Q. I am consulting on a new commercial building, specifically for indoor air quality issues. I read a good article on your Web site, “Preventing Moisture Problems” by Martha Van Geem and Medgar Marceau. I have one follow-up question regarding vapor retarders: What do you recommend for sealing joints in the vapor retarder? If we are using 10- to 15- mil-thick plastic and have to cut it, what should be done at the seams to ensure that moisture does not get into the slab at unacceptable rates?

A. We have learned a lot about vapor barriers from the folks at Stego Industries (www.stegoindustries.com). They are keen on the idea of a completely sealed barrier. To that end their product is thick and tough, and they have numerous warnings that penetrations and joints must be sealed. They also have products for doing so, and guidelines. For joints they suggest a 6-inch overlap sealed with high quality tape. They provide information along that line on their Web site, under installation procedure. You'll notice that they also refer to ASTM E 1643-98, “Standard Practice for Installation of Water Vapor Retarders Used in Contact with Earth or Granular Fill Under Concrete Slabs,” which would be good for you to know about.

On the other hand, some people aren't so adamant. The following observation is excerpted from the Building Science Corporation's Web site (www.buildingscience.com/resourcesfoundations/sand_layer_under_slab.htm), where the author is explaining why the barrier should go directly under the slab (instead of covering it with a sand layer). His comments on holes and effectiveness are interesting. Note the caveat that air flow/leakage must not be occurring.

“The fourth reason, ‘puncture protection,' is based on incorrect physics. A sand layer is not necessary to protect polyethylene vapor barriers. Vapor diffusion is a direct function of surface area. Rips, holes, tears, and punctures in sheet polyethylene vapor barriers constitute a very small surface area of vapor transmission compared to the total floor slab area. If 95 percent of the surface area of the slab is protected by a vapor barrier, then that vapor barrier is 95 percent effective. This holds true only if air flow or air leakage is not occurring through the vapor barrier. Where concrete is in direct contact with the polyethylene vapor barrier, this is the case. Air flow is not occurring. The concrete slab is an ‘air-barrier' and the polyethylene is the ‘vapor barrier'—and an effective vapor barrier even if the polyethylene has numerous punctures.”

Finally another article you should (and may already) be aware of is the detailed explanation of the vapor transmission problem provided by Peter Craig in the March 2004 issue of Concrete Construction. You can read that article on the Web here.

Explaining odd test results

Q. We have had a problem with the compressive strength of some grade 25 concrete. The 7-day cylinder break results indicated that the strength was 3090 psi, while the 28-day test result was 3178 psi. Is this possible?

A. It is not possible that the true strength of properly cured concrete could be about 3100 psi at 7 days and 3200 psi at 28 days, but it is possible that such results might genuinely appear on a testing machine. There are many variables in the testing procedure that can produce inaccurate results. For example, the 7-day strength could be inflated if the test cylinder was heated, perhaps by leaving it in the sun, or if it was tested dry. It is possible for a cylinder to be poorly compacted, badly capped, or not centered in the testing machine. The explanation might be apparent if you have previous test results on this grade to which you can make a comparison.

Another thing to look at, though, is air-void clustering around the coarse aggregate particles. This can occur without any outward indications. Evidence of this can be observed by examining the coarse aggregate sockets of the fractured specimen using a hand lens. Using a microscope to examine a thin section, however, is better. Check this even if the concrete was supposed to be non-air entrained. There have been cases where the 7-day strength is pretty normal, but does not progress further as a result of this problem.

A talk at ASTM's “Second Symposium on Petrographic Techniques for Examining Hydraulic Cements and Concrete” in June 2005 described some research under way on air-void clustering. Preliminary results indicate that retempering, which is the late addition of water to the concrete, leads to air-void clustering. Longer mixing times also seem to exacerbate the problem.