EXPERIMENTAL OBSERVATIONS OF TOOL WEAR IN MULTILAYER COATED INSERTS

Jorge A. Olortegui-Yume and Patrick Y. Kwon

Department of Mechanical Engineering, Michigan State University, East Lansing, MI ABSTRACT This paper deals with understanding the evolution of tool wear of multilayer coated inserts using Confocal Laser Scanning Microscopy (CLSM). Steady-state turning experiments were carried out on 1045 steel bars with commercially available multi-layer coated inserts consisting of TiN /Al2O3 /TiCN deposited on a C6 carbide substrate provided by Kennametal, Inc.. Topographical images of the crater wear as well as orthogonal sections of the flank and crater wears as functions of machining time are obtained. A humped island of TiN coating material next to a growing crater of Al2O3 and steel traces were found. The maximum crater depth value and location respect to the edges of the insert are gathered for all machining times. The maximum crater depth location was observed to move across the rake face as machining time increased. These unexpected features of crater wear are suspected to have their origin in the competition of the abrasive and dissolution mechanisms, and the difference in wear-resistant properties between the TiN and Al2O3 coatings. Validations of the confocal results are also performed using the Scanning Electron Microscope (SEM) and the Atomic force microscope (AFM). and time-consuming.[3,4] .Traditional methods of surface measurement have also involved stylus profilometers. However, this method has a number of disadvantages such as the limitations on the detail they can resolve, the slow scanning rate, and the potential damage to the surface investigated. Also, severe difficulties arise when 3D information of surfaces is required [4] The main advantages of Confocal laser scanning microscopy (CLSM) include the fast and convenient means of obtaining surface features and the determination and analysis of the true surface topography. These capabilities provide the unique characteristics of the CLSM such as the versatile depth of field, the minimal sample preparation, the adequate image quality for engineering purposes, and the savings in costs relative to SEM technology [4]. Even though CLSM has been established as a powerful tool in biological and medical sciences since the early 80s, only in the past few years some comparative studies on the performance of traditional methods versus CLSM have been carried out in wear measurement, tribology, fractography, and biomedical sciences [3, 4, 5, 6]. A superior performance of CLSM surface profiles compared to those of a profilometer was always found. Further, CLSM measurements were in good agreement with Atomic Force Microscope (AFM) values.

1. INTRODUCTION
Up to date many efforts have been made to understand the nature of tool wear, however, it continues to have unsolved problems. Solutions to these problems would bring a great economical impact in machining industry. However, a thorough understanding of the micro-mechanisms involved in tool wear and a meaningful method to measure it must precede such solutions. So far, the traditional methods used to measure tool wear have included reflected light interference microscopy, scanning electron microscopy, and stereo microscopy [1, 2]. However, due to the limitation on the depth of field, the difficulty while obtaining and interpreting images and the high cost involved, these methods have proven to be burdensome

2. BACKGROUND
The confocal principle The power of the CLSM resides in its ability to get an image only from the light in the focal plane. This means that all the out-of-focus light is blocked from the final image allowing a sharper image compared to the conventional light microscope. This feature is achieved by positioning a pinhole in the light path. Light that is exclusively coming from the focal plane is focused at a point right in the position of the pinhole (Figure 1). The pinhole is positioned so that it passes light from the focal plane to a detector. Light rays originally coming from outof-focus regions have a focal point either in front

or behind the pinhole, therefore, few of these rays make their way through the pinhole. As a consequence, the image is formed mostly by rays coming from the region of the sample located in the focal plane. This process is the main concept of optical section. An optical section can be better understood as an electronical slice of the sample coinciding with
Light Source
(LASER)

the focal plane level. The advantage is that an actual sectioning is not necessary to perform, the sample remains intact and repetition of the process can be done. Once light of the region not in focus has been discarded in the final image, resolution and contrast are drastically improved respect to the conventional light microscopes.

MIRROR DICHROIC (Beam splitter) COLLECTING LENS

PHOTODETECTOR (PHOTOMULTIPLIE

PMT

OBJECTIVE

FOCAL PLANE or OPTICAL

FOCUS

DETECTOR PINHOLE (Confocal

SCREEN DISPLAY

IN-FOCUS LIGHT RAYS OUT-OF-FOCUS LIGHT RAYS

SPECIME

S T A G E Out-of focus region Figure 1 The confocal principle The CLSM can perform operations such as the gathering of a series of optical slices called a zstack which could be overlaid to give sharp 2D image. Also, the 3D reconstruction feature allocates for the acquisition of a height value (z) that identifies the surface position and stores it a matrix , z(x,y), that can be manilpulated to obtain different types of rendering including surface topography, isolines (contour lines), and roughness parameters .

3 EXPERIMENTAL PROCEDURE
3.1 Machining tests Dry turning tests were carried out in 8033 Clausing/Colchester lathe with the control for infinitely variable speed. The experiments are carried out at the constant feed of 0.3175 mm/ rev, depth of cut of 1.905 mm and constant cutting speed of 250 m/min (820 sfpm). These conditions are used based on the following criteria: the range of optimal speeds provided by the insert manufacturers, the need for obtaining

as much measurable wear as possible in the shortest time, and the torque limits of the lathe while obtaining minimal chattering, good surface finishing and optimal area size for imaging. After few trial runs the approach angle was set to 5 degrees and the toolholder shank was machined to provide a back rake angle of 2o. These angles set a compromise between minimal chattering and proximity to orthogonal cutting conditions. Machining times were interrupted at: 60 s, 120 s, 180 s, 240 s, 300 s, 480 s and 720 s. in order to obtain a cutting edge at each of these times. Thus, allowing for the wear evolution to be observed as function of time. Steel rounded bars of 1045 steel, 6 diameter and 18 length, were used as work material. 3.2 Inserts and Work material The square inserts are provided by Kennametal, Inc., Latrobe, PA , having a SNMA 190612 ISO designation The multi-layer insert is designated as a KC9315 grade insert, which corresponded to a top layer made of TiN, an intermediate layer of Al2O3, and a bottom layer of TiCN deposited

on a C6 cemented carbide substrate. TiN thickness is found to be 2.5 m, Al2O3 7.1 m, and TiCN 7.8 m. 3.3 Confocal microscopy data acquisition A Zeiss LSM 210 confocal microscope is used in reflection mode to obtain 2D images and 3D data sets z(x,y) matrices of all the corners with the different machining times. Then a Zeisss LSM 310 software was used to manipulate the z(x,y) matrices for measurement , analysis and topographical rendering. Finally Adobe photoshop 7.0 is employed to post-processs the images and give the final pictures. 3.4.SEM data acquisition A JEOL JSM-6400 Scanning microscope is used to obtain BSE images and X-ray spectroscopy chemical analysis are performed in all the corners to identify the evolution of the wear associated to the limits of the two first layers (TiN and Al2O3). In addition, four SE (secondary electrons) images are merged to validate the shape of the crater for the corner with 300s machining time. 3.5 AFM data acquisition This is the main validation for the depth values obtained with the confocal microscope. Nanoscope III Scanning probe microscope manufactured by Digital Instruments Inc. was used to perform this part. 3. RESULTS AND DISCUSSIONS The confocal pictures reveal an unexpected hump starting approximately at 190 m from the edge, a growing crater next to the hump and a thin depression that run along with the edge-side limit of the crater (Figures 2 (a) and (b)). The BSE images and X-ray spectroscopy showed the hump composed of TiN (bright zone), the growing crater composed of Al2O3 (black zone) and some traces of steel (very bright zones). It turned out that the thin depression was coinciding with a strip of Al2O3 running along with the edge limit of the crater (Figure 3). In addition, the maximum crater depth locations and its values are obtained out of the z(x,y) confocal matrices (Figure 4). An increasing trend of the maximum crater depth with machining time and constant tendency between 300 and 720s is observed (Figure 5). This location moved across the tool face as the machining process continues (Figure 6), suggesting that the peak temperature location, which occurs in the vicinity of maximum crater depth, is also moving [7]. It is believed that this

behavior is the result of the competition between the dissolution and diffusion mechanisms, which are greatly influenced by the temperature [8]. Regarding flank wear, phi-z sections are obtained in the locations correspondent to the maximum crater depth for all machining times. Although gradual flank wear is observed, no clear trend is observed when measuring the slopes in different parts of the flank profiles. The secondary electrons image corroborated the shape of the worn tool face obtained in the topography view of the confocal pictures. A good agreement with the AFM depth values was found during the validation.
Inland side limit of the crater

2D VIEW

Edge side limit of the crater

700m 190 m Trailing edge

HUMP

Edge side limit of the crater Inland side limit of the crater Depression

Trailing edge

33D TOPOGRAPHICAL IMAGE (Left view orientation)

Figure 2 (a) Confocal 2D image of crater wear , (b) Confocal 3D image of crater wear

4. CONCLUSIONS
The usefulness of confocal microscopy to obtain reliable surface profiles and topography was confirmed. SEM and AFM microscopes aid was relevant as validation tools for the CLSM information. Also, the optical properties of steel are found to have a distorting influence in the results of the topography rendering. The particular behavior of coated tools wear was also verify [9, 10] and in particular the wear of

multi-layered coated has unexpected features in the tool face and maximum crater depth behavior. However, extended work and analysis are currently being carried out to interpret these results in terms of wear mechanisms.

Al2O3 exposed in the crater TiN hump

Steel attached to the TiN coating

Al2 O3

Figure 3 BSE image of crater wear

[5] Hanlon, D.N., Todd, I., Peekstok, E., Rainforth, W.M., Van der Zwaag, S., The application of laser scanning confocal microscopy to tribological research, Wear 251, pp. 1159-1168, 2001 [6] Wessel, S., Pagel, S., Ritter, M., Hohenberg, H., and Wepf, R., Topographic measurements of real structures in reflection confocal laser scanning microscope (CLSM), Microscope Microanalysis, 9, suppl. 3, p. 162, 2003. [7] Childs,T.H.C., Maekawa , K. ,Obikawa, T.,and Yamane, Y., Metal Machining, Arnold Publishers, pp.76-132, [8] Kim, W.S., A new methodology for predictive tool wear, PhD dissertation, Michigan State University, p. 15, 2000 [9] Chubb, J.P. and Billingham, J., Coated cutting tools- A study of wear mechanisms in high speed machining, Wear, 61, pp. 283-293 [10] Kountanya, R.K., Predictive Tool wear of coated tools systems, M.S. Thesis, Michigan State University, 1998
Maxim um depth location as a function of m achining tim e
1400 1200 1000 800 600 400 X Y

Figure 4 Maximum crater depth location REFERENCES

[1] Devillez, A. , Lesko, S. and Mozer, W., Cutting tool crater wear measurement with white-light interferometry, Wear, 256, pp. 56 65, 2004 [2] Dawson, T.J. and Kurfess, T.I., Wear trends of PCBN cutting tools in hard turning, http://www.hardinge.com/hardium/PDF/Dawso n_NAMRC_2002.pdf [3] Gee, M.G. and McCormick, N.J., The application of confocal scanning microscopy to the examination of ceramic wear surfaces, Journal of Applied Physics, 25 , A230-A235, 1992 [4] Anamalay, R.V., Kirk, T.B. and Panzera, D., Numerical descriptors for the analysis of wear surfaces using laser scanning confocal microscopy, Wear, 181-183, pp. 771-776, 1995

200 0 0 200 400 600 800

M a c h i n i n g t i m e ( se c )

Figure 5 Maximum depth locations

Maximum crater depth as a function of machining time
Maximum crater depth (microns) 2.5 2 1.5 1 0.5 0 0 200 400 600 800 Machining time (sec)

DEPTH

Figure 6 Evolution of crater depth