Asbestos and floor adhesive: the black layer that stays behind
Lifting an old floor does not settle the question of what is underneath it. The adhesive is its own product, with its own history, and it is still stuck to the boards when the tiles are in the skip.
The black bituminous glue under an old Australian floor (usually called cutback adhesive, sometimes black mastic) is one of the products that can contain asbestos. And it is entirely independent of the tile or sheet above it. An unremarkable modern covering can be stuck down with an old adhesive. And a suspect tile can sit on glue that has nothing in it.
Neither can be settled by eye: only analysis by a NATA-accredited laboratory establishes what a given material contains. What this page is about is the practical consequence of that independence. The adhesive is the layer still there after the floor has gone. And it is the layer that drives people towards the worst possible methods.
What is cutback adhesive?
Cutback is a bitumen-based flooring adhesive, black or very dark brown, spread with a notched trowel so it dries in ridges. Intact, it is hard and brittle and looks like nothing much: a thin, dull skin welded to the timber or the concrete. It softens and goes tacky with heat, and it smears rather than lifts when it meets a solvent. Anyone who has taken up a floor from the 1960s or 1970s in Australia has met it.
“Asbestos adhesive” is a broader phrase than that, though, and worth pinning down. It gets used for flooring cutback, for some tile and sheet adhesives, and for mastic-type sealing and jointing compounds elsewhere in a building.
They are different products used in different places. If you want to know whether a specific named product was one of them, search the federal Asbestos Product Guide rather than a forum thread. It is built for exactly that question.
Why the glue is a separate question from the floor
An old floor is a stack, and asbestos can sit in any layer of it on its own terms: the tile or sheet, the paper or millboard backing bonded to the underside of sheet vinyl, and the adhesive holding the lot down. This site covers reading a floor layer by layer in its own guide. The layer that most often turns out to be the problem is the backing, not the surface.
The adhesive is the third question, and the one people forget, because it is the only layer that is not visible until the work has already started. Taking the covering off does not answer it. It exposes it.
The consequence for testing is direct: a laboratory result for the tile says nothing about the glue. They are separate materials and they need separate samples. Being told “the floor came back clear” is only useful if you know which layer was sent.
Why the adhesive is where the job goes wrong
Nobody sets out to disturb old glue. People set out to get back to a clean, flat substrate, because that is what the next floor needs. That goal, not carelessness, leads straight to the methods that must not be used on material that has never been identified.
Getting residue off a floor means scraping it dry, taking a grinder, a wire wheel or a hired floor sander to it, warming it with a heat gun until it lets go, or flooding it with an adhesive remover. Power tools, abrasive cutting and high-pressure water are prohibited on asbestos regardless of the quantity involved or who is doing the work. Every one of those approaches is a version of one of them.
The other thing worth noticing is the arithmetic. A skin of glue looks like nothing next to a sheet of fibro. But it covers the whole room, and the work is happening indoors, in an enclosed space, where anything released settles on surfaces rather than dispersing. In terms of area worked and dust generated, the thinnest layer in the stack is the largest job in it.
Is old adhesive bonded or friable?
Hard, intact cutback behaves like a bonded material. The content, if any, is locked into the bitumen and is going nowhere while the layer is undisturbed. That is the honest starting position, and it is why an intact residue under a covered floor is not an emergency.
Degraded adhesive is a different proposition. Glue that has gone powdery at the edges, been heated, been part-scraped by a previous owner or broken down into whatever is left of a failed backing is no longer a neat, sealed skin.
Material soft enough to be crumbled by hand pressure when dry is described as friable, and that classification changes who may lawfully do the work. Friable asbestos is Class A licensed work in every Australian jurisdiction, at any quantity. The boundary itself is set out in bonded versus friable.
In practice the classification argument is usually beside the point. What you are about to do decides it, not what the layer is. Grinding a hard bonded material produces exactly the dust the category was invented to prevent.
The option that keeps the adhesive out of it
There is a reason so many Australian houses have three generations of flooring stacked on top of each other, and it is not laziness. An intact, well-bonded residue left in place and covered over is not doing anything. Leaving it is a legitimate decision rather than an avoidance, and it is frequently the better one.
Two conditions make it honest. The new floor system must not need aggressive preparation of the old surface: if the build-up only works after the substrate has been ground flat, the option has quietly turned back into removal. And you have to write down what is under there, in the house records and in what you hand to the next owner. Then the person who eventually lifts it is not ambushed the way you nearly were.
If you do need to know what is in it
Sampling adhesive is more awkward than sampling a sheet. The layer is thin, stuck fast, and getting a piece off is itself the disturbance. The sample also has to be of the glue specifically, not of the tile sitting on it. That combination is a decent argument for having somebody competent take it rather than attacking a corner of the room with a chisel on a Saturday morning. The page on how sampling actually works covers what the process involves and who does it.
Until a laboratory has looked at it, the working assumption that costs you nothing is that the layer might contain asbestos and is therefore left alone. Regulators call that presuming, and it produces exactly the same behaviour as a positive result would.
What people do to old floor glue
| Approach | What it does to the adhesive | Where it stands |
|---|---|---|
| Dry scraping | Breaks a brittle layer into chips and fine dust indoors | Not a method to use on unidentified material |
| Sander, grinder or wire wheel | Works the whole floor area at once and generates the most dust of anything on this list | Power tools and abrasive cutting are prohibited on asbestos |
| Heat gun | Softens and smears the layer, and adds fume to the problem | Not a method to use on unidentified material |
| Solvent adhesive remover | Spreads the layer into a slurry over a wider area; removes nothing | Whatever was in the glue is still there, and now harder to contain |
| Sweeping or a household vacuum afterwards | Moves fine dust around the house | A domestic vacuum is not a containment device |
| Leaving it and covering over | Nothing: the layer stays sealed and undisturbed | Legitimate, provided no grinding is needed and it is recorded |
Frequently asked questions
Does old black floor glue contain asbestos?
Some bitumen-based cutback flooring adhesives were manufactured with asbestos and some were not, and the colour, hardness and age of the layer do not separate the two. It is also independent of the covering above it. Only analysis of a sample of the adhesive itself by a NATA-accredited laboratory establishes what is in it.
Can I scrape or sand old adhesive off the floor?
Not on material that has not been identified. Dry scraping, floor sanders, grinders and wire wheels all work the residue into airborne dust, and power tools, abrasive cutting and high-pressure water are prohibited on asbestos regardless of quantity. Heat guns and solvent removers soften and spread the layer rather than removing what is in it.
If the tiles tested clear, is the glue clear too?
No. The tile, the backing and the adhesive are three separate products and each needs its own sample. A result for one layer says nothing about the others, which is why it is worth asking which layer was actually sent to the laboratory.
Can I lay a new floor straight over old adhesive?
Covering an intact, well-bonded residue and leaving it undisturbed is a recognised approach and often the safer one. It stops being safe if the new floor system needs the old surface ground or sanded flat first, because that is removal by another name. Either way, record what is underneath for whoever lifts it next.
Sources
What this page explains is based on these official sources. If you check one and it does not match what you read here, send it to us and the article gets corrected.
- Asbestos Safety and Eradication Agency. Asbestos Product Guide. What products contained asbestos and where they were used
- NSW Government. Asbestos in NSW. The NSW Government asbestos portal