You’ve seen the wet leakage current test fail. The module passes everything else—visual inspection, insulation resistance, ground continuity—then at 85°C and 85% RH, after 96 hours, the current spikes. The backsheet blisters, or you find a carbonized track under the edge seal. The flame retardant gets blamed first. But it’s not the flame retardant itself. It’s what came with it.
I spent three shifts last quarter tracking down a batch of EVA encapsulant that kept failing IEC 61730 wet leakage. The formulation was standard: 50 phr of magnesium hydroxide flame retardant (MDH), a silane coupling agent, and the usual crosslinking package. The only variable was the MDH source. We swapped from a general-purpose grade to a solar-grade MDH, and the failure rate dropped from 12% to under 0.5%. The difference wasn’t the particle size or the decomposition temperature. It was the ionic impurity profile.
Let me walk through what we found, because this is one of those cases where a specification sheet doesn’t tell you the whole story.
The Ionic Impurity Threshold in Damp Heat Testing
The first thing we did was run inductively coupled plasma (ICP) on the failing MDH batches. The general-purpose grade had sodium at 180 ppm, chloride at 250 ppm, and calcium at 400 ppm. The solar-grade MDH from Zhongyao Chemical—our product—runs sodium ≤50 ppm, chloride ≤100 ppm, and calcium ≤100 ppm. On paper, those numbers look like a factor of 2-4x difference. In practice, the effect on dielectric breakdown is nonlinear.
We set up a controlled experiment: same EVA formulation, same processing conditions (extrusion at 280°C, 5 m/min line speed), same backsheet construction. We spiked the solar-grade MDH with controlled amounts of NaCl and CaCl₂ to simulate the impurity levels of the general-purpose material. At 50 phr loading, when the ionic impurity concentration crossed a threshold equivalent to ~150 ppm sodium and ~200 ppm chloride, the dielectric breakdown voltage dropped from 18.2 kV/mm to 14.7 kV/mm. That’s a 19% drop—enough to cause failure in the wet leakage test.
The mechanism is straightforward: under damp heat, water vapor permeates the backsheet and dissolves the mobile ions. Those ions migrate under the electric field, forming conductive paths. The higher the impurity level, the faster the leakage current rises. The catch is that you don’t see this in dry testing. It only shows up after 85°C/85% RH aging. So if your QC only checks dry dielectric strength, you’ll miss it.
Particle Morphology and Dispersion in Thin Films
The second pain point we solved was pinhole formation. Thin-film backsheets—especially PET-based ones—are typically extruded at 280°C. If the MDH has a broad particle size distribution or irregular morphology, you get agglomerates that don’t break down during compounding. Those agglomerates become stress concentrators. During extrusion, they can cause local thinning or even pinholes.
Our solar-grade MDH has a platy hexagonal morphology. The D50 is controlled to 0.8-1.5 μm, and the specific surface area (BET) runs 6-12 m²/g. The platy shape means the particles orient parallel to the film surface during extrusion. That gives two advantages: first, the flame retardant path length is maximized in the through-plane direction, which improves flame retardancy. Second, the particles don’t create sharp stress risers. In production trials, we saw pinhole defect rates drop from 3.2% to 0.1% when we switched to the controlled morphology grade.
The downside of high surface area MDH is that it can increase melt viscosity. At 55 phr loading, we saw Mooney viscosity at 100°C jump from 42 to 58 units when the BET exceeded 35 m²/g. Our grade stays in the 6-12 m²/g range, so that’s not an issue. But if you’re sourcing MDH from a supplier that doesn’t control morphology, you could see both dispersion problems and viscosity shifts.
Decomposition Temperature Window for PET Extrusion
PET backsheet extrusion runs hot. Typical melt temperatures are 270-290°C. Magnesium hydroxide flame retardant starts decomposing to MgO and water at around 320°C. That’s the nominal onset. But the actual onset depends on the crystal structure and impurity content.
We tested decomposition onset using thermogravimetric analysis (TGA) at a 10°C/min ramp in nitrogen. The solar-grade MDH showed onset at 342°C—well above the extrusion window. The general-purpose grade started decomposing at 318°C. At 280°C, the general-purpose material had already lost 0.4% of its mass as water vapor. That water vapor, trapped in the melt, causes surface haze, bubbles, and in severe cases, die drool.
The difference comes down to the calcination process. Our material uses multi-zone kilns with automated temperature control, ensuring complete conversion without over-burning. The result is a consistent decomposition onset above 340°C. If your supplier can’t guarantee that, you’ll see surface defects on the backsheet that don’t show up until the film is laminated and tested.
Whiteness and Color Shift in Multi-Layer Films
This one is cosmetic but costly. Multi-layer backsheets often have a white layer to reflect UV and improve module efficiency. If the MDH has high iron content—above 50 ppm—it can cause a yellow shift after UV aging. We keep iron ≤30 ppm and whiteness ≥95. In a 12-month accelerated UV exposure test (per IEC 61730), the solar-grade MDH showed a ΔE of 0.8. The general-purpose grade with 80 ppm iron showed a ΔE of 3.5. That’s enough to fail a customer’s visual acceptance criteria.
Technical Specifications & Performance Data
The following table summarizes the key specifications for Solar Film Grade Magnesium Hydroxide based on internal QC batch testing and production analysis:
| Parameter | Value | Test Method |
|---|---|---|
| Purity (Mg(OH)₂) | ≥99.5% | XRF / Titration |
| Particle Size (D50) | 0.8 – 1.5 μm | Laser Diffraction (ISO 13320) |
| Specific Surface Area (BET) | 6 – 12 m²/g | BET (ISO 9277) |
| Loss on Ignition (LOI) | 30 – 31% | 1050°C, 2 hours |
| Chloride (Cl⁻) | ≤100 ppm | Ion Chromatography |
| Sodium (Na) | ≤50 ppm | ICP-OES |
| Iron (Fe) | ≤30 ppm | ICP-OES |
| Calcium (Ca) | ≤100 ppm | ICP-OES |
| Decomposition Temperature (TGA onset) | ≥340°C | TGA, 10°C/min in N₂ |
| Whiteness (Hunter L*) | ≥95 | Colorimeter |
In practice, these numbers translate to a flame retardant that hits V-0 at 50 phr in EVA, maintains dielectric strength above 18 kV/mm after 1000 hours of damp heat, and doesn’t cause pinholes or color shift. The cost premium over general-purpose MDH is roughly 15-20%, but the scrap reduction alone pays for it in a production run of 100,000 modules.
Frequently Asked Questions
Q: How does calcium oxide content affect dielectric strength in EVA encapsulant?
Calcium oxide isn’t typically a direct additive in MDH, but calcium ions from the ore can carry through. In our testing, calcium content above 200 ppm correlates with a measurable drop in dielectric strength after damp heat aging—about 1.5 kV/mm per 100 ppm increase. The mechanism is the same as sodium and chloride: calcium ions are mobile under humid conditions and contribute to leakage current. If your MDH supplier doesn’t control calcium, you’re taking a risk.
Q: Can I use general-purpose magnesium hydroxide in PET backsheet extrusion if I increase the coupling agent loading?
You can try, but it’s a band-aid. Silane coupling agents improve the interface between MDH and the polymer, but they don’t fix the ionic impurity problem. The mobile ions are in the bulk of the particle, not on the surface. Increasing coupling agent might improve dispersion and mechanical properties, but it won’t prevent dielectric breakdown under damp heat. The root cause is the impurity profile, not the adhesion.
Q: What’s the maximum MDH loading I can use in EVA without losing transparency?
For thin-film applications (0.2-0.5 mm), loading above 60 phr starts to scatter light noticeably. At 50 phr, with a D50 of 0.8-1.5 μm and controlled morphology, you can maintain >85% transmittance in the visible range. The platy hexagonal particles scatter less than irregular or spherical particles because they orient parallel to the film surface. If you need higher loading for flame retardancy, consider a co-additive like zinc borate or a halogenated flame retardant to reduce the MDH loading.
Q: How do I verify the decomposition temperature of MDH before it goes into production?
Run TGA with a 10°C/min ramp in nitrogen. Look for the onset temperature where 1% mass loss occurs. For solar-grade MDH, that should be above 340°C. Also check the derivative curve for a single, sharp peak—multiple peaks suggest incomplete calcination or mixed phases. If you don’t have TGA in-house, request a certificate of analysis from your supplier that includes the TGA curve, not just the onset number. The shape of the curve tells you more than the single value.
