Concrete Roof Deck Issues

• Organic (cellulose based) material susceptible to mold and decay with prolonged exposure to RH above 80%.
• High RH within the roofing system prone to condensation when the outside temperature drops.

• Curing compounds are used to slow or reduce the evaporation of moisture from the concrete to prevent cracks due to drying shrinkage.
• Applied immediately after finishing while the surface is damp.
• Curing compounds may affect adhesion of products to the concrete.
• Sealers are applied after the concrete is cured to protect concrete from exposure to external constraints. 
• Sealers will allow moisture vapor from the concrete to vent outwards.

Depending upon the amount of moisture that got into it, it can start to deteriorate in easily 2-3 months’ time. With gypsum there is also concern that if we start to add load to it (such as wind load or just walking on it), that will also break down gypsum structure. Other products will take a little bit longer but depends on how much moisture gets into system.

We may see some horizontal venting, however it will be minimal. Generally we see an upward movement of the water vapor. Trying to get sufficient lateral movement without using mechanical means will probably be very difficult.  

On metal decks with new pours, the issue is that roofing too soon will not allow the moisture to leave the system. We have also looked at the time of year such as pouring in late winter/ early spring and then doing roofing in spring and early summer which will allow vapor to dry down into the building and not the system. By insulating the system, putting insulation on concrete as well as putting the finished roof cover on, we have now brought concrete slab at an equilibrium. If there is not enough insulation on the concrete deck which will bring the concrete deck to approximate temperature equilibrium, we may see moisture in the roof system. Also, as the second case study shows, if there is constant wetting of an existing concrete deck, we most likely will see moisture or condensation issues.  

Concrete add mixes can cause adhesion concerns, acting as an adhesive bond breaker, which affects the wind uplift performance. With FM insured projects, if the concrete add mix is not included in the RoofNav assembly, FM Global typically will not accept the assembly. We had a project where the consultant monitoring the project noticed the concrete sealer was used. Being a FM insured project he questioned the application, FM rejected the surfacing causing the GC to remove the sealer. The sealer was done without the knowledge of the roofing contractor or us. All manufacturers will have the same issue with this scenario.

Many water base adhesives are bonding type, where the water is to flash off to a tacky consistency before mating. Our water based adhesive has just enough water to allow for applying/spreading the adhesive. The small amount of water in our adhesive cures and dries with no effect on the components or system. We have been using this water base adhesive since 1979.  

In areas of the country with high wind uplift pressures, such as eastern seaboard and the Gulf coast areas, the concrete deck provides the platform that allows for high uplift rated systems. Steel, wood and other types of decks have uplift some limitations. We also believe when the new ASCE 7-16 wind design criteria takes effect, the design pressures will also increase, especially hurricane prone coastal areas.

While there are some systems that claim venting can dry out wet roof systems, these typically have more components than just vents. The vents used in the old LWC systems may not work as well as people think. There was a research paper done by Wayne Tobiasson for CRREL suggests venting built up most likely will not dry out the components or systems.

A temporary roof, one utilizing asphalt based products adhered to the deck, will not allow the concrete to dry. If a temporary roof is installed on a newly poured concrete deck, there is a good possibility the vapor pressure will push against the temp roof, de-bonding it from the deck. Northern climates this is more of an issue during the colder months. Even in the southern areas there is a chance for blistering to occur. A good approach is to adhere an asphalt vapor retarder, mop, torch or self-adhered to the deck. Insulate and place the roof cover. This should keep the top and bottom of the concrete deck at a similar temperature, thus minimizing or stopping the vapor drive.

If the water stays in the concrete with little to no means to condense, and if oxygen is not present (or very minimal) there should be little chance for corrosion to take place. The structural engineers and concrete supplier should be consulted.

The benefit of the fiberglass facers include no absorption of moisture that an organic based facer has, which eliminates the cohesive failure of the facer as seen in the case study photos. Also the probability of mold is far less with the fiberglass facer. The roofing industry is taking the position with moisture and concrete, the glass facer is a better choice of product.

If hot mopping a VR, it is recommended that the asphalt show if the moisture is far enough down in the concrete to allow good adhesion. Another possible method is to use moisture meter, which should provide the condition of the top ½ to ¾ inch of the concrete. As noted above, after adhering the VR, the roof system will be completed by adding the insulation and roof cover immediately.

Using a heavy mod bit type VR that is attached by hot mopping, torching, cold applied or self-adhered. Based on the system, the concrete should be primed. Adding the insulation layers will keep the PVC or any single ply roof cover away from the asphalt.

Curing concrete depends on what concrete mix is and how it’s designed. You can accelerate a concrete structure and cure in 7-14 days. Typically concrete will cure in 28 days. Cure means it’s structurally sound and you can use it as concrete is expected to be used. However 99% of water is still in there through which the water will want to come out. The big things is to separate cure and water. Curing and drying are two separate things because drying is getting the water out of the concrete. Normal weight structural concrete may take anywhere from 3 to 9 months to “dry enough”, while lightweight structural concrete can be twice as long.

Unfortunately, there is no good test method to determine the moisture content of an exposed concrete deck. All current test methods require a time span for the test area to be at equilibrium, for temperature and humidity. The problem with an exposed deck, equilibrium can never be reached. Just from day to night the surface temperatures vary along with sun and shade.

The first case study was a concrete pour into a steel form while the second was a normal weight concrete with the forms removed. That older deck was continually rewetted due to the leaks from the original BUR system.

We have made the change from the word barrier to retarder for all of our VR sheets. IBC definition of a vapor retarder: a measure of a material or assembly’s ability to limit the amount of moisture that passes through that material or assembly. Vapor retarder class shall be defined using the desiccant method of ASTM E 96 as follows:
Class I: 0.1 perm or less.
Class II: 0.1 < perm <1.0 perm.
Class III: 1.0 < perm <10 perm.

IBC examples of vapor retarders from section 1405.3.3 Material vapor retarder class. The vapor retarder class shall be based on the manufacturer’s certified testing or a tested assembly. The following shall be deemed to meet the class specified:

Class I: Sheet polyethylene, non-perforated aluminum foil with a perm rating of less than or equal to 0.1.

Class II: Kraft-faced fiberglass batts or paint with a perm rating greater than 0.1 and less than or equal to 1.0.

Class III: Latex or enamel paint with a perm rating of greater than 1.0 and less than or equal to 10.0.

Be sure to design the concrete deck with normal weight concrete with a removable form.   

This is the issue with moisture and concrete, it takes a long time to move excess water out of the concrete, and as you note this is without adding additional water from precipitation.

Lightweight concrete (LWC) and lightweight insulating concrete (LWIC) will allow for the excess water to move through the pour much quicker than structural concrete. The slide with the graph from SGH shows the LWC losing most of the water in two weeks or less. It is typical to adhere a roof cover to LWC within 10 to 14 days after pouring. The reason for the faster drying is the cell structure of the LWC as compared to structural concrete.

As long as the VR is sealed, which includes through all deck penetrations, walls, etc both should work. A venting sheet will have cavities or pathways where the vapor can move and if conditions are right condense. The attachment of the VR is important. With the vented sheet there will be partial attachment whereas with an adhered VR there should be as close to possible full adhesion. My preference is using an adhered VR. The insulation on top of the VR should be designed to minimize or eliminate as much as possible the vapor drive pushing against the VR and roof system. In my opinion the roof system should be installed at the same time as the VR.

We are offering this presentation to bring the wet concrete issue to anyone involve with the roofing industry. We will be more than willing to present to any group you have in mind. There are also numerous papers written on the subject, such as the one done for this presentation as well as another I wrote for RCI Interface in 2014. The consulting firms SGH and SRI among others have also written a number of papers as well as presented at RCI and NRCA. If you want copies of papers please contact us.