Effects of Moisture, Location, and Angle on Automotive Paint System Appearance During Natural Weathering
Henry K. Hardcastle III
Atlas Material Testing Technology LLC
This paper presents results from a simple designed experiment (DOE) in natural outdoor weathering. Results show comparative effects of three weathering variables: moisture, exposure angle, and exposure location. The experiment design reveals rank importance of the study variables and links characteristics of the outdoor weathering environment with appearance degradation of coatings. The DOE data follows gloss retention of four automotive paint systems over 96 months in subtropical Florida and desert Arizona. The paper discusses some of the root causes and co-variables which may explain automotive paint appearance degradation.
1. Introduction and Background
1.1. The Moisture Variable
Researchers know moisture plays a key role in weathering of many materials and often place moisture in a set of three primary weathering variables with sunlight and temperature. Researchers use southern Florida as a weathering reference environment because of its relative humidity, rain, condensation, and the important effect moisture plays in weathering. Researchers also make great efforts to include moisture variable control in artificial weathering tests for both simulation and acceleration of natural outdoor weathering degradation.1 Moisture represents an important focus for weathering studies, exposure standards, service life prediction methodology, and weathering device design.2,3,4,5 Several recent studies also show the important effect of moisture on material degradation rates and characteristics of automotive coating systems.6,7 Because of these historic and recent perspectives of moisture
1.2. The DOE Approach to Weathering Experiments.
Traditional natural outdoor weathering studies ordinarily consist of simple exposures of material specimens in outdoor environments for periods of time while making intermittent measurements of material characteristics throughout the exposures. Sometimes the measured characteristic (also known as
2. Design of Experiment
Researchers planned a simple natural weathering DOE to help understand the effects of moisture on gloss degradation of automotive coatings. The study objectives included obtaining information which compared the effects of moisture to two other weathering factors; exposure location and exposure angle. These three variables; moisture, exposure location, and exposure angle naturally fit into a 23 full factorial DOE. Figure 1 shows this weathering DOE modified from Montgomery.12
2.1. Weathering DOE Trials
This DOE included eight long term weathering exposures varying the three factors (moisture, exposure angle, and location) simultaneously. The experiment design varied the three factors in an orthogonally balanced manner. Contrasting trials varied each factor to a low (-) and high (+) setting independently of the other factor settings. All eight exposures began within three days of November 20, 1998 and continued throughout the 96 months reported here within. Every three months, the exposed automotive coating specimens were measured and the measurements were plotted against exposure time in order to obtain the degradation curves for each of the eight exposures. In this manner the DOE characterized the long term weathering degradation for each variable setting. All specimens were exposed in the backed condition.
2.2. Exposure and Independent Variables
Varying independent environmental factors in a natural outdoor weathering experiment may seem paradoxical.
2.2.1. The Moisture Variable
In order to characterize the effect of moisture in this DOE, four of the eight trials exposed specimens on racks with water spray. Essentially identical to non-spray racks except for the moisture, the spray racks applied a light rain-like water spray on exposed specimens. A spray nozzle applied high purity de-ionized water to wet specimens exposed on spray racks. A single spray event lasted 60 seconds. Eight spray events occurred at the beginning of each hour from 08:00 hours until 16:00 hours during each day of the exposure. Trials shown in figure 1 with the moisture spray variable set high (+) exposed specimens on spray racks while trials with the moisture spray variable set low (-) exposed specimens on racks with no spray. In this way, the DOE characterized the effect of an artificially introduced moisture spray on the weathering degradation of the auto paint systems.
It is interesting to consider some of the co-variables associated with the moisture spray factor in this experiment. The controlled eight-minute sprays represent one source of the moisture in this experiment. However, another important source of moisture co-varies with the location factor and the angle factor. Figure 2 shows naturally occurring approximate time of wetness for Arizona and Florida at 5
2.2.2. The Exposure Angle Variable
In order to characterize the effect of exposure angle in this DOE, four of the eight trials exposed specimens on racks oriented 45
Exposures at 45
2.2.3. The Exposure Location Variable
In order to characterize the effect of exposure location in this DOE, four of the eight trials exposed specimens in southern Florida while the other four trials exposed specimens in desert Arizona. Trials shown in figure 1 with the location variable set high (+) exposed specimens at Atlas Material Testing Technology LLC
2.3. Materials and Dependent Variables
2.3.1. Automotive Paint Systems
Within each exposure trial, the DOE exposed four
2.3.2. Appearance Measurements
Gloss represents an important characteristic of automotive paint systems and the auto industry widely studies gloss weathering degradation of automotive coatings.14,15 Technicians performed 20
3. Results and Analysis
Figures 5 through 12 show the results of the eight trials in this weathering DOE as the degradation curves through 96 months of exposure. Gloss measurements were not performed on the specimens until 30 months after the exposures started. Some of the curves show a portion of the function indicative of rate dependent degradation (slope) followed by a leveling off of the curve once some level of degradation had been achieved. Because of this characteristic of the degradation curves, analysis of the data at different points in time can drastically affect the results. For example, analyzing data after all the trial specimens have completely degraded will show no effects of the study variables. For this reason the DOE included analysis at two points in the exposure period before the degraded condition had been reached in all the trials. Having the degradation curves to visually inspect, the 45 and 60 month intervals were chosen for analysis.
The results of this DOE are shown in figures 5 through 12.
DOE is primarily a logic tool. The logic includes comparing, or contrasting the set of trials with the specific factor of interest set low to the set of trials with the same factor set high. Figure 13, adapted from Montgomery, illustrates this logic for this weathering DOE.10 The analysis calculates the effect of factor A (spray) by determining an average of the four trials with no spray (low, A-), determining an average of the four trials with spray (high, A+), and finding the difference between these two averages. The analysis uses the same procedure to calculate the effect of factor B (angle), however, contrasts different sets of trails to determine the effect of angle. The analysis contrasts factor C (location) along a third axis of the experimental volume. In this manner, the analysis reveals information regarding the effects of each variable, as well as interactions in a robust manner from only eight trials in this experiment.
The mechanics of DOE analysis is fairly standardized and widely published and it will be left to other texts to outline standard analysis procedures. The Barrentine reference fully documents the analytical procedures used in this study.9 The analysis tables in figure 14 for 45 months and figure 15 for 60 months follow Barrentine
3.2.1. Forty Five Month Analysis
3.2.2. Sixty Month Analysis
4.1. Observations at the 45 Month Interval
The analysis reveals several critical observations shown in figure 14: Spray (A) had, by far, the largest effect on the gloss. Location (C) also showed a significant effect. Angle (B) showed only a marginal effect. Spray (A) interacted significantly with location (C). Spray did not effect gloss degradation in southern Florida as it did in Arizona! A similar interaction also appeared between spray (A) and angle (B). The analysis also indicates the possibility of a three-way interaction between spray (A), angle (B), and location (C) (however, the level of this effect appears very close to the decision limits for significance). Additionally, spray (A) and location (B), as well as their interaction, show a significant effect on the variation observed in the data.
4.2. Observations at the 60 Month Interval
Several observations can be made from the analysis shown in figure 15: Spray (A) had, by far, the largest effect on the gloss. Location (C) showed only a marginal effect. The analysis also indicated the possibility of an interaction between spray and location. Additionally, spray (A) and location (C), as well as their interaction, show a significant effect on the variation observed in the data.
The results of the main effects analysis for 60 months are graphed in figure 16 showing the effect of the spray, angle, and location factors studied in this experiment.
The results of the interaction effects for 60 months are graphed in figure 17 showing the interaction between the spray (A) and location (B) factors in this experiment.
5. Conclusions and Considerations
Natural weathering DOEs offer an efficient and robust approach to characterizing natural weathering material degradation.
Application of only eight daytime moisture sprays of one minute duration showed significant and important effects on the degradation curves of the automotive paint systems exposed in both Arizona and Florida.
Application of water sprays dramatically accelerated gloss loss and the effect of these moisture sprays far outweighed the effects due to location and exposure angle.
The significant differences in radiant exposure due to different exposure angles did not appear to cause significant effects in this experiment since exposure angle did not show a significant effect on the results.
The significant differences in naturally occurring time of wetness due to different exposure angles, did not appear to cause significant effects in this experiment since the exposure angle did not appear to have a significant effect on the results.
The moisture spray factor appeared to interact with the exposure location factor in this experiment.
One-Factor-At-A-Time (OFAT) experiments may not have characterized the interaction between the moisture spray factor and the exposure location factor.
Applying moisture sprays in artificial weathering methods may have dramatic effects on the results and conclusions.
The author would like to thank the technicians at Atlas Weathering Services Group
- Pickett, J.E., Umamaheswaran, V. "Highly Predictive Accelerated Weathering of Engineering Thermoplastics" SAE technical paper 2003-01-1192, Society of Automotive Engineers, March, 2003.
- Bank, R., "Study Shows Differences in Weatherability of Universal Grades in White Paint," Paint and Coatings Industry, September, 1997.
- SAE J1960 Accelerated Exposure of Automotive Exterior Materials Using a Controlled Irradiance Water-Cooled Xenon Arc Apparatus. 2003 SAE Handbook, vol. 1. Warendale, PA: Society of Automotive Engineers, Inc., 2003.
- Hardcastle, H.K., "Considerations for Relating Artificial Laboratory and Natural Outdoor Weathering Durability Testing" Proceedings of the Annual Technical Conference of The Society of Plastics Engineers (ANTEC) Chicago, IL, May (2004).
- ASTM G90-94 Practice for Performing Accelerated Outdoor Weathering of Nonmetallic Materials Using Concentrated Natural Sunlight. 1994 Annual Book of ASTM Standards, vol. 14.02. Philadelphia, PA: American Society for Testing and Materials, 1994.
- Nguyen, T., et al, "Relating Laboratory and Outdoor Exposure of Coatings: II. Effects of Relative Humidity on Photodegradation and The Apparent Quantum Yield of Acrylic-Melamine Coatings" Journal of Coatings Technology, 74, No. 932, 65 (2002).
- Misovski, T., Nichols, M.E., Hardcastle, H.K. "The Influence of Water on the Weathering of Automotive Paint Systems", currently in proceedings publication, presented at The 4th International Symposium on Service Life prediction, National Institute for Standards and Technology, Key Largo, FL. December 2006.
- Pianka, E. Evolutionary Ecology, 2nd ed.; Harper & Row: New York, NY 1978.
- Barrentine, L.B. An Introduction to Design of Experiments - A Simplified Approach; ASQ Quality Press: Milwaukee, WI 1999.
- Hardcastle, H.K. "Applying Taguchi Designs to EMMAQUA Experiments", Proceedings of Materials Life Society, Japan: 4th International Symposium on Weatherability, Kanagawa University, Japan, 2000.
- Hardcastle, H.K., "Understanding the Effects of Weathering Variables on Plastics Using Fractional Factorial Experiments" Proceedings of the Annual Technical Conference of The Society of Plastics Engineers (ANTEC) Orlando, FL, May (2000).
- Montgomery, D.C., Design and Analysis of Experiments, 3rd ed., (1991) John Wiley and Sons: New York.
- Hardcastle, H.K. "Characterizing the Effect of Weathering Variables Using Accelerated Fractional Factorial Experiments", in Natural and Artificial Ageing of Polymers: Reichert, T., Ed. Publication No.5, Gesellschaft fur Umweltsimulation:Germany, 2004.
- Bauer, D.R., "Chemical Criteria for Durable Automotive Topcoats," Journal of Coatings Technology, 66, No. 835, 57 (1994).
- Adamsons, K., "Chemical Depth Profiling of Automotive Coating Systems Using Slab Microtome Sectioning with IR/UV-VIS Spectroscopy and Optical Microscopy," Journal of Coatings Technology, 74, No. 924, 47 (2002).