Wear 267 (2009) 568575

Contents lists available at ScienceDirect

Wear
journal homepage: www.elsevier.com/locate/wear

Short communication

The correlation of material characteristics and wear in a laboratory scale cone crusher
P. Kivikyt-Reponen a, , S. Ala-Kleme b , J. Hellman a , J. Liimatainen a , S.-P. Hannula b
a b

Metso Materials Technology, POB 237, 33101 Tampere, Finland Laboratory of Material Science, Helsinki University of Technology, POB 6200, 02015 TKK, Espoo, Finland

a r t i c l e

i n f o

a b s t r a c t
Wear of materials is a widely recognised industrial problem. The direct costs of wear failures, such as wear part replacements, loss of productivity as well as indirect losses of energy, and the increased environmental burden, are real problems in everyday work and business. The mineral industry is particularly susceptible to wear part consumption resulting in high expenses. In this study, the performance of various steel-based wear resistant powder metallurgical metal matrix composites (MMCs) was studied for mineral industry applications. The matrix of the composites consisted of various tool steels, manganese steel and martensitic steel and the reinforcements used were tungsten carbides (WC), titanium carbides (TiC) or cemented carbides (WC/Co). Hot isostatic pressing (HIP) was used for compacting the materials. The performance of composite materials was studied in a laboratory scale cone crusher wear test. Correlations between the wear behaviour and microstructure parameters, such as volume fraction and the size of the hard particles and spacing between hard particles were investigated. It was found that the most important parameters in this wear environment were the total volume fraction of the hard phase and the type of the hard phase. A necessary feature of the matrix material is that it is also resistant to abrasion. 2009 Elsevier B.V. All rights reserved.

Article history: Received 28 August 2008 Received in revised form 19 December 2008 Accepted 19 December 2008 Keywords: Abrasive wear Cone crusher Mineral crushing Steel matrix composites

1. Introduction Cone crushers are widely used in mineral processing industries. The cone crusher can be classied as a compressive crusher among jaw and cyratory crushers. In the cone crusher, the feed is crushed into wanted size fractions in compression in the cavity between concave and mantle, which is in eccentric rotating motion. Typically, cone crushers are secondary crushers (feed size 100400 mm) or tertiary/ne crushers (feed size 10100 mm) in the crushing process, while jaws, gyratories and impactors are typically primary crushers with a feed size of 5002000 mm [1]. The geometry of the liners or wear parts (mantle and concave) has been optimised in the crusher in terms of the nal product, crushed stone. When liners are worn and the geometry is changed, the shape of the cavity is also changed. How to predict the wear-induced changes in geometry has been extensively studied by Lindqvist [24], while Ruuskanen [5] has studied the dependence of liner wear on crusher operating parameters. In both cases the crusher liner materials were manganese steels.

Corresponding author. Tel.: +358 40 5155041; fax: +358 20484121. E-mail address: Paivi.Kivikyto-Reponen@metso.com (P. Kivikyt-Reponen). 0043-1648/$ see front matter 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.wear.2008.12.055

According to these studies, the wear system in a cone crusher is three-body abrasive wear, where many external load directions and variable loads exist. Wear is caused by the motion of the rock bed and consists of both sliding and indentation/compression (squeezing). Lindqvist has described wear as squeezing wear with a small sliding motion [3]. Therefore, all the forms of abrasive wear, such as microploughing, microcutting, microcracking and microfatigue, contribute their relative importance to wear mechanisms. The composite materials dealt with in this study are new in the present application. In other industrial crushing applications the powder metallurgical composite materials have proven to be more wear resistant than conventional materials [6]. The relationships between wearing motion and wear mechanisms are affected by reinforcement-related parameters and matrix material. The importance of internal microstructural parameters in composite wear has been noted, e.g., by Berns et al. [79]. The impact wear of similar types of MMCs than in this present study has been studied by Osara [10,11] using an impact hammer device. Osara made several kinds of wear tests and showed how the wear environment affects the material wear performance and the ranking of the materials. As a result of the differences in the wear systems of the two crushing concepts, those results cannot be directly applied to cone crushers. In the present study, the performance of the steel matrix composites is studied in the cone crusher environment.

P. Kivikyt-Reponen et al. / Wear 267 (2009) 568575 Table 1 Chemical contents of the matrices presented as nominal compositions given by the manufacturer. Symbol WR4 WR6a 9980a 2218a Hadac 120a
a a

569

C (wt.%) 1.8 2.90 0.03 0.1 1.11.3

Cr (wt.%) 5.25 5.25 15.3 2.3 2.03.0

V (wt.%) 9.0 11.5 0.15 0.40.5

Mo (wt.%) 1.3 1.3 0.9 1.0 <2.0

Mn (wt.%) <0.5 <1.0 0.75 10.012.0

Ni (wt.%)

Co (wt.%)

Si (wt.%) 0.9 <1.0 0.45 <1.0

Fe (wt.%) Bal Bal Bal Bal Bal

Other (wt.%)

5.3 0.7 <1.0

0.1

W 0.1; Cu 0.40 P < 0.050; S 0.040; Cu < 0.20

By Metso Materials Technology.

Table 2 Symbols and compositions of the materials, SET1. Symbols WR6WC WR4WC 9980aWC 9980bWC 2218WC WR6
a b

Matrix WR6a WR4a 9980a 9980a 2218a WR6a

Matrix (wt.%) 75 75 75 65 75 100

Reinforcement WC/Cob WC/Cob WC/Cob WC/Cob WC/Cob (10%) (10%) (10%) (10%) (10%)

Reinforcement (wt.%) 25 25 25 35 25

Reinforcement size (m) 200400 200400 200400 200400 200400

HV matrix nominal 695 575 320 320 170 695

By Metso Materials Technology. By Tikomet Oy.

2. Materials and test procedures The MMCs studied here are powder metallurgical (PM), materials. Composites tested in the laboratory scale cone crusher were divided into two sets. In SET1 the steel matrix type was varied, but each composite was reinforced with recycled and crushed WC/Co hardmetal. Tool steels (WR4 and WR6), martensitic steels (9980 and 2218) and manganese steel (Hadac) were used as matrix materials. Their nominal composition is presented in Table 1. The composites in SET1 are introduced in Table 2. This set was designed to evaluate the role of the matrix material type in the composite structure, while the reinforced particles were kept the same. In the SET2 the role of the reinforcements was the focus of interest and in all these composites the matrix was tool steel WR6. Reinforcements were either tungsten carbide particles (WCS), recycled and crushed WC/Co-hardmetal, dense coated WC/Co reinforcements or TiC particles. Symbols of the composites in SET2 are presented in Table 3. The production of the PM materials was carried out by hot isostatic pressing (HIP). HIP will normally offer distinctive advantages

in terms of ner grain size and homogeneous microstructure with less segregation and no porosity, compared to conventional casting technology. On the other hand, HIPed products can be easily tailored by microstructure, i.e., manufacturing evenly distributed double dispersions, or tailored by macrostructure as gradient or multimetal constructions. HIP processing was carried out at 1150 C, at a pressure of 100 MPa and for a holding time of 3 h. The heat treatments were carried out according to the recommendations for the particular matrix material (Table 4). The interfaces between tool steel and WC/Co particles has been studied by Lou et al. [12] and selected microstructures in SET1 and SET2 by Ala-Kleme [13,14]. Generally, the shapes of the reinforcements vary from the at and long to the more round or square depending on their prior shape and quality, Fig. 1af. Some defects, fractures or clustering of the reinforcements have been observed in the composites. The prior powder boundaries were visible in many composites, e.g., Fig. 1a. The small-scale Nordberg laboratory cone crusher B90 was used in order to obtain information on the wear rates of the mantle. The stones are crushed in compression in the cavity between concave

Table 3 Symbols and composition of the materials in SET2, testing the role of the particle and particulate reinforcements. Symbol 30WCSf 20WCSf 30WCSc 20WCSc 30WCf 30522dcf 30522dcc 30TiCf
a b c

Matrix WR6a WR6a WR6a WR6a WR6a WR6a WR6a WR6a

Matrix (vol.%) 70 80 70 80 70 70 70 70

Reinforcement WCb WCb WCb WCb WC/Coc (10%) WC/Cob WC/Cob TiCb

Reinforcement type after manufacturer WCS cast WCS cast WCS cast WCS cast Recycled Dense coated Dense coated

Reinforcement (vol.%) 30 20 30 20 30 30 30 30

Reinforcement size (m) 4590 4590 250425 250425 100200 4590 200300 75250

By Metso Materials Technology. By H.C. Starck GmbH. By Tikomet Oy.

Table 4 Heat treatments of composites. Materials Tool-steel-based composites 9980-steel-based composites 2218-steel-based composites Manganese-steel-based composites Annealing Hold 1 h 1080 C Hold 2 h 1010 C Hold 2 h 900 C Hold 2 h 1040 C Quench N gas quenching Air cooling Air cooling Air cooling Temper Double tempering 2 h 570 C air cooling Hold 3 h 580 C air cooling Hold 3 h 580 C air cooling

570

P. Kivikyt-Reponen et al. / Wear 267 (2009) 568575

Fig. 1. Optical micrographs of (a) WR4WC and (b) 2218WC and SEM micrograph of (c) 20WCSf, (d) WR6WC, (e) 30TiCf and (f) pure WR6.

and mantle. The loss of material in the liners (concave and mantle) affects the product (rock) quality and the crusher efciency. The basic principle of the cone crusher is presented in Fig. 2. Stones used in the wear tests were micagneiss for the SET1 and granite for both SET1 and SET2. Both stones are multi-minerals in composition. According to Metso Minerals the UCS1 of the granite used is 193.9 MPa and UCS of micagneiss is 63.7 MPa. The tests were carried out using 5 mm close side setting (CSS), 4.0 mm stroke and 1020 mm feed in every test. Both stones, granite and micagneiss, had been sieved out into the standardised size fraction. About 25004500 kg of abrasives were used for each test.

3. Results The measured wear rates of the materials of the SET1 obtained for two stone types, micagneiss and granite, are presented in Table 5. The average coefcient of variation in the SET1 was 11.7%. Some test results are omitted in Table 5, because the test parameters did not remain stable during the experiment. Additionally, the results marked with ()* were clearly not steady state wear rates, i.e., the rst wear result (or also the second) of the set of six tests were different from the others. The cumulative wear losses of selected materials are presented in Fig. 3. One of the most widely used parameters in studying the effect of the microstructure on the wear behaviour is the volume fraction of the hard phase. Fig. 4 presents the wear rate as a function of the total hard particle content of the materials in SET1. Decreasing

Ultimate compressive strength.

P. Kivikyt-Reponen et al. / Wear 267 (2009) 568575 Table 5 The wear rates with micagneiss and granite stone on the steady state region of wear, SET1. Symbol WR6WC WR4WC 9980aWC 9980bWC 2218WC HadacWC WR6
a

571

Wear rate average (g/ton) (micagneiss) 1.5 3.8 7.8 10.7 3.1

Wear rate (micagneiss) (S.D.) 0.1 0.3 0.4 1.0 0.4

Wear rate average (g/ton) (granite) 2.0 5.6 (18.2)a 16.2 (19.2)a 16.9 (11.6)a 9.6 4.7

Wear rate (granite) (S.D.) 0.2 0.2 (5.9)a 1.6 (4.7)a 1.6 (3.2)a 1.1 0.2

Initial state of wear is included. Table 6 SET2 wear rates with granite stone. Symbol 30WCSf 20WCSf 30WCSc 20WCSc 30WCf 30522dcf 30522dcc 30TiCf Wear rate average (g/ton) (granite) 3.6 5.5 4.2 2.4 2.8 3.6 3.6 3.6 Wear rate (granite) (S.D.) 1.3 0.8 0.4 0.2 0.5 0.4 0.7 0.4

Fig. 2. Principle of cone crusher.

wear rate with increasing total hard particle content can be clearly seen for both rock types used. In SET1 all the composites are WC/Co reinforced, but their wear rates seem to vary widely, although WC/Co contents with different materials varies only between 10 and 15 vol.% according to image analysis. This also implies that the type of matrix plays an important role in the wear of the composites. The ne VC precipitates in the matrix seem to reduce the composite wear rate: materials WR6, WR4WC and WR6WC, which all contain ne VC precipitates in the matrix, have a lower wear rate compared to the other tested steels. Vanadium carbides increase the hardness

of the matrix and its resistance to groove formation. The thickness of the scars varies considerably, and can be up to 100 m. The measured wear rates of the materials of the SET2, tested with granite stone, are presented in Table 6. The COV of these tests was 14.9 on average. This is higher than in the SET1, but the wear rates are also considerably lower. The materials in the SET1 had a larger variation in the wear rates than the materials in the SET2. In SET1 the lowest surface roughness (and wear rate) was associated with WR6WC and WR6 materials. In stereomicroscopy, the macroscale roughness of the surfaces and the rock indentation scars are clearly seen on most of the materials in both sets. Both upraised hard particle hills as well as valleys in the surfaces of composites can be seen. SEM micrographs of the wear surface replicas of the WR4WC composite are presented in Fig. 5a and b. Because these pictures are taken from the replicas, the hills are truly the valleys. No differences in the wear surfaces were noticed on different parts of the laboratory crusher mantle (from top to bottom) by stereomicroscopy. The geometrical dimensions of the mantles were not changed during the test time of this study, but it has been previously observed that liners may wear differently at different parts [3].

Fig. 3. Cumulative weight losses as a function of the amount of used abrasive, for selected materials of the SET1.

572

P. Kivikyt-Reponen et al. / Wear 267 (2009) 568575

Fig. 4. The wear rate in the B90 test versus the total carbide content of the materials in SET1. Stones granite and micagneiss.

In SET2 the ranking of the materials does not change as a function of the reinforcement volume fraction or the total carbide content because the matrix material is same. The wear rates of the few SET2 materials are presented in Fig. 6 as a function of the reinforcement volume fraction and the results of each composite type are connected by a line. It appears that the smaller volume fractions of coarse particles, both recycled hardmetal reinforcement and cast WC carbide, seem to perform better than ner structures of the same reinforcements, although wear rate increases in coarse solid cast WC reinforced composite with an increase of WC content from 20 to 30 vol.%.

The wear rates of the SET2 materials are shown as a function of the average size of the carbides of the reinforcement particles and the average size of the reinforcement particles, Figs. 7 and 8. The composites with intermediate sizes of the carbides were the worst in respect of wear. However, the volume fraction of the carbides has a strong effect on the wear rate; the study of composites with different volume fractions may therefore partly disguise the possible trends. When considering the materials with 30 vol.% (in material symbols value 30 in the plot), there may be a shallow gradient for increasing wear rate with true carbide size. The relationship between wear rate and average reinforcement particle size is presented in Fig. 8. The reinforcement size is considered to be the average size of the reinforcement particles, i.e., solid carbide size or hardmetal size (carbide cluster size). After drawing in the gure the minimum and maximum wear rate borderlines, it seems that, in general, larger reinforcements perform better than the smallest ones. The inuence of the nominal spacing between reinforcement particles on the wear rate can be seen in Fig. 9. Nominal spacing between reinforcements, i.e., solid carbides or hardmetals (carbide clusters), is determined by the amount of the carbides and the size of reinforcements. Composites with the largest average spacing (large reinforcement size and low volume fraction) have the lowest wear rates in the tested set. This suggests that the increase of material toughness will increase the wear resistance, although considerable scatter exists in the results.

4. Discussion A comparison of the current wear results with those in the literature is difcult, since similar tests have not been made for these types of composites. Some general trends have been observed for good wear resistance in a multi-mode wear system or in a wear system where a proper combination of hardness and toughness is needed [69,15]. Wear properties of composites depend on the size, volume fraction and distribution of the reinforcement, the properties of the matrix and the compatibility of matrix and reinforcement. Relatively large reinforcement sizes are favourable, and in rough abrasion, metal matrix composites with a 50 100 m reinforcement size have performed well [7]. However, small true particle sizes are also benecial. It is known that the increase in hard particle size will reduce bending strength, Rm [9]. Also an increase in spacing will increase fracture toughness, KIc [9]. Large carbide sizes and particle clustering are common reasons, which tend to

Fig. 5. SEM micrographs of the wear surface replicas of the WR4WC composite (a) after crushing granite stone and (b) after crushing micagneiss stone. Note different magnications.

P. Kivikyt-Reponen et al. / Wear 267 (2009) 568575

573

Fig. 6. Wear rates in the cone crusher, B90 wear test as a function of tungsten carbide content of the composites in SET2. Granite stone as a wear medium.

Fig. 7. Wear rate versus true carbide size in logarithmic scale, WR6-based composites.

Fig. 8. Wear rate versus average reinforcement size, i.e., solid carbide size or hardmetal size (carbide cluster size) of the composites (WR6 based).

favour the formation of voids. The actual crack initiation has been investigated earlier [16] with MMCs. In a cone crusher wear environment, several micromechanisms of wear are present simultaneously. In the case of pure sliding,

the cutting and ploughing of the soft metallic matrix is dominant while ceramic reinforcements are only ground. In the case of pure compression/indentation, the microcracking of ceramic reinforcement is a crucial phenomenon, while the matrix will be plastically

574

P. Kivikyt-Reponen et al. / Wear 267 (2009) 568575

Fig. 9. Wear rate versus average spacing between reinforcements in WR6-based composites.

deformed. The relationships between wearing motion and wear mechanisms are affected by reinforcement-related parameters and matrix material. The importance of internal microstructural parameters in composite wear has been noticed inter-alia by Berns et al. [79]. Reinforcement particles may contribute to different micromechanisms in the rock contact on the surface. Based on Fig. 6, it can be concluded that at smaller volume fractions coarse particles, both recycled hardmetal reinforcement and cast WC carbide, seem to perform better than ner structures of the same reinforcements. Wear resistance of the composites with ner reinforcements is increasing with the increase of hard particle content in the tested range. On the other hand, wear rate increases in coarse solid cast WC reinforced composite with an increase of WC content from 20 to 30 vol.%. However, a limited number of different volume fractions were tested. Comparing WC/Co and cast WC, hardmetal is tougher, i.e., the possibility of particle cracking is higher with cast WC particles. Based on the analysis of the experiments, it seems that each individual composite must be evaluated separately. Different types of reinforcements have different effects on composite wear rate as a function of reinforcement volume fraction. However, the basic characteristics of the composites combining WC and VC are the following: vanadium carbide particles are evenly distributed in the matrix material and therefore increase the matrix hardness, while WC particles are more clustered resulting in a more localised effect on wear. The reason for the small effect of WC/Co on the wear rate may be the easy pull-out of the reinforcements in wear. In spite of the fact that it was not observed from wear surface replicas, it has been observed before in abrasion [13,17]. In conclusion, the wear resistance of the matrix is more crucial than that of the reinforcements for the wear rates in SET1. 5. Conclusions The microstructural parameters of the composites have a signicant inuence on the wear rate and wear resistance of the composites. This inuence and its strength vary with the wear conditions. These parameters inuence material properties, such as hardness and toughness, which are commonly connected to the wear performance of the material. The important ndings from the studied reinforcement-related parameters in a cone crusher wear environment were: (i) The primary parameters which inuenced the wear behaviour of the composites were the total hard particle content and the

type of the hard particles. The effect of the increasing total carbide content was signicant in decreasing wear rates. When the reinforcement content between the composites was relatively similar, the wear rate of the different composites varied widely depending on the matrix selection. The role of the matrix characteristics and the matrix carbide content, i.e., the total carbide content, was more crucial than that of the added reinforcement particles for the formation of wear rates. (ii) The reinforcement type seems to be a noticeably strong factor in material wear. It overshadows such internal parameters as size and spacing of reinforcement particles. The size of the reinforcement and the spacing between reinforcements has earlier been found to affect the wear rates. However, in the present testing conguration, no general trends between the size and spacing between reinforcement particles and the wear rate were found. In the present study, the WR6WC composite yielded the lowest wear rate for both rock types tested. (iii) The wear of each composite has to be evaluated separately, because different types of reinforcement have different effects on composite wear rate as a function of reinforcement volume fraction. Acknowledgements The nancial support from the Graduate School of the Academy of Finland and Metso Materials Technology in Finland is gratefully acknowledged. References
[1] J. Eloranta, Inuence of crushing process variables on the product quality of crushed rock, Ph.D. Thesis, Tampere University of Technology, Publication 168, Finland, 1995, 118 pp. [2] M. Lindqvist, C.M. Evertsson, Linear wear in jaw crushers, Minerals Engineering 16 (2003) 112. [3] M. Lindqvist, D.M. Evertsson, Prediction of worn geometry in cone crushers, Minerals Engineering 16 (2003) 13551361. [4] M. Lindqvist, Wear in cone crusher chambers, Doctoral Dissertation, Department of Applied Mechanics, Chalmers University of Technology, Gteborg, Sweden, 2005, 36 p. and publications. [5] J. Ruuskanen, Inuence of rock properties on compressive crusher performance, Doctoral Dissertation, Tampere University of Technology, Publication 585, Finland, 2006, 236 pp. [6] W. Theisen, A novel PM-wear protection method to meet high comminution demands, Wear 250 (2001) 5458. [7] H. Berns, Microstructural properties of wear-resistant alloys, Wear 181183 (1995) 271279. [8] H. Berns, N. van Chuong, Hartverbundwerkstoff mit zweistugem Dispersionsgefge in Hartlegierungen und Hartverbundwekstoffe, Befge, Eigenschahten, Bearbeitung, Anwendung, Hrsg. Hans Berns, Springer-Verlag, Berlin/Heidelberg, 1998 (in German).

P. Kivikyt-Reponen et al. / Wear 267 (2009) 568575 [9] H. Berns, A. Melander, D. Weichert, N. Asna, C. Broeckmann, A. Gross-Weege, A new material for cold forging tools, Computational Materials Science 11 (1998) 166180. [10] K. Osara, Characterization of abrasion, impact-abrasion and impact wear of selected materials, Doctoral Dissertation, Tampere University of Technology, Finland, 2001. [11] K. Osara, T. Tiainen, Three-body impact wear study on conventional and new P/M + HIPed wear resistant materials, Wear 250 (2001) 785794. [12] D. Lou, J. Hellman, D. Luhulima, J. Liimatainen, V.K. Lindroos, Interactions between tungsten carbide (WC) particulates and metal matrix in WC-reinforced composites, Materials Science and Engineering A340 (2003) 155163. [13] S. Ala-Kleme, WC/Co -lujitettujen tersmatriisikomposiittien kulumis- ja sitkeysominaisuudet, Diplomity, Espoo, Finland, 2004, p. 109 (in Finnish).

575

[14] S. Ala-Kleme, P. Kivikyt-Reponen, J. Liimatainen, J. Hellman, S.-P. Hannula, Abrasive wear properties of tool steel matrix composites in rubber wheel abrasion test and laboratory cone crusher experiments, Wear 263 (2007) 180 187. [15] H. Berns, Comparison of wear resistant MMC and white cast iron, Wear 254 (2003) 4754. [16] C. Broeckmann, A. Pyzalla-Schiek, Microstructural aspects of failure in particle reinforced metal matrix composites, Computational Materials Science 5 (1996) 3244. [17] S. Ala-Kleme, P. Kivikyt-Reponen, J. Liimatainen, J. Hellman, S.-P. Hannula, Abrasive wear properties of metal matrix composites produced by hot isostatic pressing, Proceedings of Estonian Academy of Science and Engineering 12 (4) (2006) 445454.