How is this different from blushing?

By where the haze sits, and the test takes two minutes. Amine blush is a reaction product that forms on the surface of a curing coating. It is water soluble, so warm water and a scrub pad lift it and the floor underneath is clear. That is a genuinely good outcome and it is the most common cause of a cloudy floor by a wide margin. Whitening from moisture reaction is inside the film. The gas that caused it was generated within the coating while it cured, so there is nothing lying on top to remove. Wash it and you get a clean, still cloudy floor. So wash a patch first, always, before assuming the worse of the two. If it clears, you have blush and a page that tells you how to prevent it next time. If it does not, you are here. The third possibility is that the haze is air whipped in at the bucket rather than gas made in the film. That is micro foam, and the cloudy clear coat router separates all of these by what you can see.

Why does a polyaspartic react with moisture at all?

Because one half of the chemistry is an isocyanate, and isocyanates react with water. A polyaspartic is built from an aliphatic polyisocyanate and a polyaspartic ester. The intended reaction is between those two. But the isocyanate will also react with any water it can reach, and that unintended reaction produces carbon dioxide. In a slow coating that gas has time to rise and escape while the film is still open, and you may never see it. A polyaspartic does not give it that time. The film is skinning within minutes, so the gas gets caught, and millions of trapped bubbles too small to see individually read as whiteness or a pearl sheen. This is why the fault is so much more common on polyaspartics than on epoxies, and why it gets worse rather than better as you move to faster products. The same speed that lets you return a floor to service the same day is what stops the gas escaping.

Where is the moisture coming from?

Three places, and they are worth ruling out in this order because they need different responses. Humid air is first and most common. The coating pulls moisture from the air it is curing under, and above a certain relative humidity there is simply enough of it to matter. This is the one the cure time calculator flags, and it is independent of how warm or cool the slab is. Condensation on the slab is second. If the surface is at or below the dew point of the air, water forms on it, and the coating is laid onto a wet floor whether or not anyone can see it. The dew point rule exists for this, and the cloudy clear coat router works through it in detail. A green or damp slab is third. Concrete that has not dried, or that is passing vapor from below, supplies moisture from underneath for as long as the coating stays open. That is a different problem with a different fix, and it is worth testing for rather than guessing at. All three end in the same place chemically, which is why one appearance has three separate preventions.

What humidity is too high?

The cure time calculator raises a blushing and hazing flag once air relative humidity is above 85%, because several polyaspartics start to haze there regardless of what the slab temperature is doing. Above 90% it calls the risk high. Treat both as prompts to check your own data sheet rather than as the answer. Products differ more on this than on almost anything else, and manufacturers publish a humidity ceiling precisely because they know it varies. Some tolerate conditions that ruin others. It is also worth being clear that this is air humidity, measured with a hygrometer in the space you are working in. It is not the in situ relative humidity of the slab, which is a probe reading from a hole drilled in the concrete and belongs to the moisture question above. The two get confused constantly and they send you to different fixes. And humidity is a second, independent check rather than a substitute for the dew point spread. Passing one does not excuse the other.

Can I fix a floor that has already cured white?

Not by cleaning it, and that is the practical consequence of the haze being inside the film. Washing, buffing and solvent wiping all act on the surface. The whiteness is distributed through the coating, so removing a surface layer does not remove it and polishing only changes how the cloudiness catches the light. A clear coat over the top is usually disappointing for the same reason. The hazy layer is still under it, so you get depth over cloudiness rather than clarity. What works is removing the affected coat and recoating in conditions that will not repeat the fault. If the whitening is confined to the topcoat and the layers below are sound, that can mean abrading back rather than grinding to concrete, which is a substantially cheaper job. Check the recoat window before you plan on it. The honest summary is that this defect is much cheaper to prevent than to repair, which is why the conditions checks are worth doing even when they feel like a delay.

How do I stop it happening on the next floor?

Four things, and none of them is a product change unless the first three fail. Measure the conditions rather than judging them. Air temperature, relative humidity and slab temperature, taken in the space you are about to coat rather than outside it. Conditions are the one piece of evidence that disappears within a day, so recording them also protects you if the floor is later disputed. Keep the slab clear of the dew point, with margin. A floor that is comfortably clear in the morning is often not clear by late afternoon, so the check that matters is the one for the conditions you will actually be finishing in. Control the space rather than the coating. Dehumidification, ventilation and warming the slab all move the numbers. Closing a building up and running a dehumidifier the night before is more effective than anything you can do once the material is mixed. And read the humidity ceiling on the data sheet. If the conditions you get regularly sit above it, that is when the product choice becomes the answer.

Is a whitened floor still sound?

Usually structurally, but with a genuine caveat that separates it from blush. The coating has still cured. It will still be hard, chemically resistant and bonded, and on a floor where appearance does not matter it may be acceptable to leave. The caveat is that the gas which caused the haze occupies volume. A film full of trapped micro bubbles is not quite the solid film that was specified, and where the bubbling is heavy it can mean slightly reduced abrasion resistance and more paths for liquids to sit in. That is a real if modest difference, and it is worth mentioning to a client rather than describing the floor as cosmetically affected only. The other consideration is that it is a symptom. A coat that reacted with moisture cured in conditions outside what the product wanted, and whatever went into that coat may have affected the bond as well. If the floor is also showing adhesion problems, treat the whiteness as the visible part of a bigger conditions failure rather than as the whole of it.

Polyaspartic Whitening: A Haze That Will Not Wash Off

Most cloudy clear coats are blush sitting on the surface, and warm water takes them off. This is the one that does not come off, because the whiteness is not on the coating, it is in it. Polyaspartics react with moisture as they cure, and that reaction makes gas inside a film that is already closing.

Quick answer

Wash a test patch with warm water and a scrub pad before anything else. If the haze lifts and the floor underneath is clear, you have amine blush and the much better outcome, covered on blushing. If the patch comes up clean and still cloudy, the whiteness is cured into the film: the coating reacted with moisture while it was curing and made gas inside itself. That is this page.

Where the haze sits decides everything

Three unrelated things make a clear coat look milky, and the cloudy or milky clear coat router separates them by what you can see. This page is the fourth, and it is the one that branch describes without having anywhere to send you.

On the film

Warm water and a scrub pad lift it, and the coating under it is clear
Cause: amine blush, a reaction product formed on the surface
The common case by a wide margin, and the cheap one
Go to blushing

In the film, from the bucket

Tiny bubbles you can resolve individually if you look closely enough
Cause: air whipped in during mixing and never released
Mechanical, not chemical. It happened before the material reached the floor
Go to bubbling

In the film, made by the film

An even milky or pearl sheen with nothing you can resolve, and it does not wash off
Cause: the coating reacted with moisture and generated gas inside itself
Worst where the floor was coolest or the air dampest
This page.

Why a polyaspartic does this and an epoxy mostly does not

A polyaspartic is built from an aliphatic polyisocyanate and a polyaspartic ester, and the intended reaction is between those two. But an isocyanate will react with any water it can reach as well, and that second reaction produces carbon dioxide inside the coating.

In a slower material the gas has time to rise and leave while the film is still open, and often nothing shows. A polyaspartic gives it no such time. The film is skinning within minutes, so the gas is caught where it formed, and bubbles far too small to pick out individually read as whiteness or a pearl sheen across the whole surface.

That has an uncomfortable corollary worth stating plainly: the faster the product, the worse this gets. The same speed that returns a floor to service the same day is exactly what stops the gas escaping, and it is the reason this defect is a polyaspartic problem rather than a general coatings one. If the material was also going off before you could spread it, the two complaints have one cause and curing too fast covers the other half.

The three places the moisture comes from

All three end in the same chemistry, which is why one appearance has three separate preventions. Rule them out in this order.

Humid air, first and most common. The coating takes moisture from the air it is curing under. The cure time calculator raises a hazing flag once the air is above 85% relative humidity, because several polyaspartics start to haze there whatever the slab is doing, and above 90% it calls the risk high. Both are prompts to check your own data sheet rather than answers in themselves, because products differ on this more than on almost anything else
Condensation on the slab, second. If the surface sits at or below the dew point of the air, water forms on it and the coating goes onto a wet floor whether anyone can see it or not. The dew point calculator gives you the spread you actually had, and the cloudy clear coat router works the rule through in full rather than repeating it here
A green or damp slab, third. Concrete that has not dried, or that is passing vapor from below, feeds moisture upward for as long as the coating stays open. Test for it with the concrete moisture calculator rather than guessing, because the fix is a different one again

Two different humidities, and mixing them up sends you to the wrong fix. The percentages above are air humidity, read off a hygrometer held in the room. Slab readings are in situ relative humidity, taken with a probe in a hole drilled into the concrete. They measure different things and they belong to different branches.

Why cleaning it does not work

Every instinct here acts on the surface, and the whiteness is not on the surface. It is distributed through the thickness of the coating, so taking a layer off the top removes some coating and no haze.

Washing is the right first test and the wrong repair. If it were going to work you would already know, because that is the blush test
Buffing or polishing changes how the cloudiness catches the light rather than removing it
A clear coat over the top is usually disappointing, because the hazy layer is still underneath. You get depth over cloudiness rather than clarity

What works is taking the affected coat off and recoating in conditions that will not repeat it. If the whitening is confined to the topcoat and the layers below are sound, that can mean abrading back rather than grinding to concrete, which is a very different bill. Settle the recoat window first with intercoat delamination, because a new coat that does not bond has not improved anything.

Preventing it on the next floor

Measure the conditions rather than judging them. Air temperature, relative humidity and slab temperature, taken in the space you are about to coat. Conditions are the one piece of evidence that disappears within a day, so writing them down also protects you if the floor is disputed later
Keep the slab clear of the dew point with margin, and check it for the conditions you will be finishing in rather than the ones you started in. A floor comfortably clear at nine is often not clear at four
Control the space, not the coating. Dehumidification, ventilation and warming the slab all move the numbers you are fighting. Shutting a building up and running a dehumidifier overnight does more than anything available to you once the material is mixed
Read the humidity ceiling on the data sheet. If the conditions you routinely work in sit above it, the product choice is the answer and no amount of technique substitutes

Is the floor still sound?

Structurally, usually yes. The coating cured, and it will be hard, chemically resistant and bonded. On a floor where appearance is not the point, leaving it may be reasonable.

There is a real caveat that separates this from blush, though, and it is worth saying to a client rather than describing the floor as cosmetically affected only. The gas that caused the haze takes up volume, so a film full of trapped micro bubbles is not quite the solid film that was specified. Where it is heavy that can mean slightly less abrasion resistance and more places for liquids to sit.

The other thing to weigh is that it is a symptom. A coat that reacted with moisture cured outside the conditions the product wanted, and whatever affected that coat may have affected the bond too. If the floor is showing adhesion problems as well, treat the whiteness as the visible part of a larger conditions failure rather than the whole of it.

Still not sure which of these you are looking at? Start from what you can see and work back with the defect router.