Are fibreglass batts safe? What the makers’ own documents say.

Once it is laid and left alone most of the fibre stays where it was put, and the trouble starts the moment somebody disturbs it, which in a real house is every electrician and every plumber and every antenna bloke for the next thirty years. Cutting a batt or dragging one aside or pulling one out puts fibre into the air, and 10 to 20% of the fibres in glasswool are under 3 microns across, which is the size the code itself says reaches the deepest part of the lung. This is not a cancer claim — IARC moved insulation glasswool down to Group 3, not classifiable, back in 2001, and I take them at their word on that. My argument is about the size of the particle and what that means for how much of it ends up in a lung rather than caught by a nose and throat on the way through.

That is not a typo and it catches everybody out the first time. A fibre is a needle, so it has two measurements, and the definition uses both of them. Thinner than 3 microns across is what lets it ride the air past your nose and your throat instead of being caught on the way through, and longer than 5 microns is what stops the cleaning cells in your lung from wrapping it up and shifting it back out again. Thin enough to get all the way in, long enough to be hard to get back out. The 1990 national code words it as a particle with a diameter less than 3 micrometres and a length greater than 5 micrometres and a length to width ratio greater than 3 to 1, and that same definition is used by the World Health Organization and by NIOSH.

It is the same base chemistry and a different physical form, and the form decides everything. Fibreglass is essentially glass, and glass is a silicate melted out of sand, so at the element level fibreglass is silica. But glasswool is amorphous silica, meaning the molecules sit in no particular order, and silicosis is caused by respirable crystalline silica — the dust off cutting and grinding stone, concrete and engineered stone, where the molecules sit in a rigid repeating pattern. Safe Work Australia says plainly that crystalline is generally the more hazardous form, and the glasswool maker's own composition sheet lists the fibre itself as amorphous, non-crystalline, at over 85% of the product — the remaining share on that sheet is an organic binder, not a crystalline fraction of the glass. So fibreglass does not cause silicosis. That is different from fibreglass being harmless: the maker's own sheet still documents nose, throat and upper respiratory discomfort from installation dust, and that may matter more for anyone with asthma, bronchitis or other breathing issues.

Not the part that matters, and this is the bit that catches people. The code itself gives the scale, saying a strand of human hair is approximately 50 micrometres in diameter, so a 3-micron glass fibre is about a seventeenth the width of a hair off your own head — small enough that no naked eye will ever pick out a single one. What you can see in a shaft of sunlight through the manhole is the coarse stuff. The fine fraction is up there with it, doing nothing at all to announce itself.

On the fibreglass-vs-fibreglass numbers that exist, yes — laying glasswool batts averaged 0.14 respirable fibres per cubic centimetre for the installer in a field survey of 107 houses, and pumping loose fibreglass that DID have a binder averaged 0.55 for the installer, roughly four times higher. Loose fibreglass with no binder at all ran highest of the lot, averaging 7.67. That 1993 study measured the general category rather than any specific current product, so the 7.67 figure should not be hung on any one brand. What I can tell you about a product like Jetmax specifically is what its own maker's datasheet says: it is unbonded, meaning it has no binder holding the fibre together at all, which puts it closer to the highest-exposure category than the lowest one — though nobody has published a respirable-fibre figure for that exact product, so I have not put a number against it on this page.

Because they made the fibre finer, and finer cuts both ways. Owens Corning's own patent says conventional residential fibreglass has fibre diameters typically greater than 4 microns and that its new low-itch product runs 2.03 to 3.04 microns, measured against the wool industry's own prickle standard on a Wool Comfort Meter. The old specimens scored between 150 and 267. The new fine-fibre ones still scored between 57 and 140 — roughly half the old prickle number, not zero. The catch is in the government's own carcinogen profile: thinner fibres are deposited deeper into the lung, and it cites work finding that operations making smaller-diameter fibre put airborne fibre concentrations many orders of magnitude higher into the air, with more of those fibres of respirable size. A more comfortable batt is a finer batt, and that means more of it can be inhaled deep into your lungs when you use batts like this in your home.

Yes, and it carried the higher rating for thirteen years. IARC, the World Health Organization's cancer agency, classified glasswool in 1988 as Group 2B, possibly carcinogenic to humans, on sufficient evidence in animals and inadequate evidence in humans. In October 2001, published as Monograph 81 in 2002, IARC re-evaluated and moved insulation glass wool down to Group 3, not classifiable as to its carcinogenicity to humans, since the animal evidence was downgraded from sufficient to limited. What changed between the two evaluations was the product: manufacturers reformulated the glass to dissolve faster in lung fluid, and the new biosoluble fibres caused no fibrosis and no tumours in rodents even after lifetime high-dose exposure. Worth reading properly though — IARC's own preamble says a Group 3 evaluation is not a determination of non-carcinogenicity or overall safety, and any batt made before the reformulation is a Group 2B product by its own era's paperwork, sitting in plenty of roofs today looking identical to the current stock.

It is backwards, and it is the single most common thing I correct on the phone. People picture cellulose as a loose heap of paper blowing everywhere and picture a batt as tidy because it comes in a pack. The measured numbers go the other way. The NIH generated cellulose insulation dust the way an install generates it and found under 0.1% of the particles were of respirable size at all, and that the respirable fraction held no cellulose material in it. The same field survey that measured batt installation found loose fibreglass with a binder averaging 0.55 respirable fibres per cubic centimetre for the installer against 0.14 for laying batts. Ours is a heavier, coarser plant fibre that drops where it is put; theirs is a 2 to 8 micron glass filament that goes where the air goes. I have the full kilogram-by-kilogram breakdown of what that means for a real roof on a separate page.

It depends what is up there, and I will price both ways rather than push one. If the batts are sunken, gappy or full of rodent nests then removal and a pump of Cellulose Fibre Insulation is the job I would do — after more than 6,000 houses I can tell you the removal is the part nobody enjoys, because the product is itchy and normally stinks of rat urine. If the batts are still in reasonable shape then a top-over with cellulose keeps whatever value they have, fills every gap, covers the joists, and puts a seamless blanket over the old fibre so it stops getting stirred up every time somebody goes up there. Send me your address and I will price both.