As UV coatings continue to be applied to a broader range of substrates, the demand for scratch and abrasion resistance continues to grow. This is especially true with thin-film applications over flexible, semi-porous or hard substrates with varying types of gloss or haptic properties. There are a variety of ways in which damage can occur due to various geometries and forces of the objects scraping over the coating surfaces. There are many factors, such as resin composition, surface uniformity and irregularities, etc., that will influence the scratch and abrasion resistance performance of the coating. Different test methods highlight different aspects of a coating’s integrity, and often there is not consensus across test methods in regards to consistent performance. All these variables combined can result in a wide performance response from slight deformation, which may not be visually observable, to fractal failure of the polymer itself that leaves visually apparent damages.
To obtain continual improvements in scratch- and abrasion-resistant coatings, a variety of additive technologies have been developed to address this issue. In this study, the article examines additive technologies ranging from surface-active siloxanes and nanocomposite technology to synthetic amorphous silica and co-binders. These technologies are then evaluated side-by-side in a urethane UV-curable coating. The study will look at compatibility of these products in the coating formulation and provide relative ratings of their impact on scratch resistance. The scratch and abrasion resistance are measured using several common test methods. The results will provide a comparable overview of how these various technologies perform in improving scratch and abrasion resistance of the UV coating, and the variations that can occur across testing methods.
Additive Technologies
Surface slip agents can markedly lower the damage to a coating by increasing its surface slip. These additives allow objects to slip off rather than penetrate the coating matrix. They affect the surface tension of the coating, resulting in a smoother/higher slip surface, with an improved capability to deflect force across the surface, avoiding a scratch. These additives are surface-active siloxanes that exhibit weak interactions with each other and with other materials. The modified polysiloxanes migrate to the surface of the coating during cure, reducing the slip resistance of the cured film and making it possible for solid objects scraping across the surface to slide easier. The general structure of these additives is illustrated in Figure 1.
Slip agents based on polyether siloxane copolymer.
FIGURE 1 Slip agents based on polyether siloxane copolymer.
These slip agents can by modified with hydrophilic-hydrophobic polyether to control its compatibility in the coating and the behavior of the polysiloxanes. Other functionalities in the chemistry of the slip agent help with anti-cratering effects, and these liquid-based technologies can be formulated for varying ranges of recoatability.
Synthetic, amorphous silicas are produced with varying wet and high-
temperature-based processes that yield similar chemical compositions but significantly different particle types and morphologies, with a wide range of physical-chemical properties. The three types of silica used in this study are shown relative to the overall synthetic amorphous silica market in Figure 2. Fumed silica is derived from a high-temperature process. Whereas, both precipitated and colloidal silica are derived from a wet process.
Types of synthetic, amorphous silica.
FIGURE 2 Types of synthetic, amorphous silica.
A typical function for most amorphous silica is to improve the overall hardness and reinforcement of the coating. Structure-modified fumed silica is manufactured using the high-temperature flame hydrolysis that produces primary particles that are irreversibly fused to form sub-micron aggregates. These aggregates are then processed to form structure-modified particles. The wet process produces precipitated and colloidal silica. The precipitated silica is a unique, spherical particle that differs from conventional precipitated silica. The colloidal silica is functionalized to provide better compatibility in coating formulations while maintaining excellent dispersion stability in water. The colloidal silica in water is designated nanocomposites in this study.
Fumed silica TEM photos in Figure 3 show the conventional structure of fumed silica on the left and the structure-modified fumed silica on the right. This silica has higher bulk density and more compact structure, allowing for higher loading level without adversely affecting viscosity.
Conventional vs structure-modified fumed silica.
FIGURE 3 Conventional vs structure-modified fumed silica.
Figure 4 shows a new spherical precipitated silica in the center surrounded by TEM images of other conventional precipitated silica, as well as other similar products on the market such as natural ground silica, diatomaceous silica and synthetic microspheres. The particle porosity and sphericity of the novel precipitated silica is controlled by the manufacturing process, resulting in linseed oil absorption of 40 mL/100 g silica and BET surface area of <15 m 2 /g.
TEM of precipitated and natural silica.
FIGURE 4 TEM of precipitated and natural silica.
The nanocomposites are aqueous dispersions of colloidal silica. Like the fumed and precipitated silica, the colloidal silica improves the overall hardness of the coating while providing improved mechanical properties among other attributes, and due to its very small size, achieves very high clarity. Figure 5 is an example of mono-dispersed, discrete silica nano-particles of 20 nm distributed uniformly throughout a cured film. These particles are functionalized and stabilized in the aqueous dispersion. For UV-curable systems, dispersions in various monomers are also available. Particle-based dispersions in solvent and water are also available.
TEM of colloidal nano-silica in cured sample.
FIGURE 5 TEM of colloidal nano-silica in cured sample.
The last group of property enhancers are co-binders that can improve the overall hardness of the coating by increasing the overall glass transition temperature (Tg) when combined with the main resins in the final coating. Here, two co-binders are included – a high-Tg polyester and high-Tg ketone resin, shown in Figure 6. The Tg of both the polyester and ketone resin co-binders are at 90 °C.
High-Tg polyester and ketone resin.
FIGURE 6 High-Tg polyester and ketone resin.
Test Formulation
The test formulation is shown in Table 1. The additives were added at 0.1-10.0% on formulation. The silica content was added at 10.0% above the total formulation.
UV coating test formulation.
TABLE 1 UV coating test formulation.
Sample Preparations
Wet samples were prepared by weighing 100 grams of test formulation into a beaker, then incorporating various levels of the additives, then mixing for 3 min at 1,000 rpm using a 30-mm diameter dissolver blade. These samples were then applied on different substrates and cured using UV light at a speed of 20 m/min.
Testing Methods
The scratch and abrasion resistance were tested using various methods listed in Table 2.
Scratch/abrasion test methods.
TABLE 2 Scratch/abrasion test methods.
Results and Discussion
The additives from each of the technology areas chosen for this study are listed in Table 3. When the additives are added and mixed into the formulation, its compatibility to the system is checked.
Additives evaluated in this study.
TABLE 3 Additives evaluated in this study.
Some of the additives were incompatible and flocculated in the beaker after a short time. Others showed incompatibility at lower concentrations but incorporated uniformly at higher concentrations. Finally, some resulted in an inflexible film at higher concentrations that cracked and could not be used for testing. Some examples of these are shown in Figure 7.
Examples of incompatibility at lower concentration and inflexible film at higher concentration.
FIGURE 7 Examples of incompatibility at lower concentration and inflexible film at higher concentration.
The formulation for this study was not fully optimized to take into consideration some of the incompatibility or concentration effects. A summary of the test results are listed in Table 4.
Summary of test results.
TABLE 4 Summary of test results.
In general, the slip additives and co-binders had good compatibility, with the exception of Slip 500 at lower concentrations. The slip additives are surface-active agents, therefore only low dosages are needed. The structure-modified fumed silica (SMF silica) showed incompatibility due to the need to grind these particles in a high-energy mill and pre-stabilized before addition into a coating formulation. Therefore, we were unable to obtain any results with this material in this study. Finally, in regards to the nanocomposites, at higher concentrations some of these property enhancers resulted in inflexible films that could not be used in measuring their scratch and abrasion resistance.
Since the structure-modified fumed silica (SMF silica) was incompatible, it is left out in the following analysis. The remaining blanks in the graphs are because the measurements could not be obtained from the coating applied.
Martens hardness in Figure 8 shows that some additives can reduce the hardness of the coating. This was especially true of the co-binders. The nanocomposites did not form films that were acceptable for measurements, except for the NC 153. However, even this reduced the MH of the coating. Here, the best performance was seen with the Slip 496, where at 0.1%, results were better than the blank coating itself.
Martens hardness (MH).
FIGURE 8 Martens hardness (MH).
The Martindale abrasion test (Figure 9) also showed the best performance was with the slip agents, where both the Slip 410 and 496 both performed well. Here the co-binders gave good performance at the higher loading levels. Of the nanocomposites, the NC 153 had more consistent performance.
Martindale abrasion at 20° gloss.
FIGURE 9 Martindale abrasion at 20° gloss.
When we look at the Crockmeter abrasion results at both 20° and 60° gloss in Figure 10 and Figure 11, the nanocomposites showed best performance, followed closely with Slip 496.
Crockmeter abrasion at 20° gloss.
FIGURE 10 Crockmeter abrasion at 20° gloss.
Crockmeter abrasion at 60° gloss.
FIGURE 11 Crockmeter abrasion at 60° gloss.
In the TABER® Shear test shown in Figure 12, the slip agents allow for higher weight on the shear arm before it begins to scratch the coating. Here, the best performance is shown with the slip agents, specifically the Slip 410 providing the best performance. Among the nanocomposites, the NC 153 had the best performance.
TABER shear test.
FIGURE 12 TABER shear test.
Finally, when the additive’s impact is evaluated using the TABER abrader test, the best performance is between the Slip 496 and the nanocomposite NC 153, as shown in Figure 13 on the next page.
TABER abrader test.
FIGURE 13 TABER abrader test.
If we consider lowest dosage for slip additives, due to their surface activity and highest dosage for the silica and nanocomposites, their impact on the coating can be summarized in Figure 14 on the next page.
Summary of additive performance.
FIGURE 14 Summary of additive performance.
From these results, the slip additives generally performed best across these tests in this UV-curable coating formulation. The Slip 496 was the best performer among the slip additives. Following the slip additives is the nanocomposite technology, of which the NC 153 offered the best performance.
As stated, this study was to provide an overview of the comparable technologies. However, further developments would be necessary to optimize formulation and incorporation of the various technologies to provide definitive results on performance.
Conclusions
The study provides an overview of comparable technologies to improve scratch and abrasion of UV-curable coatings. The results showed that slip additives perform best in this study, with the Slip 496 giving the best overall performance. Following the slip additives would be the nanocomposites, of which NC 153 had the best performance.
This study also demonstrates that scratch resistance performance can vary significantly depending on which method is used, and often there is not a common improved performance across each and every scratch test.
Further work would be needed to examine the incorporation of additives such as the novel precipitated spherical silica, or the structure-modified fumed silica. In addition, there is a need to optimize nanocomposites to obtain best compatibility and performance.
Acknowledgement
Special thanks to Mr. Marco Heuer, Mr. Roger Reinartz and Ms. Aline Skotarczak for their work in this study.
References
1. Evonik Report PLO2018-03 – Scratch-Resistance UV-System by Mr. Reinartz and Ms. Skotarczak.
2. “Testing of Scratch and Abrasion Resistance of Coating Systems for Permanent Protection of Plastics and Other Substrates” by M. Heuer, A. Skotarczak, M. Schaepermeier, R. Reinartz, and F. Eichenberger; presented January 2020.
For more information, e-mail [email protected].
Tolerancing: The Key to Accurate Color
Tolerances are used to control color, ensure consistency within a production run and minimize lot-to-lot variability. Even when using spectral data and tolerances to quantify color, customers and suppliers can still disagree. To ensure expectations are clear and everyone is aligned, it’s important to select and use the right tolerancing method for your application. X-Rite’s Color iQC quality control software makes it easy to establish realistic tolerances and evaluate color quality. To make educated pass/fail decisions you need to know how to select the right tolerancing method and set an achievable tolerance.
Selecting a Color Model
The most common color models are L*a*b* and L*C*h°. The concept behind them is similar to longitude, latitude and altitude. With three coordinates, you can describe the exact location of any place on the planet, or in this case, any color in color space.
L*a*b*
CIELAB, or L*a*b*, was the first internationally accepted color space definition. L*a*b* values are calculated from the tristimulus values (X,Y,Z), which are the backbone of all color mathematical models. The location of a color in the CIELAB color space is defined by a three-dimensional, rectangular coordinate system (Figure 1):
The L*a*b* color space.
FIGURE 1 The L*a*b* color space.
• L* indicates the color’s lightness or darkness;
• a* is the color’s position on the red-green axis;
• b* is the color’s position on the yellow-blue axis.
Once the L*a*b* position of a color is determined, a rectangular tolerance box can be drawn around it to indicate the acceptable color difference. But since visual acceptability is the shape of an ellipse, not a rectangle, there are some places in L*a*b* color space where using a tolerance box may cause problems. Some colors may pass that shouldn’t, and some acceptable colors may fail.
L*C*hº
L*C*hº color difference calculations are derived from L*a*b* values, but mathematics converts the rectangular coordinate system to a cylindrical polar coordinate system (Figure 2).
The LCh polar coordinate system.
FIGURE 2 The L*C*hº polar coordinate system.
• L* is the same as in L*a*b* and represents the lightness plane.
• C* is the calculated vector distance from the center of color space to the measured color. Larger C* values indicate higher chroma or saturation.
• Δh° is the calculated hue difference between two colors.
Using the L*C*hº polar coordinate system to set up tolerancing allows a tolerance box to be rotated in orientation to the hue angle. This more closely matches human perception of color, which reduces the chance of disagreement between human observer and instrument readings or values.
Selecting a Tolerancing Method
If you can plot two colors on a color model, you can calculate the distance between them (the Delta) and select a range of acceptable color (the tolerance), similar to calculating the distance between two cities on a map.
DE* (CIELab Delta E)
Delta E is spherical tolerance that uses a single number to measure the difference between two colors (Figure 3). Consider you are taking a trip from point A to point B. There are different routes you could take – that’s like using L*a*b* or L*C*h°. Delta E* is a measure of the total distance between the two points “as the crow flies.”
Delta E tolerancing method.
FIGURE 3 Delta E tolerancing method.
DEcmc, DE94, DE2000
These three elliptical tolerances are progressions in the mathematics and algorithms used to calculate tolerances by measuring the distance between two colors (Figure 4). They are unique because the tolerances change in shape and size based on the color to better agree with human vision.
DEcmc, DE94, DE2000 tolerancing method.
FIGURE 4 DEcmc, DE94, DE2000 tolerancing method.
Aligning Your Tolerance with Visual Assessment
Since the goal is to provide visually pleasing color, most manufacturers want to choose a tolerancing method that most closely agrees with visual assessment. Your eyes will agree with a L*a*b* tolerance pass/fail assessment about 75% of the time. You will get about 85% agreement with L*C*H, and about 95% or better with CMC2:1. Using DE2000, you can reach about 98% visual agreement.
Sometimes customers specify which tolerancing method you must use to win their business, but we suggest looking at multiple metrics to ensure you’re producing good color. Just because you’re assessing pass/fail with one method doesn’t mean you can’t evaluate using them all, and it’s always a good idea to visually evaluate colors that fall close to the outside edge of the tolerance.
Setting an Achievable Tolerance
Using Color iQC, you can set tolerances and evaluate them using different metrics. The examples in this section illustrate the most important steps to set an achievable tolerance using actual color data from Color iQC.
Build a Tolerance Using Real Samples
If you are just getting started and have not set up tolerancing parameters, start by capturing both good and bad samples. Choose samples that your customer did and did not approve, then measure them and evaluate at least 10-20 samples to determine the proper limits.
For example, Figure 5 shows a plot from Color iQC using DEcmc (an elliptical tolerance) with several plotted samples. Those that plot with a green circle are within the current tolerance, and the red circles are outside the tolerance. Red circles that appear to be in the center but are failing are actually outside of the three-dimensional ellipse. If you look at the data on the right, you see these red circles are either too dark or too light. Graphically, it is easy to see which samples passed and which failed.
A plot using DEcmc (an elliptical tolerance) with several plotted samples.
FIGURE 5 A plot using DEcmc (an elliptical tolerance) with several plotted samples.
For this example, let’s assume samples 0001 and 0004 are visually acceptable and approved by the customer. Since they fall at 1.02 and 1.04, a tolerance of 1.0 is probably too small. Adjusting the tolerance to 1.05 will plot more samples with a green circle and a passing green checkmark.
Of course, it’s never this straightforward. It’s not unusual to have a customer accept a color with a higher DEcmc than a color they rejected. That’s why visual evaluation is also important. To set an achievable tolerance, find the DE of the lowest sample you want to fail, and make sure the tolerance is less so you don’t pass anything the customer will reject.
Make Your Tolerance Actionable
The difference between good and bad color is not a knife’s edge. For example, if your tolerance is 1.00, that doesn’t mean that .99 is great color and 1.01 is awful – visually they are essentially the same color. Setting a margin in Color iQC can help you take action before you get too close to the edge of the tolerance.
In the example in Figure 6, the default tolerance is 1.0 with a 10% margin. Colors that fall below 0.9 will pass with a green circle, colors that fall above 1.0 will fail with a red circle, and colors that fall between will appear with a yellow circle.
Setting a margin can help you take action before you get too close to the edge of the tolerance.
FIGURE 6 Setting a margin can help you take action before you get too close to the edge of the tolerance.
If you’re watching production trends and see the Delta E enter the yellow margin, you can stop and take action. The margin will also allow you to pass green circles with confidence because you know they are not close to failing. The amount of margin you set is up to you. It simply makes the tolerance a little tighter so you can visually evaluate anything that is in danger of not passing customer inspection.
Employ Multiple Metrics
While elliptical tolerances like Delta E or Delta E CMC are very good at showing you colors that pass and colors that fail, they don’t tell you how to fix issues. We recommend pairing L*a*b* or L*C*H* – whichever three-dimensional color space you are most comfortable with – to correct issues.
In Fig 6 we can see the second sample (0002) is in the margin, nearly failing, but why? And what should we be doing to correct this color before it starts to fail? When you look at DL*, Da* and Db* you can see that it is 2.45 units too dark, and nearly spot on for Da* and Db*. This gives us the direction we need to adjust this color to bring it back to passing the tolerance.
Using Color iQC makes it is easy to tolerance in both Delta L*a*b* (or L*C*H*) and Delta E. Figure 7 shows toler-ancing in Delta Ecmc with instructions to also test DL*, Da* and Db*. Although it appears to be plotting the same things, it is actually plotting two tolerances. For a sample to pass, it has to be inside the box AND inside the ellipse, three-dimensionally.
Tolerancing in Delta E CMC with instructions to also test DL*, Da* and Db*.
FIGURE 7 Tolerancing in Delta E CMC with instructions to also test DL*, Da* and Db*.
While the first two samples passed with DEcmc, they both fail when DL*, Da* and Db* are added. Leveraging this additional tolerancing method not only stops you from accepting a color that your customer will likely reject, it also shows why the color failed.
Conclusion
While Color iQC can take the guesswork out of tolerancing, it does not replace the need for human interaction. Be sure to fully understand the choices that are selected in the software and incorporate visual evaluation for the most effective tolerancing program.
For more information, visit www.xrite.com.
Recent Comments