News & Articles

Case Studies

eNews

Industry News

Industry Standard Updates

Product News

Regulatory News

Contact Us

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 Blistering and Disbonding Filter Bed Tank Coatings

By David S. Leyland, KTA-Tator, Inc.

A northeastern municipality constructed a new water treatment facility that incorporated four filter bed tanks, whose walls were constructed of concrete. The walls of the tanks were coated with a two coat polyamide epoxy coating system, suitable for contact with potable water. The recommended dry film thickness for each coat was 4 to 6 mils. The concrete walls were originally coated in May, 1992; additional coating was applied during July, 1993, to repair failing areas on the concrete walls. The initial method of surface preparation involved acid etching the concrete wall surfaces. Several months after the acid etching was performed, the concrete walls were power washed to remove dust and debris that accumulated during the construction of the facility.

Two months after the water treatment plant had been in service, it was reported that the coating system was covered with liquid filled blisters. The presence of the liquid Mod blisters produced the need for extensive repairs to the concrete walls of the filter bed tanks. Repair procedures included water blasting the entire surface to remove all contaminants, as well as hand tool cleaning the surfaces in accordance with. SSPC-SP 2, to remove all the blisters and feather the edges of the remaining intact coating. As a result, large areas of previously applied, intact coating remaining on the concrete substrate. The identical coating system was then applied to the walls.

Several days after the coating system was applied and aproximately two days before the tanks were filled with water, a representative from the paint manufacturer visited the jobsite and noted a strong solvent odor in the building. The epoxy coating system was rubbed with Q-tips soaked with methyl ethyl ketone (MEK) and a significant color transfer was noted after only two double rubs. Mobile heating equipment was then moved into the building and the coating system was forced cured for the next forty-eight (48) hours. The paint manufacturer's representative found the coating system to remain moderately solvent sensitive, even after forced curing.

Approximately one month after the tanks were refilled with water, the coating system was beginning to peel from the filter bed tank walls. After two unsuccessful attempts to apply the coating system to the concrete walls, KTA-Tator, Inc. was contracted to determine the cause of the coating failure and to recommend appropriate repair procedures.

Field Investigation

The field investigation involved examining the filter bed mks that had blistered and been repaired as previously described. Filter bed tank "A" had areas of peeling paint that measured several square feet in size. The coating system was only peeling on the walls facing south and east. The walls facing west and north did not exhibit any signs of failure. The largest areas of spontaneous peeling were below the water-line. The peeling paint appeared to be only the repair coat. The originally applied coat which exhibited areas of previous blistering, was exposed after the repair coating peeled.

Filter bed "B" had not been repaired. An extensive number of blisters were observed. The blisters measured approximately 1/4" in diameter. Approximately 25 to 30% of the total surface area below the water-line was covered with blisters. Blisters were liquid filled, and samples of blister liquid were removed with a hypodermic needle. Portions of the walls above the water-line had not blistered and appeared to be unaffected. In areas where the coating had not blistered or areas immediately adjacent to blisters, the coating system was intact and tightly adherent to the concrete walls. Samples of failing and non-failing materials were removed from both Tank "A" and Tank "B" and returned to the KTA laboratory for analysis.

Laboratory Investigation

Infrared analysis of the soft, peeled coating removed from filter bed tank "A" revealed amide without the characteristic bands of an epoxy. Apparently, the peeled paint was the polyamide component of the coating system without the addition of the epoxy component.

The blister liquid that was removed from the blisters in Tank "B" were analyzed using gas chromatography and ion chromatography. The analysis performed using gas chromatography revealed that organic solvents were not present; the blister contained only water.

An analysis of the blister liquid using ion chromatography revealed the presence of water soluble salts. The ions detected and corresponding concentrations are presented in the chart below.

Nitrites and phosphates were not detected. Analysis of concrete removed from a wall in filter bed Tank "B", revealed the following concentrations of soluble salts also listed in parts per million:

A second sample of concrete was also analyzed, yielding the following results:

Cause of the Failure

There were two separate applications of polyamide epoxy coatings to the concrete walls of the water filter bed tanks. Both coating applications failed, but for different reasons. The first coating application failed because the coating system developed liquid filled blisters. The concentration of water soluble salts detected in the blister liquid was sufficiently high to result in the development of osmotic blisters. The quantities detected of either chlorides or sulfates alone would have been sufficient to result in osmotic blistering.

Osmotic blistering occurs when water soluble salts or contaminants are on one side of a coating layer and ample moisture is present on the other side of the coating. Because all coatings are semi-permeable, molecules of water will slowly permeate the coating, dissolving the water soluble salts. Once the water soluble salts begin to dissolve, an osmotic cell has been created and osmotic pressures will force additional molecules to more readily permeate the coating in an attempt to dilute the concentrated solution of soluble salts. This process will continue until the concentration of soluble salts on both sides of the coating are equal. Before equilibrium is obtained, large liquid filled blisters are created.

The source of the water soluble salts appeared to be the concrete itself. The presence of chlorides may indicate inadequate rinsing after acid etching. Sulfates may have resulted from sulfur trioxide, which is an ingredient in the manufacturing of cement. However, the sources of nitrates, nitrites, bromides and phosphates were unknown. They may have been contaminants in the water that was used to mix the concrete slurry or contaminants in the aggregate mix used to prepare the concrete. It was unlikely that these contaminants resulted from dirt and debris accumulation on the walls, deposited during construction of the facility.

The repair coating system did not fail by blistering. Most of the surface on the concrete walls still contained the original coating system, therefore, there was a barrier between the contaminants on the concrete substrate and the repair coat. The repair system failed by peeling.

An analysis of the disbonded coating from filter bed "A" revealed that the converter had not been mixed with the polyamide component of the epoxy coating system. As a result, the coating system was unable to cure, and disbonding occurred after being subjected to immersion service. The coating system did not peel from all of the walls, as in some cases the material was properly mixed and applied.

Recommendations

Because the concrete walls appear to be the source of water soluble contaminates, there is no way to guarantee successful repairs. Perhaps the contaminants could be substantially removed from the concrete surfaces by steam cleaning or high pressure water washing after complete coating removal. If distribution of the contaminants is only skin deep, abrasive blast cleaning in combination with high pressure water washing could be an effective means of preparation. The distribution of soluble contaminants could be determined by analyzing core samples removed from the concrete walls. The same polyamide epoxy coating system used during the first two coating applications could be used again.

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