Skip to contentUnited States Department of Transportation - Federal Highway Administration FHWA Home
Research Home
Report
This report is an archived publication and may contain dated technical, contact, and link information
Publication Number: FHWA-HRT-11-046
Date: June 2011

Performance Evaluation of One-Coat Systems for New Steel Bridges

508 Captions

Figures

Figure 1. Photo. Composite of small panels. This photo shows the 4-by-6-inch (10-by-15-cm) panels. The panels are placed in 2-column sets, with each column consisting of 7 panels, making a total of 14 panels for each coating system. The individual coating systems are labeled below the columns. Panels are shown for all coating systems except glass flake reinforced polyester. All panels are covered with white tape around the edges and on the top in the center, where they are identified by numbers.

Figure 2. Photo. Composite of large panels. This photo shows the 6-by-12-inch (15-by-30-cm) panels. The panels are placed in 2-column sets, with each column consisting of 6 panels, making a total of 12 panels for each coating system. The individual coating systems are labeled below the columns. Panels are shown for all coating systems except glass flake reinforced polyester. All panels are covered with white tape around the edges and on the top in the center, where they are identified by numbers.

Figure 3. Photo. Example of a scribing tool. This photo shows a scribing tool that was designed and built at the Coatings and Corrosion Laboratory. The tool is made of stainless steel consisting of a 14.5-by-11-by-1-inch (368.3-by-279.4-by-25.4-mm) support base to rest the test panel while mechanically scribing it. The test panel is held in place by raised hinges. The surface of the base consists of two fixtures: a guide rod fixture and a v-notch. A handle consisting of a scribing tip fixed at an angle slides back and forth on support guide rods to scribe the test panel. When the handle is not in motion, it rests in the v-notch.

Figure 4. Photo. Salt-fog chamber. This photo shows a salt-fog chamber that consists of a unit to house the test panels while using cyclic corrosion of fog and dry-off processes. The unit is made of plastic and includes a custom controller to program the salt fog and drying cycles. A solution reservoir, which feeds salt solution for the fog cycles, is next to the housing unit. Test panels can be loaded into the chamber from the top by opening a pyramid shaped door. During the operation of the chamber, the door is secured using plastic hinges. The unit has the capacity to shut off by itself at the end of the test cycle. The fog chamber is installed on the floor and cannot be moved.

Figure 5. Photo. UV weathering tester. This photo shows an ultraviolet (UV) weathering tester, which is a large apparatus on four wheels. Test panels are loaded into the UV weathering tester from the sides. Test panel specimens are installed in the test machine so that they form the sidewalls of the test chamber. The chamber also consists of a water reservoir which is heated to produce the water vapor for condensation. The source of UV is light from 1.12 x 10-6 ft (340 nm) lamps. Condensation and UV cycles are programmed using an automated controller. The unit has the capacity to shut off by itself at the end of the test cycle.

Figure 6. ME exposure rack in Sea Isle City, NJ. This photo shows a marine environment (ME) exposure rack in Sea Isle City, NJ. The test panels are on wooden racks inclined at 45 degrees facing south. These racks consist of horizontal wooden runners spaced apart to accommodate the length of the test panels and supported on slanted vertical wooden runners. The horizontal runners have nonmetallic hinges to support the test panels as they are being deployed.

Figure 7. Graph. Mild NW exposure rack at TFHRC. This photo shows mild natural weathering (NW) exposure racks at Turner-Fairbank Highway Research Center (TFHRC). The racks are similar in design and structure to those in figure 6, except that they are inclined at 30 degrees. These racks face south and have nonmetallic (Teflon®) hinges to support the test panels. Two racks are shown: one for NW with salt spray and the other for NW only. A solution of 15 weight percent sodium chloride was manually sprayed every 24 h onto these panels.

Figure 8. Photo. Simple grinding tool and hand press kit. This photo shows a simple grinding tool and hand press kid. The simple grinding tool consists of a mortar and a pestle, and the hand press kit is a stainless steel enclosure consisting of a nut to house the powder and bolts pressing from both sides to compress the powder and make the pellets.

Figure 9. Photo. Drill press to score a test area around a dolly. This photo shows a test panel with a dolly affixed on the surface. Adhesion strength testing was performed by hydraulically pulling on test dollies affixed to the surface of test panels using an epoxy strength adhesive. A drill press with a bit scores around the dolly to isolate the adhesion test area.

Figure 10. Photo. Hydraulic adhesion tester. This photo shows a hydraulic adhesion tester. The tester is a cylindrical enclosure that transfers the pumping action of the handle. The hydraulic force is transferred to a coupling fixture which attaches to the dolly to pull it away from the surface of the coating system.

Figure 11. Graph. FTIR spectrum of three-coat system (top coat) before ALT. This graph shows the Fourier Transform Infrared Analysis (FTIR) of the three-coat control before accelerated laboratory testing (ALT). Wave number is shown on the x-axis, and absorbance is shown on the y-axis. Major peaks can be observed at wave numbers of 3,400, 3,200, 1,700, 1,400, and 1,100. Highest absorbance was observed for the peak at wave number 1,700 (0.42), and lowest absorbance was found at wave number 3,400 (0.05).

Figure 12. Graph. DFT data for the 10 coating systems. This bar graph shows dry film thickness (DFT) of all coating systems on the x-axis and film thickness in mils shown on the y-axis. Standard deviation is represented on the primary y-axis, and the coefficient of variance is represented on the secondary y-axis. Polyaspartic (ASP) and glass flake reinforced polyester (GFP) had the highest dry film thickness values (greater than 20 mil (508) while polysiloxane (SLX) and urethane mastic (UM) had the lowest values (less than 10 mil (254)).

Figure 13. Graph. Mean gloss reduction data. This bar graph shows mean gloss reduction (percent) of all coating systems in four exposure conditions: accelerated laboratory testing (ALT), marine exposure (ME), natural weathering (NW), and natural weathering with salt spray (NWS). Coating systems are represented on the x-axis, and gloss reduction is shown on the y-axis. Epoxy mastic (EM) had the highest gloss reductions (greater than 90 percent) in all exposure conditions, while urethane mastic (UM) had the lowest gloss reductions in all outdoor exposure conditions (less than 5 percent) and moderate gloss reductions in ALT (25 percent).

Figure 14. Graph. Δ E after exposure tests. The graph shows mean color differences (E) of all coating systems in four exposure conditions: accelerated laboratory testing (ALT), marine exposure (ME), natural weathering (NW), and natural weathering with salt spray (NWS). Coating systems are represented on the x-axis, and color change is shown on the y-axis. Epoxy mastic (EM) had the highest color change (greater than 9) in all exposure conditions, while urethane mastic (UM) had the lowest color changes in all outdoor exposure conditions (less than 1) and moderate gloss reductions in ALT (less than 4).

Figure 15. Graph. Pencil scratch hardness before and after exposure tests. This bar graph shows pencil scratch hardness before and after exposure in accelerated laboratory testing (ALT), marine exposure (ME), natural weathering (NW), and natural weathering with salt spray (NWS). Coating systems are represented on the x-axis, and pencil scratch hardness is shown on the y-axis. All coating systems except two-coat, polyaspartic (ASP), polysiloxane (SLX), and urethane mastic (UM) had no change in pencil scratch hardness values after outdoor exposures and ALT. Also, the two-coat system had unchanged pencil scratch hardness in ALT and increased pencil scratch hardness in the three outdoor exposures. The other three coating systems, ASP, SLX, and UM, all showed increases in pencil scratch hardness in all exposure conditions.

Figure 16. Graph. Comparison of initial adhesion strength data using pneumatic and hydraulic methods. This bar graph shows initial adhesion strength using pneumatic and hydraulic adhesion methods. Coating systems are represented on the x-axis, and the adhesion strength in psi is represented on the y-axis. Both methods resulted in similar adhesion strength values for all coating systems except the two-coat system and glass flake reinforced polyester (GFP), with coefficients of variation of 28 percent and 22 percent, respectively.

Figure 17. Graph. Changes in mean adhesion strength after ALT and outdoor tests. This bar graph shows changes in adhesion strength (percent) before and after exposure in accelerated laboratory testing (ALT), marine exposure (ME), natural weathering (NW), and natural weathering with salt spray (NWS) for both scribed and unscribed panels. Coating systems are represented on the x-axis, and variation in adhesion strength is shown on the y-axis. The results were mixed in that two-coat system, polyaspartic (ASP), calcium sulfonate alkyd (CSA), and high-build waterborne acrylic (HBAC) showed increased adhesion strength, while the three-coat system, epoxy mastic (EM), waterborne epoxy (WBEP), polysiloxane (SLX), and urethane mastic (UM) showed a decrease in adhesion strength after lab testing and outdoor exposures.

Figure 18. Photo. Cohesive failure modes of ASP and EM. This photo shows the coating surface on polyaspartic (ASP) and epoxy mastic (EM) panels after adhesion strength testing was performed. Both these coating systems demonstrated cohesive failure modes, meaning that the coating system had an adhesively well-bonded primer to the steel substrate in comparison to the adhesion bonding between the coating layers. ASP had a remaining dry film thickness (DFT) of 4.2 mil (106.68 μm), and EM had a remaining DFT of 1.5 mil (38.1 μm) on the surface.

Figure 19. Photo. Progressive changes of panel 4 (three-coat: ALT).This figure shows a series of photos of scribed three-coat panel 4 at time periods of 0, 1,080, 2,160, 3,240, 4,320, 5,400, 6,120, and 6,840 h of accelerated lab testing (ALT). The photos do not show any holidays or surface deterioration from the time of exposure to 6,120 h; however, one holiday is shown at the termination of testing at 6,840 h. Rust creepage growth appears minimal at all time periods.

Figure 20. Photo. Progressive changes of panel 11 (three-coat: ME). This figure shows a series of photos of scribed three-coat panel 11 at time periods of 0, 6, 12, 18, and 24 months of marine exposure (ME). The photos do not show any holidays or surface deterioration through the time of exposure. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 21. Photo. Progressive changes of panel 18 (three-coat: NW). This figure shows a series of photos of scribed three-coat panel 18 at time periods of 0, 7, 13, and 18 months of natural weathering (NW) exposure. The photos do not show any holidays or surface deterioration. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 22. Photo. Progressive changes of panel 24 (three-coat: NWS). This figure shows a series of photos of scribed three-coat panel 24 at time periods of 0, 7, 13, and 18 months of natural weathering exposure with salt spray (NWS). The photos do not show any holidays or surface deterioration. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 23. Photo. Progressive changes of panel 30 (two-coat: ALT). This figure shows a series of photos of scribed two-coat panel 30 at time periods of 0, 1,080, 2,160, 3,240, 4,320, 5,400, 6,120, and 6,840 h of accelerated lab testing (ALT). The photos do not show any holidays or surface deterioration. Rust creepage growth appears minimal for all time periods.

Figure 24. Photo. Progressive changes of panel 36 (two-coat: ME). This figure shows a series of photos of scribed two-coat panel 36 at time periods of 0, 6, 12, 18, and 24 months of marine exposure (ME). The photos do not show any holidays or surface deterioration. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 25. Photo. Progressive changes of panel 44 (two-coat: NW). This figure shows a series of photos of scribed two-coat panel 44 at time periods of 0, 7, 13, and 18 months of natural weathering (NW) exposure. The photos do not show any holidays or surface deterioration. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 26. Photo. Progressive changes of panel 51 (two-coat: NWS). This figure shows a series of photos of scribed two-coat panel 51 at time periods of 0, 7, 13, and 18 months of natural weathering exposure with salt spray (NWS). The photos do not show any holidays or surface deterioration. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 27. Photo. Progressive changes of panel 65 (ASP: ALT). This figure shows a series of photos of scribed polyaspartic (ASP) panel 65 at time periods of 0, 1,080, 2,160, 3,240, and 4,320 h of accelerated lab testing (ALT). This test panel shows 2 holidays at 1,080 h, 16 holidays with slight blistering at 2,160 h, 21 holidays with significant blistering at 3,240 h, and 22 holidays with significant blistering at the termination of testing at 4,320 h. Rust creepage growth appears at 2,160 h and continues to grow through the rest of ALT.

Figure 28. Photo. Progressive changes of panel 62 (ASP: ME). This figure shows a series of photos of scribed polyaspartic (ASP) panel 62 at time periods of 0, 6, 12, 18, and 24 months of marine exposure (ME). The photos do not show any holidays or surface deterioration. Rust creepage growth appears to be very minimal or zero for all time periods.

Figure 29. Photo. Progressive changes of panel 70 (ASP: NW). This figure shows a series of photos of scribed polyaspartic (ASP) panel 70 at time periods of 0, 7, 13, and 18 months of natural weathering (NW) exposure. The photos do not show any holidays or surface deterioration. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 30. Photo. Progressive changes of panel 76 (ASP: NWS). This figure shows a series of photos of scribed polyaspartic (ASP) panel 76 at time periods of 0, 7, 13, and 18 months of natural weathering exposure with salt spray (NWS). The photos do not show any holidays or surface deterioration. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 31. Photo. Progressive changes of panel 84 (EM: ALT). This figure shows a series of photos of scribed epoxy mastic (EM) panel 84 at time periods of 0, 1,080, 2,160, 3,240, 4,320, 5,400, 6,120, and 6,840 h of accelerated lab testing (ALT). The photos do not show any holidays or surface deterioration. Rust creepage growth appears at 3,240 h and continues to grow through the rest of ALT.

Figure 32. Photo. Progressive changes of panel 88 (EM: ME). This figure shows a series of photos of scribed epoxy mastic (EM) panel 88 at time periods of 0, 6, 12, 18, and 24 months of marine exposure (ME). The photos do not show any holidays or surface deterioration until 18 months and a holiday at the termination of exposure at 24 months. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 33. Photo. Progressive changes of panel 96 (EM: NW). This figure shows a series of photos of scribed epoxy mastic (EM) panel 96 at time periods of 0, 7, 13, and 18 months of natural weathering (NW) exposure. The test panel initially had two defects on the surface and developed an additional holiday after 13 months of exposure. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 34. Photo. Progressive changes of panel 102 (EM: NWS). This figure shows a series of photos of scribed epoxy mastic (EM) panel 102 at time periods of 0, 7, 13, and 18 months of natural weathering exposure with salt spray (NWS). The test panel had an initial defect and developed an additional defect at the end of testing. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 35. Photo. Progressive changes of panel 113 (HRCSA: ALT). This figure shows a series of photos of scribed high-ratio calcium sulfonate alkyd (HRCSA) panel 113 at time periods of 0, 1,080, 2,160, 3,240, 4,320, 5,400, 6,120, and 6,840 h of accelerated lab testing (ALT). The photos do not show any holidays or surface deterioration until 4,320 h when one holiday developed. Rust creepage growth appears minimal for all time periods.

Figure 36. Photo. Progressive changes of panel 111 (HRCSA: ME). This figure shows a series of photos of scribed high-ratio calcium sulfonate alkyd (HRCSA) panel 111 at time periods of 0, 6, 12, 18, and 24 months of marine exposure (ME). The photos do not show any holidays or surface deterioration. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 37. Photo. Progressive changes of panel 122 (HRCSA: NW). This figure shows a series of photos of scribed high-ratio calcium sulfonate alkyd (HRCSA) panel 122 at time periods of 0, 7, 13, and 18 months of natural weathering (NW) exposure. The test panel developed two holidays after exposure for 7 months and an additional holiday at the end of testing at 18 months. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 38. Photo. Progressive changes of panel 129 (HRCSA: NWS). This figure shows a series of photos of scribed high-ratio calcium sulfonate alkyd (HRCSA) panel 129 at time periods of 0, 7, 13, and 18 months of natural weathering exposure with salt spray (NWS). A holiday developed on the surface after 13 months of exposure. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 39. Photo. Progressive changes of panel 134 (GFP: ALT). This figure shows a series of photos of scribed glass flake reinforced polyester (GFP) panel 134 at time periods of 0, 1,080, 2,160, 3,240, 4,320, 5,400, 6,120, and 6,840 h of accelerated lab testing (ALT). The photos do not show any holidays or surface deterioration. Rust creepage growth appears at 3,240 h and progressively increases through the rest of the test period.

Figure 40. Graph. Photo. Progressive changes of panel 167 (HBAC: ALT). This figure shows a series of photos of scribed high-build waterborne acrylic (HBAC) panel 4 at time periods of 0, 1,080, 2,160, 3,240, 4,320, 5,400, 6,120, and 6,840 h of accelerated lab testing (ALT). The photos do not show any holidays or surface deterioration. Rust creepage growth appears at 3,240 h and grows significantly through the rest of the test period.

Figure 41. Photo. Progressive changes of panel 163 (HBAC: ME). This figure shows a series of photos of scribed high-build waterborne acrylic (HBAC) panel 163 at time periods of 0, 6, 12, 18, and 24 months of marine exposure (ME). The photos do not show any holidays or surface deterioration until 18 months, and two holidays appeared at the end of test period. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 42. Photo. Progressive changes of panel 174 (HBAC: NW). This figure shows a series of photos of scribed high-build waterborne acrylic (HBAC) panel 174 at time periods of 0, 7, 13, and 18 months of natural weathering (NW) exposure. The photos do not show any holidays or surface deterioration. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 43.Photo. Progressive changes of panel 181 (HBAC: NWS). This figure shows a series of photos of scribed high-build waterborne acrylic (HBAC) panel 181 at time periods of 0, 7, 13, and 18 months of natural weathering exposure with salt spray (NWS). The photos do not show any holidays or surface deterioration. Rust creepage growth appears to start at 13 months and grows until 18 months.

Figure 44. Photo. Progressive changes of panel 186 (WBEP: ALT). This figure shows a series of photos of scribed waterborne epoxy (WBEP) panel 186 at time periods of 0, 1,080, 2,160, 3,240, 4,320, 5,400, 6,120, and 6,840 h of accelerated lab testing (ALT). The photos do not show any holidays or surface deterioration through the time of exposure. Rust creepage starts at 2,160 h and progressively grows, with severe creepage at the end of testing at 6,840 h.

Figure 45. Photo. Progressive changes of panel 191 (WBEP: ME). This figure shows a series of photos of scribed waterborne epoxy (WBEP) panel 191 at time periods of 0, 6, 12, 18, and 24 months of marine exposure (ME). The test panel appears to have certain imperfections on the surface before exposure. The imperfections grow into defects at 6 months, with two additional defects developing at the end of testing at 24 months. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 46. Photo. Progressive changes of panel 202 (WBEP: NW). This figure shows a series of photos of scribed waterborne epoxy (WBEP) panel 202 at time periods of 0, 7, 13, and 18 months of natural weathering (NW) exposure. The photos do not show any holidays or surface deterioration. The test panel had initial defects on the surface but did not develop any further defects through the test period. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 47. Photo. Progressive changes of panel 208 (WBEP: NWS). This figure shows a series of photos of scribed waterborne epoxy (WBEP) panel 208 at time periods of 0, 7, 13, and 18 months of natural weathering exposure with salt spray (NWS). The test panel initially had multiple coating defects. Further coating defects and rusting developed around these areas. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 48. Photo. Progressive changes of panel 214 (SLX: ALT). This figure shows a series of photos of scribed polysiloxane (SLX) panel 4 at time periods of 0, 1,080, 2,160, 3,240, 4,320, 5,400, 6,120, and 6,840 h of accelerated lab testing (ALT). The test panel initially had about 10 defects that were present through the test period, and no new holidays developed. Rust creepage growth started at 2,160 h and got significantly worse toward the end of testing, with a large area around the scribe rusted and peeled off.

Figure 49. Photo. Progressive changes of panel 218 (SLX: ME). This figure shows a series of photos of scribed polysiloxane (SLX) panel 11 at time periods of 0, 6, 12, 18, and 24 months of marine exposure (ME). The test panel had two initial defects. Rust creepage started to grow at 6 months and increased rapidly at the end of testing, with a large area around the scribe rusting and peeled off.

Figure 50. Photo. Progressive changes of panel 226 (SLX: NW). This figure shows a series of photos of scribed polysiloxane (SLX) panel 226 at time periods of 0, 7, 13, and 18 months of natural weathering (NW) exposure. The test panel initially had 7 defects and did not increase through the rest of the test period. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 51. Photo. Progressive changes of panel 232 (SLX: NWS). This figure shows a series of photos of scribed polysiloxane (SLX) panel 232 at time periods of 0, 7, 13, and 18 months of natural weathering exposure with salt spray (NWS). The photos do not show any holidays or surface deterioration. Rust creepage started to grow at 13 months and continued through the rest of the test period, with rusting around the scribed area and at the bottom edges.

Figure 52. Photo. Progressive changes of panel 239 (UM: ALT). This figure shows a series of photos of scribed urethane mastic (UM) panel 239 at time periods of 0, 1,080, 2,160, 3,240, and 4,320 h of accelerated lab testing (ALT). The test panel had about one defect before testing. Blistering of the test panel started at 2,160 h and progressively increased at the end of testing at 4320 h. Rust creepage growth started at 1,080 h and got significantly worse at the end of testing, with a large area around the scribe rusting.

Figure 53. Photo. Progressive changes of panel 247 (UM: ME). This figure shows a series of photos of scribed urethane mastic (UM) panel 247 at time periods of 0, 6, 12, 18, and 24 months of marine exposure (ME). The photos do not show any holidays or surface deterioration through the time of exposure. Rust creepage started at 12 months with slight rusting around the scribed area that increased progressively with more rusting at the end of the test period.

Figure 54. Photo. Progressive changes of panel 252 (UM: NW). This figure shows a series of photos of scribed urethane mastic (UM) panel 252 at time periods of 0, 7, 13, and 18 months of natural weathering (NW) exposure. The test panel initially had 10 defects and did not increase throughout the rest of the test period. Rust creepage growth appears to be minimal or zero for all time periods.

Figure 55. Photo. Progressive changes of panel 258 (UM: NWS). This figure shows a series of photos of scribed urethane mastic (UM) panel 258 at time periods of 0, 7, 13, and 18 months of natural weathering exposure with salt spray (NWS). The photos do not show any holidays or surface deterioration. Rust creepage started to grow at 13 months and continued throughout the rest of the test period, with rusting around the scribed area and at the bottom edges.

Figure 56. Graph. Development of coating defects during ALT. This graph shows the development of coating defects during accelerated laboratory testing (ALT). Duration in hours is shown on the x-axis, and number of coating defects is shown on the y-axis for all coating systems, which are represented as various series. The polysiloxane (SLX), urethane mastic (UM), and polyaspartic (ASP) coating systems developed a many defects (greater than 30) in the initial phases of testing (1,440 h), while two-coat and epoxy mastic (EM) developed multiple defects toward the end of the testing. The rest of the coating systems either had few (less than 10) defects or no defects through the test period of 6,840 h.

Figure 57. Graph. Development of coating defects during ME This graph shows the development of coating defects during marine time of exposure in months for marine exposure (ME). Duration in months is shown on the x-axis, and the number of coating defects is shown on the y-axis for all coating systems, which are represented as various series. Epoxy mastic (EM) and urethane mastic (UM) developed many defects (greater than 35) after 18 months of exposure, while the remaining coating systems had minimal (less than 10) or no defects through the test period of 24 months.

Figure 58. Graph. Development of coating defects during NW. This graph shows the development of coating defects during natural weathering (NW). Duration in months is shown on the x-axis, and number of coating defects is shown on the y-axis for all coating systems, which are represented as various series. Polysiloxane (SLX) and urethane mastic (UM) had many coating defects (greater than 60) before exposure, which did not increase with time of exposure. The rest of the coating systems either had very few or no coating defects through the test period of 18 months.

Figure 59. Graph. Development of coating defects during NWS. This graph the development of coating defects during natural weathering exposure with salt spray (NWS). Duration in months is shown on the x-axis, and number of coating defects is shown on the y-axis for all coating systems, which are represented as various series. Waterborne epoxy (WBEP) initially had many defects (greater than 100) before exposure, and these defects increased slightly (greater than 150) throughout the test period. The rest of the coating systems either had very few or no coating defects through the test period of 18 months.

Figure 60. Photo. Surface coating failure by cracking (two-coat system). This photo shows an optical microscopic image of the two-coat system in accelerated laboratory testing magnified 20 times. The surface shows micro-cracks that seem to originate from a central location for each set of cracks and propagate through the surface of the coating.

Figure 61. Photo. Surface condition of the three-coat system. This photo shows an optical microscopic image of the three-coat system magnified 20 times after 6,120 h in accelerated laboratory testing, which showed small holes on the panel surface. However, the micro-sized holes were not holidays and thus did not affect the coating performance during the test.

Figure 62. Photo. Large panels with defects from coating application deficiency in NWS. This figure shows two large panels of waterborne epoxy (WBEP) coating system. The panel on the left shows WBEP at 0 h with some initial film defects and small dent areas on the surface. The right panel shows the resultant surface after 18 months of exposure in natural weathering exposure with salt spray (NWS), where the small dents grew into rusted blisters, and all of the thin dry film thickness areas were rusted. Many defects were detected by a holiday detector on the test panel surfaces.

Figure 63. Graph. Development of rust creepage during ALT. This graph shows the development of rust creepage during accelerated laboratory testing (ALT). Duration in hours is shown on the x-axis, and growth of rust creepage is shown on the y-axis for all coating systems, which are represented as various series. Polysiloxane (SLX), urethane mastic (UM), and polyaspartic (ASP) were removed after 4,320 h, while waterborne epoxy (WBEP) and high-build waterborne acrylic (HBAC) were removed after 5,040 h of exposure due to excessive rust creepage growth. The rest of the coating systems were exposed to the complete 6,840 h of the ALT period. SLX, UM, ASP, WBEP, and HBAC had more than 0.23 inches (6 mm) of rust creepage growth, while the rest of the coating systems had moderate or very minimal rust creepage. Calcium sulfonate alkyd (CSA) had creepage equal to or less than the three-coat control system.

Figure 64. Graph. Development of rust creepage during ME. This graph shows the development of rust creepage during marine exposure (ME). Duration in months is shown on the x-axis, and growth of rust creepage is shown on the y-axis for all coating systems, which are represented as various series. Polysiloxane (SLX) showed high rust creepage growth (greater than 1.2 inches (30 mm)), and urethane mastic (UM) showed moderate creepage growth of 0.20 inches (5 mm) toward the end of testing. The rest of the coating systems had moderate or minimal rust creepage. Calcium sulfonate alkyd (CSA) had creepage equal to or less than the three-coat control system.

Figure 65. Graph. Development of rust creepage during NW. This graph shows the development of rust creepage during natural weathering (NW). Duration in months is shown on the x-axis, and growth of rust creepage is shown on the y-axis for all coating systems, which are represented as various series. All coating systems had moderate or minimal rust creepage growth toward the end of the test period. Polysiloxane (SLX) had the highest creepage at 0.086 inches (2.2 mm) at the end of the test period.

Figure 66. Graph. Development of rust creepage during NWS. This graph shows the development of rust creepage during natural weathering exposure with salt spray (NWS). Duration in months is shown on the x-axis, and growth of rust creepage is shown on the y-axis for all coating systems, which are represented as various series. All coating systems except polysiloxane (SLX) and urethane mastic (UM) had moderate or minimal rust creepage growth toward the end of the test period. SLX had the highest creepage at 0.476 inches (12.2 mm), followed by UM at 0.27 inches (7 mm).

Figure 67. Photo. Rust creepage of HRCSA during ALT. This figure shows a series of photos of high-ratio calcium sulfonate alkyd (HRCSA) at 0, 720, 2,160, and 6,840 h of accelerated laboratory testing (ALT) showing creepage growth with time. CSA developed low initial creepage after 360 h; however, the creepage did not grow much during the entire test period.

Figure 68. Graph. Positive linear regression analysis between color and gloss in NW This graph shows a positive linear regression analysis between color and gloss in natural weathering (NW). Gloss reduction in percent is shown on the x-axis, and color reduction in percent is shown on the y-axis for all coating systems. Linear regression analysis was performed to correlate color to gloss, and the resultant R-squared value was 0.6918.

Figure 69. Graph. Poor linear regression analysis between color and gloss in ME. This graph shows a poor linear regression analysis between color and gloss in marine exposure (ME). Gloss reduction in percent is shown on the x-axis, and color reduction in percent is shown on y-axis for all coating systems. The R-squared value was 0.1927.

Figure 70. Regression analysis of color versus gloss for one-coat systems and control coating systems in NW and NWS. This graph shows a regression analysis of color versus gloss for one-coat and control coating systems in natural weathering (NW) and natural weathering with salt spray (NWS). Gloss reduction in percent is shown on the x-axis, and color reduction in percent is shown on the y-axis for all coating systems. Nonlinear regression analysis was performed using an exponential fit resulting in an R-squared value of 0.7046. Data from both NW and NWS were used for this correlation. This offers better correlation with increased sample size and lower standard deviation for the variable data.

Figure 71. Graph. Improved regression analysis results from figure 70. This graph shows improved regression analysis results from figure 70. Gloss reduction in percent is shown on the x-axis, and color reduction in percent is shown on the y-axis for all coating systems. The correlation between color and gloss for all one-coat systems without controls resulted in an increase in the R-squared value from 0.7046 to 0.7726.

Figure 72. Graph. Regression analysis of adhesion strength versus coating defects for one-coat and control coating systems in NW. This graph shows the regression analysis of adhesion strength versus coating defects for one-coat and control coating systems in natural weathering (NW). Adhesion strength variation in psi is shown on the x-axis, and the number of coating defects is shown on the y-axis for all coating systems. The R-squared value was 0.5511.

Figure 73. Graph. Improved regression analysis result of adhesion strength versus coating defects for one-coat systems in NW. This graph shows an improved regression analysis result of adhesion strength versus coating defects for one-coat systems in natural weathering (NW). Unscribed adhesion strength variation in psi is shown on the x-axis, and the number of coating defects is shown on the y-axis for all coating systems. When the control coating systems were not included for the analysis performed in figure 72, the R-squared value increased from 0.5511 to 0.8264.

Figure 74. Graph. Gloss reductions in NW versus gloss variations in NWS. This graph shows gloss reductions in natural weathering (NW) versus gloss variations in natural weathering with salt spray (NWS). Gloss reduction in NW in percent is shown on the x-axis, and gloss reduction in NWS in percent is shown on the y-axis. The nonlinear polynomial correlation yielded an R-squared value of 0.9437. This correlation demonstrates that NW and NWS are similar types of outdoor exposure.

Figure 75. Graph. Color variations in NW versus NWS. This graph shows color variations in natural weathering (NW) versus natural weathering with salt spray (NWS). Color variation in NW is shown on the x-axis, and color variation in NWS is shown on the y-axis. The nonlinear polynomial correlation yielded an R-squared value of 0.9561. This correlation demonstrates that NW and NWS are similar types of outdoor exposure.

Figure 76. Graph. Relationship between adhesion strength variations of scribed panels in NW and NWS. This graph shows the relationship between adhesion strength variations of scribed panels in natural weathering (NW) and natural weathering with salt spray (NWS). Adhesion strength variation in NWS in psi is shown on the x-axis, and adhesion strength variation in NW in psi is shown on the y-axis. Regression analysis was performed using a polynomial equation of second order with an R-squared value of 0.7518.

Figure 77. Graph. Relationship between adhesion strength variations of unscribed panels. The graph shows the relationship between adhesion strength variations of unscribed panels. Adhesion strength variation in percent is shown on both the x-axis and the y-axis for all exposure conditions. This figure shows the strong relationships among adhesion strength variations of scribed panels and the resultant R-squared values, as follows:

  • Accelerated laboratory testing (ALT) (dependent) versus natural weathering (NW) (independent): R-squared equals 0.86.
  • ALT (dependent) versus natural weathering with salt spray (NWS) (independent): R-squared equals 0.90.
  • Marine exposure (ME) (dependent) versus NWS (independent): R-squared equals 0.84.

Figure 78. Gloss reduction as a function of aromaticity. This graph shows gloss reduction as a function of aromaticity. All one-coat systems with aromatic (AR)/aliphatic (AP)*100 is shown on the x-axis, and gloss reduction in percent is shown on the y-axis. The regression correlation resulted in an R-squared value of 0.6107.

Equations

Equation 1. Delta E. Delta E equals bracket parenthesis delta L asterisk end parenthesis squared plus parenthesis delta a asterisk end parenthesis squared plus parenthesis delta b asterisk end parenthesis squared end bracket superscript 1 divided by 2.

Equation 2. Rate of creepage. Rate of creepage equals delta creepage divided by delta time.

Equation 3. Final average. Final average equals parenthesis 0.35 times creepage end parenthesis plus parenthesis 0.25 times holidays end parenthesis plus parenthesis 0.1 times adhesion end parenthesis plus parenthesis 0.15 times color end parenthesis plus parenthesis 0.15 times gloss end parenthesis divided by 5 all times 100.

ResearchFHWA
FHWA
United States Department of Transportation - Federal Highway Administration