The Sand Problem Goes Global: How Australia and New Zealand
Forced the World to Pay Attention
Sand is the definition of “low risk” in most people’s minds: harmless, inert, and frankly a bit boring. Which is exactly why it’s such a problem when the lab results say otherwise. When dyed craft sands and polymer-bound “kinetic” sands start showing asbestos minerals, it forces an uncomfortable reset: this isn’t a demolition-site issue anymore – it’s a consumer product, supply-chain, and testing-method issue.
 |
| Loose dyed “craft” sand recovered from a school setting. Because it’s loose, dry, and easily disturbed, it should be treated as a potentially friable dust source (i.e., higher likelihood of becoming airborne during handling/clean-up) if asbestos is present. |
Australia and New Zealand “picked the scab”
Australia and New Zealand didn’t create the problem – but they were early to treat it like a real one.
Once recall activity started, both countries moved with a precautionary mindset that’s worth recognising. In New Zealand, the messaging and response quickly cut through the usual fog of “let’s wait for more information” and treated contaminated coloured sand as something needing immediate control. Australia’s consumer safety response similarly leaned into rapid recall communication and public guidance, rather than assuming it was a one-off anomaly.
That decisive approach did something important: it made sand-based products visible as a legitimate asbestos pathway. When a product sits in classrooms, early learning centres, therapy settings, and homes, the tolerance for uncertainty drops fast – and it should.
Let’s be precise: tremolite is the main asbestos type in this discussion
Across the sand results we’ve been dealing with in our work and the broader “asbestos in sand” narrative, tremolite has been the predominant asbestos type identified. That matters because tremolite in sands often points to source geology and processing/QA problems upstream – not something that can be fixed with a nicer label or a stronger marketing claim.
That said, it hasn’t been tremolite only. In the ACT, WorkSafe confirmed a decorative coloured sand product imported from China contained traces of chrysotile. Again: not a building, not demolition, not a factory – a consumer product in ordinary settings. The key takeaway isn’t “which fibre wins the worst award”; it’s that more than one asbestos type has shown up across the category, reinforcing why the response needs to be methodical, not reactive.
“Asbestos in sand” isn’t new – we just keep acting surprised
A big part of why this story resonates is that it punctures a comfortable myth: asbestos risk lives in old buildings and nowhere else. It doesn’t.
Concerns about tremolite in children’s play sand were raised decades ago. Langer & Nolan’s 1987 correspondence on “play sand” is often referenced because it captured a simple, enduring truth: when source geology, crushing, and quality controls go wrong, percent-level amphibole contamination in a consumer sand product is possible. The headlines change; the lesson doesn’t.
So why does today feel different? Two reasons:
- Modern products are more complex. Dyed sands and polymer/binder-rich “kinetic” sands are not chemically or physically identical to plain mineral sand.
- Modern supply chains amplify spread. The same product category can be imported, re-batched, rebranded, and distributed internationally at speed.
What we actually see in the lab: numbers that matter
At Aerem, we started with IANZ-accredited Polarized Light Microscopy (PLM) screening across both kinetic and craft sand products to triage findings and guide early decisions. But sands – especially dyed or organics-rich matrices – can be heterogeneous and method-challenging. So we escalated the same materials for robust confirmatory testing in the United States, specifically to verify results and quantify asbestos content using quantitative Transmission Electron Microscopy (TEM), specifically NY ELAP Method 198.4 (TEM-NOB).
In one coloured craft sand set, the confirmatory TEM-NOB results reported tremolite at trace to percent levels:
- Black sand: Trace tremolite (Organic fraction 51.7%)
- White sand: 3.3% tremolite (Organic fraction 44.7%)
- Green sand: 2.6% tremolite (Organic fraction 47.5%)
The companion PLM point count screening results for the same set reported:
- Black sand: None detected
- White sand: PC 3.5% tremolite
- Green sand: PC 1.8% tremolite
Those numbers are doing real work here. They show this isn’t always a “trace-only” conversation, and they show why escalation is not optional when decisions have real-world consequences. A “none detected” result by PLM in a fine, complex matrix can reflect method limits rather than true absence – and that’s not a criticism of PLM; it’s simply being honest about what each method can and cannot do well.
 Microscope view showing asbestiform tremolite present within the sand matrix (white, needle-like / fibrous bundles amongst the dyed grains). This aligns with the laboratory confirmation of tremolite asbestos detected in the sand sample set. |
 |
| Isolated/picked fibre from the sand submitted for confirmatory analysis and confirmed as tremolite asbestos by TEM (definitive fibre ID rather than “looks-like” microscopy). |
Why craft sand and kinetic sand behave differently
Dry craft sand is free-flowing. It is easy to disturb and easy to spread – and it is also easy to sample poorly if someone assumes one scoop represents a batch, especially across colours and lots.
Kinetic sand is polymer-bound and cohesive. It may appear less dusty in-hand, but analytically it can be more difficult: binders, dyes, and higher organic content can obscure optical identification and complicate preparation. The organic fraction figures above (around 45–52%) are a strong reminder that some “sand” products behave more like an organically bound matrix than a simple mineral.
From a risk-management standpoint, this means we should stop talking about “sand” as if it’s a single uniform material. It’s a product category, and the matrix matters.
Children, lungs, and the exposure reality we don’t routinely measure
Here’s the uncomfortable part for the profession: this is a children’s product issue, and our day-to-day asbestos hygiene work is still heavily construction-based.
Most asbestos exposure measurement, controls, and standards have been built around demolition, refurbishment, removal, drilling, cutting, and industrial disturbance – where the job tasks are defined, the controls are known, and the monitoring strategies are mature.
A sensory sand product in a classroom is not that world.
Children are not little adults. Their lungs and bodies are still developing, they spend more time closer to the floor, and their play behaviour can include vigorous disturbance and hand-to-mouth contact. Even where asbestos content is “low,” the exposure context is different – and it’s not one we have decades of routine occupational monitoring data for in the way we do for construction tasks.
This doesn’t mean we should panic. It means we should be cautious about pretending we can confidently quantify risk based on assumptions designed for demolition sites. Where the setting involves children, we should expect scrutiny to be higher – and our testing and control decisions need to match that reality.
The China supply chain point, without lazy conclusions
Undoubtedly people will ask: “Where is it coming from?”
A significant number of the recalled coloured/decorative sand products connected to early activity in Australia and New Zealand have been reported as manufactured in China, and investigations have pointed upstream toward quarrying and processing as the most plausible control points. That doesn’t mean “China equals asbestos.” It means upstream sourcing and QA are the real levers – and where supply chains repeatedly surface in recall activity, regulators and importers should focus verification efforts there.
“Country of manufacture” is a blunt instrument. Quarry source, processing steps, batch traceability, and method transparency are where prevention actually lives.
Why the world is catching up now
Once Australia and New Zealand treated the issue seriously, it became easier for other jurisdictions to recognise similar product categories in their own markets. Whether it presents as craft sand, decorative sand, sand-filled toys, or sensory products, the underlying theme is consistent:
- These are high-contact consumer settings.
- The products are often imported and re-distributed widely.
- The matrix can be fine, dyed, polymer-rich, or heterogeneous – exactly the conditions where method and sampling strategy decide what you see.
And as global testing increases, the picture is still developing. What we know today will sharpen as more lots, colours, brands, and supply chains are verified.
Actions: Moving from detection to exposure evidence (respirable risk)
Detection and quantification of asbestos in the product is critical – but it is only one part of the risk picture. The next step, and the part we currently lack robust evidence for, is respirable exposure risk in realistic (and worst-credible) use scenarios.
In plain terms: does the sand generate respirable airborne dust and fibres during typical handling and play, and under what conditions? We also need to answer a related question that often gets lost in the asbestos conversation: what is the respirable crystalline silica (RCS) risk in the same disturbance scenarios? Sensory play environments are not demolition sites – but they can still be enclosed, repetitive, and close to the breathing zone.
A defensible risk assessment therefore requires an exposure study design that can measure:
- Airborne respirable dust generation during representative handling (pouring, tipping, sifting, tray play, clean-up activities).
- Asbestos fibre release potential under controlled disturbance (including the influence of dyes/coatings, particle size, and matrix type).
- RCS exposure potential where silica-bearing sands are involved (because silica doesn’t need asbestos to create a respiratory hazard).
- The effect of moisture and handling controls (e.g., dampening, containment, cleaning method) on airborne release.
Practically, this points to a controlled chamber study (with appropriate safety and containment) designed to simulate realistic use, capture airborne samples in the breathing zone, and quantify outcomes using fit-for-purpose methods. It also allows us to test the key question stakeholders keep asking: can certain sands behave “more friably” under disturbance, meaning they generate a respirable fraction that is meaningfully higher than expected for “normal sand”?
This is where the profession needs to lean forward. Our historical asbestos hygiene muscle memory is construction-based – and this scenario is not. To close the gap, exposure science must catch up to product recalls.
What “good practice” looks like right now
This is solvable but only if we stop treating sand products as low risk by default.
- Sampling like you mean it: multiple increments across colours/lots/batches with traceability. One scoop is not a dataset.
- Two-tier testing with triggers: PLM for screening, with defined escalation to quantitative TEM-NOB for fine, dyed, polymer/organic-rich matrices or any suspected amphiboles.
- Supply-chain QA that’s real: certificates that state method, preparation pathway, and decision criteria – not just “tested.”
- Controls that match the setting: calm, practical communication – but precautionary handling/disposal pathways where recalls occur, especially in child-focused environments.
- Close the evidence gap: progress respirable exposure studies (including chamber-based work) to inform proportionate, defensible risk assessments for asbestos fibres and RCS in real use scenarios.
Australia and New Zealand didn’t start this. They made it visible – and visibility is what forces the global market to improve. The long-term win isn’t more headlines; it’s better upstream control, better method selection, and fewer families learning what “tremolite” means from a recall notice.
Submitted February 24, 2026/ Uploaded March 3, 2026
_______
1 Ben Alford can be reached by email at: ben@aerem.co.nz
