Printer Friendly

Evaluation of low volatile organic compound coating systems for steel bridges.

Introduction

Under the November 1990 Amendment to the Clean Air Act, the volatile organic compound (VOC) content for maintenance coatings on steel bridges will be regulated throughout the country within a few years. This VOC restriction will reduce smog formation t improve air quality. An architectural and industrial painting regulation could be proposed by the U.S. Environmental Protection Agency this year. Promulgation could follow as early as 1993. In anticipation of this regulation, many States already have limited 340 g/l (2.8 lb/gal) of VOC for industrial and architectural maintenance paints. The South Coast and Bay Area Air Quality Management Districts of California have established a 250 g/l (2.08 lb/gal) limit for VOC in traffic paints. The classification of lead and chromate anticorrosive pigments as hazardous materials and the mandated reduction of solvent in bridge paints have resulted in an immediate need for cost-effective and environmentally acceptable replacement materials for both bridge construction and maintenance painting. To this end, researchers are now developing high solid solvent-borne resins with the required viscosity and waterborne formulation to reduce solvent content while maintaining acceptable application characteristics. Due to the desperate need for replacement products, many of the new, reformulated low-VOC products are being marketed without thorough laboratory and field performance testing. To ensure against failures in the field and prevent any unnecessary repair and repainting, these new systems must be evaluated in depth. In response to this need, the Federal Highway Administration (FHWA) initiated the present study.

This study evaluated candidate low-VOC coating systems for steel bridges. Two accelerated weathering tests -- a cyclic salt-frog/freeze test and an ultraviolet (UV)/condensation test were used in this evaluation. Currently used high-VOC bridge coating systems were used as controls. The corrosion results obtained by the FHWA's accelerated laboratory testing were carefully evaluated to predict long-term field performance. Additionally, because direct prediction of field performance is not possible, the candidate systems were also exposed to a natural marine environment at Ocean City Research Corporation's Sea Isle Test Site in Ocean City, New Jersey.

The findings of this study will be valuable in providing guidelines for selecting long-lasting low-VOC coating systems for both new construction and maintenance of steel bridges. Complete study details are reported elsewhere. [1] (1)

Experimental Procedure

The coating systems evaluated in this study are listed in table 1. Eleven low-VOC coating systems were chosen for evaluation; for high-VOC coating systems (code nosl 2, 4, 6, and 8) corresponding to four of the reformulated low-VOC products were used as controls. Two

[TABULAR DATA OMITTED]

California Department of Transportation (CALTRANS) bridge coating systems were also included as controls. A 50.8 mm (2-in) diagonal scribe was made on one side of each of the salt-fog and Sea Isle panels to evaluate film undercutting.

Test methods

The following tests were performed:

1. Cyclic salt-fog/freeze test (Singleton corrosion test cabinet)

* Test cycle: 6-day salt-fog/1-day freeze cycle.

* Salt-fog method: American Society for Testing and Materials (ASTM) method B117.

* Freeze temperature: -18 [degrees] C (0 [degrees] F).

* Panel position: vertical instead of 15 to 30 [degrees] from vertical.

[TABULAR DATA OMITTED]

2. UV/condensation test (QUV Weaterometer)

* Test cycle: 4-hour UV/4-hour condensation cycle.

* UV lamp: QFS-40.

* UV temperature: 70 [degrees] C (158 [degrees] F).

* Condensation temperature: 40 [degrees] C (104 [degrees] F).

3. Outdoor exposure

* Exposure site: southern New Jersey coast, near Sea Isle City; approximately 30.5 m (100 ft) from Atlantic Ocean and bordered on the west by Ludlum's Bay.

* Panel position: 45[degrees] angle on wooden racks, facing directly south.

* Each panel was sprayed three times daily with seawater.

Physical properties

Adhesion pull-off tests were made according to ASTM method D4541. Gloss was measured using ASTM method D525. Other phenomena such as chalking, color fading, and peeling were also examined and noted.

Evaluatin of performance

All panels for accelerated tests were evaluated and photographed every 500 hours. The evaluation parameters used were surface rusting, surface blistering, scribe blistering, and scribe undercutting determined according to ASTM methods D1654 and D714. Blisters were described in terms of size and frequency. Blistering numbers 6, 4, and 2 represented progressively larger sizes. Frequency of occurrence was presented as dense (D), medium dense (MD), medium (M), few (F), and very few (VF).

The test panels for outdoor exposure were examined and graded by Ocean City Research Corporation personnel. Blistering was rated in terms of both blister size and density. To obtain this rating, blister size was rated as designated in ASTM method D714. Blister density was rated as follows: none = 10, few = 8, medium = 6, medium dense = 4, dense = 2. The blister size and density ratings were summed and divided by 2. This method produced a rating scale of 0 to 10.

Results and Discussion

Cyclic salt-fog/freeze test

Most of the tested coating systems exhibited failures from the attack of salt and moisture after 3,000 hours of cyclic salt-fog/freeze exposure. They also generally lost some adhesion strength and gloss. The extent of surface failure and scribe undercutting for all the tested coating systems are presented in table 2. Good data repeatability between samples was obtained in all cases; for example, the standard deviation for undercuttings was calculated to be 0.5 mm (0.02 in).

Surface failure. Only three coating systems developed significant surface blistering or rusting after 3,000 hours of cyclic salt-fog/freeze exposure. The high-ratio water-based inorganic zing potassium silicate with a topcoat of water-based acrylic showed significant delaminations of the topcoat. Several circular areas of acrylic topcoat delamination started to appear after 2,000 hours of cyclic salt-fog/freeze exposure, demonstrating a poor adhesion of the topcoat to the primer (figure 1). The intercoat adhesion failure is probably the result of either the moisture sensitivity of the acrylic polymer or a sensitivity of the acrylic topcoat to the surface chemistry of the high-ratio zinc potassium silicate primer. This failure, however, did not result in rusting or undercutting because the inorganic zinc primer protected the panel's scribed area cathodically.

The low-VOC epoxy mastic/acrylic epoxy system exhibited surface blistering of 6M after 3,000 hours of exposure. When the epoxy mastic failed cohesively during the adhesion pull-off test, extensive subfilm corrosion and rust staining of the coating was found on the exposed plane area and the pulled plug (figure 2). This subfilm corrosion is similar to that found by Smith using an epoxy polyamide primer after 1,000 hours of salt-fog exposure. [2] Trimber found residual salt to be a major contributor to the failure of a similar paper mill maintenance coating system through undercutting. [4] The adverse effect of sodium chloride contaminated steel on the coating life was also reported recently by the Steel Structure Painting Council. [4]

Severe subfilm corrosion was observed for the CALTRANS high solids phenolic system after 3,000 hours of exposure. The phenolic resin primer disbonded completely from the steel surface during the adhesion test. When pulled off, the primer film had a rusted color instead of the original red brick color of the iron oxide pigment.

Scribe undercutting. Most of the coating systems developed blistering and rusting at the scribes after 3,000 hours of cyclic salt-fog/freeze exposure.

The coating performance of silicone rubber was extremely poor; severe blistering and rusting developed at the scribes after 2,000 hours of cyclic salt-fog/freeze exposure. The film has been undercut to depths of 11 and 15 mm (0.43 and 0.59 in) for code nos. 14 and 15 systems, respectively. On the stripping of the coating, a large rust nodule was noted under the film. Rust was also observed at the edges. The rust pattern on the steel surface is shown in figure 3. Of all the coating systems tested, silicone rubber specimens exhibited the greatest tendency to corrode at the scribes and edges.

Of the bridge coating systems presently being used, the CALTRANS high solids phenolic system and the high-VOC epoxy mastic/polyurethane systems had the poorest resistance to scribe undercuttings (9.5 mm [0.37 in] and 9.3 mm [0.36 in], respectively). The low-VOC epoxy mastic/acrylic epoxy system developed an undercutting that was 6.8 mm (0.27 in) which is also relatively poor.

Seven coating systems -- the low-VOC and high-VOC modified inorganic zinc/epoxy/acrylic epoxy, the low-VOC and high-VOC organic zinc/epoxy/polyurethane, the low-VOC and high-VOC solvent-based inorganic zinc/alkyl silicate/epoxy/polyurethane, and the CALTRANS waterborne acrylic system--showed a moderate degree of undercutting at the scribes (3.0 to 4.0 mm [0.12 to 0.16 in]) after 3,000 hours of the cyclic salt-fog/freeze exposure.

The best corrosion performance was provided by the water-based inorganic zinc potassium silicate/water-based acrylic/water-based acrylic (code no. 10), which did not undercut at the scribe or develop any rusting or blistering. The high-ratio water-based inorganic zinc potassium silicate/water-based acrylic (code no. 11) did not undercut, but did--as noted above--exhibit a poor topcoat adhesion. A different topcoat for the water-based inorganic potassium silicate may improve overall performance. The water-based inorganic zinc potassium silicate/epoxy/urethane (code no. 9) developed only minute undercutting at the scribe 1.5 mm (0.06 in). The superior rust and scribe undercutting resistance in accelerated salt-fog/freeze testing of the water-based inorganic zinc primer system reconfirmed the idea that a zinc-rich primer is an essential coating system element for high performance in a salt environment.

Thw two low-VOC solvent-based inorganic zinc systems did not protect against underfilm corrosion at the scribes as well as did the water-based inorganic zincs. The low-VOC solvent-based inorganic zinc (code no. 7) developed 3.3-mm (0.13-in) undercutting; the low-VOC solvent-based modified inorganic zinc (code no. 1) developed 4.0 mm (0.16 in) of undercutting.

An overall comparison of the average undercutting (U) for the candidate coating systems is shown in figure 4. A dividing line is drawn at an undercutting of 6.35 mm (0.25 in), which is generally considered as a standard criterion for a pass/fail classification: Undercutting of greater than 6.35 mm (0.25 in) is considered a failure in many rating systems. Using this criterion, the low-VOC epoxy mastic/acrylic epoxy, high-VOC epoxy mastic/polyurethane, and CALTRANS high solids phenolic system (code nos. 3, 4, and 13) failed afte 3,000, 2,000, and 2,500 hours in the cyclic salt-fog/freeze exposure, respectively. The high undercuttings of silicone rubber (code nos. 14 and 15) are not shown here because they are off the scale.

The underfilm corrosion as measured at the scribe of four generic, similar low- and high-VOC coating pairs was compared to study the effect of the reduction of solvents and increase of solids. The average undercutting from the scribes of duplicated panels was correlated with cyclic salt-fog/freeze exposure time (figures 5, 6, 7, and 8). For the epoxy mastic/polyurethane and the solvent-based inorganic zinc/epoxy/polyurethane, low-VOC systems outperformed the corresponding high-VOC systems. A comparison of the low- and high-VOC organic zinc/epoxy/polyurethane coating systems produced almost superimposed curves; this indicates similar durability. Only the low-VOC modified inorganic zinc/epoxy/polyurethane system performed less well than its counterpart high-VOC coating system at exposure times greater than 1,300 hours; it did however, have a later initial failure time. In general, the new low-VOC coating systems performed better than--or at least as well as--their high-VOC counterparts.

UV/condensation test

Virtually all the candidate coating systems showed color fading, chalking, and gloss reduction after 2,000 hours of the UV/condensation exposure. These changes are due to the light sensitivity of organic polymers: these tend to degrade or oxidize to form unsaturated or carbonyl compounds in the presence of ultraviolet light. [5] Of all the coatings, only one--the CALTRANS waterborne acrylic coating system--actually failed after 2,000 hours of exposure; the entire coating rusted and peeled (figure 9). The precursor for this drastic failure was the severe blistering of 2MD which developed after 1,500 hours of exposure (figure 10). This blistering suggested a poor resistance of this coating system to moisture transfer in high-temperature, high-humidity environments. The moisture permeated through the coating resulting in loss of adhesion and formation of blisters.

Composite test rating for accelerated

weathering tests

The cyclic salt-fog/freeze test using a scribed panel is a more effective accelerated testing regimen for evaluating coating systems than is the UV/condensation test; this is because the former test provided more and quicker blistering, rusting, and scribe undercutting. In most cases, differences in performance could be noted after 1,500 hours of cyclic salt-fog/freeze exposure. However, a combination of testing regimens should provide more information on field performance and failure modes. Therefore, a composite test rating was developed from the results of all of the accelerated testings to better define the comparative performance of the candidate materials.

For this composite rating, coating systems were rated based on the three types of failures evaluated in this study:

* Undercutting at the scribe of the cyclic salt-fog/freeze exposed panels (R1).

* Blistering and rusting of the scribed side of the cyclic salt-fog/freeze exposed panels (R2).

* Blistering and rusting of the UV-exposed side of the QUV panels (R3).

For each test, a rating of 10 was considered to be a perfect performance; a rating of 0 was considered a total failure. These ratings were summed to yield a composite rating. A rating of 26 to 30 is good, 20 to 25 is medium, and below 20 is considered poor. These composite test rating results, along with comparative ranks, are presented in table 3.

The water-based inorganic zinc potassium silicate/water-based acrylic/water-based acrylic, water-based inorganic zinc potassium silicate/epoxy/polyurethane, and high-ratio water-based inorganic zinc potassium silicate/water-based acrylic all performed well (ratings of 26 to 30), even though the high-ratio water-based inorganic zinc potassium silicate/water-based acrylic exhibited poor topcoat performance.

That coating systems that performed moderately well were the low- and high-VOC solvent-based inorganic zinc alkyl silicate/epoxy/polyurethane, low- and high-VOC modified inorganic zinc/epoxy/polyurethane, and low- and high-VOC organic zinc/epoxy/polyurethane (ratings of 24 to 25).

Poorly performing systems were the CALTRANS waterborne acrylic, CALTRANS high solids phenolic, low-VOC epoxy mastic/acrylic epoxy, and high-VOC epoxy mastic/polyurethane systems (ratings of 15 to 18).

Outdoor exposure

Accelerated corrosion testing--with judiciously selected controls and exposure regimens--can provide significant information to predict coatings' general comparative field performance. Because such performances may vary significantly in specific field environments, in this study, marine exposure at Sea Isle City, New Jersey was selected as a way of comparing the accelerated test results with those obtained in a salt-rich natural environment.

After 22 months of outdoor marine exposure at Sea Isle, the general appearances of all the coating systems were fairly good with regard to peeling, checking, cracking, surface blistering, and surface rusting. They all faded and lost gloss to some degree. The major failures observed were those of rusting and blistering at the scribes. All panels developed rust at the scribes, but only six coating systems exhibited blistering at the scribes. The blistering results at 15 and 22 months of outdoor exposure are presented in table 4. The undercutting results were not measured.

The preliminary failure trends agree with the performance obtained in the accelerated cyclic salt-fog/freeze results. In particular, the high-VOC epoxy mastic/polyurethane system developed severe blistering (size 0 blisters with high density, see figure 11) at the scribe extending as far as 16 mm (0.63 in) after both 15 and 22

[TABULAR DATA OMITTED]

Notes: Rating conducted in accordance with ASTM method D714 (R1), "Evaluating Degree of Blistering of Paints" and ASTM method D1654, "Evaluation of Painted or Coated Specimens Subjected to Corrosive Environments" (R2 and R3).

[TABULAR DATA OMITTED]

months. However, two silicone rubber systems did not develop any blistering afte 15 months of outdoor exposure. They exhibited slight blistering (3.72 mm [0.15 in]) after 22 months of exposure, whereas blisters started to form at the scribes after only 1,000 hours of cyclic salt-fog/freeze exposure. This difference in behavior is not well understood and demonstrates that salt-fog testing is not always a good indicator of field performance. [6,7]

Conclusions

Based upon accelerated test data generated and the short-term (22 months) outdoor test results, the following conclusions can be made:

* Coating formulations with reduced solvent content performed as well as similar higher VOC-containing materials.

* The water-based inorganic zincs performed better than the solvent-based inorganic and the epoxy organic zinc-rich candidates tested; this low-VOC coating system is strongly recommended for a high salt marine environment.

* Both the low- and high-VOC epoxy mastic did not perform well in the salt-fog evaluation; this was evidenced by excessive undercutting at the scribe and general subfilm corrosion on surfaces away from the scribe. Poor performances with identical failure modes have been reported for similar formulations on some bridges exposed to high salt environments. Both these epoxy mastic materials developed blistering at the scribe. The high-VOC product exhibited extreme blistering at the scribe that extended much farther than 6.35 mm (0.25 in) after 22 months of Sea Isle exposure.

* Both CALTRANS low-VOC systems did not perform well in the accelerated testing. However, these materials are reported to perform reasonably well in the California field environments where they are used. Other investigations have also found that the performance of organic waterborne materials in the field is better than that predicted by salt-fog results.

References

[1] S-L. Chong and J. Peart. Evaluation of Volatile Organic Compound (VOC)--Compatible High Solids Coating System for Steel Bridges, Publication Number FHWAR-RD-91-054, Washington, DC: Federal Highway Administration, August 1991.

[2] T. Smith, "New Generation Epoxies for the Pulp and Paper Industry," Journal of Protective Coatings and Linings, Vol. 6, No. 8, August 1989, p. 25.

[3] K.A. Trimber. "Detection and Removal of Chemical Contaminants in Pulp and Paper Mills," Proceedings of the Pulp and Paper II Industry Seminar to the Steel Structures Painting Council. New Orleans, LA, July 28-29, 1988.

[4] S.K. Boocock, et. al. Effect of Surface Contaminants on Coating Life. Publication Number FHWA-RD-91-011, Washington DC: Federal Highway Administration, November 1990.

[5] G. Gupta and C.P. Chiang. "Correlating Accelerated with Natural Weathering Through FT-IR Spectroscopy," Proceeding of the ACS Division of Polymeric Materials: Science and Engineering, Vol. 63, Washington, DC, August 26-31, 1990, p. 667.

[6] B.R. Appleman. "Survey of Accelerated Test Methods for Anti-Corrosion Coating Performance," Journal of Coatings Technology, August 1990, pp. 57-67.

[7] J.A. Ellor and R. Kogler. "Evaluation of Selected Maintenance Coatings for Hand and Power Tool-Cleaned Surfaces," Journal of Protective Coatings and Linings, Vol. 7, No. 12, December 1990, p. 46.

Shuang-Ling Chong is a research chemist in the Materials Division, Office of Engineering and Highway Operations Research and Development, Federal Highway Administration. Her 20 years' research experience covers photolysis, ion-molecule reactions, fractionation and characterization of organic materials in fossil fuels and petroleum, and identification of toxic organics and metals in coal combustion residues. Dr. Chong is currently conducting staff research in paint testings and evaluation of low volatile organic compound coating systems.
COPYRIGHT 1992 Superintendent of Documents
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Chong, Shuang-Ling
Publication:Public Roads
Date:Mar 1, 1992
Words:3189
Previous Article:Intermodal Surface Transportation Efficiency Act of 1991.
Next Article:Testing of a concrete sealer using the rapid chloride permeability technique.
Topics:


Related Articles
Steel bridge coatings research.
Laboratory testing of the performance of moisture-cured urethanes on new steel.
Polyester Replaces Solvent-Based Paints.
Internal mold release.
Laboratory evaluation of waterborne coatings on steel Publication No. FHWA-RD-03-032.
Internal mold release.
PU coatings.
Are two coats as effective as three?
Multiple corrosion protection systems for reinforced concrete bridge components publication no. FHWA-HRT-07-043.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters