The wear mode identification of hard materials as means for concurrent estimation of their abrasion and fracture resistance

Transcription

1 S.F. Scieszka, J. Krajzel HM The wear mode identification of hard materials as means for concurrent estimation of their abrasion and fracture resistance S.F. Scieszka*, J. Krajzel * Technical University of Silesia, Akademicka 2, Gliwice, Poland Węgliki Spiekane BAILDONIT, Sp. z o.o., Katowice, Poland Summary The fracture toughness is often the major limiting parameter governing the use of hard materials tools. Hence there is a need for research aimed at increasing toughness without sacrificing wear resistance. To aid in this objective, a simple and reliable integrated testing method, in which a conjoin action involving both fracture and abrasion occur, is needed. One such method currently developed is presented in this paper. The method is based on the concept of edge chipping during the initial transition stage of abrasion wear which is controlled by brittle fracture process. The limitation of the method for tougher materials such as tool steel as well as for very brittle materials such as ceramics was also indicated. The empirical relationship between mass loss as a result of edge chipping during the initial transition stage of abrasive wear and fracture toughness in the form of formulae was calculated and discussed. Keywords Wear resistance, Fracture toughness, Hard materials, Edge chipping. Introduction The properties of hard materials have been extensively studied [-4], due to its importance in industrial applications. Hardmetals for example have a variety of applications, and the requirements of their properties are in many cases directly related

2 00 HM 83 S.F. Scieszka, J. Krajzel to the mechanical properties, such as wear resistance and fracture toughness. Hence fracture toughness and wear resistance are two of the major materials characteristics to take into consideration when designing components such as hardmetals or ceramics tools. This is mainly due to the risk of brittle fracture in tools made of these materials and due to the importance of resistance to abrasion action where the content pressure between the component and abrasive is high, such as in mining and rock drilling [4]. Rock drilling and cutting produce both impact and abrasion in various relative amounts at the tool/rock interface mainly because the rock fragmentation itself is a discrete process rather than a continuous one [4]. Since the fracture toughness is often the major limiting parameter governing the use of drilling, cutting and other tools, there is a need for research aimed at increasing toughness without sacrificing wear resistance. To aid in this objective, a simple and reliable integrated testing method, in which a conjoint action involving both fracture and abrasion occur, is needed for quick assessment of progress in such research. One such method currently developed [5, 6] is proposed for further, more advanced study in this paper. The method is based on the concept of edge chipping during the initial transition stage of abrasion wear which is controlled by brittle fracture process. Although the method of testing is not yet well established it is seen as a promising and pragmatic way of ranking hard materials for fracture toughness and wear resistance. The limitations of the method for tougher materials such as tool steels as well as for very brittle materials such as ceramics were not yet investigated and determined. Over the last decade there have been developments in methods of testing specifically for advanced technical ceramics and hardmetals. There is a considerable body of published information on these methods, e.g. comprehensive reviews are included in NPL Measurement Good Practice Guides [2, 3, 7]. This wide variety of methods is suitable rather exclusively for low toughness material. Problems associated with the effective and reliable use of the above presented experimental methods stimulated effort towards the development of empirical and semi-empirical formulae describing the relationship between the critical stress intensity factor (K IC ) and other mechanical and physical properties that are much easier to measure and which finally led to the alternative techniques for fracture toughness evaluation.

3 S.F. Scieszka, J. Krajzel HM Toughness evaluation based on edge damage pattern When a load is applied along the sharp edge of hard and brittle tools or various other construction elements cracks may initiate, propagate and eventually spall from the surface. On the other hand, in grinding or machining of brittle materials such as ceramics, optical glasses and hardmetals etc, flaking or chipping are often seen at the work edge where the cutting edge comes into contact with or separates from the workpiece. This sudden controlled by fracture edge damage on the tools and on the workpieces has been identified as a technologically significant problem in e.g. edge machining, edge mounting etc, and was named as edge cracking, edge spalling, edge flaking, edge chipping [8]. A number of attempts to devise novel methods of testing were made for a pragmatic way of ranking materials for toughness [5, 6, 8] based on edge damage controlled by fracture. 2.. Edge chipping in transition, wearing-in process in granular abrasive medium The controlled combined action of a granular, hard abradant both as multipoint source of loading of the sample of hard material in the vicinity of edge and as hard, angular abradant can be used for ranking materials for both toughness and for wear resistance (Fig. ). This combined action takes place within a cylindrical chamber filled with granular abradant inside which normal stress, σ n, P σ n = () 2 πr (where R is the internal radius of chamber) is controlled by external loading P. External torque, T, applied to the drive shaft is used to overcome the shear resistance of granular abradant 3 T τ = (2) 3 2 πr The method which enables simultaneous brittle fracture and abrasion testing allows for simulation of stress and sliding speed conditions found e.g. between drills and rock. But in fact any material combination (i.e. material of bar-sample or granular abradantmineral sample) can be used under any operating condition (i.e. normal stress, sliding velocity of temperature). Another advantage of the tribotester itself is that the ground abradant granular mineral is allowed to leave the attrition area through the gap between a disc and a wall of the cylindrical chamber as occurs in actual drilling or

4 02 HM 83 S.F. Scieszka, J. Krajzel grinding. The apparatus consists of the disc rotating in the cylindrical chamber under normal force. The specimen bars are attached to the underside of the disc. The specimen bars are made from hardmetals being tested. The test procedure, overall set-up of the tribotester, the method of calculation and presentation of the results were described in details elsewhere [, 5, 6] Test system and experimental details The tests were performed in a purpose-built testing machine. The overall view in Fig. shows the upright drilling machine assembly with mounted cylindrical tribotester and loading pulley. Specification of the apparatus and experimental details are shown in Table. a) b) Fig. a Fig. b Schematic diagram of apparatus, and interpretation of interaction between the particulate mineral and the bar within the shear zone:. Drill chuck; 2. Drive shaft; 3. Cylindrical chamber; 4. Thermocouple; 5. Torque indicator; 6. Thrust bearing; 7. Force indicator, 8. Displacement indicator; 9. Recorder Overall set-up of the tribotester:. Upright drilling machine providing controlled rotary motion and normal loading to the drive shaft; 2. Torque indicator; 3. Cylindrical chamber and shaft of the tribotester; 4. Test controller; 5. Transducer and recorder; 6. Computer data processing system; 7. Load Test samples (Fig. 2) were prepared in accordance with design specification []. The hardmetal, ceramic and tool-steel testpieces were prepared by careful grinding. Test edges were not chamfered. The materials and their properties are summarised in Table 2.

5 S.F. Scieszka, J. Krajzel HM The preliminary results [] clearly revealed that the edge chipping process (the fracture controlled transition wear process) was taking place only during the first 5 revolutions and that the upper and lower limits of the method had been determined by tool steel and Si 3 N 4 ceramic respectively. Fig. 2 Apparatus design specimen holder, drive shaft and specimens Name Edge abrasion tribotester Load range used (N) 000 Drive shaft speed (rpm) 30 Test duration (rev) 5 Mean sliding distance (m) 0.47 Abradant Fused alumina ( µm) Table Specification of the apparatus and experimental details Material (Hardmetals) ρ (Mgm -3 ) E (GPa) K IC (MPa m ½ ) Hardness HV 30 TSM30 (WC-0% Co) TSM30BV2 (WC-Co) TSM30BV (WC-Co) Shm3 (WC-3.3% Co) CW20C (WC-20% Co) CW25C (WC-25% Co) Table 2 Materials compositions and mechanical parameters

6 04 HM 83 S.F. Scieszka, J. Krajzel 2.3. Experimental results and discussion In the final part of the investigation all the materials listed in Table 2 were tested. The test procedure developed during the previous study and presented in [] was applied. The procedure consisted of three consecutive tests lasting 5 revolutions, but only the first one starting with a sharp sample edge. This test procedure was repeated at least six times for every material tested. In every test only one normal load equal 000N was applied. In this part of the investigation the following notations and calculations were used: m - mass loss during the first test (the first of three consecutive tests starting with unworn sharp edge) lasting 5 revolutions, m - mean mass loss during the first tests repeated each time with new sharp samples edge, m = n n m ( mg) where n is the number of edges tested, a m - mean mass loss during tests other than the first ones (tests two and three) a each lasting 5 revolutions. m represents mass loss controlled by stable-abrasion mode of wear after the transition-fracture controlled mode of wear was completed, 2n a m = m ( mg) 2n F m - mean mass loss solely as a result of fracture (edge chipping) during the first tests, F m = m - m W = ( mm 3 ) ρ m ρ m ρ a W a = 3 ( mm ) F W F = 3 ( mm ) a m (3), where ρ is density AR - abrasion resistance number rev mg

7 S.F. Scieszka, J. Krajzel HM AR = a m rev AR v - volumetric abrasion resistance number 3 mm 5 AR v = W a rev FR - fracture resistance number mg 5 FR = F m rev FR v - volumetric fracture resistance number 3 mm 5 FR v = W F (4) (5) (6) (7) FI - fracture index FI = F m = m W W F (8) The procedure presented above is based on a concept of precise distinction between initial processes (tribological transition) controlled by fracture edge damage and the steady stage of wearing predominantly controlled by microabrasive wear mechanisms. The steady stage of wearing yields a rating of the wear resistance of materials. In steady-state wear, the spread of results reported in previous investigations [5, 6] was remarkably small, which when combined with the very accurate mass-loss measurement, makes the method very discriminating. The method presented above does not produce the fracture toughness value (e.g. K IC ) as directly as the standardised plain strain fracture toughness tests does. The integrated testing method requires the theoretical or empirical models (formulae) which describe the relationship between fracture toughness and other mechanical properties. These formulae are usually based on well established assumptions and laws such as

8 06 HM 83 S.F. Scieszka, J. Krajzel energy balance, the Hertzian stress distribution, or on an empirical and semi-empirical relationships which were proved to be valid in the specified range [5, 6, 9, 0]. In accordance with the objectives of the present work shown in the introduction, the relationship [] based on study of abrasive wear by brittle fracture which consists of the removal of material by lateral cracking. Hence equation (9) is proposed for further study and the least square fit evaluation: K c 2 2 E = H S γ H The results from the final part of the investigation were summarised in Tables 3 and 4. The results show that by using one apparatus, one shape of test specimen and only one relatively easy testing procedure it is possible to obtain reliable rating of hardmetals according to both their edge fracture toughness as well as their resistance to abrasive wear in rubbing contact with particulate alumina. Examination of the selected new and worn edges, on various stages of test procedure, had been conducted. SEM observations of the new, unworn samples revealed the quality of the surface finish obtained by grinding and the surface scars representative for the edge tip area. The morphology of ground surfaces together with unworn edges are depicted in Fig. 3. SEM micrographs shown microploughing furrows and microcuttings both parallel to the edge. Occasionally abrasion grits from the grinding wheel were found embedded into the specimen s surface. 5 4 p 9 4 (9) a) b) Fig. 3 SEM micrographs of the unworn samples showing grinding scars on both sides of the edge and the edge tip width: CW20C (x000), (b) shm3 (x200)

9 S.F. Scieszka, J. Krajzel HM Material m (mg) Test results: arithmetic mean values a m (mg) F m (mg) W (mm 3 ) W a (mm 3 ) W F (mm 3 ) TSM TSM30BV TSM30-BV Shm CW20C CW25C Table 3 Results from tests run at normal load, F N = 000N and individual consecutive test duration, i = 5 revolutions Material Properties: arithmetic mean and standard deviation AR AR v FR FR v rev rev ( ) FI rev rev ( ) mg 3 mm mg 3 mm ( ) TSM TSM30BV TSM30BV Shm CW20C CW25C ( ) Table 4 Property indicators calculated from test results run at normal load, F N = 000N and individual consecutive test duration, i = 5 revolutions

10 08 a) Fig. 4a,b a) Fig. 5a,b HM 83 S.F. Scieszka, J. Krajzel b) SEM micrographs of the submicron grained hardmetal after only two revolutions showing the early stage of edge fracture controlled wear: (a) shm3, a typical edge chipping and abrasion damages on leading (left) side of the edge (x00), (b) shm3, surface created by intergranular fracture partly reshaped by abrasion action of alumina particles on leading side of the edge, (x000) b) SEM micrographs of the hardmetal CW20C worn sample after the full test (45 revolutions): (a) the edge at its extremity (x00) showing the difference between the leading (left) and trailing (right) side, (b) abrasion damages on the leading side (x5000)

11 S.F. Scieszka, J. Krajzel HM As the surface roughness in the vicinity of the edge and particularly the crack-like defects generated by the grinding process such as sharp microcutting and microcracking contribute to the fracture controlled mass loss it is essential that all the samples have statistically similar surface morphology. The original, almost intact surface morphology was found of the test procedure on the trailing side of every sample. This was recorded on the hardmetal CW20C (Fig. 5a). Fig. 4 presents a typical fracture surface of the submicron grained hardmetal (shm3) with some embedded carbide grains protruding from the phase and the other being removed, exposing the sockets that contained them. For this hardmetal the crack propagation was primarily along WC-Co interfaces with partially transgranular fracture through the cobalt matrix. Surfaces worn by abrasion action of alumina particles such as the hardmetal CW20C sample after the full test presented in Fig. 5 resemble the wear mechanisms described in [2]. These wear mechanisms include such abrasion controlled stages as: extrusion of cobalt binder, cracking of the carbide grains, WC grains breaking into small fragments and finally the small fragments gradual removal. The above wear mechanisms were investigated separately in stepwise abrasion tests and are described in []. The proposed simultaneous abrasion and fracture testing as well as the final experimental results presented and discussed in this paper clearly indicate that the method offers some potential advantages when it is used in hardmetals development programme to rank a large number of materials in term of wear and fracture resistance. For such the programme conventional evaluation methods (e.g. ISO 2962/ASTM B6, bulk fracture toughness methods) are less convenient as they require two different shapes of the specimens. There is a need for a reliable and cost effective ranking in terms of performance properties of the novel hard materials emerging as a result of the new innovative PM processing technique introduction. Over the last 0 years there has been a strong development of the novel approach leading to hardmetal microstructure with much higher resistance to fracture than normal, without sacrificing wear resistance [3-20]. One developed approach leads to the microstructure with cellular architecture with the ability to stop or delay the propagation of microcracks [3]. The interior of the cells has a low-carbon abrasion-resistant WC/Co composition, whilst the relatively thin walls are of high-cobalt, coarser, more fracture-resistant carbide. The above microstructure is an example of several functionally designed composite cemented carbide with distinctively

12 0 HM 83 S.F. Scieszka, J. Krajzel anisotropic properties [4]. To this category belong the established DC carbide, a socalled double cemented carbide in which pre-sintered granules of WC/Co hardmetal, of 60-30µm grain size, are embedded in a pure cobalt matrix. DC carbide can be described as a composite within a composite exhibits a superior combination of fracture toughness and high-stress wear resistance then conventional cemented carbide [5-7]. But perhaps the most pronounced trend occurring during the past years in the hardmetal industry has been a strong tendency towards finer and finer grained hardmetals [9-20]. An assumption can be made that further development in hardmetals technologies will take place as well as the need for their reliable evaluation will be made even more urgent. The method and the indicators of functional wear resistance and toughness presented above have the potential to meet the development s test demands. In the last stage of the investigation the attempt was made to evaluate the empirical relationship between volume loss as a result of edge chipping during the initial transition stage of abrasion wear and fracture toughness in the form of formulae. Material S (mm 2 ) K IC (MPa m ½ ) E (GPa) H (MPa) TSM TSM30BV TSM30BV Shm CW20C CW25C Table 5 Test results and materials properties The experimental data taken for the least square fit evaluation are presented in Table 5. The value of equivalent fracture surface S in mm 2 were calculated from W F (mean volume loss solely as a result of fracture edge chipping during the first tests) assuming that the crack surface produced during the test is flat and parallel to one side of the sample (prism), i.e. 2W 2 Fl S = sin60 where l is the length of the sample.

13 S.F. Scieszka, J. Krajzel HM 83 The calculations revealed that the best fit was obtained with formulae (9): K IC = 88,9 E H 2 2 S For this correlation the coefficient of determination R = 0,9749 was quite satisfactory. H 5 4 p Conclusions The proposed simultaneous abrasion and fracture resistance testing method and procedure offers potential advantages when used in hardmetals development programme to rank a large number of materials in terms of their above mentioned functional properties. The method enables the evaluation of abrasion and fracture resistance using only one apparatus, one shape of specimen and one testing procedure. The method is based on the finding that the wear transition stage, typical for the early and unsteady stage of the wearing process is controlled by brittle fracture while the following steady-state stage is controlled by abrasion process. Acknowledgements The authors gratefully acknowledge the financial support of this research by the European Commission, Marie Curie Action Proposal: MCF , Contract G5TR-CT The authors also express his sincere appreciation to Drs. Mark Gee, Roger Morrell, Brian Roebuck and Andrew Gant from National Physical Laboratory, Teddington for their kind assistance. Authors would also like to thank the NPL Engineering Services for building the tribotester and for supplying hardmetals and ceramics samples used in the study. References [] S.F. Scieszka, Simultaneous Abrasion and Brittle Fracture Testing by Identification of the Wear Mode of Hardmetal Specimens, NPL Report MCF , Teddington, [2] B. Roebuck, M. Gee, E.G. Bennett and R. Morrell, Mechanical Tests for Hardmetal, NPL Good Practice Guide No 20, Teddington, 2000.

14 2 HM 83 S.F. Scieszka, J. Krajzel [3] R. Morrell, Flexural Strength Testing of Ceramics and Hardmetals, NPL Good Practice Guide No 7, Teddington, 997. [4] J. Larsen-Basse, Wear of hardmetals in rock drilling, Powder Metallurgy, Vol 6,, (973). [5] S.F. Scieszka and K. Filipowicz, An integrated testing method for cermet abrasion resistance and fracture toughness evaluation, Wear, 26 (998) 202. [6] S.F. Scieszka, Wear transition as a means of fracture toughness evaluation of hardmetals, Tribology Letters,, 3-4 (200) [7] B. Roebuck, E. Bennett, L. Lay and R. Morrell, Palmqvist toughness for hard and brittle materials, NPL Good Practice Guide No 9, Teddington, 998. [8] R. Morrell and A.J. Gant, Edge chipping of hard materials, J. J. of Refractory Metals and Hard Materials, 9 (200) [9] S.F. Scieszka, Testing mechanical properties of materials for mining tools. Abrasive wear measurement methods a review. ZN Pol. Sl. Gliwice, 599 (2003) [0] S.F. Scieszka, Testing mechanical properties of materials for mining tools. Fracture toughness measurement methods a review. ZN Pol. Sl. Gliwice, 599 (2003) [] I.M. Hutchings, Tribology, Friction and Wear of Engineering Materials, Arnold, London (992) p 53. [2] J. Larsen-Basse, Binder extrusion in sliding wear of WC-Co alloys, Wear, 05 (985) [3] K. Brookes, Novel approaches lead to better wear resistance in hard materials, Metal Powder Report, October (2003) [4] K. Brookes, Functional design puts the bite into hard and refractory metals, Metal Powder Report, November (2003) [5] X. Deng, B. Patterson, K. Chawla, M. Koopman, Z. Fang, G. Lockwood and A. Griffo, Mechanical properties of a hybrid cemented carbide composite, I. J. of Refractory Metals & Hard Materials 9 (200) [6] Z. Fang, A. Griffo, G. Lockwood and J. Bitler, Properties of a high toughness cemented tungsten carbide composite, Conference Proceedings Advances in Powder Metallurgy and Particulate Materials, MpiE, New Jersey (999) 0, 23. [7] Z. Fang, A. Griffo, B. White, G. Lockwood, D. Belnap, G. Hilmas and J. Bitler J, Fracture resistant super hard materials and hardmetals composite with functionally designed microstructure, I. J. of Refractory Metals & Hard Materials, 9 (200)

15 S.F. Scieszka, J. Krajzel HM 83 3 [8] R. Toth and J. Keanne, Tough coats on hard powders a revolution in the making? Metal Powder Report, September (2003) [9] G. Gille, B. Szesny, K. Dreyer, van den Berg H, Schmidt J, Gestrich T and Leitner G, Submicron and ultrafine grained hardmetals for microdrills and metal cutting inserts, I. J. of Refractory Metals & Hard Materials, 20 (2002) [20] L. Prakash, Application of fine grained tungsten carbide based cemented carbides, I. J. of Refractory Metals & Hard Materials, 3 (995)