Your New Mattress Smells Strong. Here's What to Do Tonight, This Week, and When to Stop Waiting.

You are at the top of the curve tonight. Beckett et al. (2022) put two new memory foam mattresses in a simulated consumer use environment and watched airborne 2-propanol, acetone, chloromethane and toluene peak on the first day after installation, then fall for the next 31 days. Peer-reviewed The paper calls that setup a simulated consumer use environment, and a sleeping body adds heat, which is Step 2 — but the direction of travel from tonight is down.

Before you count days: every other page tonight gives you a number, usually 24 to 72 hours. We could not find the measurement behind it. What has been measured is the emissions, never the smell, so the steps below tell you what to check rather than what you will smell.

Here is the plan. Four steps, and the last one has a date on it.

1. Tonight. Open the room up, and decide where the sleeper goes.
2. Days 1 to 7. Ventilate, keep the room cool, wash the bedding. Only one of those does what most people think it does.
3. Day 14. Check whether it has changed, not whether it has gone. This is the day people decide something is wrong.
4. Day 30. Check your return window and run the escalation rule below. That is where the plan ends, and where the data ends too.

Step 1. Tonight

Sleep somewhere else if you have somewhere else. If not, you do not need to leave your house over this. Open the bedroom window a few inches and leave the door open too.

A shut bedroom is often poorly ventilated. Across twelve analysed nights in one Oakland house, doors and windows shut and no mechanical ventilation, Molinier et al. (2024) calculated the bedroom's air changing a median of 0.54 times an hour on the lower-bound estimate and 0.72 on the upper. Peer-reviewed In that same 37 m³ bedroom, half an air change an hour with two sleepers works out to one-third of the minimum ASHRAE Standard 62.2-2016 recommendation per person. Peer-reviewed One house, twelve nights, not a national average.

Most of what you can smell is probably on a fast clock. Probably, because what Beckett et al. (2022) timed was the emissions, not the odour: the fast phase halves in hours, roughly 4 or 12 depending on the chemical, while the slow one takes about 24 days. Peer-reviewed Reading a smell off an emissions curve is our inference, not their finding. Inferred

Don't spray anything on it: whatever you spray adds its own chemistry to the room you are trying to clear, and no mattress-emission study we could find tested a deodorizer. Inferred

Step 2. Days 1 to 7

Ventilate. It lowers what you breathe. Whether it makes the mattress finish sooner is genuinely open: Domhagen et al. (2024) open by stating that previous research on this is inconsistent, and identify a theoretical ventilation threshold beyond which enhanced ventilation has no further influence on emission rates. Peer-reviewed Their model covered gypsum board, particle board and carpet, with no mattress in it, so reading it across to foam is our inference. Inferred Ventilate anyway: lowering the concentration in the room is the part that reaches you.

Keeping the room cool matters more than it sounds, and the heated blanket stays off this week. In a chamber study of eight types of polyurethane mattress, Oz et al. (2019) found emissions rose significantly under sleeping conditions, with heat rather than humidity or carbon dioxide as the major contributor. Peer-reviewed Vaezafshar et al. (2025) saw the same direction in different chemicals: of 45 target compounds, 12 came off eight children's mattresses at room temperature, rising to 20 at body temperature, and 21 at body temperature with a child's weight applied. Peer-reviewed For the flame-retardant group ∑TCPP, warming one end of the mattress to body temperature raised the median emission from 74 to 380 picograms per square centimeter per hour. Peer-reviewed Turning the thermostat down is our reading of that, not something anyone has tested. Inferred

Washing the bedding is the cheapest thing here and the one with the thinnest evidence under it. Vaezafshar et al. (2025) expect bedding and children's clothing to act as a protective barrier by sorbing these chemicals, particularly if frequently washed. Peer-reviewed Keep the word expect: they wrote likely, citing other work, and did not test it here. Inferred

Step 3. Day 14, the day people look this up again

Do not ask whether the smell has gone. Ask whether it has changed since the first night, and write the answer down: that comparison is the only evidence you will have if you call the retailer. If someone who has been out of the house can give you a second opinion, take it. Inferred

Expect it to still be there, and treat that as an expectation rather than a finding. The slow half-life Beckett et al. (2022) fitted was about 24 days, so two weeks is not one full halving of the emissions. Peer-reviewed Nobody has drawn the equivalent curve for odour. Inferred Still smelling it on day 14 is not evidence of a defective mattress.

Step 4. Day 30, and when to stop waiting

Two clocks are running and only one is chemistry. Find your retailer's return or trial window tonight and write the date down. It is the only hard deadline here, and the one people miss while waiting for a smell to fade.

If any of these four is true, stop waiting and call the retailer.

1. The smell has not changed at all by two weeks. The published emission curve falls from day one; a smell holding steady is not tracking it. Inferred
2. Symptoms track the room. Headache, sore throat or watering eyes that start in the bedroom and clear after a day away. A doctor can use a record with dates on it. Inferred
3. An infant sleeps on it. Oz et al. (2019) estimated exposure for a sleeping child or infant and reported concerning exposure levels for few compounds, which is the authors' own phrasing. Peer-reviewed Air a baby's mattress fully first, and the parents' page sorts the rest.
4. You already unzipped the cover. That is a different problem with a different page: what to do if you've already removed the cover.

What nobody has measured: how long the smell lasts

Three independent groups have measured what comes off a new mattress, and one of them fitted a curve for how fast it falls. Not one has measured how long you can smell it.

We went looking in August 2026, searching Europe PMC and CrossRef for mattress crossed with odour, odor and emission, and found nothing on point. We found no regulator publishing guidance on mattress off-gassing either. The 24-to-72-hour figure everyone prints traces back to no measurement we could locate. Inferred

They are not the same quantity: an emission is micrograms leaving a surface every hour, an odour is whichever of those compounds sits above the threshold your nose can detect. So every day-count here, ours included, is read across from one to the other and labelled where that happens. Documenting the gap is the whole method.

Who did the measuring, and who paid for it

Beckett et al. (2022) concluded that the concentrations they measured and modeled were well below available health-based benchmarks, Peer-reviewed and all five of those authors worked for an industry consulting firm. That may well be right, and it is also what you would expect that authorship to conclude. Oz et al. (2019), with no such affiliation, reached a narrower verdict: for a sleeping child or infant, a few compounds reached levels the authors called concerning. Peer-reviewed

Every number here about the mattress itself came out of a chamber, a simulated room or a model; the one real-bedroom figure, about half an air change an hour, was calculated from occupant CO₂ in somebody else's house in Oakland.

CertiPUR-US conditions a foam sample for 72 hours at 23 ± 2 °C before it measures anything, Industry which tests new foam rather than the years you sleep on it; what that certification does and does not cover is its own page.

Questions people ask on the first night

How long should you let a mattress expand before sleeping on it?

Two different clocks, and people run them together. Expansion is mechanical, the foam recovering its loft after being rolled, and the only figure that applies to yours is on the instruction sheet in your box. There is a published timeline for the emissions and none for either the smell or the loft. A fully expanded mattress is not a finished one.

Does a mattress encasement prevent off-gassing?

Nobody has measured it. The closest thing in the literature is Vaezafshar et al. (2025) writing that bedding and children's clothing likely act as a protective barrier by sorbing these chemicals, particularly if frequently washed. Peer-reviewed They expected it, did not test it, and were describing semivolatile chemicals rather than the volatiles behind the new-mattress smell. Treat an encasement as unproven here. Inferred It is worth having anyway: you never need to unzip the cover.

My new pillow smells bad. Is it the same thing as the mattress?

Same material, same packaging, and almost certainly the same chemistry, but the published emission curves are for mattresses, so applying the timeline to a pillow is a read-across rather than a measurement. Inferred The difference that helps you is practical: a pillow can leave the bedroom. Unwrap it fully, air it somewhere else, and use an old one for a few nights. An old pillow that smells musty rather than chemical is a different problem entirely.

I just unboxed my foam mattress and it smells strong. When should I worry?

Four things, and none is the strength of the smell on night one: no change at all by two weeks; symptoms that start in the bedroom and clear after a day away; an infant sleeping on it; or a cover you have already unzipped, which is a fiberglass question rather than a smell question. A strong smell that is fading is what the emission curve would predict, though the odour itself has never been tracked. Inferred

Can I sleep on it tonight?

If you are a healthy adult with nowhere else to go, yes, with the window open. If the sleeper is a baby or a small child, give the mattress the airing time first. Next time, the question worth asking before it ships is whether the fire barrier is fiberglass.