Fall 2001 Vol. 3, No. 1

Most powder coatings have superior adhesion and mechanical properties compared with air-dried or baked waterborne and solution coatings. With the exception of those containing acrylic or silicone resins, powder coatings exhibit high impact resistance, direct and reverse, good flexibility and excellent adhesion as measured by the cross-hatch tape pull test. Attempts to determine a more quantitative measure of adhesion by a tensile pull-out test are usually unsuccessful with powder coatings because the adhesion of the coating to the substrate is greater than the adhesion of the probe to the coating. Therefore, tests other than those normally used need to be developed to discriminate between the adhesion and mechanical properties of powder coatings.

This paper reports on adhesion and related properties for three generic types of powder coatings at two variations in pigment-to-binder ratio (P/B), based on test equipment used in other industries for characterizing the adhesion of thin films to substrates. Test results are reported for the force required to separate the coating from the substrate using a computer-controlled knife edge using the Hesiometer and the force required for a spherical diamond stylus to progressively penetrate the coating and subsequently remove it from the substrate, utilizing the Stylometer.

Background

The relationship between the dynamic mechanical properties of coatings and physical properties, such as hardness, flexibility, mar resistance and others, has been studied and is generally well known (1). However the role of adhesion in influencing the properties of the coating on a substrate is less well known. Adhesion is not a fundamental property of a material, and measures of adhesion are rather subjective and a function of the test method used to make the measurements. However, most coating chemists will agree that a coating with "good" adhesion to the substrate will have superior mechanical properties and corrosion resistance and will outperform a coating with "poor" adhesion. Adhesion is usually measured by the cross-cut tape test (2) method B and sometimes by the tensile pull-off method (3).

For the vast majority of powder coatings, none of the coating in the cross-cut area is removed by the tape pull, and a perfect classification of 5B is obtained. That is, this test does not discriminate between different degrees of adhesion of powder coatings. In the case of the tensile pull-off test, where a probe is bonded to the surface of the coating with an epoxy adhesive, meaningful values of adhesion of the coating to the substrate are usually not obtained because the adhesive failure is between the probe and the coating surface. These tests are useful for conventional paints and distinguish between various levels of adhesion. However, in the case of powder coatings, the adhesion of the coating to the substrate is much greater, and these test methods cannot discriminate between various levels.

At the Coating ‘99 show in Dallas, the author visited the booth of the Quad Group Inc. (Spokane, WA), a supplier of instruments used to measure the adhesion of thin films, such as plated, vacuum-metallized and tribological coatings to various substrates. These instruments are computer-controlled, and it appeared that they might be able to measure the adhesion of powder coatings to a substrate, because the force required to cause the coating to be separated from the substrate could be accurately controlled and measured. A joint effort to determine if the Quad Group instruments might be valuable in distinguishing between the adhesion and mechanical properties of various types of powder coatings was agreed upon. The initial results of this study are reported in this paper.

The Experiment

A series of nine powder formulations was prepared for this study: three formulations each in polyester/TGIC, polyester/urethane and epoxy/polyester binders. The three formulations consisted of:

• a low pigment/binder (P/B) ratio (PVC = 4.45)
  
• a typical level of 60 parts per hundred parts of binder resins (PHR), pigment and filler, (PVC = 14.21), with barium sulfate as a filler
  
• the same PVC as the second formulation, but using calcium metasilicate as gloss-reducing filler.

The formulations are listed in Table 1. They were prepared in a twin-screw extruder using standard techniques, ground and screened through 140-mesh (105-micron) sieve and electrostatically applied at a thickness of 2.5 mils ±0.5 mil. Test panels were 0.035-inch cold-rolled steel, cured 10 minutes at 400°F.

Adhesion tests were performed on two instruments: the Hesiometer (4) and Stylometer (5). The Hesiometer is a software-driven, blade-cutting-type adhesion tester that can test the bond strength of coatings, bonding agents, paints, polymers, laminates, composites and other thick coatings. The operating principle is to establish a wedge-shaped splitting plane at a layer interface, project that opening forward and measure the energy required to separate the coating from the structure below it (Figure 1). Hookian materials, this condition of progressive splitting ahead of the blade, can be identified by the high amplification of the splitting noise made. For visco-elastic materials, such as some paints and polymers, interfacial splitting does not occur, and only the cutting energy can be evaluated. This is caused by the inability of this type of material to support the projection of a rigid member forward of the cutting tip due to plastic or inelastic deformation. In fact, when the interfacial splitting condition initiates, the "cutting" action stops, and the blade contact moves from the blade tip to a location on the sloping blade face above the tip. The coating is then "pried" from the surface. This prying action involves only "intrinsic adhesion energy" and does not include the energy consumed in deformation of the adherend, which is a major contributor to the "practical adhesion energy." This mechanism is clearly shown in Figure 2.

There are two levels of interest in using the Hesiometer: practical adhesion measurement and scientific studies of adhesion. This instrument has effective application to both, although this experiment only used the practical adhesion capability.

There are two different methods for addressing the blade to the sample:

• Cutting through the upper layers and to the interface of interest. This is employed in the practical adhesion method.
  
• Starting on an uncoated portion of the substrate surface and advancing toward the coating edge. This is used for scientific applications where the coating is shaped in a distinct pattern to eliminate blade edge spurious energy consumption.

In the first method, there is a typical rise in cutting force. When the interface is reached and the interfacial splitting begins, a drastic reduction in cutting force is displayed. In the second, the splitting plane (at the coating/substrate interface) is created abruptly on initial contact with the coating edge. There is a special combination of blade angle and blade hold-down force, which enhances the prospect of layer separation rather than blade-cutting. The optimum conditions are specific to each coating/substrate combination, but in general, similar settings can be used for related sample types. The instrument is capable of being reset to nearly identical conditions by reference to a precision-angle measuring goniometer attached to the blade structure. Thus, the desired test conditions can be duplicated even if long periods of time have elapsed between tests. Finally, it is possible to test each layer of a multilayered structure by careful adjustment of the blade angle and blade hold-down force.

Output of the Hesiometer consists of a plot of the cutting force, net normal force and acoustic output. As the blade cuts through the coating, the cutting force increases. When the blade contacts the substrate, an elastic "rigid beam" segment of the coating projects a crack forward of the blade tip with a simultaneous abrupt decrease in the cutting and net normal forces. Initiation of interfacial cracking is signaled by an abrupt increase in acoustic output. The crack propagation manifests itself as a cyclical buildup and reduction of both cutting and normal forces. Major bursts of acoustic energy are associated with the incremental advance of the crack. Figure 3 shows the Hesiometer plot typical of the formulations studied.

The Stylometer is a software-driven coating adhesion tester, generically identified as a scratch tester. A scratch tester introduces mechanical stress to a coated substrate and provides numeric measurements of coating adherence, or substrate cohesive strength. The basic principle of operation is to apply a controlled perpendicular force to a spherical diamond against a coating. The coating is moved at a constant rate of travel, and the force is increased at a constant rate of loading until the coating fails. In some few cases, the substrate fails or crushes. This condition indicates the limit of the operating range of the system, since the substrate is too weak to support a greater adherence. Throughout the test, the normal force, transverse force (diamond drag force), effective friction and stylus acoustic signal are displayed, recorded and plotted real-time. The primary purpose of the test is to identify the "critical force," or the force at which the coating fails. Certain "events" are identified so as to simplify the search for the critical force, but their magnitude and their location on the plot have no further use, at present, after the critical force has been identified.

The primary use of the Stylometer is in the measurement of relative adherence of thin films, particularly for the evaluation of ultra-adherent coatings, such as diamond, nitride or other forms of tribological or wear coatings. In these cases, there is virtually no other method capable of measuring adherence in the required range, which is above 90 Mpa (>13,000 psi).

Secondly, the test can be used for any thin-film application where a quick adherence measurement is required, because there is no sample preparation or substantial testing time involved. Lastly, the test may offer substantial information about friction coefficient, coating tensile strength, coating elastic properties, micro hardness and other physical properties, which should be of substantial interest to fundamental science applications. There is substantial current interest in the use of this tool and the measurement of micro-physical properties, and it is anticipated that techniques will progressively become available for the quantitative measurement of various physical properties. By employing constant load techniques, it is also possible to measure practical scratch resistance, surface durability and relative plastic deformation energy in visco-elastic materials.

Test Results

The energy required to progressively separate the various powder coatings from the substrate using the Hesiometer are listed in Table 2. In the case of formulation 1 of each of the chemistries, (the lowest P/B ratio), the hybrid and TGIC-polyester formulations show a significantly higher value than the urethane. It should be noted, however, that the urethane formulation used in this study is at the lower end of the cross-link density range, based on a relatively low OH number polyester resin. The higher P/B ratio formulations appear to show no consistent pattern compared with the lower P/B control. In the hybrid formulation, there is a slight decrease in adhesion energy in the filled system, while with the urethane formulations there is a significant decrease. The TGIC-based formulations show a slight decrease with one filler and an increase with the other.

Although the data for the samples did not show a fundamental trend, the method was able to reproducibly measure the practical adhesion energy of each sample. This practical adhesion energy measurement not only included the adhesion of the coating but also the adherend deformation energy and the frictional force between the blade and the coating. It appears a practical adhesion energy measurement could be useful in the day-to-day quality control of powder coated surfaces but may not provide a meaningful comparison of the relative adhesion of different coatings.

The Stylometer data for the various coatings is shown in Table 3. In the case of the control (low P/B ratio) coatings, the TGIC formulation shows significantly higher values for the critical force than the hybrid and urethane formulations, which exhibit close to the same values. For the filled formulations, the hybrid shows a slight increase over the control, while the TGIC and urethane formulations show a decrease. Scratch adhesion measurements again do not seem to show a consistent trend for the coatings tested. Since the Stylometer is purely a qualitative adhesion tool, it appears to be best suited to testing the relative scratch resistance of powder coatings but not the quantitative differences between the coatings as initially hypothesized.

A typical Stylometer plot for formulation is shown in Figure 4, and a photograph of a tested panel is illustrated in Figure 5.

Conclusions

It is difficult to determine the significance of the Hesiometer and Stylometer test results generated in this study for various powder coatings. However, these tests may be valuable in research studies to elucidate the effect of formulation variable on certain physical properties, such as the effect of increasing cross link density on adhesion. Since the Hesiometer can separate two coating layers as well as a coating from a substrate, it may have utility in evaluating the adhesion of various topcoats to a powder primer, for example.

It is possible that Hesiometer and Stylometer test results may also be useful as a quality-control tool to measure the consistency of properties for various lots of the same material over an extended period of time.