Elsevier

Wear

Volume 256, Issues 11–12, June 2004, Pages 1037-1049
Wear

The influence of particle rotation on the solid particle erosion rate of metals

https://doi.org/10.1016/S0043-1648(03)00536-2Get rights and content

Abstract

It has long been recognised that particle spin may have a significant effect on the impact erosion rate, particularly of ductile metals. However, no work has previously been carried out to quantify this effect, partly due to the practical difficulty of measuring the magnitude of the rotational speed. Particle spin is a feature of the centrifugal accelerator erosion tester. In this tester it has proved possible to examine the effect on erosion of particle spin direction by varying the target orientation. The results indicated a strong effect of the spin direction on erosion rate at low impact angles when the targets were impacted by angular particles. A quantitative model was developed to explain the effect of particle spin direction on the observed differences. The model is a modification of the Finnie–Bitter model [Wear 3 (1960) 87; Wear 6 (1963) 5; Wear 6 (1963) 160], and is the first to explicitly incorporate the effect of rotating particles on the subsequent erosion rate when the particles impact a metal target. The model supposes that the effective impact velocity, the contact velocity between the particle and the target, is altered due to spin of the particles. The predictions of the model were validated through actual measurement of particle rotational speed by high-speed photographic techniques; the first such measurements. Experimental erosion results conformed to the predictions of the model. An effect of particle spin on the peak erosion rate is also predicted by the model and confirmed by the experimental results.

Introduction

Few studies have considered the possible interaction between a spinning particle and a target during erosion, and yet, an understanding that particle rotation during impact may have a significant influence on erosion has been recognised for many years [1]. The importance of particle rotation should not be dismissed. However, practical considerations have effectively prevented the experimental quantification of such effects. Irregular-shaped particles accelerated in a fluid tend to rotate as a result of unbalanced forces acting on them. Only homogenous spherical particles will not rotate under such circumstances. It is possible that rotation of particles may affect the erosion rate and even the erosion mechanism when the particles strike a target surface. The first reported discussion of the potential significance of particle rotation was by Finnie [1]. He presented theoretical results, based on a hypothetical distribution of rotational velocities, that indicated a pronounced effect of particle rotation on erosion rate.

Work by Hutchings [2], [3] was highly influential in suggesting that particle rotation may even affect the actual mechanism of material removal during erosion of ductile metals. Although his model does not explicitly refer to the effect of rotating particles striking a surface, it does consider the effect of a particle rotating forwards (top-spin) or backwards (back-spin) after impact. Hutchings proposed three possible kinds of material removal mechanism [2], [3] during the erosion of ductile metals, and suggested that particle rotation occurring after impact had a significant effect on the mechanism in operation. It was suggested that an angular-type particle striking a surface may generate ‘Type I’ cutting or ‘Type II’ cutting (micro-machining) depending on the rotational direction of the particle following impact. For a spherical particle, it is generally assumed that ploughing deformation is more likely. It was shown that these three different erosion mechanisms might lead to very different erosion rates [2], [3]. More recently, Papini and Spelt [4], [5] developed a rigid–plastic model of impact and used it to predict crater volume in the case of symmetric angular particles. The model is shown to conform to Hutchings [2], [3] experimental data for square steel plates impacting smooth steel surfaces.

Indirect evidence that particle spin during flight could have significant effects on the subsequent erosion rate was presented by Burnett et al. [6]. Studies using the centrifugal accelerator erosion tester did not show the commonly observed peak in the erosion rate versus impact angle curve. Instead, erosion rate continued to rise with decreasing impact angle to the lowest angle at which it was measured (8°). With respect to this anomaly, Burnett surmised that particle rotation induced by the laboratory tester increased the efficiency of cutting at low angles. It was demonstrated that under otherwise identical test conditions, target orientation with respect to the particle velocity vector strongly affected erosion rate. This experimental result led to an investigation of particle spin in the centrifugal accelerator erosion tester. Consideration of particle dynamics in this tester indicated that particles emerging from the acceleration tubes always exhibited spin in the same direction as a result of the frictional interaction between the particles and the walls of the acceleration tube [7]. Thus, particles striking the target did so with top-spin or back-spin depending on the orientation (clockwise or anti-clockwise) of the targets. The clockwise and anti-clockwise terminologies refer to target rotations with respect to the 90° impact position. Hutchings [2], [3] suggested that particle top-spin after impact leads to “Type I” cutting and that back-spin after impact results in “Type II” cutting (micro-machining). These different types of cutting mechanism result in different erosion rates. An alternative (and perhaps simpler) way of making sense of the results is to consider that the speed and direction of rotation of the particles will change the effective impact velocity of the particles (and impact angle) during point contact with the target surface.

The present paper describes work in which some of these issues are addressed. Further erosion studies have been carried out in the centrifugal accelerator type erosion tester, in which the orientation of the target is varied in order to control the spin direction. This results in the targets being impacted by either top-spinning or back-spinning particles. Both angular and spherical particles have been investigated because previous work [7] demonstrated that in this tester the particle spin characteristics depend on particle shape. Furthermore, for the first time, experimental measurements of particle rotation have been made. This was only possible due to the recent availability of advanced measurement technology. In addition, a quantitative model to predict erosion rate was developed based on the Finnie–Bitter erosion model [8], [9], [10] modified to include the effects of particle spin. The particle rotation measurements allowed a comparison of the predictions of the model with experimental results.

Section snippets

Experimental procedure for erosion testing

The methodology of erosion tests in the centrifugal accelerator type erosion tester has been detailed previously, e.g. [6], [11]. In the present work, each test was carried out under identical erosion test conditions except for the variables under consideration, so that comparable test results could be obtained.

Two types of abrasive particle were chosen for the present studies, as previous work had shown that the shape of a particle affected its frictional interaction with the wall of the

Erosion test results and discussion

The erosion test results for the EN-24 steel targets, showing the variation of erosion rate with impact angle, for angular and spherical particles are presented in Table 1 and Fig. 2, Fig. 3. It may be observed that angular particles (Fig. 2) generate a far higher erosion rate than spherical particles (Fig. 3) under the same test conditions. The low erosion rates due to the spherical glass beads meant that perceived trends were rather inconclusive.

The large difference in erosion rate between

Measurement of particle velocity and particle spin

Since the erosion rates of metals display a power-law dependence on particle velocity, accurate measurement of particle velocity is of paramount importance. Photographic techniques are most commonly used to determine particle velocity and are recommended in ASTM G76-83 [15]. Several techniques have been developed including multiple-flash photography, the streaking camera method and high-speed framing photography [2], [3]. In the present work, it was decided to use a multiple-flash photography

Development of an erosion model incorporating particle spin

Currently, there is no predictive erosion model that satisfactorily takes into account particle rotation. Analytical models have been developed that predict the crater volume created by the impact of individual particles of given angularity and orientation [2], [3], [4], [5], [6] but these do not incorporate a means of averaging over the range of orientations expected in multi-particle impact. In addition, these models predict an increase in crater volume with increasing impact angle. This is

Predictions of the modified Finnie–Bitter model

Fig. 11 shows the predicted erosion rate given by the modified Finnie–Bitter erosion model developed in this paper, in which the effects of particle rotation are taken into account. The predictive model has been fitted to the present set of experimental erosion data, as given in Table 1 for crushed glass of size 150–250 μm. The angular particle shape was expected to provide the maximum difference in erosion rate for the two target orientations. The data points are included in the figure. The

Conclusions

  • (1)

    Experimental work using metal targets positioned at different orientations with respect to spinning particles allowed the effect of particle spin direction on erosion rate to be investigated. A clear trend was apparent for angular particles, although the results for spherical particles were less conclusive. The results indicated a higher erosion rate if particles strike the target with ‘back-spin’ rather than ‘top-spin’ or ‘no-spin’.

  • (2)

    The difference in the erosion rates of the targets was more

Acknowledgements

The authors would like to acknowledge the technicians in the Department of Materials Science and Metallurgy, University of Cambridge, UK, for the development of the eight-channel triggered flashgun controller unit. Thanks are also extended to the technicians at the Wolfson Centre for Bulk Solids Handling Technology, University of Greenwich, UK, for their support during the experimental work.

Cited by (45)

  • Optimization of a convex pattern surface for sliding wear reduction based on a definitive screening design and discrete element method

    2021, Powder Technology
    Citation Excerpt :

    It should be noted that the sliding wear does not remove the meshes, and the sample keeps intact during the simulation. In terms of operational conditions, the normal load [41] and the relative velocity [42,43] have significant effect on the sliding wear. The normal load is directly related to the particle bed thickness, which is defined by the depth of the particle bed, so particle bed thickness is selected as an operational condition.

  • Surface erosion due to particle-surface interactions - A review

    2021, Powder Technology
    Citation Excerpt :

    There is relatively less data on the variation in erosion due to particle angular rotation during the impact. Deng et al. [144] pioneered the experimental investigation of this effect. They reported that the angular particle impacting the target with ‘back-spin’ (backward rotation) erodes the surface more as compared to the ‘top-spin’ (forward rotation) or ‘no-spin’.

  • Experimental and computational analysis of the erosion behaviour of unidirectional glass fiber epoxy composites

    2020, Wear
    Citation Excerpt :

    It is seen that the transition is smooth from the micromodel to the homogeneous part and the stresses are in the range of 1 MPa, which is well below any kind of failure in the homogeneous part. It was observed from previous studies [28–30] that the particle shape can be responsible for the change in erosion magnitude and behaviour. It is observed from the literature that most of the numerical models are confined to a particular shape.

View all citing articles on Scopus
1

Tel.: +44-20-8331-8646; fax: +44-20-8331-8647.

View full text