Radon and Soil Gas Entry Paths

Why this matters

A plumber cuts a trench through a basement slab for a new bathroom, a waterproofing crew installs an open sump, an electrician core-drills the floor for a conduit run. None of them touched anything they think of as an air system, and any of them can raise the radon level in that house. Radon is a colorless, odorless radioactive gas, and the EPA ranks it as the leading cause of lung cancer among people who have never smoked. The trade that opened the path is usually the only one who knows the path exists, so this card is about seeing the paths, ranking them, and closing the ones your own work creates. Testing strategy belongs to Radon Zones and Who Should Test; mitigation system design belongs to Radon Mitigation Basics Reference.

Where the gas comes from and why it comes inside

Radon forms continuously in soil and rock as uranium decays. It moves through the air spaces between soil particles, so the gas sitting in the soil under a house carries a concentration commonly in the hundreds of picocuries per liter (pCi/L, the unit US test results use), against an outdoor background the EPA puts at about 0.4 pCi/L. A house does not need much soil gas to reach a problem level. It needs a small, steady stream of it.

That stream is pressure-driven. Soil gas moves into the house only where the house is at a lower pressure than the soil beneath it, and the lower floor of a house usually is:

• Stack effect in the heating season. Warm indoor air is lighter than cold outdoor air, so it pushes out through the top of the house and pulls replacement air in at the bottom. The lowest floor runs negative to the outdoors and to the soil. See related: Stack Effect and the Neutral Pressure Plane.
• Exhaust devices. A range hood, bath fans and a clothes dryer all pull air out of the house, and the house replaces it through whatever holes exist, including the ones in the slab. See related: Counting Every Exhaust Device in the House.
• Leaky returns in the basement. A return duct that leaks in the basement pulls the basement negative to the rest of the house as well as to the soil.

The pressures involved are a few pascals, tiny by any trade standard (see related: What a Pascal Is and Why House Pressures Are Tiny), but a few pascals applied steadily across a slab with open joints is enough. This is why radon readings tend to run higher in winter and on the lowest level, and why the same house can test differently from one season to the next.

The entry paths, drawn

 SECTION: basement slab beside a crawlspace addition

 lower floor runs NEGATIVE to the soil in winter
 +-----------------------------+--------------------+
 |        BASEMENT             |  CRAWLSPACE        |
 |                             |  floor gaps open   |
 | [1]     [3]     [4]    [2]  |  into the house    |
 |==gap====drain===pipe===sump=|                    |
 |  ^       ^       ^      ^   |  [5] bare soil     |
 |  |       |       |      |   |   ^    ^    ^      |
 +-----------------------------+--------------------+
   gravel bed and drain tile below the slab
   SOIL GAS rises into every opening above it

 [1] slab-to-wall joint and slab cracks
 [2] sump basin with an open or loose lid
 [3] floor drain with a dry trap
 [4] slab penetration: pipe, conduit, post base
 [5] exposed soil in a crawlspace

Two more paths do not fit the drawing. Hollow block foundation walls carry soil gas up through their cores and release it at the top course, at utility holes through the block, and at any crack in a block face. Well water releases dissolved radon when it is sprayed or agitated indoors, which is a separate path a well-water specialist should assess; for most houses it is the smaller contribution.

Area is not the whole story, connection is

The instinct is to rank openings by size. Size matters, but what the opening connects to matters more. Work one basement through, a 30 by 40 ft footprint with a floating slab (a slab poured inside the foundation walls rather than on top of the footings, which leaves a joint all the way around).

The total is 105 + 254 + 9 + 3 + 19 = 390 sq in, and the open sump alone is 254 of it, about 65 percent. But the sump also outranks everything else on connection. A sump basin sits at the low point of a perimeter drain tile system, perforated pipe in a gravel bed that runs around the whole footing. Gravel is far more permeable to gas than packed soil, so the drain tile behaves like a collection manifold for soil gas from the entire footprint, and an open sump is that manifold's outlet into the basement.

The floor drain is the same story at a smaller size. Three square inches looks trivial, but some floor drains discharge into the drain tile or straight to the gravel rather than to the sewer, and a dry trap on one of those is a direct line to the sub-slab gravel bed. The shrinkage crack is the opposite: nine square inches over packed fill passes far less gas than its area suggests.

So the working ranking for that basement is the sump first, the slab-to-wall joint second because it is long and sits over the gravel at the footing, the new sleeve third, and the crack and drain after that unless the drain turns out to connect to the tile. A ranking by area alone would have put the slab-to-wall joint and the sump in the same tier and missed why the sump is worse.

The paths your trade opens, and how to close them

Cutting or trenching a slab (plumbing, electrical, waterproofing). The trench opens the sub-slab gravel to the basement for as long as it is open and leaves a cold joint (the seam where new concrete meets old) around the patch for good. Restore the vapor retarder under the patch, lapped over the existing one where you can reach it and taped, so the patch is not sitting on an open gravel channel. Seal the patch perimeter and any pipe or conduit through it with a polyurethane sealant rated for concrete, over backer rod where the gap is wide. Sawing and coring concrete releases respirable crystalline silica: for employees on construction work that is 29 CFR 1926.1153, and its Table 1 sets the controls for each task (water delivered at the blade, or a shroud with dust collection, and a respirator where the table calls for one under a written program per 29 CFR 1910.134).

Installing or servicing a sump. A clear-water sump should get a gasketed, sealed lid with the discharge pipe, cord and any check valve sealed where they pass through it, so it stays serviceable without being open. A sewage ejector basin is already required by plumbing codes to be sealed and vented, so an ejector pit with a cracked lid or an unsealed cord grommet is a code defect as well as a soil gas path. If a mitigation system already draws on the sump, the lid is part of that system; do not swap it for an unsealed one.

Leaving a floor drain dry. A floor drain that never sees water loses its trap seal to evaporation. A trap primer, a waterless trap seal insert, or a note to the customer to pour water in seasonally all work; the note is the weakest of the three.

Core-drilling for a post, a pipe or a conduit. Every hole through the slab is a penetration that needs the same sealant treatment as the trench patch.

Crawlspace work. Pulling back or tearing the ground vapor retarder to reach a pipe or a duct leaves bare soil exposed. Put it back lapped and taped. Ducts running through a crawlspace that leak on the return side pull crawlspace air, soil gas included, straight into the air handler; see related: Duct Leakage Outside the Envelope.

Adding exhaust. A bigger hood, a new bath fan or a second dryer makes the lower floor more negative to the soil. The trade that adds the device is changing the radon driving force even though it never touched the floor. See related: Which Trade Jobs Change House Pressure and Moisture.

What the 4 pCi/L number means and does not

The EPA recommends fixing a home whose radon level is at or above 4 pCi/L, and recommends considering a fix between 2 and 4. That 4 is an action level for a long-term average in the lived-in space, not a safe line: the risk falls with the number, and there is no level the EPA calls risk-free. Four things keep a tech from misusing it.

• It is a guideline at the federal level, not a law for most houses. Disclosure at sale, radon-resistant new construction (the IRC's Appendix F, where a jurisdiction adopts it), and licensing of testers and mitigators come from the state and local layer. Several states license or certify radon professionals (Illinois and Pennsylvania among them); many have no such rule. Check yours.
• A short-term test is a snapshot. The EPA's protocol has the house closed up for 12 hours before and during a test that runs under four days, and readings swing with weather and season. A single short-term number near the line is a reason for a follow-up test, not a verdict.
• The lowest lived-in level is where it is read. A finished basement counts. An unfinished one used only for storage is usually not where the test belongs.
• Your work can move it either way. Sealing a sump lowers the reading; cutting a trench and patching it loosely, or adding a large exhaust fan, can raise it.

A trade doing slab work in a house with a known radon history can offer the customer a retest after the job, using a test kit or a licensed tester where the state requires one. Do not quote a predicted reading; nobody can predict one.

When to stop and route

• There is an active mitigation system (a PVC pipe from the slab or sump to a fan, usually with a U-tube gauge on the pipe). Do not cut, cap or reroute any part of it. If your work has to cross it, the mitigator who installed it or another qualified one makes the change. Note the U-tube reading before and after your work; a gauge that reads level after you leave means the system lost suction.
• The customer asks whether the house is safe. That is a test result and a mitigator's judgment, not a field call. Route them to a test, and to a licensed mitigator if the result calls for one.
• The customer asks you to seal everything as a fix. Sealing alone rarely brings a high reading under the action level; it supports a sub-slab depressurization system rather than replacing it. Seal what you opened and route the rest.

How to verify you closed what you opened

• The patch or penetration is sealed with no visible gap at the perimeter, and the sealant is a type rated for concrete, not a general-purpose caulk that will release from the slab as it shrinks.
• Puff smoke at the finished seam with the house at its normal operating condition (see related: How to Use a Smoke Pencil to Read Air Direction). Smoke that moves at all means the seam is still open: with the basement negative to the gravel, the usual winter case, soil gas coming up through the seam pushes the smoke away from it.
• The sump lid gaskets seat all the way around and every pass-through is grommeted or sealed.
• Any mitigation system gauge reads the same as before you started.
• The job record notes which slab openings you created and how each was closed, so a future radon test that climbs can be traced to something specific.

References

• US EPA, A Citizen's Guide to Radon and Home Buyer's and Seller's Guide to Radon (action level 4 pCi/L, short-term test conditions)
• 29 CFR 1926.1153, Respirable Crystalline Silica in construction, Table 1, and 29 CFR 1910.134, Respiratory Protection
• IRC Appendix F, Radon Control Methods, enforceable only where a jurisdiction adopts it; appendix lettering changed in recent editions, so check the one in force
• See related: Radon Mitigation Basics Reference; Radon Zones and Who Should Test; The House as One Pressure System
• See related: The Radon Number That Climbed After the Basement Bath