Magnesium Alloy Corrosion: Negative Difference Effect, Galvanic Coupling, and Protective Coatings

Magnesium alloys offer the lowest density of any structural metal — 1.74 g/cm³, approximately one-quarter the density of steel and two-thirds that of aluminium — making them highly attractive for weight-critical applications in automotive body panels, aerospace brackets, portable electronics casings, and biomedical implants. Their corrosion susceptibility, however, remains the primary engineering barrier to wider adoption. Understanding why magnesium corrodes so readily, what electrochemical anomalies govern its behaviour under polarisation, how micro-galvanic interactions within multi-phase microstructures accelerate attack, and which protective coating systems provide reliable long-term protection is essential for any materials engineer specifying or qualifying magnesium alloy components.

Thermodynamic Basis: Why Magnesium is so Reactive

Magnesium occupies one of the most electronegative positions in the electrochemical series of structural metals. Its standard electrode potential of −2.37 V versus the standard hydrogen electrode (SHE) compares with aluminium at −1.67 V, zinc at −0.76 V, and iron at −0.44 V. This extraordinary negative potential reflects the strong electron-donating character of the Group II metal: the two outer s-electrons are lost easily, and the Mg²+/Mg couple has a large negative Gibbs free energy of formation.

In practical electrochemical terms, the open-circuit corrosion potential (E_corr) of magnesium alloys in neutral NaCl solution ranges from approximately −1.45 V to −1.75 V versus SCE (depending on alloy composition, surface condition, and electrolyte). This means that virtually every metal magnesium contacts in a conductive electrolyte will act as a cathode — making galvanic coupling the dominant life-limiting corrosion mode for most magnesium alloy components.

The Pilling-Bedworth Ratio and Passive Film Quality

The Pilling-Bedworth (PB) ratio compares the molar volume of the oxide film formed to the molar volume of the metal consumed in its formation:

PB ratio = V_oxide / V_metal = (M_ox × ρ_M) / (n × M_M × ρ_ox)

where:
  M_ox  = molecular mass of oxide (g/mol)
  M_M   = atomic mass of metal (g/mol)
  ρ_M   = density of metal (g/cm³)
  ρ_ox  = density of oxide (g/cm³)
  n     = number of metal atoms per oxide formula unit

For MgO:
  PB = (40.3 × 1.74) / (1 × 24.3 × 3.58) = 70.12 / 86.99 = 0.81

For Al2O3 (comparison):
  PB = (102 × 2.70) / (2 × 27 × 3.99) = 275.4 / 215.5 = 1.28

Interpretation:
  PB < 1.0:  Film is in tension → porous, non-protective (Mg, Ca, Na)
  PB = 1–2:  Film is compact, coherent → protective (Al, Cr, Fe at high T)
  PB > 2.0:  Film buckles and spalls → poor protection (Mo, W)

The PB ratio of 0.81 for MgO means the oxide occupies only 81% of the volume of the metal it replaces — the film is porous, under tensile stress, and cannot prevent electrolyte penetration. This is in stark contrast to the aluminium oxide film (PB = 1.28), which forms a compact, largely defect-free barrier. The magnesium hydroxide Mg(OH)₂ that forms in aqueous environments is similarly non-protective, being soluble above pH 10.5 and porous below that value.

The Negative Difference Effect (NDE)

The negative difference effect is the most distinctive and counterintuitive electrochemical behaviour of magnesium, and understanding it is essential for correctly estimating galvanic corrosion rates in engineering structures containing magnesium.

In classical electrochemistry, the Butler-Volmer equation predicts that as the electrode potential of a metal is raised anodically (above the corrosion potential), the anodic partial current — metal dissolution — increases, while the cathodic partial current — hydrogen evolution and/or oxygen reduction — decreases. The total measured current is the sum of these two partial currents. For most metals, this means the measured current increases monotonically with anodic polarisation.

For magnesium, the opposite happens: as the electrode is polarised anodically, not only does metal dissolution increase as expected, but the rate of hydrogen evolution — a nominally cathodic reaction — also increases. The net result is that the actual rate of magnesium dissolution (measured by mass loss) is significantly higher than the rate predicted from the electrical current alone, because a substantial fraction of the corrosion involves an electrochemically anomalous hydrogen-generating dissolution pathway that is not captured in the electrical circuit.

Mechanistic Explanation: The Mg+ Intermediate

The most widely accepted mechanistic explanation for the NDE involves the generation of a monovalent magnesium ion (Mg⁺) as an unstable reaction intermediate:

Mg dissolution pathway including NDE mechanism:

Step 1 (primary anodic reaction):
  Mg → Mg⁺ + e⁻             [monovalent intermediate]

Step 2a (further oxidation):
  Mg⁺ → Mg²⁺ + e⁻           [normal anodic current]

Step 2b (NDE pathway — chemical reaction with water):
  Mg⁺ + H₂O → Mg²⁺ + OH⁻ + ½H₂↑  [generates H2 from ANODIC surface]

Net anodic dissolution (combining Step 1 → 2a):
  Mg → Mg²⁺ + 2e⁻           [normal Faradaic dissolution]

NDE contribution (Step 1 → 2b):
  Mg → Mg²⁺ + OH⁻ + ½H₂    [no electrical current but mass lost!]

Practical consequence:
  Mass loss = Faradaic loss + non-Faradaic NDE loss
  Actual corrosion rate = 1.5 to 4× the Faradaic prediction
  
Under galvanic coupling, as anodic current increases (more coupling),
NDE pathway accelerates → corrosion severely underestimated by
standard electrochemical kinetic models.

Micro-Galvanic Corrosion in AZ Series Alloys

The AZ series (Mg-Al-Zn) alloys, particularly AZ91 (nominally 9% Al, 1% Zn, 0.3% Mn), are the most widely used magnesium alloys in die casting. Their corrosion behaviour in chloride solutions is dominated by internal micro-galvanic cells between the microstructural constituents.

Microstructure of AZ91

As-cast AZ91 contains three main constituents:

• α-Mg matrix (HCP, major phase): the aluminium-lean, magnesium-rich solid solution. Electrode potential approximately −1.65 to −1.75 V vs SCE in NaCl — the anodic phase.
• β-Mg₁₇Al₁₂ (BCC intermetallic): forms at grain boundaries during solidification and eutectic reaction. Electrode potential approximately −1.20 to −1.35 V vs SCE — significantly more positive than the α matrix, making it the cathodic phase. Volume fraction and morphology depend strongly on cooling rate and Al content.
• Al-Mn intermetallics (AlMn, Al₈Mn₅): form during solidification; electrode potential approximately −0.5 to −0.8 V vs SCE — highly cathodic, but typically present at low volume fractions in high-purity alloys.

The Dual Role of β-Mg₁₇Al₁₂

The β-phase plays the counterintuitive dual role in AZ91 corrosion — it can either accelerate or retard corrosion depending on its morphology, which is controlled by solidification conditions and heat treatment:

Impurity Tolerance Limits and High-Purity Alloys

The most significant advance in magnesium alloy corrosion control in the past four decades has been the recognition and implementation of impurity tolerance limits. Iron, nickel, and copper form cathodic intermetallic particles with activities far higher than the β-phase, and even trace concentrations above the tolerance limits increase corrosion rate by one to two orders of magnitude.

Impurity tolerance limits for Mg-Al alloys (AZ91, AM60):

Iron (Fe):
  Tolerance: Fe/Mn mass ratio < 0.032  (or Fe < ~50 ppm absolute)
  Why: Fe forms FeAl3 and Al-Fe particles; high cathodic exchange current
       density for H2 evolution; even 20 ppm Fe increases corrosion rate 5×
  Control: Mn addition (0.15–0.5%) ties up Fe as Al8(FeMn)5 with lower
           cathodic activity than free Fe particles

Nickel (Ni):
  Tolerance: < 20 ppm (0.002 wt%)
  Why: Mg2Ni particles are highly cathodic; Ni is a very effective
       catalyst for hydrogen evolution in neutral/alkaline solutions
  Control: Primary melt cleanliness; avoid Ni-containing tools in contact
           with Mg melt; use high-purity feed stock

Copper (Cu):
  Tolerance: < 300 ppm (0.030 wt%) for Mg-Al alloys
  Why: Al2Cu particles are cathodic; Cu contamination from copper alloy
       tools or die components
  Control: Dedicated Mg alloy tooling; avoid Cu-containing alloy additions

Effect on salt spray corrosion rate (AZ91, ASTM B117 test):
  Low-purity AZ91 (Fe/Mn > 0.032, or Ni > 50 ppm): 10–50 mg/cm²/day
  High-purity AZ91D (all impurities below limits): 0.1–1 mg/cm²/day
  Improvement factor: 10–100× by impurity control alone

Galvanic Corrosion: Magnesium Coupled to Other Metals

Because of its extremely negative corrosion potential, magnesium is susceptible to galvanic corrosion when in electrical contact with any other structural metal in a conductive electrolyte. Galvanic corrosion is the dominant failure mode for magnesium alloy structural components in automotive, aerospace, and marine applications where multi-material assemblies are the norm.

Galvanic Series in NaCl Solution and Coupling Severity

Filiform Corrosion

Filiform corrosion is a specific form of galvanic-driven under-film corrosion that produces characteristic worm-track filaments beneath organic paint coatings on magnesium alloy surfaces. It initiates at coating defects (scratches, cut edges, drill holes) where the electrolyte contacts the bare metal, and propagates laterally beneath the intact coating as a network of advancing corrosion filaments.

The mechanism involves an electrochemical cell within each filament: the active head contains anodic magnesium dissolution (pH ~4–5, high Mg²+ concentration) advancing ahead of an alkaline tail (pH ~10–11, Mg(OH)₂ precipitate). The pH differential drives further dissolution at the head and maintains the propagation direction. Filiform corrosion is particularly prevalent on coated AZ31 and AZ91 components in automotive under-hood environments where temperature cycling and condensation create periodic electrolyte availability.

Prevention requires: proper surface preparation before coating (degreasing, pickling, and conversion coat application per applicable standard); selection of coatings with low water vapour transmission rate (WVTR); sealing of all cut edges and drill holes; and avoidance of thin organic coatings on magnesium without a conversion coat or anodic undercoat.

Protective Coating Systems for Magnesium Alloys

Given the inherent thermodynamic reactivity of magnesium, adequate corrosion protection for engineering service almost always requires a multi-layer coating system. The choice of system depends on the severity of the service environment, the required salt spray life, dimensional tolerances, weight, processing constraints, and regulatory requirements (particularly the restriction on hexavalent chromium under EU REACH Regulation 1907/2006 and US EPA guidelines).

Surface Preparation

All coating systems for magnesium begin with surface preparation — the most critical and most frequently compromised step. Magnesium alloy surfaces contain native oxide, machining oils, casting release agents, segregated surface phases, and sometimes heat tint or corrosion products that must all be removed before any coating is applied. Standard preparation sequence:

1. Alkaline degreasing (sodium orthosilicate or phosphate bath, 60–80°C, 5–15 min) to remove organic contamination without attacking the metal.
2. Water rinse — deionised water preferred; tap water can introduce chloride contamination.
3. Acid pickle (chromic-free: 15–20% phosphoric acid, 20–30°C, 1–3 min; or HNO₃/HF mixture) to dissolve the native oxide and expose a fresh, reactive surface. This step is time-critical — the freshly pickled surface oxidises within seconds in ambient air.
4. Immediate transfer to conversion coat or anodising bath — within 30–60 seconds of acid pickle completion to prevent re-oxidation.

Chemical Conversion Coatings

Conversion coatings form in-situ on the magnesium surface by chemical reaction of the metal with the bath solution, producing a thin (1–5 μm) inorganic film. They provide moderate corrosion protection themselves and, critically, serve as adhesion-promoting primers for subsequent organic topcoats.

Chromate conversion coatings (CCC) using hexavalent chromium (Cr⁶⁺) baths (DOW No.1 process, Koch process) historically provided the best corrosion performance (200–500 h salt spray) for a single thin layer, but are now restricted under EU REACH Annex XIV and equivalent regulations in most jurisdictions due to the carcinogenic and mutagenic properties of Cr⁶⁺. Trivalent chromium (Cr³⁺) conversion coatings, developed as direct REACH-compliant replacements, provide similar adhesion promotion with 100–300 h salt spray performance and are now the industry standard for chromate-type treatments.

Phosphate-permanganate coatings (ASTM D1732 Type II) provide Cr⁶⁺-free conversion coating through reaction of the surface with dilute phosphoric and permanganic acid. The resulting green-brown film has good adhesion properties and 100–250 h salt spray when topcoated.

Anodising

Anodising uses electrochemical oxidation to grow a thick (5–30 μm) oxide layer on the magnesium surface. Unlike aluminium anodising (where the oxide grows in sulphuric or chromic acid), magnesium anodising uses alkaline fluoride-containing or phosphate-containing electrolytes (HAE process: KOH + Al(OH)₃ + KF + Na₂Cr₂O₇; DOW 17: NH₄HF₂ + Na₂Cr₂O₇; modern Cr⁶⁺-free equivalents use silicate or phosphate-vanadate electrolytes).

The anodic layer is characterised by a thin, dense inner barrier layer (~1–3 μm) and a thicker, porous outer layer that must be sealed with polymer or wax to provide effective barrier protection. Unsealed anodic layers provide poor salt spray performance; sealed and topcoated systems achieve 500–1000 h.

Plasma Electrolytic Oxidation (PEO)

PEO (also called micro-arc oxidation, MAO, or spark anodising) represents the state of the art in magnesium surface treatment for demanding applications. It is the REACH-compliant technology that has largely replaced hexavalent chromate-based anodising in aerospace and automotive applications.

PEO Process Parameters:
  Electrolyte:   Aqueous alkaline solution
                  Silicate-based: NaOH (5–20 g/L) + Na2SiO3 (10–30 g/L) [most common]
                  Phosphate-based: NaOH + Na3PO4 + NaAlO2
  Temperature:   15–30°C (electrolyte cooled by heat exchanger)
  Voltage:       200–600 V (AC or pulsed DC)
  Current density: 5–30 A/dm²
  Time:          5–60 minutes (layer thickness ~0.3–1 µm/min typical)
  Resulting layer: 5–50 µm total thickness

Layer structure (from substrate outward):
  1. Dense inner zone (~1–5 µm): amorphous MgO + crystalline MgO;
     hardness HV 500–800; provides corrosion and wear resistance
  2. Porous outer zone (~10–45 µm): MgSiO3 + Mg2SiO4 + MgO;
     provides adhesion base for topcoat; must be sealed for corrosion

Phase composition (silicate electrolyte):
  MgO (periclase):       ~40–60 vol%
  Mg2SiO4 (forsterite):  ~20–35 vol%
  MgSiO3 (enstatite):    ~15–25 vol%
  Amorphous silica:      remainder

Typical properties:
  Hardness:     HV 400–800 (vs AZ91 base HV 65–80)
  Thickness:    15–30 µm typical for corrosion applications
  Bond strength: > 30 MPa (cross-cut adhesion, ISO 2409)
  Salt spray:   500–2000 h alone; 2000–5000+ h with sealer + topcoat

Organic Coating Systems (Primers and Topcoats)

For most engineering and automotive applications, PEO or anodic coatings are used as a base for multi-layer organic coating systems. The typical automotive exterior system for magnesium closures consists of:

Alloy-Specific Corrosion Considerations

AM60B (Mg-Al 6% Die Cast)

AM60B is preferred over AZ91D for automotive structural applications requiring high ductility and energy absorption (instrument panel beams, seat frames, steering column brackets). Its lower aluminium content (6% vs 9%) means less β-Mg₁₇Al₁₂ at grain boundaries, reducing the barrier effect but also reducing the brittle continuous β network. Corrosion resistance is slightly inferior to AZ91D in salt spray testing but acceptable in coated applications. High-purity grade (B suffix) maintains Fe/Mn < 0.032 and Ni < 20 ppm.

AZ31B (Mg-Al 3% Sheet)

AZ31B is the primary wrought (rolled, extruded) magnesium alloy. With only 3% Al, it has very little β-phase and relies almost entirely on high-purity practice for corrosion resistance. AZ31B sheet is used in aerospace interior panels, electronics housings, and sport equipment. Its corrosion resistance in bare form is comparable to AZ91D high-purity, but it is more susceptible to filiform corrosion under organic coatings due to the absence of the β-phase barrier network.

Rare Earth-Containing Alloys (WE43, ZE41)

Rare earth additions (Y, Nd, Ce, La) to magnesium alloys improve elevated-temperature creep resistance (used in aerospace gearboxes at 150–250°C) and also significantly improve corrosion resistance compared to AZ series alloys. WE43 (Mg-4Y-2.25Nd-1Zr) exhibits salt spray corrosion rates 2–5× lower than AZ91D because the rare earth elements form more corrosion-resistant intermetallic compounds and suppress the negative difference effect through surface film modification. The Zr addition (in place of Mn for Fe scavenging) refines the grain structure and further reduces cathodic particle density.

Biodegradable Magnesium for Medical Implants

In a unique application reversal, the high corrosion rate of magnesium is deliberately exploited in biodegradable orthopaedic implants (bone screws, pins, plates) where the implant is designed to dissolve in the physiological saline environment of the body over a period of 6–24 months as the bone heals. Alloy compositions for biodegradable implants must be biocompatible (no toxic alloying elements — Al and rare earths are restricted for permanent implants); the Mg-Zn and Mg-Ca binary alloys and ternary MgZnCa systems are extensively studied for this application. Corrosion rate control — achieving a predictable dissolution rate while maintaining adequate mechanical properties during the healing period — is the central design challenge.

Corrosion rate estimation from weight loss (ASTM G1):
  CR (mm/year) = (K × W) / (A × T × D)

where:
  K = 8.76 × 10⁴ (for mm/year, g, h, cm², g/cm³)
  W = mass loss (g)
  A = specimen area (cm²)
  T = immersion time (h)
  D = metal density (g/cm³)   [Mg: 1.74 g/cm³]

Example: AZ91D coupon (25×25 mm), 168 h in 3.5% NaCl, mass loss 0.050 g:
  CR = (8.76×10⁴ × 0.050) / (6.25 × 168 × 1.74)
     = 4380 / 1827 = 2.40 mm/year

High-purity AZ91D in same test: typical 0.05–0.50 mm/year
Low-purity AZ91 (Fe/Mn > 0.032): typical 5–50 mm/year
Biodegradable Mg-2Zn in Hanks' solution: 0.2–2.0 mm/year (design range)

Standards for Magnesium Alloy Corrosion Testing and Coating Qualification

Recommended Reference Books

Disclosure: MetallurgyZone participates in the Amazon Associates programme. If you purchase through these links, we may earn a small commission at no extra cost to you. This helps support free technical content on this site.