Quick answer: If you run your own twin-screw line, the flame retardant question is not which chemistry works — it is which grade your extruder can actually process at the loading you need. Particle size, surface treatment and active content decide whether a package disperses at 60% filler or torques out your line. Two products with the same chemical name can behave completely differently on your equipment.
Most flame retardant content is written for people buying finished material. This one is for the compounder: what to specify when you are the one feeding powder into a twin-screw, and what goes wrong when the spec is incomplete.
You Are Buying a Different Product Than the Moulder Next Door
A moulder buying masterbatch asks about let-down ratio and dosing accuracy. A processor buying finished compound asks about the rating and the datasheet. Neither of them cares about particle size distribution — someone upstream already dealt with it.
When you compound in-house, that someone is you. The chemistry is the easy half; the hard half is that you are pushing 40–65% of a mineral or a crystalline powder through a machine that was designed to melt polymer. Four specifications decide whether that works, and only one of them usually appears on a supplier’s first quotation.
The Four Numbers That Decide Processability
| Specification | Typical Range | What It Changes on Your Line |
|---|---|---|
| Particle size (D50) | Mineral hydroxides commonly 0.5–2.5 µm | Finer disperses better and gives smoother surface, but raises melt viscosity and torque. Coarser runs easier and hurts elongation and surface finish. |
| Specific surface area (BET) | Roughly 1–15 m²/g | Higher surface area means more polymer wetting the filler — more torque, more coupling agent needed, more heat generated in the melt. |
| Surface treatment | Silane, stearate, titanate — or none | The single biggest lever on mechanicals at high loading. Treated grades lower melt viscosity and let the line run faster; untreated is cheaper per kilo and costs you elongation. |
| Active content | Varies by product and blend | Decides your real dosage and your real price. A blended package at 70% active is not comparable to a pure powder on price per kilo. |
There is a published data point worth keeping in mind on the first row: for ATH in polyolefin, a D50 around 2.3 µm with a moderately wide distribution has been reported as a good balance — fine enough to disperse and bond at the interface, coarse enough not to wreck the melt. That is a starting point for a trial, not a specification. Your screw configuration and throughput move it.
What Actually Goes Wrong on a Twin-Screw
Four failure modes, in the order we see them raised.
Feeding, not mixing. High-loading mineral packages are bulky and low-density, and side feeders starve before the barrel does. If you are hitting a throughput ceiling well below the machine rating, the constraint is usually volumetric feed rate rather than melt capacity — and the fix is a feeder change or a denser grade, not more screw speed.
Shear damaging what you paid for. Aggressive kneading blocks disperse agglomerates and also break down particles that were sized deliberately. With expandable graphite this is dramatic — over-shear it and the expansion that makes it work is gone before the part leaves the machine.
Moisture. Mineral hydroxides carry surface moisture, and some phosphorus systems are hygroscopic. Porosity, silver streaks and a compound that mysteriously fails burn testing at one production run and passes at the next usually trace back to drying discipline rather than formulation.
Temperature versus decomposition. This is the constraint people discover last and it is not negotiable: ATH begins releasing water around 180–200 °C, MDH around 300–330 °C. Run ATH in a polymer that needs 230 °C and it decomposes in your barrel — you get foam, not compound. This single number rules out more filler choices than price ever does.
Choose the System Your Line Can Actually Run
| Your Situation | Workable Route | Watch For |
|---|---|---|
| Polyolefin, LSZH target, torque headroom available | Surface-treated MDH or ATH at 55–65% | Elongation. At 60% filler some mechanical loss is physics, not formulation skill |
| Engineering resin above 250 °C | Phosphinate with MPP, or MDH — not ATH | Decomposition temperature is the gate. Check it before anything else |
| Styrenics, cost-driven, no halogen policy | DBDPE with antimony synergist at 15–25% | Lowest loading, easiest processing. Weaker on tracking and glow wire |
| Limited torque, older machine | Treated grades and a coarser D50; consider a pre-blended package | You will trade some surface finish. Better than not making rate |
How to Compare Quotes Without Getting Caught
Three things, and the price per kilo is not one of them.
Normalise to price per kilo of active. A blended package at 70% active against a pure powder is not the comparison the quotation implies. Ask for active content on every quote and do the arithmetic yourself — rankings reverse often enough that it is worth the five minutes.
Ask for the particle size distribution, not just D50. Two products can share a median and behave differently because one has a long tail of fines that raises viscosity. If a supplier cannot produce a PSD curve, they are reselling someone else’s material without knowing its spec.
Ask what the surface treatment is. “Coated” is not an answer. Silane, stearate and titanate behave differently with different base resins, and an untreated grade quoted 15% cheaper can cost you more than that in elongation and line speed.
What We Would Actually Advise
If your line is torque-limited, do not start by looking for a cheaper powder. Start with a treated grade at the same loading — lower melt viscosity often buys back more throughput than the price difference costs you. That is the cheapest experiment available to you.
If you are moving from masterbatch to buying powder, budget for the qualification rather than assuming a drop-in. Dispersion on your line will not match whoever pre-dispersed it before, and the first trial usually needs a screw configuration change. Plan two trials, not one.
If a supplier quotes you a drop-in match for a grade you already run, ask what evidence they have. Identical chemical names still differ in particle size, surface treatment and active content, and those differences land in your dispersion rather than in the quotation. We will not claim a drop-in without a trial — and if someone does claim it without asking what resin you run, that tells you something.
And the case where you should not buy powder at all: if your loading is above roughly 40–50% and your line is already running near its limit, the honest answer is that finished compound is cheaper per acceptable part once scrap is counted. We would rather tell you that than sell you a package you will fight with.
FAQ
What particle size should I specify for a flame retardant filler?
For mineral hydroxides in polyolefin, D50 in the 0.5–2.5 µm range covers most compounding work, with something around 2 µm a reasonable starting point. Finer improves dispersion and surface finish but raises torque; coarser runs easier and costs elongation. Specify it against your screw configuration and target throughput rather than copying a datasheet.
Does surface treatment actually justify the price?
At high loading, usually yes. Treated grades lower melt viscosity, which either raises your line speed or lets you hold the same rate at lower temperature, and they recover a meaningful share of the elongation that filler loading costs. At loadings below roughly 30% the benefit is smaller and untreated may be the economical choice.
Can I switch from masterbatch to powder additives to save money?
If you own twin-screw capacity, the cost per kilo of active is almost always lower with powder. What you take on is dispersion responsibility, dust handling, and a qualification cycle. Budget for two trials and a possible screw configuration change before comparing the savings.
Why did my compound pass burn testing once and fail later?
Three usual causes, in order: moisture in the filler or the resin, active content drift between lots, and dispersion variation from a feeder or screw change. Check incoming ash content against the COA first — it is the cheapest test that catches loading drift.
What minimum order applies to powder additives?
Trial quantity and production MOQ are different conversations, and it is worth asking for both figures in one enquiry so you can plan the qualification. Standard grades and custom blended packages carry different minimums.
Can you match the particle size distribution I already use?
Often, provided you share the current PSD or the supplier’s technical data sheet. What we will not do is claim a match without a trial on your own line — and you should treat any supplier who does so without asking about your resin and screw configuration with some caution.
Final Thoughts
Compounding your own flame retardant material is the lowest cost per kilo of active available to you, and the reason it stays cheaper is that you absorbed the difficult part yourself. So specify like it: particle size, surface treatment and active content on every enquiry, not just chemistry and price. Send us your resin, target rating, loading and line details and we will propose a system with a starting specification rather than a product list — see the additive range or the chemistry overview to start from a system rather than a product.
If you do not own compounding capacity, the equivalent decision on the masterbatch side has its own sequence — carrier, active content, process fit and incoming checks. It is set out in how to choose a flame retardant masterbatch.
Halogen-free packages behave very differently from one another on a twin-screw: mineral systems at 55–65% versus phosphinate at 15–25% is not the same processing problem. The families and their loadings are compared in the full list of halogen-free flame retardants. At that loading the filler coating stops being a dispersion detail and starts deciding outdoor life — the case is made in filler treatment in UV resistant LSZH compound. For the most demanding case — mineral at 60 % — the feed sequence zone by zone is laid out in LSZH cable compound production process. Before the process question comes the grade question: same d50 does not mean same material, as set out in choosing flame retardant fillers. Buyers comparing finished halogen-free grades rather than running their own line will find the quote-normalisation checklist in halogen free compound.
—— Rectivas Materials 团队 技术工程师老陈