Material design method for the functionally graded cemented carbide tool

  • Published on

  • View

  • Download

Embed Size (px)


<ul><li><p>Material design method for the functionally graded cemented carbidetool</p><p>Toshio Nomura a,*, Hideki Moriguchi a, Keiichi Tsuda a, Kazutaka Isobe a,Akihiko Ikegaya a, Kiyoko Moriyama b</p><p>a Itami Research Laboratories, Sumitomo Electric Industries Ltd., 1-1-1, Koya-kita, Itami, Hyogo 664-0016, Japanb Information Systems Department, Sumitomo Electric Industries Ltd., 4-5-33, Kitahama, Chuo-ku, Osaka 541-0041, Japan</p><p>Abstract</p><p>The aim of this study is to apply the concept of functionally graded materials (FGMs) to tool materials and to develop high-</p><p>performance cutting tools. The requirement of the graded structure is that the surface is highly wear resistant cermet, and the inside</p><p>is tough cemented carbide. Compressive residual stress was introduced to the material surface by grading the composition. To</p><p>develop the new material, the cutting condition of broken cermet was investigated and their cutting temperature distribution was</p><p>measured by a newly developed measuring method. Then Computer Aided Engineering (CAE) analysis was performed to calculate</p><p>the generated thermal stress. The new material was developed with the aim to introduce the compressive residual stress over the</p><p>calculated thermal stress. As a result developed tools demonstrated higher wear resistance, breakage resistance, thermal crack</p><p>resistance and peeling resistance over those of conventional tools in the market. 2000 Elsevier Science Ltd. All rights reserved.</p><p>Keywords: Functionally graded cemented carbide; Cermet; Thermal crack; Cutting temperature distribution; Compressive residual stress</p><p>1. Introduction</p><p>Materials obtained by continuously grading theirconstituent elements (metals, ceramics, etc.) to matchthe usage conditions and by optimally controlling thecoecient of thermal expansion and coecient of ther-mal conductivity are known as functionally gradedmaterials (FGMs) [1,2]. These materials are attractingattention as new materials that provide functions sur-passing the characteristics specific to each element.</p><p>A typical FGM has the functionally graded heat-re-sistant properties on one side with the composition ofthe material providing high mechanical strength andthermal conductivity on the other side as shown inFig. 1. Active research and development is being con-ducted on such FGMs in various material and applicationfields, although there are few examples of practical ap-plication with manufactured products.</p><p>The aim of this study is to apply the concept of thefunctional design of FGMs to cemented carbide tools[3,4] and to develop high-performance cutting tools by</p><p>grading the composition using practical manufacturingprocesses [5,6].</p><p>Sintered hard materials used for cutting tools includeWCCo cemented carbide, TiCNNi cermet, and thin-film ceramic coatings of a maximum 10 lm thickness onthe surface of a cemented carbide (see Fig. 2). Whilecemented carbide oers excellent toughness, wear re-sistance is relatively poor. Conversely, cermet oersexcellent wear resistance, but toughness is poor. Coatedmaterials can provide both excellent toughness and wearresistance, but are high-cost because of the additionalprocess required to apply a ceramic coating on thesurface of the cemented carbide.</p><p>Ways of enhancing the characteristics have beenstudied in order to improve cutting performance bydividing the functions of wear resistance and toughnessbetween the materials surface and its interior by grad-ing the composition as shown in Fig. 3. The features ofthis material are described below.</p><p>1. A graded material of cemented carbide and cermet inwhich the highly wear resistant cermet structure islocated at the material surface, and the cementedcarbide structure providing excellent toughness islocated inside.</p><p>International Journal of Refractory Metals &amp; Hard Materials 17 (1999) 397404</p><p>* Corresponding author. Tel.: +81-0727-72-4805; fax: +81-0727-70-</p><p>6727.</p><p>0263-4368/99/$ - see front matter 2000 Elsevier Science Ltd. All rights reserved.PII: S 0 2 6 3 - 4 3 6 8 ( 9 9 ) 0 0 0 2 9 - 3</p></li><li><p>2. By grading the composition, compressive residualstress was introduced to the material surface.</p><p>3. The ability to manufacture this FGM by sinteringalone.</p><p>The additional function in (2) above is an importantaspect in material design. In general, hard materials areeasy to break by tensile stress as shown in Fig. 4(a), andthe occurrence of cracks and crack propagation leadingto breakage are aected by the magnitude of tensilestress acting on the material. As shown in Fig. 4(b),however, if high compressive residual stress can be im-parted to the material surface, it should also be possibleto make the material resistant to the occurrence ofcracks and crack propagation.</p><p>The introduction of compressive residual stress in thesurface by grading the composition can be achieved by</p><p>adjusting the linear expansion coecient of the surfaceso that it is smaller than that of the interior. This dif-ference in the coecients of linear expansion introducescompressive residual stress in the surface while it iscooling from the sintering temperature to room tem-perature. However, if the correct amount of compressiveresidual stress to be introduced has not been determined,material development must depend on trial and error.</p><p>This report describes the analysis method used todetermine the amount of compressive residual stress tobe introduced when designing a FGM, together with thematerial characteristics and cutting performance of thefunctionally graded cemented carbide (FGCC) tool de-veloped using this analysis method.</p><p>2. Material design</p><p>2.1. Tool damage analysis</p><p>This research was intended for the use of a cermetcomposition, which has superior wear resistance butinferior breakage resistance, on the surface of the FGM.Therefore the causes of cermet breakage were investi-gated first. Used tools were obtained from an auto-mobile manufacturer where cutting was performedalong the shape of the work piece as shown in Fig. 5,which is known to cause cermet to damage easily, andthe damage pattern was analyzed.</p><p>As the results in Fig. 6 show, fracture or chippingoccurred on approximately 50% of the used corners. Theresults also reveal that fracture started from the thermalcracks generated in the straight cutting edge as shown inFig. 7. These thermal cracks are assumed to propagatein back-facing from their originating point as indi-cated by the arrow in Fig. 5.</p><p>Fig. 2. Comparative characterization for hard metal tool materials.</p><p>Fig. 1. Graded structure of ceramics and metal.</p><p>398 T. Nomura et al. / International Journal of Refractory Metals &amp; Hard Materials 17 (1999) 397404</p></li><li><p>Based on this assumption, the occurrence of thermalcracks during a cutting test that simulated profiling wasconfirmed when changing from back-facing to outside-diameter cutting.</p><p>The mechanism whereby thermal cracks are gener-ated during the back-facing is shown in Fig. 8 sche-matically [7]. The back-facing caused an extremelylong, narrow area near the straight cutting edge to</p><p>Fig. 5. Example of profiling.</p><p>Fig. 6. Analysis result of damaged tool.</p><p>Fig. 7. Mechanism of thermal crack formation.</p><p>Fig. 8. Schematic diagram of measurement system.</p><p>Fig. 3. Schematic design of functionally graded cemented carbides.</p><p>Fig. 4. Reinforcing by residual compressive stresses.</p><p>T. Nomura et al. / International Journal of Refractory Metals &amp; Hard Materials 17 (1999) 397404 399</p></li><li><p>heat up. But since the straight cutting edge was notused by the outside diameter cutting immediately afterthe back-facing, only the straight cutting edge wasquickly cooled by the cutting oil. This resulted in alow-temperature area forming on the surface of thestraight cutting edge and a high-temperature areaforming inside of it. The resulting dierence in ther-mal shrinkage caused tensile thermal stress to form onthe cutting edge surface, which seemed to causethermal cracks perpendicular to the straight cuttingedge.</p><p>2.2. Measuring cutting temperature distribution</p><p>If the thermal stress generated in a cutting conditioncausing thermal cracking can be measured, then themagnitude of the compressive residual stress required onthe surface of a FGCC can be anticipated. In order tocalculate the thermal stress, a detailed temperature dis-tribution on the surface and inside of the tool needs tobe accurately measured. However, there was nomeasurement or simulation method to enable this.Therefore we developed a new cutting temperaturemeasurement method [8] using an infrared thermal videosystem. By combining this measurement method withComputer Aided Engineering (CAE) analysis methodusing the finite element method (FEM), we attempted todetermine the temperature distribution in the tool and toquantify the generated thermal stress.</p><p>The principle of the developed cutting temperaturedistribution measurement method is shown in Fig. 9.This measurement method has two major distinguishingfeatures.</p><p>1. In order to expose the rake face of the tool that isnormally covered with chips and cannot be observed,a portion of the work being cut is notched.</p><p>2. A thermal image of the tools exposed rake face is ob-tained by using an infrared thermal video system andused to obtain the temperature distribution of therake face through computer processing.</p><p>The measurement results of the cutting temperaturedistribution during back-facing and following outside-diameter cutting using this measurement method areshown in Fig. 10. They reveal that the maximum cuttingtemperature in back-facing is approximately 1200C,and heat is generated in an extremely long, narrow areanear the straight cutting edge. Furthermore, they alsoindicate that maximum cutting temperatures in outside-diameter cutting reach approximately 1100C, althoughthe generated heat concentrates near the nose radius andthe temperature of the straight cutting edge is onlyabout 400C.</p><p>2.3. Thermal stress analysis</p><p>Based on the temperature distribution data of therake face of the tool obtained in Fig. 10, two assump-tions were made when analyzing the thermal stressgenerated after back-facing:</p><p>1. The highest temperature occurs where the tool comesinto contact with chips due to heat generated by fric-tion between the two, which in turn raises the tem-perature inside the tool.</p><p>Fig. 9. Cutting temperature distribution.</p><p>Fig. 10. Thermal stress distribution.</p><p>400 T. Nomura et al. / International Journal of Refractory Metals &amp; Hard Materials 17 (1999) 397404</p></li><li><p>2. The amount of heat generated by friction is unaect-ed by the use of cutting oil.</p><p>By employing the tool used for the temperature dis-tribution measurement, the position and area size onthe rake face in contact with the chips in back-facingand outside-diameter cutting was measured first. Next,using the temperature data obtained for the area incontact with the chips from the temperature distribu-tion data in Fig. 10 as the boundary condition, thechanges in the temperature distribution within the toolwere calculated using MARC k6.1 finite element anal-ysis software, which supports transient analysis. Thethermal stress generated by this temperature distribu-tion change was then analyzed. The physical constantsused to perform the analysis were density: 7 g/cm3;Youngs modulus: 430 MPa; Poisson ratio: 0.24; heattransfer coecient between the tool and the holder:7:5 105 W/m2 K; and heat transfer coecient betweenthe tool and the cutting oil: 3:9 105 W/m2 K. Otherheat transfer, specific heat and thermal expansion co-ecients are shown in Table 1.</p><p>As expected, the results of the analysis showed thatthe greatest tensile stress was generated when changingfrom back-facing to outside-diameter cutting. Thethermal stress distribution generated at that time isshown in Fig. 11. The greatest tensile stress was gener-ated in the straight cutting edge, which agreed with thelocation of actual thermal cracks. These analysis resultsagreed with actual tool damage qualitatively. Next, inthe profiling shown in Fig. 10, the cutting conditionsduring back-facing were varied to determine the cuttingconditions that caused thermal cracks in conventionalcermet.</p><p>Using the above analysis method, the thermal stressin that cutting condition was calculated to be approxi-mately 0.8 GPa. Furthermore, the thermal stress underseverer conditions shown in Fig. 10, aiming for nobreakage of the developed material was calculated to beapproximately 1.3 GPa in a similar way. From theabove analysis results we engaged to develop the ma-terial imparting a compressive stress exceeding 0.5 GPa( 1:3 0:8 GPa) to the material surface in order toprevent the cutting tools from breaking under the targetcutting conditions.</p><p>3. Evaluation test details</p><p>TiCN, WC, Co and Ni powders were combined to amixture of TiCN, 40% WC, 10% Co and 5% Ni byweight, wet-mixed in a ball mill for 24 h, and then pressmolded to 10 10 5 mm3 samples at a pressure of98 MPa. The samples were heated in a vacuum forsintering to 1673 K, at which the liquid phase of thematerial appeared and held at that temperature for onehour under a nitrogen atmosphere. The samples werethen cooled at a rate of a, b, c (a &lt; b &lt; c) to makegraded structures.</p><p>To evaluate the structure of the sintered alloy, a crosssection of the alloy was polished with a #200 GCgrindstone, and then mirror-polished using a #3000-diamond grid and observed using optical microscopy,SEM analysis was also used. Compressive residual stressin the sintered alloy surface was measured by the X-raysin2 / method [9]. The samples were lapped with a#3000 diamond grid from the surface to the inside ofthe sintered alloy, stress was measured at variousdepths, and the stress distribution was determined fromthe surface to the inside of the sintered alloy.</p><p>Table 1</p><p>Thermal characteristics used for calculation</p><p>Temperature (K)</p><p>273 473 673 873 1073 1273 1473 1773</p><p>Thermal conductivity (W/m K) 34.3 28.5 23.9 20.9 18.0 14.7 13.8 11.7</p><p>Specific heat (cal/kg K) 126 153 166 174 180 185 188 194</p><p>Coecient of thermal expansion (106/K) 7.72 7.81 7.98 8.20 8.49 8.89 9.31 10.0</p><p>Fig. 11. Compressive stress distribution and microstructure of func-</p><p>tionally graded cement carbide.</p><p>T. Nomura et al. / International Journal of Refractory Metals &amp; Hard Materials 17 (1999) 397404 401</p></li><li><p>To evaluate performance when used as a cutting tool,a CNMG120408 indexable insert was manufacturedusing the same material composition and cutting testswere performed.</p><p>4. Results and discussion</p><p>4.1. Characteristics of FGCC</p><p>The changes in the material characteristics for vary-ing cooling rates as indicated by a, b and c (wherea &lt; b &lt; c) have already been reported in detail [5].Table 2 therefore only lists the measurement results ofthe surface compressive residual stress imparted on thesurface of the sintered body fabricated by grading thecomposition. This table indicates that when the coolingrate is a, a FGCC that introduces a compressive residualstress onto the material surface exceeding 0.5 GPa,which was established as the target using CAE stressanalysis as described above, can be fabricated. Thecompressive residual stress from the surface to the inside...</p></li></ul>