Insights from a Recent Meeting: Current Status and Future Directions in Magnesium Corrosion Research

Magnesium and its alloys are of interest for a diverse range of applications, including structural components to achieve vehicle lightweighting, biodegradable/bioresorbable medical implants, anodic protection systems, functional use for hydrogen production and storage, and electrodes in fuel cells and batteries.^1-5  A key challenge for Mg is the control of its high reactivity to achieve desired levels of corrosion resistance.^6-8  Understanding, controlling, and mitigating corrosion of Mg has emerged as a major research topic, with the Web of Science™^,† database yielding over 4,500 papers alone from 2010 to 2015 in a key word search of “magnesium” and “corrosion.”

In April 2016, the United States Department of Energy (DOE) Vehicle Technologies Office held a one day Technical Review Meeting entitled “Current Status and Future Directions in Magnesium Corrosion Research.” This meeting was devoted to discussion of the understanding and mitigation of Mg corrosion, with a focus on automotive lightweighting applications. The event drew industrial, academic, and national laboratory researchers primarily from the United States and Canada. The morning session was centered on industrial perspective presentations, with the afternoon session featuring presentations devoted to fundamental aspects of Mg corrosion (Table 1). The goal of this paper is to highlight select topics from the meeting for the wider corrosion and Mg research communities, with an emphasis on the authors’ interpretation of the key challenges, future technology needs, and research directions that emerged from the meeting presentations and discussions. As such, this paper represents a snapshot of only one segment of the Mg corrosion research community, and is not intended to be a comprehensive overview of the state of the literature, nor is it a consensus representation of the views of the meeting participants. Rather it presents opinions for some key Mg corrosion areas of interest for future research aimed at achieving more widespread use of Mg alloys in vehicles.

Magnesium is the lightest structural metal and is therefore of great interest for vehicle lightweighting.^1,2,9-11  New U.S. fuel economy and emission regulations for model years 2017 to 2025 require car and light truck performance equivalent to 54.5 miles per gallon and reductions in greenhouse gas emissions to 163 grams per mile in 2025.¹²  As a result, the pressure to lightweight vehicles is stronger than ever before.¹³  Vehicle lightweighting represents one of several design approaches automakers are currently evaluating to improve fuel economy. The next few years will likely see considerable lightweighting across the automotive industry. A maximum 20% vehicle mass reduction has been considered in the proposed rulemaking for 2017 to 2025 light-duty vehicle greenhouse gas emission standards and corporate average fuel economy standards. However, some automotive executives believe that a 30% mass reduction (without adding major costs) will be required to meet corporate average fuel economy (CAFE) standards.¹⁴ 

In the DOE funded Multi-Material Lightweight Vehicle project, a 2013 Ford Fusion^† baseline vehicle (1,559 kg) was reduced 23.3% in weight to 1,195 kg, primarily by the use of high-strength steels, Al, and composites (Figure 1).^15-16  The fuel economy was increased from 28 to 34 miles per gallon (21.4%), representing a 9.13% increase in fuel economy for every 10% in mass reduction. It also resulted in a 16% reduction in CO₂ emissions and life-cycle primary energy usage.^15-16  That design and build project was known as Mach I. A second, yet-to-be-built design, Mach II, achieves a full 50% mass reduction for the vehicle (761 kg) (Figure 2).¹⁶  The Mach II design contains, by weight, 6% Mg alloys in the body in white, 62% Mg alloys in the closures, 100% Mg alloys in the sub-frames, and 54% Mg alloys in the chassis.¹⁶  For a comparable vehicle to be feasible for future automotive production, technical obstacles to widespread Mg alloy introduction in vehicles must be overcome.

Despite a component weight savings potential that often exceeds 45%, Mg alloys find only limited use in today’s production vehicles. The average North American model year (MY) 2013 vehicle was less than 0.5% Mg by weight.¹⁷  In almost every case, Mg alloys are currently used as die castings. For example, the MY2010 and newer rear lift gate (“trunk door”) inner of the Lincoln MKT^† is a single piece Mg alloy die casting weighing more than 8 kg and meeting the various crash safety requirements.¹⁸  Similarly, the MY2017 Chrysler Pacifica^† will feature a large, die cast Mg alloy rear lift gate inner.¹⁹  Historically, Mg die castings have been used as powertrain components, under-hood structures, and interior structures.^1-2,9-11,20  Mg sheet products have been minimally applied in production vehicles, finding use as an interior console cover and, more recently, the roof of specialty Porsche products.²¹  Even after successful implementation, Mg is sometimes subsequently replaced by an incumbent material such as in the case of the Chevrolet Corvette^†, which featured a die cast Mg engine cradle for MY2006 through MY2013 and an Al engine cradle as MY2014.²²  Advancing beyond low and medium volume vehicles to have a greater impact on average weight across the North American fleet will require dedicated research and development (R&D) to address the key challenges and barriers to automotive Mg use.

The limited degree to which Mg alloys are applied in vehicles stands in stark contrast to the many promising applications.^1-2,9-12  Mg alloys generally cast well because of a low melting temperature and amenable flow properties of the liquid metal.²³  Figure 3 provides examples of how these characteristics offer tremendous design flexibility and enable complicated, single piece structures such as instrument panel beams and even complete engine blocks.^24-25  The process and equipment characteristics for Mg die casting are also reasonably similar to the characteristics for Al die casting. Recent investment into new Al die casting infrastructure throughout the automotive industry thereby presents an opportunity for growth in Mg die casting by reducing the capital investment hurdles in the near term.

Although Mg sheet faces greater technical and commercial hurdles than Mg die castings, the vast majority of passenger vehicles use stamped sheet metal architecture; viable Mg sheet products are therefore necessary for achieving significant adoption. Because of their low density and relatively low strength, Mg sheet alloys are particularly well suited for stiffness-limited applications such as panels, closure inners, and planar underbody components. In the long term, future high-strength Mg sheet alloys could also be considered in crash critical and strength-limited components; however, this compounds the existing technical hurdles with the added challenge of high-strength alloy design and should be considered a lower near-term R&D priority.

Several recent R&D programs have worked to assess the potential of Mg in automotive structures. Of particular note is the “Magnesium Front End R&D Project” (MFERD), a joint effort between the United States, Canada, and China.²⁶  This project focused on developing and demonstrating new Mg technologies in the context of a “demonstration structure” that emulates the shocktower/rail assembly in a vehicle front end. Examples of demonstration structures, from among the many different variants, are shown in Figure 4.

The MFERD program was very successful in developing and demonstrating new Mg technology. However, the project results also highlight the magnitude of the corrosion challenge for automotive Mg structures.²⁶  Despite a large experimental matrix of different cleaning, pretreating, and coating approaches, corrosion protection was a significant issue even in the final variants of the demonstration structure. For example, corrosion bloom was observed at self-piercing rivets in nearly all cases as a result of the complicating interactions of the piercing process, rivet materials, and pretreatment layers. Additionally, coatings on the warm-formed Mg sheet components generally performed poorly, perhaps because of residual forming lubricant or other artifacts from the forming process. Overall, corrosion results from the MFERD project align with recent experience in other Mg demonstration activities: corrosion protection is a substantial challenge and will require sustained effort in fundamental corrosion science, coating and chemical research, development of new processing techniques for coating and protection, and application-based demonstration efforts for validation and testing.

A long list of technical and commercial challenges introduces barriers to wider use of Mg, and properly addressing all barriers for all components throughout a vehicle would require enormous resources (and may not be ideal anyway). However, applying Mg in the components and structures with the greatest potential benefits is generally limited by three major barriers: the cost of primary Mg metal, the formability of wrought Mg alloys, and the stability of Mg structures against corrosion in an automotive environment. The cost of primary Mg is a result of a complicated interaction of technical and market influences. Although varying considerably over the past decade, the U.S. market spot price of primary Mg is approximately double the U.S. market spot price of primary Al.^27-28  This cost differential introduces an obvious commercial barrier that could be addressed in part through the development of new extraction technologies.

The limited formability of Mg sheet is mostly a result of the anisotropy of the available deformation mechanisms and the propensity for strong texture in processed sheet.^29-30  Alloy development investigations have revealed several elements that improve formability such as Nd and Ca.^31-35  For example, successful warm forming of a door panel inner from a Zr and Nd containing ZEK 100 type Mg alloy (ZE10A [UNS M16100⁽¹⁾]) in sheet form was recently reported.³⁶  However, modifications to alloy chemistry can also dramatically affect corrosion behavior and must be considered carefully. Protection of Mg components from corrosion is a tremendous barrier. As discussed in this article, despite excellent ongoing work in the corrosion community, much remains to be done. Gaps in the fundamental understanding of Mg corrosion, the capability of predictive tools, and practical coating approaches compatible with automotive manufacturing must be addressed in order to leverage the tremendous weight saving potential of Mg alloys.

Magnesium exhibits relatively good ambient temperature corrosion resistance under dry atmospheric conditions. Corrosion resistance becomes degraded in the presence of high humidity and/or aqueous exposures, especially in the presence of salt species.^6-8  Galvanic corrosion is a major concern for Mg, particularly in automotive structures as candidate Mg parts frequently need to be joined to steel or Al.^26,37  The key themes to emerge from the industrial perspective presentations and discussions related to the need for more corrosion resistant coatings and more realistic corrosion test methods. Points of emphasis included:

• (1)

  Large expansion in the use of Mg in automotive applications will require joined parts with steel and/or Al, not simply monolithic Mg.

• (2)

  Cleaning, pretreatments, and coating processing must be amenable to multi-material steel/Al/Mg joined parts.

• (3)

  Testing of coated, flat, monolithic Mg coupons does not adequately represent real-world automotive exposure conditions.

Corrosion protection schemes for Mg involve multi-layered coatings, typically a chemical or electrochemical surface pretreatment (conversion coatings, anodization, etc.); an intermediate layer (electrocoatings, platings, powder coatings, organic coatings, etc.); and a final, outer layer(s) of sealant and/or paint.^37-42  The current coating practice for Mg automotive components is typically conversion coating followed by powder coating and sealant/paint.⁴²  It is performed offline, separate from other components and the body in white (body in white refers to the unpainted frame shell of the car joined sheet metal structure; it does not include moving parts, motor, chassis sub-assembles, trim, etc.).

Several participants commented that electrocoating of Mg was favored over powder coating because of lower cost, no line-of-sight limitations, and better throw power and coverage into corners and confined areas, but that the resultant electrocoated structures currently exhibit too much variability in corrosion resistance for widespread automotive use. One comment speculated that degradation of the initial conversion coated Mg surface from the electrocoating process may be one contributing factor to this variability;³⁸  other comments emphasized the critical aspects and challenges with surface preparation and cleaning of Mg components to achieving quality coatings. Concerns were also expressed whether existing, non-Cr⁶⁺, Ti/Zr conversion coatings conveyed sufficient corrosion resistance for widespread automotive use. Development of conversion coatings for specific Mg alloys (many were originally developed for Al), and continued exploration and development of electro-ceramic and micro-arc oxidation coatings^41,43  were viewed as important topics for future work.

Automotive body-in-white processing involves cleaning and pretreatments, Zn phosphating, electro-coating, curing, sealing, primer, and topcoat (Figure 5).⁴²  These processes work effectively with steel, and can also generally accommodate Al. A key point of discussion was that Mg presents several key compatibility challenges to conventional body-in-white processing. Mg requires the use of acid etch cleaning, whereas current body-in-white practices utilize alkaline cleaning. Mg also poisons the Zn phosphating bath.⁴⁴  Furthermore, it typically requires different (lower voltage) electrocoating parameters than does steel and Al. Multiple presentations and participants emphasized the need for development of universal cleaning, pretreatment, and coating processing approaches to enable more widespread use of multi-material joined steel/Al/Mg structures. Coating surface finish aesthetic quality was also brought up as a key consideration for use of Mg in automotive exposed external surface applications.

The other major consideration that emerged from the industrial perspective session was corrosion evaluation methods. Cyclic testing³⁷  was generally favored as better simulating real-world conditions over ASTM B117 salt spray testing, which several audience comments indicated was too aggressive. Most academic coating development efforts utilize small, monolithic, flat Mg coupons and evaluation by electrochemical, salt spray, or cyclic methods. Although discussion recognized such work as a reasonable first step, good coating performance in flat coupon form was not considered sufficient to move forward with automotive evaluation. For example, Ford (Figure 6) utilizes a multi-feature test coupon design to evaluate coated Mg, which includes a scribe mark, a punched hole for a sharp interior edge, knife/rounded/sheared coupon edges, and a coated bolt-washer assembly inserted over a scribed region to promote a galvanic couple.

Presentations and audience discussion also emphasized the need for corrosion testing to consider loading/residual stress/thermal expansion mismatch factors which would be encountered with multi-material steel/Al/Mg joined parts, both in coating processing (e.g., body-in-white curing) and in automotive end use. Assessment of such considerations would benefit from modeling and advanced ex situ/in situ residual stress characterization methodologies (x-ray, neutrons, etc.), and are important topics for future research. Improved joining methods (friction-stir weld, self-piercing rivets,⁴⁵  etc.) and improved joint isolation/coating approaches were also widely recognized as a key need to mitigate galvanic effects of dissimilar materials contact with Mg components.

The afternoon session of the meeting transitioned to presentation of more fundamental focused Mg corrosion research topics from academic and national laboratory researchers, including advanced electrochemical and microstructural characterization techniques, film growth, dissolution and dealloying phenomena, microgalvanic effects of second phases and impurities, and modeling of corrosion processes.^46-58  A key audience comment that received wide acknowledgement was that the state of Mg corrosion understanding is 20 years behind that of Al and 40 years behind that of steel. Points of emphasis included microgalvanic effects of second phases and impurities,^49-52,59  and the current vigorous discussion in the Mg corrosion literature regarding the mechanistic source of the “negative difference effect” (NDE) encountered in Mg corrosion.^6,49,60-65  Conventional H₂ evolution behavior is observed for Mg under cathodic conditions. However, H₂ evolution is also observed under anodic conditions, with the atypical, apparent behavior of increasing H₂ evolution with increasing potential, frequently referred to as NDE.^6,49,60-65 

This paper will not seek to recount or assess the relative merits of proposed NDE mechanisms. A number of recent papers have done this from multiple viewpoints, and the reader is referred to that literature for a more thorough treatment of the NDE effect (e.g., references^60-65 ). Rather, this paper will highlight some of the research needs identified in presentations and audience discussion that underpin eventual achievement of better fundamental understanding of Mg corrosion, including the NDE. It should be noted that the NDE is not simply an academic curiosity. A better fundamental understanding of Mg corrosion mechanism(s), particularly dissolution behavior, is needed to provide a basis for the design of Mg alloys and coatings with improved corrosion resistance. Such understanding is also needed to guide input for critically needed modeling and predictive capabilities of Mg corrosion.

The key themes to emerge from presentations and discussions in the Mg corrosion fundamentals session included:

• (1)

  Complexity in assessing corrosion rates for Mg and characterization of corrosion products.

• (2)

  Dependence of dissolution/film growth features on exposure conditions and local environment.

• (3)

  Application of new characterization techniques and modeling.

Aqueous corrosion of Mg involves dissolution and film growth processes, both direct film formation from the substrate and re-precipitation and film deposition from the fluid phase.^6-8,60,62,64  These processes are highly non-uniform for Mg, can involve extensive local H₂ evolution, and can be significantly impacted by local conditions in both the Mg metal (impurities, alloying additions, second phases, grain structure and orientation, prior polarization, etc.) and the fluid (salt species, local pH, dissolved species solubility and saturation/supersaturation, fluid conductivity, etc.).^{6-8,46-49,61-66} 

Several presenters and audience members commented that assessing the corrosion rate of Mg is nontrivial. The challenges result from non-uniform local corrosion phenomena characteristic to Mg, which complicates the relation of typically assessed bulk average corrosion measurements to gain meaningful insights into the corrosion mechanism(s) in play.^{6-8,46-48,62-64,66}  The simultaneous use of multiple techniques to monitor corrosion, including mass change, H₂ evolution (volumetric and gravimetric methods), thickness loss, bulk and local fluid chemistry, electrochemistry, and local electrochemical imaging techniques, was recognized as needed to better elucidate local corrosion rates and mechanisms.

A second point of discussion was that exposure conditions can critically influence the nature of the corrosion products observed, and impact interpretation of the corrosion mechanism(s) involved. An example is shown in Figure 7 for pure Mg exposed for 48 h in 5 mL of distilled deionized water vs. 1,000 mL.^57,67-68  In the small volume of water, Mg species dissolution from the sample leads to rapid pH increase, resulting in local outer surface enrichment of Mg(OH)₂ + MgO, with extensive growth of an underlying, seemingly dense MgO-based film several micrometers thick. In contrast, exposure in a large volume of water resulted in little change in pH during the course of the 48 h exposure, with corrosion behavior dominated by dissolution rather than film growth, resulting in a porous, intermixed Mg(OH)₂ + MgO surface film less than 100 nm thick. An interesting question for future work is which exposure condition better simulates real-world use, and which exposure condition better permits critical experiment design to achieve greater insights into fundamental corrosion mechanisms?

Several presenters also noted that characterization of Mg corrosion products is also nontrivial. Film products formed in water exposures generally include Mg(OH)₂, MgO, and, under some circumstances, MgCO₃.^6-8,57,68-71  Hydrogen may also penetrate in some form through the film and into the underlying Mg metal during aqueous exposures.⁶⁸  The extent and impacts of such hydrogen penetration on corrosion mechanism(s), stress corrosion cracking susceptibility, etc., are not yet clear but are an area of interest for future research. Impurities and alloying additions can also accumulate in various forms in the film and/or at the alloy/film interface region with potentially significant impacts on corrosion phenomena. The advent in recent years of focused ion beam milling techniques has greatly increased the application of cross-section analytical electron microscopy (AEM) to Mg corrosion films and products (e.g., Figure 7), which are a useful complement to thin film surface techniques such as x-ray photoelectron spectroscopy, Auger electron spectroscopy, and Rutherford backscatter spectrometry.

However, the Mg-O-H-C based film system is particularly challenging for AEM because of the potential for electron beam damage and beam/vacuum dehydration effects.^68-71  Identification of oxide vs. hydroxide regions and morphologies can be difficult, frequently requiring multiple characterization techniques to gain sufficient confidence in the results. Features such as local film porosity (which can be induced as an artifact in AEM) are likely critical to evaluation of corrosion mechanisms, many of which require access of the environment through the surface film and into the underlying Mg metal in some form to explain the observed behaviors. The concept of catalytically active Mg metal or Mg film surfaces and structures relative to chemical and/or electrochemical effects underpin several proposed Mg corrosion mechanisms^60-61,66,72  and characterization of exactly what corrosion product phase and morphology is locally formed is needed.^58,70-71  This is especially true in studies of impurities and alloying element effects, where local elemental segregation phenomena may play a role in determining corrosion resistance. Further, improved understanding of the degradation of Mg corrosion films in the presence of salt species^58,70-71  also requires improved knowledge of where and how species such as Cl are incorporated into the film and alloy/film interface regions.

Participant discussion widely recognized the limited protectiveness provided by MgO and Mg(OH)₂ aqueous-formed films, particularly in the presence of salt species.^6-8  Alloying strategies will need to fundamentally alter the corrosion mechanism, and/or result in the formation of alternative films to afford corrosion protection of uncoated Mg. Discussion included recent literature reports of cathodic reaction poisoning by additions of As and Ge to Mg with beneficial effects on corrosion resistance,^73-74  as well as the formation of a protective lithium-carbonate film on a ductile Mg-Li alloy.⁷⁵  Alternative alloy design strategies such as high entropy alloys⁷⁶  may also be of interest for future development of corrosion resistant Mg alloys, although the resultant alloys may no longer be Mg-based. This design approach may help overcome a key limitation of Mg, namely limited solid solution of alloying additions such as Al, Cr, and Ti, which otherwise might modify the Mg oxide/hydroxide surface film formation in a positive manner. For example, improved corrosion resistance was recently reported for metastable solid solution Mg-Ti alloys formed by sputtering, although the benefit was not apparent until the level of Ti exceeded that of Mg in the alloy.⁷⁷  Optimization of Mg alloy processing-microstructure features to enhance corrosion resistance has also re-emerged as a path of interest for future research.⁷⁸ 

Corrosion modeling was also a topic of several presentations and extensive audience discussion. It was universally recognized that modeling has the potential to provide immense benefits, not only for aiding understanding of key Mg corrosion phenomena, particularly with regard to evaluating competing mechanisms, but also for alloy design and automotive adoption via corrosion behavior predictive capability for Mg components under automotive-relevant conditions. However, a key challenge for the corrosion community is to achieve consensus regarding underpinning corrosion mechanism phenomena in Mg to provide the modeling community with better specific guidance to select model inputs that yield the most relevant model outcomes. The current lack of fundamental mechanistic understanding of Mg corrosion frequently necessitates assumptions and simplifications in modeling, which can limit the utility of the model outputs.

• This paper highlights key topics of interest for the wider corrosion and Mg research communities from the DOE Vehicle Technologies Office sponsored Technical Review Meeting: Current Status and Future Directions in Magnesium Corrosion Research held at Oak Ridge National Laboratory in April 2016. Magnesium is of interest for a wide range of structural and functional uses. Inadequate understanding and insufficient control of corrosion behavior severely limits structural uses of Mg and vehicle lightweighting to achieve improved fuel efficiencies. Corrosion mechanisms of Mg are arguably among the most complex and challenging of the structural metals. The meeting presentations and discussions highlighted both the great challenges to achieve more widespread use of Mg, but also the progress that has been achieved in recent years. The Mg corrosion research community is increasingly making use of a suite of new advanced electrochemical and microstructural characterization techniques to gain greater understanding of the corrosion behavior of Mg. The recent increase of interest in Mg corrosion phenomena by the modeling community is a particularly welcome development, as are the large number of recent international collaborations devoted to Mg corrosion studies. Substantial gains in the fundamental understanding of Mg corrosion will require synergistic insights from the corrosion, characterization, and modeling communities.

The authors thank M.G. Frith, D.N. Leonard, and B.A. Pint for helpful comments on this manuscript. The authors also wish to thank the speakers and attendees at the meeting for their participation and valuable insights. This research was supported by the U.S. DOE under the Vehicle Technologies Office. This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan).