Elsevier

Wear

Volume 260, Issues 7–8, 7 April 2006, Pages 751-765
Wear

Effects of normal load on single-pass scratching of polymer surfaces

https://doi.org/10.1016/j.wear.2005.04.018Get rights and content

Abstract

The effects of normal load and the resulting scratch depth on scratch force profile, scratch hardness and the mechanisms of deformation and material removal for a number of industrially important polymers are studied. Upon scratching by a 30° angled conical tip, the mean tangential or scratch force is found to be linearly related to the normal load at lower speed (0.2 mm s−1); however, at higher scratching speed (2.0 mm s−1), there is a decrease in the slope of the scratch force versus normal load curve for all polymers. The phenomenon of stick-slip is severe at higher normal loads and scratch depths for the polymers that show ductile nature. The scratch hardness for softer polymers tends to decrease with normal load, whereas for harder polymers, scratch hardness increases for intermediate loads and tends to decrease at very high loads. The deformation mechanism, to a large extent, is insensitive to the imposed normal load or the depth of scratching; however, material removal and debris formation process depends upon the scratch depth.

Introduction

Scratching of materials is a convenient and now very popular means of characterizing surface hardness and dynamic scratch resistance of materials and coatings [1], [2], [3], [4]. This technique has also been used as a model for understanding wear phenomenon in materials. For example, a sharp indenter can be modelled as a single asperity or sharp debris in abrasive type of wear mechanism [5], [6].

The use of scratch test on polymers has been increasing in recent years to understand how polymers respond to single point damage [1], [2], [3], [4], [5], [6], [7], [8]. Probably this has been because of the reason that polymers are being used as bearing or decorative surfaces and the performance of these polymeric surfaces can be severely compromised due to scratching or large-scale wear. The asperities on the surface of a hard material, if they come in contact, can easily produce scratches on the polymer surface. The occasional presence of hard debris or third-body, as they are referred to, can also be a source of scratches on the polymer surfaces. Several researchers have used scratch test to compare resistance to scratching for different polymers and also to find a proper way of presenting and comparing scratch data (scratch hardness and scratch force) when they are related to other mechanical test data such as normal indentation hardness, wear rate, elastic modulus, yield strength and so on. It is now well understood that scratching of surfaces is a very complex process which becomes even more unpredictable when polymers are involved. Important variables that can change the damage response of polymers during scratching are strain (defined by the angle of attack), scratching velocity, normal load (this determines the depth of scratch), temperature or thermal effects, state of lubrication and the stiffness property of the scratching device. In order to rationalize various effects, Briscoe et al. [9] have proposed the idea of using scratching maps while presenting scratching data. Though qualitative in nature, scratching maps give an excellent way of identifying some broad deformation behaviour of polymers such as ductile, brittle or visco-plastic for a wide range of experimental parameters. Recent work has also revealed that effects of each variable could be much greater than just a change in the deformation characteristics. For example, a change in the variables such as load or speed can also change the profile of scratch force from a continuous type to intermittent stick-slip type [10]. Therefore, the measurement of the scratch hardness of polymers becomes much more difficult than originally thought [11]. The important point here is to understand why polymers behave differently during scratching as a result of some parametric changes and how we could use the scratch data of polymers to interpret their tribological and bulk mechanical performances.

In the current paper, we have investigated the responses of a range of industrially important polymers in scratching when the depth of scratch is varied by the use of different normal loads. The responses are recorded in terms of scratch force traces, micro-images using scanning electron microscope and residual scratch depths by using non-contact laser profilometer. Several aspects such as stick-slip motion and changes in the deformation mechanism, as a result of changing normal load, are also investigated.

Conventionally, normal indentation methods are used to characterize the hardness of materials including polymers. The key difference between the normal indentation method and scratching is that for the case of normal indentation, the indenter is uniformly supported over the contact area and the hardness is a near-static response of the material. Since the indentation force is supported by the material right beneath the indenter, a high compressive stress is built in the vicinity which causes mostly plastic deformation (probably also involving some volume change for the case of polymers) with limited strain. However, in the case of scratch hardness where an indenter tip is made to move across the surface of the materials, a significant amount of elastic/plastic deformation and brittle cracking can occur due to tangential scratch force which acts as tensile force on the rear half of the tip. The consideration of the support of the normal load by the material may be important for a proper definition of scratch hardness. For a perfectly plastic material, there is no recovery after deformation, and hence there is no support for the rear half of the indenter tip as it moves forward. In contrast, for perfectly elastic material, there can be full recovery of the material, and thus the tip will be supported in the rear half as well. For visco-elastic materials, the recovered material will offer only partial support for the indenter in the trailing zone. This support will depend upon the material's rate of recovery and the imposed scratch velocity. A fast recovering material coupled with slow scratch velocity may support the tip more compared to the case when the material is slow recovering and the scratch velocity is high. For a general case of visco-elastic plastic material, a typical response for polymers, scratch hardness may be given as [9]:Hs=WA=q4W(πd2)where A is the projected area supporting normal load, W is the normal load applied on the indenter, d is the recovered scratch width and q is a parameter which varies according to the response of the material. q is approximately 2 for rigid-plastic material, whereas q > 1 for visco-elastic plastic material. For the purpose of this paper, we have selected a value of q = 1.5. The choice of q = 1.5 is rather arbitrary keeping the value in between complete rigid-plastic and elastic cases as polymers in general are neither perfectly elastic nor perfectly rigid-plastic. Experimental measurement of q is difficult and no experimental or theoretical value is available in the literature for polymers. Also, it should be noted that Eq. (1) does not take into consideration the effect of tangential force on the load bearing surface of the tip.

Polymers are known to undergo large amount of recovery due to the visco-elastic visco-plastic nature [12]. Also, for some polymers, there is the possibility of change in volume under compressive stress and due to the crazing (fine internal cracking) phenomenon. Therefore, it is important to understand the process of recovery after scratching before making any assumption in the calculation of scratch hardness which is computed using scratch width data. During scratching, the material is pushed sideways (in the width direction of the scratch) and a large hydrostatic/compressive stress is generated beneath the tip (in the depth direction of the scratch) and in front where the material prow is created supporting the tangential scratch force. Because of the presence of the free surface on the side way (this means that the material can simultaneously relax while being deformed) and the fact that the magnitude of stresses are much smaller at the sides, the percentage of post-scratching recovery in the width direction after scratching is generally small. This is in contrast to the depth where extremely high hydrostatic/compressive stress during scratching could lead to high post-scratching recovery, and thus considerable reduction in the scratch depth measured after scratching. Our previous work [13] on the scratching at very small load (10 mN) on similar thermoplastics has shown that these polymers can show approximately between 50 and 60% reduction in the depth after scratching due to recovery. PMMA, in particular, shows very high recovery; 56.8% [13] when scratched by conical tip and close to 75% [14] when scratched by Berkovich indenter. If we consider the actual depth of penetration at the time of scratching and calculate the real-time width of the scratch for these polymers by using the geometry of the conical tip, we find that the width measured after the scratching event (recovered width) is slightly more than the width during scratching. Thus, very low load scratching clearly indicates that the depth of the scratch decreases after scratching, whereas there is a possibility of some enlargement of the width of the scratch as a result of scratch depth recovery. It would require further investigation, which could be a research in its own rights, to firmly establish the relationship between actual and recovered scratch widths and depths for polymers. Therefore, due to the lack of actual scratch depth data, in this paper, we will use the scratch width measured after scratching for the computation of unrecovered scratch depth and scratch hardness.

Section snippets

Materials

A total of seven polymers were tested. The polymers, supplied by Goodfellows Inc., UK, were used. The selection was mostly based on their relative importance in industrial usages; however, we also ensured some examples from the two major classes of polymers viz amorphous and semi-crystalline. Table 1 lists the polymers used in this study along with some typical ranges of their mechanical properties as provided by the supplier. The polymers came in extruded bar form.

Scratch test apparatus

A simple lab-designed scratch

Scratch force

Fig. 2(a) and (b) give mean scratch force as a function of the normal load for all seven polymers used in this study for two sliding speeds. The normal load was varied between 0.5 and 10 N, whereas the sliding speeds used were 0.2 and 2 mm s−1. The scratch force data show very similar values regardless of the polymer and some differences in scratch force at high normal loads are also minimal. There is a linear relationship between scratch force and the applied load for 0.2 mm s−1 scratch speed

Scratch force

In the scratch test, the quantity that can be directly measured during and after the experiment is the scratch force and the recovered scratch depth and width. The scratch force is the resistance force imposed by the material on the tip movement. In terms of energy dissipation, the main components of scratch force are ploughing, frictional work between the indenter and the material, and the elastic hysteresis loss (applicable to visco-elastic solids). It was found in this study that the scratch

Conclusions

Scratch tests were conducted by a 30° included angle conical tip on amorphous and semi-crystalline thermoplastics with different bulk mechanical properties. The normal load was varied from 0.5 to 10 N. Scratch force and scratch hardness were computed for all polymers in order to study their trends with increasing normal load and the resulting depth of scratch. Further, scanning electron microscopic study was carried out on the scratched polymer surfaces to investigate the deformation and

Acknowledgements

Authors would like to acknowledge the financial help provided by the National University of Singapore (Research Grant No. R-260-000-132-112) to purchase polymers. Help with the non-contact profile measurement on scratches provided by Mr. Thomas Koh of Crest Technology Pte Ltd. (Singapore) is also gratefully acknowledged.

References (21)

There are more references available in the full text version of this article.

Cited by (60)

  • Influence of sliding velocity on microscratch response of carbohydrate polymers by Berkovich indenter

    2022, Polymer Testing
    Citation Excerpt :

    Microscratch tests, which can simulate the process of micro-cutting and mechanical polishing, have many advantages such as high reliability, simple operation, and easy preparation of the sample, has become one of the most popular technologies for evaluating mechanical properties such as fracture toughness [1–5], scratch hardness [6,7], dynamic friction coefficient [8–10], wear resistance [11–13], scratch resistance [14], and bond strength between coating and substrate [15] of various kinds of materials including metals [16–18], ceramics [19], glasses [20–22], polymers [23–31], and coatings [32].

  • Scratch and multi-pass scratch behavior of poly (methyl methacrylate) (PMMA)

    2022, International Journal of Polymer Analysis and Characterization
View all citing articles on Scopus
View full text