KTA-TATOR,
INC.
Corporate Headquarters
115 Technology Drive
Pittsburgh, PA 15275
Phone: 412.788.1300
Fax: 412.788.1306
E-mail:
info@kta.com
Instrument
Sales
1-800-KTA-GAGE
Lead
Hotline
1-800-LEAD-OFF
The Case of the Selectively Blistering Floor
By Rick Huntley, KTA-Tator, Inc.
A major United States corporation recently constructed a manufacturing facility in the continental United States. The floors in the manufacturing and storage sections of the facility were constructed of poured concrete. A plastic moisture barrier was installed before the concrete was poured. The specification called for the application of a two component epoxy amine self leveling coating system to be applied to most of the floor at a total thickness of about 30 mils. Surface preparation of the floor consisted of abrasive blast cleaning using a portable recyclable centrifugal abrasive blasting machine. Because of time constraints, surface preparation and coating operations began before the building was totally enclosed. A roof had been installed over the entire facility, but the concrete masonry unit walls had not been completed, nor had all the doors and windows been installed.
The product data sheet for the floor coating material included a requirement that the concrete floor be tested for moisture before application of the coating in accordance with ASTM D-4263, "Plastic Sheet Method for Checking Moisture in Concrete." This method involves taping an 18 inch square plastic sheet to the concrete floor and letting it stand for at least 16 hours. The sheet is then removed and inspected to determine whether moisture condensed either on the backside of the sheet or on the concrete under the sheet. The contractor performed the plastic sheet test in numerous areas before applying the coating and no moisture or condensation was found.
The coating was applied to several of the manufacturing areas, and three to four weeks after the application, blisters began to form in the coating. When the contractor removed individual blisters, they were found to be filled with water. The floor coating was removed in one area, and the floor was allowed to thoroughly dry. The floor coating was reapplied, but blistering again occurred. The contractor removed the coating for a second time, and again allowed the concrete to dry, before reinstalling the coating, but the blisters reoccurred after a few weeks.
Because of the reoccurrence of the blistering, KTA-Tator, Inc. was contracted to investigate the blistering problem to determine the cause, and to make recommendations for repair.
Field Investigation
A senior consultant examined the applied floor coatings and made the following observations.
The floor coating had blistered in random areas throughout the facility. The blistering affected approximately 25% of the coating. The blisters ranged in size up to about 1/4 inch in diameter and were filled with water. The water had an unusual odor.
In the area where the coating had been twice reapplied, the blisters formed in a distinct pattern. The coating had been removed using a scarifying machine. The machine cut shallow parallel half inch wide trenches in the concrete. The blisters formed in the coating predominately over the ridges between the parallel trenches.
On several of the concrete slabs the subject floor coating was applied to only a portion of the slab. In some of the slabs, the floor coating terminated at a straight line, and no floor coating was applied in the adjacent area. The blistering in the coating occurred as close as ¼ inch from the edge. In other areas, the subject floor coating terminated in a straight line and a different self leveling epoxy floor coating was applied at a thickness of about 20 mils, in the adjacent area. No blistering was found in the different floor coating. When coating was forcibly removed in both areas, water was found beneath the subject coating but not beneath the other coating, although they were applied to the same slab.
Several samples of the floor coating, along with liquid removed from the blisters and wet samples of the paint were removed and returned to the KTA-Tator laboratory for analysis.
Laboratory Analysis
Visual and microscopic observations were performed on the paint with a Nikon SMZ-10 stereo zoom microscope with magnification to 40X. The removed coating chips were generally thinner than specified and consisted of two coats; a primer and a topcoat. The thickness of the samples varied between 14 mils and 38 mils. There was found to be no noticeable correlation between the thickness of the coating system and the occurrence of blistering.
The blister liquid was analyzed both by gas chromatography and infrared spectroscopy. Gas chromatography revealed that there was no more than an insignificant trace of volatile organic solvent in the blister liquid.
Infrared spectroscopy was performed on samples of the blister liquid. The technique involved evaporating the water under vacuum and combining the nonvolatile portion of the blister liquid with potassium bromide powder, and forming a pellet under high pressure. The pellet was then placed in the optical path of the spectrophotometer, and a spectrum was obtained. The analysis revealed that there was a significant quantity of an amide or amine salt in the blister liquid. This material is a byproduct of the curing agent used to crosslink the epoxy resin.
Field Test Patches
As a result of the discovery of components of the curing agent in the blister liquid, it was decided to apply test patches of the subject coating material and similar self-leveling epoxy materials supplied by different manufacturers to a section of the floor that had not been previously coated. Five by ten foot test patches were applied on abrasive blast cleaned concrete slab in the facility. After three weeks, the subject floor coating began to blister, while the similar floor coatings supplied by other manufacturers showed no signs of blistering even after several months. A small piece of coating was removed from each test patch. Moisture was found beneath the subject floor coating, but not under any of the other coatings.
Cause of Failure
Usually, blistering of a coating applied to a concrete floor is considered to be either a result of excess moisture transmission through the concrete slab, or the presence of a water soluble solvent in the coating. The excess water moisture transmission causes water to accumulate under the floor and induces blistering, while the water soluble solvent actually attracts water, which subsequently accumulates and induces blistering. In this case, previous moisture transmission testing using the plastic sheet method failed to detect excessive moisture transmission. Laboratory gas chromatography eliminated solvent as the culprit.
Additionally, the pattern of failure was not consistent with excess moisture transmission. Although water was found beneath the failed coating, water was not found under a similar coating applied to the same concrete slab as the blistering coating. In fact, the blistering in the subject coating occurred within an inch of the coating that exhibited no blistering. Obviously, if moisture transmission through the slab was the cause, both coatings would have blistered, or at least moisture would be found beneath both. Additionally, on test patches applied side-by-side to the same concrete slab, only the subject coating blistered, while the other coatings showed no signs of blistering.
The laboratory did find a material in the blister liquid that had the ability to attract water. The amine or amide salts detected in the liquid were a byproduct of the curing agent of the epoxy floor coating. It is apparent that the water soluble material present in the epoxy curing agent attracted water and eventually induced the blistering. After a few weeks, the water began to accumulate under the floor coating and exerted pressure on the coating, resulting in blistering. Removal and reapplication of the same material did not remedy the problem because the same curing agent was used. Additionally, the water soluble material likely seeped into the top layer of the concrete and was not totally removed when the coating was removed.
In order to fix the problem the coating had to be totally removed and the concrete had to be thoroughly abrasive blast cleaned to remove any residual organic material in the top layer. Any remaining amine or amide salt may induce further blistering, even under a different coating. One of the coatings used in the successful test patches was then applied to the floor. That coating performed well with no blistering and continues to perform well several years later.
To keep informed and to have the KTA eNews delivered directly to your IN box with more great articles and news from KTA and the industry, please send your e-mail address to us.
Return
to Case Studies © 2004 KTA-Tator,
Inc. All Rights Reserved. Privacy
Statement
Site maintained by KTA. webmaster@kta.com
Best viewed with Microsoft Internet Explorer
