.

Co, Ni, or Fe improve the resistance of Al to local corrosion, while

Ce3+, Y3+, and Gd3+ are good corrosion inhibitors when added as salt

in aqueous solutions40,51. The amorphous alloy can be formed over a

wide range of Co and Ce compositions, as seen in Fig. 238. The alloy’s

amorphous state is very stable.

High-velocity oxygen-fuel spray (HVOF), CS and PTS methods

are possible deposition methods to apply such coatings. PTS utilizes

high heating and large quenching rates, which allows for either

amorphous or crystalline powder feedstock material (Fig. 3) to develop

an amorphous or nanocrystalline coating on a substrate material.

The structure of the feedstock material tends to dictate the resulting

structure of the coating. The coating powder and deposited layer

are shown in Fig. 4a–d. The feedstock powder is often supplied as a

controlled particle size distribution ranging from 0.5 to 20 µm, although

larger powder sizes can be attempted (Fig. 4a). PTS is distinguished

by rapid particle acceleration and heating, a minimal residence time

within a high enthalpy environment, and rapid particle quenching

upon interaction with the target substrate52,53. The quenching rate

for particles less than 20 µm in size can be as high as 106 K/s54. Also

notable is that naturally aging precipitation age-hardened aerospace

alloys, such as the AA 2024-T3 substrates, do not exceed 60 ºC during

the coating process54. Reduced substrate thermal loading enables short

stand-off coating capability that, from an application standpoint, is

desirable because it allows for localized repairs of the coating where

a physical defect exists exposing the underlying structural alloy and

gives the capability for coating complex geometries55. The pulsed spray

stream rasters over the surface of the substrate in such a way that the

resulting spray-applied coating is actually a built-up series of layers.

PTS and HVOF coatings have the advantage of accepting the feedstock

powder for these alloys in either crystalline or amorphous forms, as the

feedstock particles can melt mid-process as long as powders are small.

The HVOF process has a large quench rate, similar to that of the PTS

Fig. 3 X-ray diffration patterns for the Al88Co10Ce2 feedstock powder, an as-

deposited PTS coating, and an amorphous Al84Co7.5Ce8.5 melt-spun ribbon

(MSR) indicating an amorphous structure in the MSR but the presence of Al

face-centered cubic nanocrystals in both the sprayed coating and powder

feedstock. (Reproduced with permission from. © 2007 J. Electrochem. Soc.)

Fig. 2 Map of Al–Co–Ce alloy compositions that yield amorphous, amorphous/

crystalline, and crystalline nanostructures in the Al–Co–Ce alloy system

based on the at.% Co and Ce present in solid solution Al. The map shown was

produced at a high cooling rate and results may differ at different cooling

rates. The crystal structures were determined by both X-ray diffraction and

transmission electron microscopy analysis.(Reproduced with permission from.

© 2007 J. Electrochem. Soc.)

MT1110p14_23.indd 17MT1110p14_23.indd 17 18/09/2008 15:32:5318/09/2008 15:32:53

REVIEW Corrosion-resistant metallic coatings

OCTOBER 2008 | VOLUME 11 | NUMBER 1018

process56. High-density samples with favorable corrosion properties

can be achieved57–63. The CS process utilizes the plastic deformation of

the sprayed, solid, feedstock particles on impact with the substrate to

achieve a uniform coating55. The spray event occurs below the melting

point of the spray materials, so less microstructural change and less

oxidation should occur55. Numerous metallic coatings have also been

deposited by these methods64–70.

Al–Co–Ce alloy coatings have been produced by both HVOF and

PTS with thicknesses ranging from 75 to 600 µm often with good

adhesion and low porosity. In some studies these have been compared

to melt-spun ribbon (MSR) alloys produced by the traditional method

of streaming the molten alloy on to a spinning copper wheel to enable

rapid quench rates.

Tunable barrier properties in multifunctional amorphous Al–TM–RE coatings Coatings which act as barriers to corrosion are commonly used and

date back to the use of noble decorative and corrosion-resistant

metallic coatings71. However, as in the case of sacrificial cathodic

protection, barrier properties are often fixed by rather inflexible alloy

compositions and structures. Porosity in coatings is often minimized by

increasing coating thickness.

The advanced processing methods discussed above enable new

coating compositions which break out of old constraints on alloy

Fig. 5 Comparison of E–log(i) curves for an as-deposited PTS Al–Co–Ce coating

to the same coating with a hole drilled which exposes the AA 2024-T351

substrate. Data for AA 2024-T351 and a Al87Co7Ce6 MSR are also presented.

Tests were performed in deaerated 0.006 M NaCl solution at a neutral pH and

enable determination of pitting potential, Epit, repassivation potential, Erp, and

open circuit potential, OCP. The higher pitting potential indicates a greater

resistance to pitting. A more negative OCP than AA 2024-T3 indicates that the

sprayed coating can cathodically protect the AA 2024-T3 alloy at scratches.

The total area tested was 0.238 cm2, the area of the exposed substrate/drilled

hole was 2 × 10–3 cm2. These results show that a PTS Al–Co–Ce alloy coating

exhibits a much great pitting potential than the AA 2024-T3 substrate in the

absence of an intentional defect. (Reproduced with permission from. © 2007 J.

Electrochem. Soc.)

Fig. 4 (a) Conventional secondary electron imaging (LEI) scanning electron microscopy (SEM) image of the Al88Co10Ce2 feedstock powder used for spray-applied

coatings. (b) LEI cross-sectional SEM image of an as-deposited PTS coating showing good adhesion with AA 2024-T351 (Al–Cu–Mg alloy). (c) LEI SEM image of

cross-sectioned and polished PTS coating showing pulse-deposited layers and level of porosity. (d) Confocal scanning laser microscropy (CSLM) image of an-

deposited PTS coating surface showing inherent roughness. (Reproduced with permission from. © 2007 J. Electrochem. Soc.)

(b)(a)

(c) (d)

MT1110p14_23.indd 18MT1110p14_23.indd 18 18/09/2008 15:32:5518/09/2008 15:32:55

Corrosion-resistant metallic coatings REVIEW

OCTOBER 2008 | VOLUME 11 | NUMBER 10 19

composition. In one study of an Al–Co–Ce alloy, three parameters

were obtained from E–I plots (tests performed in NaCl) (Fig. 5) to

characterize corrosion properties over a flexible range of compositions.

These were the OCP, the pitting potential, Epit, and the repassivation

potential, Erp. Epit and Erp are indicators of the pitting resistance,

while OCP helps forecast the galvanic corrosion behavior of an alloy

(e.g. potential driving force for sacrificial cathodic protection when

coupled to a substrate). Various regression models suitable for use with

alloys where the composition is constrained to 100% were explored

and two examples are shown in Figs. 6 and 744. Plots include possible

substrate structural materials (AA 2024-T3, Aermet 100® steel, etc.). The influence of Co on the OCP is based on Co’s effect on the cathodic

kinetics of Al–Co–Ce alloys. By increasing the reaction rate for both

the O reduction reaction (ORR) and the H evolution reaction (HER),

additions of Co in solid solution results in a higher OCP. The decrease

in OCP associated with the addition of Ce is most likely due to the

suppression of the ORR by Ce. The positive influence of Co on Erp is

due to the decreased hydrolysis product of this cation72, probably

resulting in a less aggressive pit environment and reduced anodic

dissolution rates of a Al–Co solid solution. A complex trend with Co

was seen. A parabolic shape also described the effect of Al and Ce

on Erp. These three elemental behaviors, when considered together,

describe a complex balance where Co and Ce reduce the aggressiveness

of the pit solution: Co lowers the dissolution rate while Al aids oxide

formation in an aggressive acidic solution. Therefore, optimization

of alloying elements produces the highest Erp values, with a complex

dependency of Erp on alloy content. To optimize the composition so as

to provide a barrier with the highest resistance to localized corrosion,

Co amounts should be relatively high (6-8 at.%), and Ce contents

moderately low (3-5 at.%) in this particular alloy.

Tunable sacrificial anode-based cathodic protection in Al–TM–RE coatings Sacrificial anode-based cathodic protection when two or more metals

are galvanically coupled is a potent electrochemical protection

method12. A coating that can polarize an exposed substrate material

just a few hundred millivolts below its OCP can lower its corrosion

rate by a factor of 100 or more. It is often more useful to polarize

many structural materials that corrode by local corrosion mechanisms

such as pitting or intergranular corrosion above some critical threshold

potential (i.e. Epit) to levels below this threshold potential. This is

accomplished by engineering a galvanic couple potential below Epit or

Erp. This is called cathodic prevention. Unfortunately, the governing

properties needed to optimize cathodic protection or prevention

(material-dependent OCP, and electrochemical current–potential

characteristics) have previously only been optimized through trial and

error. Moreover, parameters such as OCP are often rather inflexible

due to the limited choices of alloy composition and structure available.

Fig. 6 Relationship between the repassivation potential, ERP, and alloy content

in solid solution (Co at.% and Ce at.%). The relationship shown is plotted

from polynomial expressions developed from Scheffé polynomials based

on tests conducted on amorphous MSR of various Al–Co–Ce alloys in 0.6 M

NaCl solution at neutral pH. The ERP of the Al–Co–Ce alloy is compared to

AA 2024-T3 and high-purity Al. These results show that the repassivation

potential and thus barrier resistance to pitting can be controlled by controlling

the alloying content in solid solution.

Fig. 7 Predicted values of the open circuit potential (OCP) for the solid solution

Al–Co–Ce alloy system over a range of solid solution Co at.% and Ce at.%

contents, utilizing a Scheffé model with two inverse terms, demonstrating the

range of OCP possible within the Al–Co–Ce alloy system. These OCP values are

compared to AA2024-T3 and Aermet 100® OCP values. These are alloys the

Al–Co–Ce alloy might be designed to protect. The more negative OCP of the

Al–Co–Ce alloy compared to AA 2024-T3 and Aermet 100 confirms not only

the feasibility of this material to serve as a sacrificial anode but shows that the

driving force for cathodic protection can be tuned. The calculated values were

based on tests conducted on amorphous MSR of various Al–Co–Ce alloys in 0.6

M NaCl solution deaerated, at neutral pH.

MT1110p14_23.indd 19MT1110p14_23.indd 19 18/09/2008 15:33:0318/09/2008 15:33:03

REVIEW Corrosion-resistant metallic coatings

OCTOBER 2008 | VOLUME 11 | NUMBER 1020

Extreme cathodic potentials are not desired because ‘overprotection’

can lead to adverse side effects such as H embrittlement, alkaline

attack, and paint blistering. Therefore, tunable cathodic protection is a

highly desired capability.

The flexibility in OCP achieved by selecting particular compositions

of the Al–Co–Ce alloy system (Fig. 7) is extremely beneficial, as it

enables application of varying degrees of cathodic protection. By

choosing an Al–Co–Ce composition with relatively low Co (3-5

at.%) and low Ce levels (3–5 at.%) a sacrificial anode material can

be produced with a moderately low OCP value, to avoid cathodic

overprotection. An example is Al90Co5Ce5. The unique combination

of OCP values more negative than pure Aermet 100 ® or AA 2024-T3

along with an enhancement of barrier properties in the coating makes

this amorphous Al–Co–Ce alloy particularly well suited to serve as

a protective coating material. The enhanced resistance to localized

corrosion offered by this alloy results in greater Faradiac efficiency for

the sacrificial material, extending the theoretical lifetime of protection

compared to the Al–Zn and Al–Mg alloys used as Al alloy cladding

materials.

The sacrificial anode-based cathodic protection attributes sought

have been recently investigated by computational methods42,45,73–75

focusing on galvanic couples between a metallic coating and

exposed substrate under atmospheric conditions73. Fig. 8a shows

how atmospheric exposure (a thin electrolyte layer) of a structural

Al alloy covered by a metallic coating with a scratch at the center of

the sample Fig. 1) was modeled. A metric of the cathodic throwing

power of the sacrificially anodic Al–Co–Ce alloy was taken as the

galvanic couple potential distribution along the horizontal axis

of Fig. 8a and its proximity relative to the OCP of the Al–Co–Ce

coating and the AA 2024-T3 Epit. A figure of merit used is ∆E, given as (Epit 2024 – Ecouple), where Ecouple is the interfacial galvanic couple

potential at the centerline of the scratch, CL (Fig. 8b). Effective cathodic

prevention design maximizes ∆E. The maximum ∆E that can be achieved during galvanic coupling is termed ∆Emax and is defined by Epit2024 – Eocp of the Al–Co–Ce alloy.

Fig. 9 shows the extent of polarization given by the parameter ∆E obtained for a variety of situations as a function of scratch width at

a fixed sample length. The parameter ∆E is shown as the ordinate for each case, with the value of ∆E at S = 0 being the theoretical ∆Emax (i.e. Epit-2024 – EOCP, AlCoCe). The theoretical limit is approached for

small scratches. A greater extent of protection can be achieved when

the Cl– concentration is low. The potential distribution model predicts

that substantial sacrificial cathodic prevention of an AA 2024-T3

scratch could be achieved with a nanoengineered alloy coating.

The extent of the sacrificial cathodic protection provided by the

Al–Co–Ce alloys is a function of pH, Cl– concentration, the cathodic

kinetics on the AA 2024-T3, and the Co content of the metallic

coating. The metallic coating provides the best protection (i.e. largest

cathodic polarization of the scratch) when it is tailored to contain a

low Co content, and is exposed to either high or low pH solutions

of low Cl– concentration. Finally, the combination of chemical and

electrochemical protection predicted by computational modeling

has been verified via experiments involving defects machined into

PTS Al–Co–Ce alloys over AA 2024-T3. Corrosion protection is indeed

Fig. 8 (a) Schematic of the geometry associated with the galvanic coupling modeled between an Al–Co–Ce alloy coating (left) and AA 2024-T3 substrate (right)

exposed as a scratch in a thin electrolyte (e.g. corrosive solution) of thickness δ. Symmetry considerations allow solution of half of the scratch (S) exposing AA 2024-T3 by applying a zero flux boundary at CL. The half-scratch length, S, was varied from 500 to 5000 µm in width while the total length of the coating–scratch

system was held constant at 1 cm. The JO = 0 boundary condition reflects the physical situation at the right end of a specimen. This system was used to conduct

finite element modeling of the galvanic protection provided by the Al–Co–Ce coating to the AA 2024-T3 scratch. (b) In this plot the galvanic couple potential

based on material properties, electrolyte characteristics, and physical geometry is plotted as a function of horizontal position along the coating/scratch surface of

Fig. 8a, showing the effect of scratch length (indicated at far right) on the galvanic couple potential achieved. The scratch is located on the right-hand side. Model

parameters: variable S, Co ~ 3–5 at %, idl on AA 2024-T3 of 1.6 A/m 2, in solution of pH 3 and 0.05 M Cl–. The vertical line indicates the position of the interface

between metallic coating and substrate for the 5000 µm scratch. The dotted horizontal line indicates Epit of AA 2024-T3. Any galvanic couple potential below this

line indicates successful cathodic prevention. Thus, for this situation scratches 500–2500 µm in halfwidth are protected, while half-scratch widths of 5000 and 3500

µm are not protected against pitting near the centerline of the scratch at the far right. This is because the throwing power of the coating is inadequate to drive the

potential below Epit. (Reproduced with permission from © 2007 J. Electrochem. Soc.)

(b)(a)

MT1110p14_23.indd 20MT1110p14_23.indd 20 18/09/2008 15:33:0518/09/2008 15:33:05

Corrosion-resistant metallic coatings REVIEW

OCTOBER 2008 | VOLUME 11 | NUMBER 10 21

provided in experiments when the PTS coating is electrically coupled to

the AA 2024-T3 substrate.

On-demand active-corrosion inhibition based on tunable Al–TM–RE alloy coatings Controlled-release technologies are often used to supply chemicals

at given rates76,77. Release often occurs upon contact with solution.

However, slow release only when needed is often desired. Conducting

electroactive polymers are one means to provide controlled drug

delivery77. In corrosion protection, triggered or on-demand release

of inhibitor ions is also very important. Conductive polymers have

recently been utilized as reservoirs for corrosion inhibitors whose

triggered release occurs by galvanic reduction or ion exchange78–82.

Amorphous Al–Co–Ce alloys enable pH-controlled release of corrosion-

inhibiting ions from a metallic coating. The pH change that triggers

release is a consequence of the chemistry changes brought about by

corrosion itself43. Therefore, on-demand release occurs as a result of

exposure to corrosive species when corrosion initiates, for instance,

pH. ‘Turning off’ inhibitor release also occurs when the pH returns

to near-neutral and changes the local pH of the solution over the

coating. For a metal coating to behave like a pigmented paint that

can supply active corrosion inhibitors, several attributes are desired83.

The coating must store enough inhibitor to supply the critical inhibitor

concentration once released. The aqueous solution formed over the

metal surface must accumulate enough inhibitor to achieve the critical

inhibitor concentration, Ccrit, needed to stop corrosion in reasonable

time periods. This necessitates a high inhibitor capacity in the alloy

and a finite release rate. The Ccrit must be lower than the saturation

concentration of the inhibitor ion in the solution. The Ccrit in a given

electrolyte volume must be low enough so that the alloy can readily

release enough ions to achieve it without completely depleting the

inhibitor capacity83.

Al–Co–Ce alloys releasing Ce3+ meet these requirements. The

inhibitor is stored in the alloy and in its air-formed oxide layer in

valence states that are useful for corrosion inhibition. Inhibitors are

released by pH-dependent chemical dissolution of the oxide or direct

oxidation of the metal23–25 (Fig. 10). Ce and Co oxides are highly

soluble at low pH, insoluble under neutral, and partially soluble under

alkaline conditions (Fig. 11) 13,84. Thus, a pH change can be exploited

to trigger Co2+ and Ce3+ release. A pH change occurs during corrosion

Fig. 9 Cathodic polarization obtained with AlcladTM compared to the

Al–Co–Ce metallic coating expressed as the potential difference, ∆E, between Epit and the galvanic couple potential at the center of the scratch (x = 1 cm;

at C-line of Fig. 8a) attained for different scratch widths at a fixed half sample

length of 1 cm. The theoretical ∆Emax for AlcladTM and the other coatings are shown on the ordinate at S = 0. The potential difference between the Epit

of AA2024-T3 and the galvanic couple potential at x = 1 cm are shown in

the plot for each half-scratch width. Boundary conditions for the AlcladTM

used: ipass=0.002 A/m2, EpitAlclad 45 mV below Epit2024. The conditions

include 0.05 M Cl– and a pH of 3, with idl = 0.4 A/m2 for the various alloys

compositions. The results indicate that the Al–Co–Ce coating protects better

than AlcladTM as indicated by the larger ∆E. This suggests that the Al–Co–Ce coating can protect bigger scratches by cathodic prevention. Moreover, a low

Co content (i.e. 0–1 > 3–5% Co > 7–9% Co ) in the Al–Co–Ce alloy helps to

protect AA2024-T3 to a greater extent. (Reproduced with permission from ©

2007 J. Electrochem. Soc.)

Fig. 10 Schematic illustration of Al–Co–Ce coating (dark grey) on AA2024-T3 with a defect exposing the AA2024-T3 substrate. The less noble coating has been

tuned to possess an OCP more negative (–) than the underlying structural alloy. Consequently, cationic inhibitors released from the dissolving coating can be

transported to the defect by both migration as well as diffusion, while anionic species are transported in the opposite direction. A coating could also be engineered

to enable anionic transport by migration to the scratch as long as the polarity was reversed. This could be accomplished by tailoring the coating composition and

structure to create a coating with a potential greater than that of the scratch. The behavior shown has been confirmed for both MSR and spray-applied coatings.

MT1110p14_23.indd 21MT1110p14_23.indd 21 18/09/2008 15:33:0718/09/2008 15:33:07

REVIEW Corrosion-resistant metallic coatings

OCTOBER 2008 | VOLUME 11 | NUMBER 1022

which releases metal cations that hydrolyze to produce H+ at anodic

sites. High cathodic reactions at cathodic sites often increase the pH.

Therefore, when a metal is scratched and a galvanic couple forms

between the coating and newly exposed substrate, the pH will drop

over the anodic coating and rise at the cathodic scratch site. pH-

dependent release rates were found for Al–Co–Ce–(Mo) alloys41

(Fig. 11). The inhibitor ions released diffuse and then migrate towards

the damage site, reach Ccrit and suppress corrosion (Fig. 10). When

corrosion of the structural alloy stops, the pH of the environment

returns to neutral and Ce3+ is subsequently released at much lower

rate.

Computational studies42,45,73–75, have also been used to study

inhibitor release from such metallic substrates. Studies of inhibitor

release, transport, and achievement of Ccrit were conducted with and

without migration through the electrolyte phase. Migration-based

transport of Ce3+ to the scratch occurs due to the potential difference

established between the OCPs of AA 2024-T3 and the Al–Co–Ce

metallic coating. The time needed to reach Ccrit over the scratch was

determined, accounting for both transport processes as well as the

effect of temporal and position varying local release rates. Fig. 12

shows how the released Ce3+ profile develops with time and position in

one example. Migration aids Ce transport (for OCPAlCoCe < OCPAA2024),

explaining the uphill transport of Ce3+. The Ccrit,Ce is reached at the

coating–scratch interface at various times for different scratch lengths.

Accumulation was limited by ionic transport.

In summary, corrosion inhibition of a AA 2024-T351 scratch can

be achieved with ions released from a metallic coating. Experimental

exposure of AA 2024-T3 in a corrosive solution near an electrically

isolated Al87Co7Ce6 alloy confirmed computational modeling. The

Ce3+ supplied significantly decreased the maximum corrosion pit sizes

on the AA 2024-T3 surface below a critical size43. This is important in

aerospace structural materials, such as fuselage and wing skins, because

large corrosion pits above approximately 110 µm greatly reduce fatigue

life85,86.

Summary A multifunctional amorphous alloy has been described that possesses

three corrosion protection abilities when deployed as a coating over

structural alloys. The coating (i) functions as a local corrosion barrier,

(ii) serves as a sacrificial anode, and (iii) supplies soluble ions used as

corrosion inhibitors by engineering metallurgical and electrochemical

properties. The alloy system described is just one example of many

exciting new possibilities in metallic coatings enabled by the progress

in amorphous and nanocrystalline alloy development, as well as novel

synthesis and coating deposition methods. In the case presented,

excellent inherent resistance to corrosion is achieved through

structural amorphicity and choice of composition to optimize corrosion

functions. For instance, tunable electrochemical and chemical throwing

powers to achieve long-range protection of substrate defects can be

achieved. These concepts are generic and other alloys can in principle

be developed to optimize these functions. Such alloys can also be

manipulated to enable transport of inhibiting cations by both migration

and diffusion. Inhibitor release can be triggered on-demand.

Acknowledgments A Multi-University Research Initiative (Grant No. F49602–01–1–0352)

entitled ‘The Development of an Environmentally Compliant Multifunctional

Fig. 11 Effect of pH of a NaCl solution on the release rate of Al3+, Co2+, and

Ce3+ ions from an amorphous Al87Co8.7Ce4.3 MSR alloy in comparison to

high-purity Al. This plot illustrates that a shift in the electrolyte pH from the

near-neutral range typical of a benign solution to either the acidic or alkaline

range of a corrosive solution would help trigger the release of Ce cations (as

well as Al). This plot illustrates a mechanism for triggered release where the

environmental stimulus is the pH change.

Fig. 12 Computational model calculations of Ce3+ profiles in the electrolyte

phases (see Fig. 8a) with time and position over an Al–Co–Ce coating (left)

with exposed AA 2024-T3 substrate alloy (right). The plot indicates that Ce3+

can build up over the AA 2024-T3 on the right and even continues to build up

against the concentration gradient. This is due to the migration of Ce3+ in the

electric field formed in the electrolyte. The greater Ce3+ concentration near

the interface is due to slightly elevated release at high pH as indicated in Fig.

11. Computational results shown are for the selected case of initial solution

pH 6, 0.05 M Cl–, S = 1500 µm and slow ORR kinetics on the metallic coating

and AA2024-T3 (right). The vertical line indicates the position of the scratch

with the coating at the left and exposed AA 2024-T3 on the right as depicted

in Fig. 8a. (Reproduced with permission from. © 2007 J. Electrochem. Soc.)

MT1110p14_23.indd 22MT1110p14_23.indd 22 18/09/2008 15:33:0918/09/2008 15:33:09

Corrosion-resistant metallic coatings REVIEW

OCTOBER 2008 | VOLUME 11 | NUMBER 10 23

Coating for Aerospace Applications using Molecular and NanoEngineering

Methods’ initially supported this work under the direction of Jennifer Gresham,

at air force office of scientific research (AFOSR). Coating development was

supported by AFOSR under Robert Manz and Jennifer Gresham in collaboration

with Enigmatics, Inc. and SAIC. In particular, the support and insight of Ben

Gauthier and Schmuel Eidelman is greatly appreciated. Additional financial

support was provided by the National Science Foundation DMR-NSF-0504983

under the direction of Harsh D. Chopra.

REFERENCES

1. Balazs, A. C., Mater. Today (2007) 10, 18

2. Davis, J. R., et al., ASM Metals Handbook, vol. 9, ASM International, Metals Park,

OH, (1987), 13 Corrosion

3. Newhard, N. J., Metal Finish. (1972) 7, 49

4. Korinek, K., Chromate conversion coatings. In ASM Metals Handbook, vol. 9,

Davis, J. R., et al., ASM International, Metals Park, OH, (1987), 13 Corrosion, 389

5. Cape, T. W., Phosphate conversion coatings. In ASM Metals Handbook, vol. 9,

Davis, J. R., et al., ASM International, Metals Park, OH, (1987), 13 Corrosion, 383

6. Ilevbare, G. O., et al., Corrosion (2000) 56, 227

8. Natishan, P. M., et al., J. Electrochem. Soc. (1986) 133, 1061

9. Prakash, S. S., et al., Nature (1995) 375, 431

10. Prakash, S. S., et al., Nature (1995) 374, 439

11. Tsuru, T., et al., Corrosion Prevention Mechanism for New Galvanized Steels,

Beijing, (2005) 12. Jones, D. A., Principles and Prevention of Corrosion, Prentice-Hall, (1996)

13. Pourbaix, M., Atlas of Electrochemical Equilibria in Aqueous Solutions, National

Association of Corrosion Engineers, (1974)

14. Reddy, C. M., et al., Corrosion (2000) 56, 819

15. Walton, C. J., et al., Corrosion (1953) 9, 345

16. Brown, R. H., et al., Engineering laminates. In Alclad and Clad Aluminum Alloy

Products, Dietz, A. G. H., (ed.), MIT Press, (1969), 227

17. Davis, J. R., Corrosion of Aluminum and Aluminum Alloys, ASM International,

Materials , (1999)

18. Pang, J. W. C., and Bond, I. P., Compos. A Appl. Sci. Manuf. (2005) 36, 183

19. Peyratout, C. S., and Dahne, L., Angew. Chem., Int. Ed. Engl. (2004) 43, 3762

20. Toohey, K. S., et al., Nat. Mater. (2007) 6, 581

21. Fery, A., et al., New J. Phys. (2004) 6, 18, doi:10.1088/1367-2630/6/1/018 22. Gupta, S., et al., Nat. Mater. (2006) 5, 229

23. Kabasakaloglu, M., et al., Mater. Corros. (1997) 48, 744

24. Shaw, B. A., et al., J. Electrochem. Soc. (1990) 137, 359

25. Hinton, B. R. W., et al., Metals Forum (1984) 7, 211

26. Virtanen, S., et al., Corros. Sci. (1997) 39, 1879

27. Ilevbare, G. O., and Burstein, G. T., Corros. Sci. (2003) 45, 1545

28. Hatch, J. E., Aluminum: Properties and Physical Metallurgy, American Society of

Metals, Metals, (1984)

29. Moshier, W. C., et al., J. Electrochem. Soc. (1986) 133, 1063

30. Moshier, W. C., et al., J. Electrochem. Soc. (1987) 134, 2677

31. Sanchette, F., et al., Surf. Coat. Technol. (1998) 98, 1162

32. Sanchette, F., and Billard, A., Surf. Coat. Technol. (2001) 142, 218

33. Sweitzer, J. E., et al., Electrochem. Solid-State Lett. (1999) 2, 267

34. Sweitzer, J. E., et al., Electrochim. Acta (2003) 48, 1223

35. Lucente, A. M., and Scully, J. R., Electrochem. Solid-State Lett. (2007) 10, C39

36. Lucente, A. M., and Scully, J. R., Corros. Sci. (2007) 49, 2351

37. Lucente, A. M., and Scully, J. R., J. Electrochem. Soc. (2008) 155, C234

38. Gao, M. C., et al., Metall. Mater. Trans. A (2007) 38A, 2540

39. Inoue, A., et al., Jpn. J. Appl. Phys. (1988) 27, L1796

40. Kabacoff, L. T., et al., Mater. Sci. Eng. A (1991) 134, 1288

41. Goldman, M. E., et al., Electrochem. Solid-State Lett. (2005) 8, B1

42. Presuel-Moreno, F. J., et al., J. Electrochem. Soc. (2005) 152, B302

43. Jakab, M. A., and Scully, J. R., Nat. Mater. (2005) 4, 667

44. Scully, J. R., et al., Corrosion (2008) 64, 210

45. Presuel-Moreno, F. J., et al., J. Electrochem Soc. (2006) 153, B486

46. Jakab, M. A., et al., Corrosion (2005) 61, 246

47. Jakab, M. A., and Scully, J. R., Corrosion (2008) 64, 198

48. Goldman, M., Demonstration of Tunable Corrosion Properties of the Al–Co–Ce

Alloy System, MS thesis, University of Virginia, (2005)

49. Goldman, M. E., et al., Amorphous Metallic Coatings with Tunable Corrosion

Properties Based on Al–Co–Ce-(Mo) Alloy Compositions. NACE, (2004), Paper

No. 4276

50. Goldman, M. E., et al., Selected Corrosion Properties of a New Amorphous

Al–Co–Ce Alloy System. (2003)

51. Davis, G. D., et al., Surf. Interface Anal. (1987) 10, 168

52. Gauthier, B., et al., Spray Applied Amorphous/Nanocrystalline Aluminum Alloy

Coatings as a Replacement for Aluminum Cladding, Tri Service corrosion

conference Denver (2007) 53. Gauthier, B., et al., ECS Transactions (2008) 11, 59

54. Gauthier, B., et al., Unpublished PTS data, 2007

55. Li, C. J., et al., Thin Solid Films (2005) 489, 79

56. Meng, Q., et al., Characterization and Mechanical/Corrosion Properties of

Amorphous Al–Co–Ce Coatings, Orlando, FL, (2005) 57. Kawakita, J., et al., Surf. Coat. Technol. (2003) 166, 17

58. Kawakita, J., et al., Sci. Technol. Adv. Mater. (2003) 4, 281

59. Ishikawa, Y., et al., J. Thermal Spray Technol. (2005) 14, 384

60. Kawakita, J., et al., Surf. Coat. Technol. (2006) 200, 4414

61. Kuroda, S., et al., Corrosion (2006) 62, 635

62. Wang, A. P., et al., Mater. Sci. Eng. A (2007) 449, 277

63. Connolly, B. J., et al., Corros. Eng. Sci. Technol. (2004) 39, 137

64. Agarwal, A., et al., JOM (2002) 54, 42

65. Burkle, G., et al., Metastable, Mechanically Alloyed and Nanocrystalline Materials

(2002) 386–3, 571

66. Dent, A. H., et al., Surf. Coat. Technol. (2001) 139, 244

67. Edris, H., et al., J. Mater. Sci. (1997) 32, 863

68. Guilemany, J. M., et al., Scr. Metall. Mater. (1995) 33, 55

69. He, J. H., et al., Metall. Mater. Trans. A (2000) 31, 541

70. Lau, M. L., et al., Nanostruct. Mater. (1999) 12, 319

71. Cramer, S. D., et al., Corrosion of clad metals. In ASM Metals Handbook, Baboian,

R., ASM International, (2005), 13B Corrosion Materials, 442

72. Baes, C. F., and Mesmer, R. E., The Hydrolysis of Cations, Robert E. Krieger,

Malabar, FL, (1986)

73. Wang, H., et al., Electrochim. Acta (2004) 49, 239

74. Lee, J. S., et al., J. Electrochem Soc. (2004) 151, B423

75. Stewart, K. C., Intermediate Attack in Crevice Corrosion by Cathodic Focusing, PhD

thesis, University of Virginia, Charlottesville, VA, 1999

76. Cussler, E. L., Diffusion – Mass Transfer in Fluid Systems, Cambridge University

Press, (1997)

77. Pernaut, J.-M., and Reynolds, J. R., J. Phys. Chem. B (2000) 104, 4080

78. Kendig, M., et al., Progr. Org. Coat. (2003) 47, 183

79. Kinlen, P. J., et al., Corrosion (2002) 58, 490

80. Kinlen, P. J., et al., J. Electrochem. Soc.(1999) 146, 3690

81. Kendig, M., and Hon, M., Corrosion (2004) 60, 1024

82. Cook, A., et al., Curr. Appl. Phys. (2004) 4, 133

83. Sinko, J., Progr. Org. Coat. (2001) 42, 267

84. Hayes, S. A., et al., J. Electrochem. Soc. (2002) 149, C623

85. Chen, G., et al., Pitting corrosion and fatigue crack nucleation. In Effects of the

Environment on the Initiation of Crack Growth, Van der Sluys, W. A., Piascik, R. S.,

and Zawierucha, R., (eds.) ASTM, (1997), 18 86. Rokhlin, S. I., et al., Eng. Fract. Mech. (1999) 62, 425

MT1110p14_23.indd 23MT1110p14_23.indd 23 18/09/2008 15:33:1018/09/2008 15:33:10