Scratch and mar resistance of filled polypropylene materials.
Significant cost savings could result from replacing ABS with PP-PMF for interior automotive components. However, PP-PMF's susceptibility for scratch and mar damage hinders its use for these applications. Although scratch and mar prevention during shipping/handling/assembly can be managed by the OEM, there can be no such control once the parts enter customer service. The scratchability of PP-PMF is not only a warranty issue, but a long-term customer satisfaction concern as well. As a result, the mineral-filled PP parts are usually top-coated, thus eliminating the cost savings over ABS. The purpose of this study is to evaluate the effect of different mineral fillers on the scratch and mar resistance of PP-PMF and to find an appropriate test method to quantify such data.
Many factors, such as filler type, additives, lubricant, impact modifier, polymer type, and surface morphology, affect the tribological properties of plastics. Typically, three methods are used to improve the tribological properties of a given plastic material: 1) modifying the polymer molecular structure such as its crystallinity, 2) blending polymers, and 3) producing polymer composites with various fillers and additives. However, without an appropriate test method to quantitatively evaluate the degree of scratch and mar, it is difficult to conduct material development work and to gauge the degree/extent of improvement.
There are several problems in evaluating scratch and mar for plastic materials:
1. Plastics have viscoelastic properties and the stresses in the plastics may relax during loading.
2. Plastics have considerable elastic recovery upon removal of applied stress.
3. Plastics may change structure during material flow. This change in structure can also change the rheology and mechanical properties of plastics.
As a result, the scratch performance of plastic materials is dependent on the strain, loading rate, temperature, and environment. At this time, scratch and mar is not so well defined as a physical property. Very few tools are available to produce a consistent, comparable scratch or mar for quantitative evaluation. Scratch and mar, as an expression, is used to summarize the many ways of mechanically damaging a surface with a device. It was found that the scratch and mar performance of plastic materials is dependent on the geometry of this device (1).
The scope of this work is to find a methodology to evaluate the scratch and mar resistance of PP and to investigate their mechanisms. Generally, scratch tests are related to cuts, scratches and to some extent, the wear resistance of materials (2). Wear is defined as displacement of material from the surface during relative motion against hard particles. The same factors that affect wear resistance should affect scratch resistance. Since it is believed that wear resistance is closely related to the hardness of the polymer surface, hardness measurements were used to quantify the scratch and mar resistance of plastics (1, 3). Hardness tests are used mainly as a simple, rapid, nondestructive test for quality control in production, and as a measure of mechanical properties affected by changes in chemical composition, microstructure and aging. In standard indentation hardness tests (such as Rockwell), the direction of the load is normal to the surface. However, a tangential force is applied in the presence of a normal load for scratch hardness test. In this paper, both indentation and scratch hardness tests were evaluated; they are: Rockwell hardness, Durometer hardness, microhardness, and scratch hardness. Moreover, the scratch and mar performance is believed to be affected by the stress whitening ability of the material. Scratch is more visible in the material that is more susceptible for stress whitening. In this study, the effects of stress whitening on the scratch and mar resistance of PP materials were evaluated.
Samples of PP-PMF compounds with typical mineral fillers were obtained from ATC Inc. The median particle size is 3 [[micro]meter] for the talc and 8 [[micro]meter] for the wollastonite. The Mohs hardness of the minerals is 1.0 for the talc and 4.5 for the wollastonite. The plaques (101.6 x 127.0 x 3.2 mm) used for the scratch test were injection molded. The surfaces of the plaques were smooth on one side and grained on the other. Medium graphite color concentrate was added at 4 wt%.
For comparison between PP and ABS, two grades of ABS were used in the study. The ABS plaques were smooth on both sides and were the same color as the PP plaques. Detailed descriptions of these materials are listed in Table 1.
Scratch Test and Measurement
The scratch test is based on the Resistance to Scratch Test (Ford Lab Test Method BN 108-13). A simple machine is used to create scratches that simulate customer usage. This apparatus consists of a movable platform connected to five beams (250 mm long). A scratch pin is attached to one end of each beam. On the tip of each pin is a highly polished hardened steel ball (1.0 [+ or -] 0.1 mm diameter). Each pin is loaded with a different weight to exert a force of 7N, 6N, 3N, 1N, and 0.6N respectively, on the surface of a test plaque. The beams are driven by compressed air to draw the pins across the polymer surface to generate the scratches. Sliding velocity of approximately 100 mm/s is used. All tests are performed at room temperature on test plaques conditioned at 25 [degrees] C for more than 24 h prior to testing. Although the FLTM requires that only grained surfaces are to be evaluated, both the grained and smooth surfaces of the specimens were tested for this study.
Table 1. Materials Used in This Study. Code Materials A1 Unfilled PP copolymer A2 PP copolymer with 13% talc, 5% impact modifier A3 PP copolymer with 6.5% talc, 5% impact modifier, and 6.5% wollastonite A4 PP copolymer with 13% wollastonite, 5% impact modifier A5 PP copolymer with 13% wollastonite, 5% impact modifier, and 0.5% lubricant(*) D1 ABS, Dow Magnum[TM] 541 (general purpose) D2 ABS, Dow Magnum[TM] 344 HP (high heat) * The lubricant used was obtained from Dow Coming Co.
After the plaques are scratched, they are evaluated by an image analyzer with a Sony XC-77CE monochrome camera, Seescan Custom II Lighting Unit with ND8 filter and the "Scratch Assessment System, Version 1" software. The camera objective is positioned at an angle of 90 [degrees] and a distance of 225 mm from the scratch. The objective then registers a portion of the scratch about 60 mm long. A total of 250 views are taken perpendicular to each individual scratch line at predetermined intervals. An electronic signal for each scratch line is integrated and recorded. Finally, the percentage of the visible marking of the scratch is produced and reported as scratch visibility.
The width and depth of the scratch were also measured with a laser profilometer (UBM Microfocus Measurement, MRB Research Corp.). The software used was UBSOFT System 2020-2A Optical Interface. Resolution was set at 400 pixels/min. All the measurements were done on the smooth surfaces of the plaques. The roughness on the bottom of the scratch was also measured and calculated with the profilometer. In addition, the scratches were examined using an optical microscope.
Fisher Hardness Tester H 100 was used to measure the microhardness of specimens in this study. This tester used the Vickers indentor and measured the hardness under load. An operating load of 0.1N was selected for this test. The load increased continuously from 0.04 to 0.1N with 60 steps. The loading cycle was 60 s. The measured hardness number was basically defined as the quotient of the applied load divided by the indentation surface area. The indentation surface area was derived from the measurement of the indentation depth and the well-defined size and shape of the Vickers indentor. The indentor displacement under load is measured with a resolution of 2 nm. Each sample was tested three times and the values shown are the averages of three measurements. This technique measures the hardness while the indentor is under load. In this way, we can avoid the uncertainty associated with the viscoelastic recovery of the contact that occurs during unloading. Usually, the resulting residual indentation after unloading has the appearance of an indentation made with a larger indentor. Because this test was very sensitive to surface roughness, only the smooth surface of the plaque was tested. For the Rockwell and Durometer Hardness tests, ASTM D 785 and ASTM D 2240 (type D) methods were followed, respectively.
Stress Whitening Test
A Gardner dart impact tester (ASTM D 3029) was used to perform the impact tests. The impact force applied on the samples was 1.46 kJ. The test used to measure the amount of stress whitening was according to Ford Laboratory Test Method (FLTM) BI 109-03. It was done by measuring the amount of color change between the impacted and non-impacted areas with a spectrophotometer (Macbeth Color-Eye 3000).
The scratch visibility of various materials was measured by the image analyzer. Results from the various loads are shown in Fig. 1 for grained surfaces and Fig. 2 for smooth surfaces. Unfilled PP (A1) showed the least amount of scratch visibility on both surfaces at all loads. Because of the limit of sensitivity, results were indistinguishable on the smooth surfaces for the loads of 6N and below. PP filled with 13% talc (A2) showed the most scratch visibility. Scratch visibility decreased as the content of wollastonite in the PP increased (A3, A4). Scratch visibility was slightly lower with the addition of 0.5% lubricant. This trend was more evident on the grained surfaces. For the ABS specimens, scratch visibility was measured on the smooth surfaces only. The scratch visibility of both ABS materials fell between that of the talc-filled PP (A2) and the wollastonite-filled PP (A3).
The depth and the width of the scratch were measured by the laser profilometer. A typical surface profile is shown in Fig. 3. Ridges of deformed material were produced on both sides of the scratch (groove). Extensive cracking was observed at the bottom of the scratches. Two types of measurements were calculated for the scratch depth and the scratch width. The "apparent depth" is the distance between the peak of the ridge and the bottom of the groove. The "true depth" is the distance between the plaque surface and the bottom of the groove. Similarly, the "apparent width" is the distance between the peaks of the ridges and the "true width" is the width of the groove on the plaque surface.
Scratch dimensions for each material are plotted for the 6 and 7N loads [ILLUSTRATION FOR FIGURES 4 AND 5 OMITTED]. As can be seen in Fig. 4, the unfilled PP (A1) has the smallest depth (both apparent and true depth) for both loads plotted. On the other hand, talc-filled PP (A2) has the largest scratch depth for both loads. The addition of wollastonite (A3, A4) reduced the scratch depth significantly. No distinctive improvement was observed for scratch depth with the addition of lubricant (A5). One of the ABS specimens (D1) has scratch depths comparable to the wollastonite-filled PP. The scratch depth of the other ABS specimen (D2) fell between the talc-filled (A2) and the wollastonite-filled (A3) PP. Both the apparent scratch depth and the true depth have the same sensitivity to material variation.
A different trend was observed for scratch widths [ILLUSTRATION FOR FIGURE 5 OMITTED]. Random differences were observed between the scratch widths (both apparent and true) of the various materials. This is due to the geometry of the indentor. Under 7N load, the typical scratch width is 250 [[micro]meter], and depth is 15.6 [[micro]meter]. Therefore, on a shallow scratch like this, the edges of the scratch show much higher elastic recovery than the bottom of the scratch, and thereby obscure the width measurement. Hence, the scratch depth can give a higher sensitivity for scratch dimension measurement than the scratch width. The scratch that resulted from the 1 and 0.6N loads could not be detected for any of the specimens. These results correlated with the data obtained by the image analyzer, as can be seen in Fig. 2.
In addition to scratch dimensions, roughness at the bottom of the scratch was also measured and plotted in Fig. 6. Similar to scratch depth, the untilled PP (A1) had the least amount of roughness in the scratches among all materials for the 7N load scratch. On the other hand, the highest amount of scratch roughness was found in talc-filled PP (A2). Additional wollastonite (A3, A4) can reduce the scratch roughness significantly. Some improvement in roughness was seen in PP with lubricant (A5). This same trend was found in scratches made by the 6N load. This improvement in scratch roughness by additional wollastonite can also be shown by the undetectable scratch made by the 3N load in both the A4 and A5 specimens. Similar to scratch visibility, both ABS specimens had roughness between talc-filled and wollastonite-filled PP. The scratch roughness of the D2 specimen was higher than that of the D1 specimen. The reason for this increase in roughness in D1 is unknown. For both ABS specimens, the scratch roughness of the 1N load could be detected.
The results from both the image analyzer and the laser profilometer are plotted in Figs. 7 and 8. As shown in Figs. 7a and 7b, the maximum depth of the scratch correlated well with the scratch roughness for both 7 and 6N loads respectively. The scratch roughness increased as the scratch depth increased. In addition, the maximum scratch depth was also plotted against the scratch visibility for scratches made by 7N load as shown in Fig. 8. Other than one ABS specimen (D1), good correlation was found between the visibility and depth of the scratch. The scratch visibility increased as the scratch depth increased. The reason for this deviation in the ABS specimen is unknown.
The results from the microhardness test were plotted in Fig. 9. Surprisingly, the hardness of the wollastonite filled PP (A3, A4) did not increase as expected with additional harder filler. All PP specimens with or without fillers had the same microhardness. It is possible that the additional hardness of wollastonite was offset by the softness of additional impact modifier. Both ABS specimens had about the same microhardness, which was higher than those of the PP specimens. The Rockwell hardness results are also shown in the same Figure. Unfilled PP had the highest Rockwell hardness value among all PP specimens. All the filled PP had similar Rockwell hardness regardless of the filler. However, because of the additional impact modifier, samples A2-A5 showed lower Rockwell hardness. Similar to the microhardness, both ABS specimens had about the same Rockwell hardness, which was higher than those of the PP specimens. Durometer hardness values are also shown in the same Figure. All PP specimens had similar Durometer hardness values. ABS specimens had higher Durometer hardness than the PP specimens. However, the difference between them became smaller. The difference in these hardness results was due to the difference in indentor geometry for these hardness tests. In addition, the hardness measurements are dependent on the viscoelastic behavior of the material; there is no simple conversion of the results obtained with testers of different ranges or by different methods. As shown in Fig. 9, none of the hardness measurements can accurately show the difference between the scratch performance of the various PP materials.
Stress Whitening Test
The results from the stress whitening test are shown in Fig. 10. As shown in the Figure, unfilled PP had the lowest amount of stress whitening. The amount of stress whitening increased significantly as talc was added to PP (A2). Additional wollastonite (A3, A4) can further increase the amount of whitening in PP. However, the difference in whitening was not significant among specimens with wollastonite and lubricant (A3, A4, A5). Both ABS specimens (D1, D2) had stress whitening values between filled and unfilled PP.
Figures 11a to 11e show the dark field optical micrographs of the 7N load scratches in specimens A1 (11a), A2 (11b), A3 (11c), A4 (11d), and A5 (11e). Both ABS specimens had the same morphology; the only picture taken from D1 is shown in Fig. 11f. For unfilled PP (A1), the scratch was shallow with some fine ripple marks visible at the bottom of the scratch. There was no apparent brittle cracking. The boundary between the scratch and plaque surface was not clear. This suggests that the ridges formed at the sides of the scratch are not as defined as they are in the other specimens. Material deformed mainly in an elastic manner with little plastic flow. The deformation mechanism is of a ductile plowing nature. However, it has been found that the indentor geometry has direct influence on the type of damage mode during scratching (1). Significant ripples were found in the scratch of PP with 13% talc (A2), as shown in Fig. 11b. The ripples formed chevron-shaped marks in the scratch bottom. It is believed that the marks were created by the continuous material flow around the indentor to form a uniform groove. The scratch had a sharp appearance with distinct boundaries between the scratch and unscratched area. This suggests that the PP deformed mainly in a plastic manner. The deformation mechanism was also of a ductile plowing nature. A similar morphology was found in the rest of the filled PP specimens. However, the amount of rippling seemed to decrease as the amount of wollastonite increased in PP. Instead of sharp chevron marks, the ripples formed parabolic shapes. The distance between successive parabolic or chevron marks seemed to increase as the amount of wollastonite increased (A3, A4). This may explain the decrease in roughness and whitening in the specimens with more wollastonite, as shown in Figs. 2 and 6. The same trend was also seen in the specimen with lubricant (A5). For ABS specimens (D1, D2), very distinct boundaries were found between the ridge and the unscratched area and also between the ridge and the groove. This suggests that defined ridges were formed at both sides of the scratch. Instead of the large ripples observed in the scratch of filled PP, very fine ripples were observed in the scratch of ABS specimens. In addition to the scratch, the fillers were also observed on the surfaces of the specimens as shown in the photos.
Hardness defines the difficulty with which a material can be indented, drilled, or abraded. It is a complex property related to the mechanical properties of a material such as modulus, strength, elasticity, plasticity, and the test method. The relationship to mechanical properties is not usually straightforward, but there is a tendency for high modulus and strength to correlate with higher hardness for most materials. The hardness can be measured by static penetration of material surface with an indentor exerting a known force. For plastics, this involves the formation of a permanent surface impression. The material under the indentor contains a zone of plastic deformation surrounded by a larger zone of elastic deformation. As a result, hardness should be directly related to the critical stresses required for plastic deformation. It is computed by dividing the peak contact load W by the projected area of impression A. In this study, several indentation hardness methods were used to investigate the scratch resistance of plastic materials. However, as shown in Fig. 9, these indentation hardness measurements cannot tell the difference between various materials and no correlation was found between the hardness and the scratch and mar resistance of PP. One of the reasons for this is that the results of hardness testing are influenced by many factors, e.g., shape and surface condition of the indentor, specimen surface, and material homogeneity. It can also be explained by the difference between static (indentation) and dynamic (scratch) tests. The indentation hardness tests are static and the results obtained by these test methods should not be considered as a measure of the abrasion or wear resistance of the plastic materials in question. Regarding scratch hardness, the sliding penetration depth can be 2 times the static penetration depth for some plastic materials. This is because the normal load is supported mainly on the front face of the indentor during scratching and there is no recovery in the track width for plastically deforming material. In addition, dynamic friction plays a role during the scratch test. A significant reduction in scratch width was found in materials with an internal lubricant (1).
The scratch hardness for various materials was measured in this study. The sliding contact area, [A.sub.s], is approximated by
[A.sub.s] = [Pi][(w).sup.2]/4 (1)
where w is the measured scratch width. The scratch hardness, [H.sub.w], can be shown as following
[H.sub.w] = x 4L/[Pi][(w).sup.2] (2)
where L is the load on the intentor, x = 2 for a plastic deformation, and 1 [less than] x [less than] 2 for a viscoelastic-plastic deformation. The value of x will be dependent on the viscoelasticity of the material, the sliding velocity, and the indentor geometry. In this study, the same sliding velocity and indentor were used in the test. We assumed x is equal to 1. The calculated results are plotted in Fig. 12 for scratches created by a 7N load. As with indentation hardness, no significant difference was found between the scratch hardness of various materials and no correlation was found between the scratch hardness and visibility, as we can see in the Figure. This result is also expected, because the scratch width has shown to be insensitive to the difference in material types [ILLUSTRATION FOR FIGURE 5 OMITTED].
We also calculate the scratch hardness with scratch depth, d, assuming there was negligible elastic deformation.
[H.sub.w] = L/[Pi](2rd - [d.sup.2]) (3)
where r is the radius of the intentor.
As shown in Fig. 13, good correlation was found between the scratch visibility and scratch hardness. Materials with higher scratch hardness have lower scratch visibility. As shown in Fig. 8, scratch visibility is related to the scratch depth. Deep scratches usually have higher visibility. As a result, scratch visibility has good correlation with scratch hardness, which is closely related to scratch depth.
Studies have been done to investigate the mechanisms for whitening during scratching (4, 5). In addition to the craze or crack in the matrix, voids are generated in the matrix/filler interface. Usually, a deeper scratch also produces more craze and crack than a shallow scratch. This increase in cracks can be quantified by measuring the roughness of the scratch. As shown in Figs. 7 and 8, scratch roughness is related to the scratch depth, and in turn related to the scratch visibility. The same conclusion can be reached by studying the optical micrographs of the scratches for various materials. As shown in Fig. 11, the number of ripples increased as the fillers were added into PP. Additional fillers increased the chance for matrix/filler debonding and crazing, which can create whitening. It is also suspected that some of the exposed white fillers will contribute to the whitening of the scratch. However, the additional wollastonite seems to be effective in reducing the scratch depth and thus the amount of damage to the specimen surface. In addition, wollastonite can increase the distance between successive ripples (with less damage) in the wollastonite-filled materials, as shown in the photograph. These are probably due to the increase in both hardness and particle size of wollastonite. For the same reason, the impact strength of this material decreases.
This study attempts to quantify the extent of scratches by both the image analyzer (specified by the FLTM) and the laser profilometer. The laser profilometer seems to be a suitable method to evaluate the scratch performance of the materials on a smooth surface. A predominant issue regarding the visibility evaluation of scratch and mar is that it can be affected by a number of factors. The color, gloss, and grain of the test specimens can directly affect the evaluation results. Additionally, the molding and post-molding parameters can significantly alter the surface morphology and properties. Currently, the image analyzer is a popular technique in quantifying the scratch and mar performance of materials. However, as can be seen in Figs. 1 and 2, the visibility of the scratch is dependent on the surface texture of the test specimen. The image analyzer has less sensitivity on the smooth surface than on the grained surface. This is due to the indentors' "bouncing" on and off the grained surface and "gliding" across the smooth surface. This "bouncing" exerts additional impact force at the "landing points." This phenomenon causes additional damage to the surface and creates more contrast for the image analyzer method. The image analyzer is also unable to detect the lower-load scratches on the smooth surface [ILLUSTRATION FOR FIGURE 2 OMITTED]. The scratch depth measured by the laser profilometer correlates with the image analyzer results for higher loads on the smooth surface [ILLUSTRATION FOR FIGURE 8 OMITTED]. Therefore, the laser profilometer seems to be a better technique to evaluate various materials without being affected by the uneven scratches resulting from grained surface texture.
Similarly, the laser profilometer is able to evaluate the mar damage to surface. In this study, the amount of damage decreased as the load decreased for all materials as shown in Fig. 1. Eventually, the damage changed from a white line (scratch) to a shiny line (mar) at some lower loads. The reason for mar is that the surface of the specimen has been plastically deformed and the macromolecules on the surface are aligned. However, the scratch was not deep enough to cause cracking and whitening, and some of the deformed material was able to recover elastically. This change in surface morphology altered the light reflection direction and caused a shiny mar on the surface. Usually the mar can only be seen at a certain sight angle. The laser profilometer can pick up the mar damage (as shown in [ILLUSTRATION FOR FIGURE 6 OMITTED]), whereas the image analyzer may not "see" the mar under the specified test conditions (as shown in [ILLUSTRATION FOR FIGURE 2 OMITTED]).
Stress whitening was used to indicate the ability for the material to craze or debond between the matrix and talc. The craze or debonding can change the refractive index that will result in whitening. As we expected, additional talc in PP increases the stress whitening through crazing and debonding. Slightly higher stress whitening was found in the wollastonite-filled PP. This is due to the large particle size of wollastonite. Usually, large filler particles have larger and weaker filler/matrix interfaces for each individual filler and increase the tendency for more stress whitening. However, because the scratch whitening is a combination of crazing, alignment, and morphology change, stress whitening alone does not correlate with the scratch whitening observed in the specimens.
* Good correlation was found between scratch visibility and scratch hardness based on scratch depth measurements. PP-PMF material with higher scratch hardness will have lower scratch visibility. This is because material with higher scratch hardness is more resistant to scratch, resulting in less damage to the material. The laser profilometer seems to be a good method to measure the scratch depth and scratch hardness.
* Consistent results were obtained from the three normal indentation hardness methods. However, hardness values do not correlate to the variation of scratch resistance in materials. No correlation was found between indentation hardness and the scratch visibility. This may be because a) different indentors are used for scratch and indentation hardness tests, and b) the scratch test is dynamic and the indentation hardness tests are static.
* The PP-PMF specimen with 13 wt% talc (A2) is the worst in terms of scratch visibility. Wollastonite was found to be able to reduce the scratch depth and visibility. However, the lubricant used in this study slightly reduced scratch depth and whitening.
* Because the scratch whitening is a combination of crazing, alignment and morphology change, stress whitening alone does not correlate with the scratch whitening observed in the specimens.
L: the applied normal load.
[H.sub.w]: scratch hardness.
d: scratch depth.
[A.sub.s]: the sliding contact area.
w: the scratch width.
r: radius of indentor.
1. B. J. Briscoe, P. D. Evans, S. K. Biswas, and S. K. Sinha, Tribology Inter., 29, 93 (1996).
2. Y. Yamaguchi, in Tribology of Plastic Materials, Y. Yamaguchi, ed., Elsevier, New York (1990).
3. E. K. L. Lau, K. Swain, and S. Srinivasan, in Third International Conference, TPOs in Automotive '96, Novi, Mich. October (1996).
4. R. Kody and D. Martin, Polym. Eng. Sci., 36, 298 (1996).
5. S. Xavier, J. M. Schultz, and K. Friedrich, J. Mater. Sci., 25, 2411 (1990).
|Printer friendly Cite/link Email Feedback|
|Author:||Chu, J.; Rumao, L.; Coleman, B.|
|Publication:||Polymer Engineering and Science|
|Date:||Nov 1, 1998|
|Previous Article:||Shear rate effect on the resistivity of HIPS/LLDPE/carbon black extrudates.|
|Next Article:||Melt strength of polypropylene: its relevance to thermoforming.|