Quick answer: If you switched to a cheaper MDH with the same d50 and your elongation, impact or colour dropped, the particle size was never the specification that mattered. Flame retardant fillers are chosen in three steps: the mineral by your processing temperature, the grade by surface area and purity, and the coating by your polymer. Most bad filler purchases get the first step right and skip the other two.
Mineral flame retardants look like commodities. White powder, a decomposition temperature, a price per tonne. That impression is why so many compounders buy them on d50 and price alone — and why the same compounders call us six weeks later about a property they never tested for.
Four minerals, one mechanism
Every mineral flame retardant works the same way. Heated, it decomposes endothermically, releases water (and in some cases carbon dioxide), and leaves an oxide residue. The heat absorbed cools the polymer, the gas dilutes the fuel, the residue shields the surface. What differs is when that happens and how much of it there is.
| Filler | Decomposition starts | What it releases | Where it fits | The catch |
|---|---|---|---|---|
| ATH — Al(OH)3 | ~200 °C | 34.6 % water by weight | PE, EVA, flexible PVC, rubber — processing below ~190 °C | Foams if processed hotter |
| MDH — Mg(OH)2 | ~330 °C | ~31 % water | PP and hotter-processed resins, processing up to ~300 °C | Costs more; grade quality varies widely |
| Boehmite — AlOOH | ~340 °C | ~17 % water | Engineering plastics and thermosets where stability matters more than efficiency | Half ATH’s water, so far less cooling per kilo |
| Huntite / hydromagnesite (natural blend) | ~220–250 °C (hydromagnesite) | Water + CO2, spread from ~220 to ~750 °C | Cost-driven halogen-free work where a wide decomposition range helps | Natural ore — composition moves with the deposit |
On heat absorbed: published enthalpy figures for ATH range from about 1.05 to 1.17 kJ/g depending on the source, and around 1.36 kJ/g for MDH. Treat the individual numbers loosely; the ranking between them is what holds up.
The first choice is almost always made for you by processing temperature. If your melt runs above about 190 °C, ATH is out, and the real decision is between grades of MDH. That comparison is on our magnesium hydroxide flame retardant page, so I will not repeat it here.
Boehmite and huntite: the two you get pitched
Boehmite gets offered as an “upgraded ATH” because it survives processing to around 340 °C. It is a real advantage for engineering plastics and thermosets. But boehmite carries about 17 % water against ATH’s 34.6 %, so the endothermic effect per kilogram is much smaller. It is not a drop-in replacement on flame performance. Where it earns its place is as a heat-stable partner in a system — often with phosphorus chemistry — rather than as the whole package. If someone pitches it as ATH that works hotter, ask to see the burn data at your loading.
Huntite and hydromagnesite come as a natural mixture. Hydromagnesite releases water and carbon dioxide between about 220 and 550 °C; huntite releases carbon dioxide in two stages between about 450 and 750 °C. That wide spread is useful: the filler keeps working as the fire develops rather than spending itself in one step. The drawback is that it is an ore. The huntite-to-hydromagnesite ratio depends on the deposit, and so does its impurity content. Lot-to-lot consistency is the question to ask before price.
Why the same d50 gives a different result
This is where most of the problems we see come from.
d50 tells you the median particle size. It says nothing about particle shape, crystal perfection, or how much surface each gram exposes. Two MDH grades can share a d50 and have very different BET surface areas — and the BET value is closer to what your compound actually experiences, because surface area is what the polymer has to wet, what the coating has to cover, and what adsorbs your stabilisers.
For flame retardant MDH the working targets are a d50 of roughly 0.5–10 µm and a BET surface area below about 10 m²/g, typically in the 3–10 m²/g band. A grade outside that window can still pass a burn test and still wreck your melt flow or your elongation.
Precipitated or ground MDH
MDH comes two ways. Precipitated (synthetic) grades are made chemically, which gives high purity, a uniform fine particle size and a regular crystal structure. Ground grades are milled from natural brucite ore. They are cheaper, and for the right application they are fine.
The difference that bites is not particle size, which grinding can get close to. It is impurities. Patent literature on MDH is direct about this: when a molded article made with a large amount of magnesium hydroxide loses impact strength, elongation or tensile strength on heating, the cause is chemical — the kind and amount of impurities — rather than the physical properties of the particles. Ground ore carries whatever the deposit carried.
That tells you where each one belongs. Thick sections, low processing temperature, dark colours, loose mechanical targets: ground is a sensible saving. Thin walls, light colours, high processing temperature, or a customer who tests elongation after ageing: precipitated earns its price. For cable, natural MDH is used successfully in low-smoke compounds — but in grades selected and treated for it, not in whatever ground powder is cheapest that month.
The switch that looked like a saving
A compounder in Vietnam making FR PP for appliance parts moved from a precipitated MDH to a ground grade with the same d50. On paper it was the same material at a noticeably lower price. The first lot burned exactly as before.
The complaints came from their moulding customer two lots later: parts slightly off-white, and impact values down enough to fail incoming inspection. The compounder’s first theory was a processing drift. It was not. The ground grade’s BET was well above the old one, and the impurity profile was different; at PP processing temperatures that showed up as colour and as property loss after moulding.
They went back to precipitated for the appliance grade and kept the ground MDH for a thick-walled, dark-coloured industrial part where nobody measured impact. Both decisions were right. What cost them was making the first one without asking for a BET value and an impurity analysis, and I would have liked to be asked before the switch rather than after.
Coating: the third decision
Surface treatment is chosen by polymer and end use. Stearate coatings mainly improve dispersion and melt flow. Silane coatings cap surface hydroxyls, which improves coupling to the polymer and reduces water uptake — and in outdoor cable it also decides how long the UV stabiliser survives, because untreated metal hydroxide adsorbs it, as covered in UV resistant LSZH compound. How the coating actually gets onto the particle, pre-coated or in-situ on the extruder, is part of the LSZH cable compound production process.
What we would advise
If your processing stays below 190 °C and your sections are thick: buy ATH, and do not let anyone upsell you precipitated MDH. It is the cheapest effective halogen-free filler there is, and its margin is thin for everyone in the chain, including us. We will still tell you it is the right answer.
If you are choosing between MDH grades: specify d50, BET and coating together, and ask for an impurity analysis on ground grades. A spec with only d50 on it is an invitation to receive a different material at the same particle size.
If someone offers boehmite or huntite: ask what problem it solves in your formulation. Boehmite solves processing temperature, not efficiency. Huntite solves cost and gives a wide decomposition range, at the price of deposit-dependent consistency.
If you are changing filler supplier: run your own mechanicals after ageing, not just the burn test. Filler problems almost never show up in the flame result first. The general processability checks are in compounding flame retardants.
FAQ
What are the main flame retardant fillers?
Aluminium trihydroxide (ATH), magnesium hydroxide (MDH), boehmite and natural huntite/hydromagnesite blends. All work by endothermic decomposition, releasing water and in some cases carbon dioxide. They differ mainly in decomposition temperature and in how much heat and gas they release per kilogram. How they sit alongside phosphorus and nitrogen systems is in the full list of halogen-free flame retardants.
Which is better, ATH or MDH?
Neither in general. ATH starts decomposing around 200 °C, so it suits resins processed below about 190 °C and is the cheapest route. MDH holds to around 330 °C, so it is the choice for PP and hotter processing. Pick by your melt temperature first, then by grade.
Is boehmite a replacement for ATH?
Not on flame efficiency. It survives processing to about 340 °C but carries roughly 17 % water against ATH’s 34.6 %, so it absorbs less heat per kilogram. It works best as a heat-stable component in a combined system.
What BET value should flame retardant MDH have?
Below about 10 m²/g, typically 3–10 m²/g, with a d50 around 0.5–10 µm. Always specify BET alongside d50; two grades with the same d50 can differ widely in surface area and behave differently in the compound.
Can mineral fillers be combined with synergists?
Yes, and they often should be. Zinc borate is the common partner, adding char strength and smoke suppression and sometimes letting you trim the mineral loading slightly. Test the combination rather than assuming the effects add up.
Do you supply both precipitated and ground MDH?
Yes, alongside ATH, in several particle sizes and coatings, as part of our halogen-free flame retardant range. We are a sourcing and technical partner rather than a mineral producer, so we match the grade to your application across the factories we work with, and we will tell you when the cheaper grade is the right one.
Final thoughts
Flame retardant fillers are the least glamorous part of a formulation and the part most often bought on the wrong specification. The mineral follows from your processing temperature; the grade follows from surface area and purity; the coating follows from your polymer and where the part will live. Get those three right and the burn test is rarely the problem. Send us your resin, melt temperature, wall thickness and colour, and we will propose the filler — or the finished flame retardant compound if compounding at 60 % loading is not work you want on your own line. The wider range is under flame retardant additives, and the chemistry map is on the flame retardant chemicals hub.
—— Rectivas Materials 团队 技术工程师老陈