What goes under the concrete in a slab-on-grade home? In the old days, not much — just dirt. Eventually, contractors discovered that it made sense to include a 4-inch-thick layer of crushed stone under the concrete. The crushed stone provides a capillary break that reduces the amount of moisture flowing upward from the damp soil to the permeable concrete.
Since the crushed stone layer provides a fairly uniform substrate, it also may also reduce the chance that a concrete slab will be poorly supported by random pockets of soft, easily compressible soil.
Eventually, polyethylene was invented. Concrete contractors learned that a layer of poly helps to keep a slab dry, because it stops upward vapor diffusion from the soil.
Finally, some contractors in cold climates began installing a continuous horizontal layer of rigid foam insulation under their concrete slabs. The foam layer isolates the room-temperature slab from the cold soil under the slab.
Getting the sandwich layers in the right order
At this point, we’ve got a sandwich with three or four layers, and the question arises: does the order of the different layers matter? What goes down first, and what goes down last?
According to most building scientists, here’s how the layers should go, from the bottom up: crushed stone; rigid foam; polyethylene; concrete.
Can I still use blotter sand?
Some contractors may ask: Is it a mistake to put the polyethylene lower down in the sandwich? The answer is yes. To understand why, it’s useful to study the history of blotter sand.
Beginning in 1989, the American Concrete Institute (ACI) recommended the installation of a 4-inch layer of granular material between a sub-slab vapor retarder and a concrete slab. ACI standard 302.1 R-96, Guide for Concrete Floor and Slab Construction, included this recommendation in Section 4.1.5: “If a vapor barrier or retarder is required due to local conditions, these products should be placed under a minimum of 4 inches (100 mm) of trimable, compactable, granular fill (not sand).”
Although the ACI specified “granular fill,” most residential builders found it more convenient to use sand. This layer became known as the “blotter sand” layer.
After a rash of flooring failures, the standard was revised
Because a layer of blotter sand between the polyethylene and the concrete allows for faster slab finishing and tends to reduce slab curling, especially if the concrete has a high water/cement ratio, many residential builders readily adopted the ACI recommendation. The use of blotter sand became particularly common in California.
In the late 1990s, however, some builders of slab-on-grade homes began receiving regular reports of flooring failures. Investigators discovered that many of these failures could be blamed on the inclusion of blotter sand above the poly.
Finally bowing to the evidence, the ACI eventually reversed the long-standing but controversial recommendation. In the April 2001 issue of Concrete International magazine, the ACI advised that for slabs with vapor-sensitive coatings (i.e., virtually any kind of flooring), the vapor retarder should be installed directly under the slab, with no intervening layer of granular fill. In November 2001, the ACI included the updated recommendations in a new edition of ACI standard 302.1R-96.
Concrete contractors love blotter sand
Contractors used blotter sand because it provided certain undeniable benefits. Sand helps protect the polyethylene vapor retarder from abuse. With a layer of blotter sand, excess water can drain out of the fresh concrete, making it easier to place concrete with a high water/cement ratio. In contrast, when polyethylene is installed directly under the slab, bleed water dissipates more slowly and concrete finishing is delayed. When excess water leaves the slab from the top side only, the slab may cure unevenly, in some cases resulting in slab curling, shrinkage, or cracking.
When I wrote an article on the topic of blotter sand for the May 2002 issue of Energy Design Update, I interviewed Howard Kanare, who at that time was a senior principal scientist at Construction Technology Laboratories in Skokie, Illinois. “On one side are the folks concerned with construction practices; they don’t want to see a plastic sheet directly under the slab,” Kanare told me. “On the other side are the people who have investigated failures. We almost never see a failure due to moisture infiltrating concrete when the concrete is directly on the plastic sheeting.”
A sub-slab sponge that stays wet
While a sponge under the floor may help contractors during construction, it will probably be a headache for those who have to live in the house. “The problem is that the blotter layer needs some moisture to be compacted,” Kanare said. “It might be at a moisture content of seven percent, and if the layer is four inches thick you have a couple of pounds of water under each square foot. You are essentially forming a bathtub. It’s a long-term source of moisture to recharge the concrete.”
Water added for compaction is just one of many possible sources of moisture keeping blotter sand damp. Another is rain, as I learned when I spoke with Kelley Roberts, a water-intrusion consultant at Construction Forensics in Huntington Beach, California. “At one home I investigated, they put down a bulletproof vapor barrier with two inches of sand on top, and then poured a 6-inch slab,” said Roberts. “During the 30 days between the slab pour and when they got the building dried in, there were 4.5 inches of rain. But they went ahead and installed the flooring and then they cranked the air conditioner up. The VCT blew off in just 45 days.”
Water can also reach the sand layer by capillary action from the edges of the slab, especially if homeowners have lawn sprinklers. In some cases, it may be able to come up from below through rips in the poly. “When you have sand with a moisture content of 18 percent, you are talking about waiting years, not months, for the slab to dry out,” said George Donnelly, a concrete and flooring consultant in Hemet, California.
Building scientist Joe Lstiburek agrees with Donnelly. Lstiburek wrote, “The sand layer is wetted by liquid phase wetting in a time frame measured in minutes. Whereas the sand layer can only dry upwards by vapor phase drying in a time frame measured in years.”
Slow down and be patient
Flooring failures over damp slabs have particularly plagued builders in the South. “In Southern California and Florida there is a tendency to do things quick and cheap, and concrete tends to be made with a higher water/cement ratio,” Kanare told me. Such a mix results in concrete that is less dense and more permeable than a mix with a lower water/cement ratio. “Even in a relatively dry environment, the moisture in the earth under a slab will eventually come up to 95 percent relative humidity. When the air conditioning is cranked up, the relative humidity of the indoor air is low, and so you have a terrific pump going to pull moisture up from the slab.”
In most cases, damp blotter sand eventually dries, but it may not dry as fast as the flooring contractor would like. “Everyone wants to get the flooring down in a hurry, and no one wants to take the time for a calcium box test,” said Roberts.
The flooring just popped
Until recently, homeowners with slab problems were mostly upset about their ruined flooring. But some homeowners have begun to notice that their failed flooring has mold on the underside, adding a whole new dimension to flooring litigation.
“Flooring failures over damp slabs are a huge issue in these class-action suits against builders,” said Donnelly. When flooring pops off a slab, repairs aren’t cheap. Once the flooring is demolished, the slab is usually shot-blasted in preparation for topside sealing.
After the sealer has cured, new flooring can be installed. “We just finished investigating a failure in a 1,200-square foot room,” Roberts told me in 2002. “It cost $10,000 to seal the slab. That’s not including $30,000 it cost to remove and replace the flooring.”
All kinds of flooring have been known to fail over damp slabs, including vinyl, carpet, and wood flooring. “The sheet vinyl flooring discolors, and a wood floor will swell,” said Donnelly. “Sometimes you lift an object like a child’s toy box, and underneath the carpet is damp.” The common denominator in all of these failures is a damp slab. “With VCT, the first thing you see is the white glue oozing out of the joints,” says Roberts. “With a wood floor, you can get cupping and buckling. If they find mold on the underside of the floor, now it turns into tens of thousands of dollars.”
Concrete specifications matter
Many residential builders pay little attention to concrete slab specifications, at least until they have flooring problems. But “anyone who has been successfully sued starts to pay attention,” said Donnelly. With the recommendations of the ACI 302 committee now in line with most building science experts, there seems to be a firm consensus that the vapor retarder belongs immediately under a concrete slab.
Lstiburek tells contractors, “Repeat after me: don’t ever, ever, put a layer of sand between a plastic vapor barrier and a concrete slab — don’t even think about it.”
Still, some concrete contractors may be reluctant to abandon blotter sand, fearing problems with excess bleed water, slab curling, and cracking. But by switching to higher-strength concrete, reducing the water/cement ratio, and using adequate steel to prevent curling, these problems can be controlled. “The water/cement ratio should not exceed 0.45,” said Kanare. “You should specify a 5-sack mix, and the largest possible sized aggregate. For a 4-inch slab, you should be using 1-inch aggregate.”
In an article on blotter sand, Lstiburek imagined the objections of a concrete contractor from California: “‘Yeah, but if I don’t put the sand layer in there it will take too long to finish the floor.’ Yes, that’s true if you use crappy concrete with too much water in it. The easy answer is: don’t use crappy concrete.”
Some builders who have long used blotter sand may be tempted to substitute a layer of crushed stone for the blotter sand layer. According to Lstiburek, however, this won’t work. “Large quantities of water are still held in the pea gravel — think surface area of the gravel and the fines in the pea gravel. Pea gravel does drain, but it retains huge amounts of water even though it drains.”
Test the concrete
If you are worried about flooring failures, it makes a lot of sense to perform a calcium chloride test before you install the flooring. This test will verify that the slab’s moisture level has dropped to within the specifications of the flooring manufacturer.
After all, the cost of the test is a lot cheaper than litigation.
Can I put the polyethylene under the rigid foam?
Some builders like to install polyethylene under a layer of rigid foam, so that the sandwich looks like this (from the bottom up): crushed stone, polyethylene, rigid foam, concrete. Is that OK?
Well, it’s not as bad a sandwich as one that includes a layer of blotter sand above the poly. But it’s still not optimal. If it rains after the rigid foam is installed but before the concrete is placed, the polyethylene can hold hidden puddles.
It’s always best to install the polyethylene on top of the rigid foam, so that it ends up directly under the concrete.
Is it important to make sure there are no holes in the polyethylene?
Some energy-efficient builders misunderstand the purpose of the polyethylene under a slab. They evidently think that the polyethylene is an air barrier, so they go to great lengths to seal every seam with expensive tape and to tape all pipe penetrations. These builders stay up at night worrying that the concrete contractors will put a little hole in their precious polyethylene.
They should stop worrying. The polyethylene is a vapor barrier, not an air barrier. It’s perfectly OK for the polyethylene to have a few holes.
The concrete slab is your air barrier. If you want an airtight floor — and there are several reasons you should, including radon worries — then you need to seal cracks at the perimeter of the slab with high-quality caulk, and you need to seal penetrations through the slab. This work is performed after the concrete has cured.
A vapor barrier can have a few holes and still perform perfectly well as a vapor barrier. If 5% of the surface area of a vapor barrier is missing entirely, the vapor barrier performs 95% as well as a vapor barrier that is intact.
Here’s what Lstiburek has to say about about polyethylene under a slab: “You can poke holes in it, you can puncture it, you can tear it, you can leave gaps in it, and pretty much have your way with it as long as it is in direct contact with the concrete. … Air barriers need to be continuous and free from holes, but vapor barriers do not need to be. Lots of vapor moves by air movement, not a heck of a lot of vapor moves by vapor diffusion. The concrete slab is the air barrier, and the ripped and torn and punctured polyethylene sheet is the vapor barrier. … I could wear golf shoes and march around the plastic vapor barrier and not do much damage. “
[Author’s note: Portions of this article appeared in an article (“Blotter Sand Woes”) that I wrote for the May 2002 issue of Energy Design Update.]
Martin Holladay’s previous blog: “Banish These Details From Your Plans.”