Under a freezer floor, the board is extruded polystyrene. Here is why it is not urethane.
Two rigid foams that look interchangeable on a submittal behave nothing alike once they are under a slab with frozen ground beneath them. This is a floor question only.
Article
This comes up on almost every freezer we are brought into: somebody has specified a rigid insulation under the slab and the submittal says urethane. It is a reasonable-looking substitution. It is also the one we ask to have changed, every time, and the reason has nothing to do with R-value on day one.
The difference is the cell, and under a slab the cell is everything
Urethane is an open-cell foam. Extruded polystyrene is closed-cell. On a spec sheet that reads as a manufacturing detail. Under a freezer floor it is the whole argument, because an open cell is a place for water to go.
The ground under a freezer is not dry and it is not going to become dry. It holds moisture, and the assembly above it is pulling heat out of it continuously. Give that moisture somewhere to collect and it will collect there.
Insulation that has taken up water is no longer the insulation that was specified. Its R-value is not what the drawing says, and the number on the drawing is what the whole floor assembly was designed around. Nobody finds out on day one. They find out in year three, when the ground under the slab has started to freeze anyway.
And a board that takes up water stops staying where you put it
There is a second problem that gets less attention and is just as expensive. A closed-cell board under a slab is very nearly neutral in water. It neither soaks nor floats away. That is a useful property in an assembly that sits below grade for forty years.
A board that absorbs is a board whose weight and behavior change over time, in a place nobody can inspect, under a slab nobody can lift. Whatever it does next, it does out of sight.
This is why the manufacturer literature on the boards we specify leads with water resistance and with retaining insulating properties over time, rather than with the R-value alone. Under a floor, holding the value is the specification.
This is a floor argument. Do not carry it onto a wall
We want to be exact about the scope, because the conclusion above is genuinely wrong in the wrong place.
In a wall, moisture migrates vertically rather than lying against the board, and in an insulated metal panel the core is vapor-sealed between two steel faces before it ever reaches the site. The conditions that make an open cell a liability under a slab are not the conditions inside a panel wall.
So: when somebody tells you urethane performs well in a cold-storage envelope, they are very likely right, and they are talking about a different part of the building. The floor is its own problem and it is the one this piece is about.
What we actually ask for
Closed-cell extruded polystyrene, non-scored, graded against the slab, the load and the racking that will stand on it. Scored backfill and foundation board is made to snap for a different job, and substituting it in is one of the quieter ways a floor assembly goes wrong.
The grade is a selection, not a default. That conversation is short and it happens once, before the pour.
If there is a rigid insulation on your sub-slab submittal and you are not certain which foam it is, send it over. It is a five-minute read and it is a great deal cheaper than the alternative.
Sources
Every figure above comes from one of these, with its publisher and date so you can check it. Our own experience — the failure modes and the selection decisions — is ours, and is not sourced to anybody else.
- GreenGuard GG60-LG XPS Insulation Board — Kingspan Insulation US, retrieved September 2026
D.W. Ross Company — Building envelopes that have to hold a temperature, from Cleveland, since 1921. These notes come out of our own projects.
Where this applies
Got a room that is not behaving?
Monday – Friday, 8:00 AM – 4:00 PM. 3195 West 63rd Street, Cleveland, Ohio 44102.