Diamond Tool Wear - Observation by Scanning Electron Microscopy (SEM)

Shai N. Shafrir ?? ?????

Abstract

In Deterministic Microgrinding (DMG), the wear of diamond grains is a significant problem and the cause for lowering grinding efficiency, which leads to “pull-out” (loose) of diamond grains from the binder and the appearance of scratching. It contributes to poor surface finish, increased machining forces, and increased tooling costs. At the Center for Optics Manufacturing (COM), we have been primarily interested in resin and bronze bonds diamond tools with a diamond size of ~2-4 microns. Unfortunately it is difficult to directly measure fine abrasives mechanical response. A qualitative model was introduced using Scanning Electron Microscopy (SEM) to better analyze diamond wear for different bond matrixes. Observation of the tool face after a variety of grinding and dressing operations provided characteristically different surfaces for each case.

Introduction

Tools are constructed of a wearable matrix embedded with diamond particles and can vary in diamond size, abrasive concentration, and bond material. Ultimately the matrix ability to wear can have a large impact on the tool ability to grind the material, resulting in tool failure when it is not wearing properly. Excelerated tool wear will result in a dull tool surface, and will subsecwently loss it's ability to remove work material. A better understanding of tool wear mechanism is needed in order to find an optimum condition between these two extremes, by maintaining proper grinding conditions to eliminate tool failure.

The tool wear process is often broken into three modes of removal: 1) Diamond blunting, 2) Diamond pullout, and 3) Bond wear [1].

Fig 2. Bronze tool after 6 cycles, notice the hole in the middle due to diamond pullout.
Image using the BSE detector, gold coating on surface.

Diamond blunting can occur during the grinding process and results in higher diamond forces due to the increasing contact area since both tangential and normal force components increase linearly with wear flat area [2]. These increased forces may lead to diamond pullout once they exceed the bond strength [3]. It should also be noted that tool wear is measured by the total volume to tool removed, and that bond wear clearly has the greatest impact on this value since diamond blunting and diamond pullout provide minimal volumetric change[4].

Bond materials include bronze and resin, are typically softer than the abrasives and therefore amy by susceptible to wear by erosion as swarf (glass chips) impact the diamond over time.

Fig 3. Schematics of the tool bond matrix [5]

The bond purpose is to hold diamonds and wear in a controlled way. There are three types of bonds for optics manufacturing [6]:

Metal – “Bronze bonds are in widespread used in optical shops. They do well on optical materials, especially when used with light grinding pressures. Bronze bonded diamond tools are the preferred choice when hard an brittle materials must be ground with a minimum of edge chipping” [7]

Vitrified – “Glass” matrix, often porous hard, but may be relatively “friable” (excessive backing under pressure). Pores may aid coolant penetration, and the “storage” of swarf (removed glass particles).

Resin – “Polymer” matrix, relatively soft, less wear resistance and temperature resistance. Smoother grind, since bond deform s under loading on of diamonds.

Fig 4. Bronze bond, notice the diamonds in the bond material.
Image using the BSE detector, gold coating on surface.

Back to top

Experimental Procedure

We suggest to investigate a comparison between two different diamond ring tools, used mainly on a Contour Deterministic Micro-Grind (CDMG) precision optics Moore Nanotech® 150AG [8] Computer Numerically Controlled (CNC) platform.

Fig 5. Moore Nanotech® 150AG at work with a resin tool.

Fig 6. Moore Nanotech® 150AG platform.

Observation of the tools face will take place after a variety of grinding and dressing operation on BK7 Crown glass. The tools of interest: resin and bronze bonded tools (fine tools, diamond size ~2-4 microns).

Microscopic Techniques

1. Secondary Electron Imaging (SEI): mix of two secondary electrons detectors are being used for the resin tool; in-lens and the in-chamber detectors.

In-lens SE detector is good for high-resolution topographic imaging. In general the topographical image is dependent on how many of the secondary electrons actually reach the detector.

In-chamber SE detector is good for most general purposes imagine.

2. Backscatter Electron Imaging (BSE): This method is useful when there is a need to distinguish between materials having a different atomic number (visible contrast defines the difference), and was implemented during the study on the bronze tool.

3. EDAX's Digital Imaging and X-ray mapping.

4. Coating Sample Surface: Sputtering gold @ 45-50 millitor for 30 sec, coating height is approximately 100(Angstrom) for a smooth surface. The bronze tool, while being the best sample due to its conductivity, needed to be coated because of diamonds charge up.

*the tool are very large compare to normal SEM samples, therefore mounting then required special consideration.

Non-Microscopic Techniques

1. Dressing and grinding of both tools.

2. Coloring of images (Pseudo color), looking at micrographs as art.

Back to top

Tools After Dressing

	Fig 7. Resin tool No diamonds are noticeable after dressing. Fig 8. Bronze tool One diamond is located in the center of the image.
		Fig 9. Bronze tool High magnification show cracking associated with the dressing material.
Both tools were dressed and coated with gold for 30 seconds, in 45-50 millitorr vacuumed sputter machine. surface roughness would suggested more than one run of coating, but as seen from the images, one run of coating was sufficient. All images were taking with 10kV beam and 2nd aperture, using the a mix between in-lens and in-chamber SE detectors. The dressing material behavior on the surface was unpredictable, showing non-uniform cover on the surface instead of exposing the diamond in the tools bond. Back to top

Tools After 3 Cycles

	Fig 10. Resin tool Tool surface reveils no diamond sticking out from the bond material. Fig 11. Resin tool Diamond flash with the surface.
		Fig 12. Bronze tool Diamonds in the bond "sticking out".
Both tools after 3 grinding cycles of BK7 crown glass. Coated with gold for 30 seconds, in 45-50 millitorr vacuumed sputter machine. Resin tool images were taking with 5kV beam and 2nd aperture in Fig 10 and the 1st aperture in Fig 11 (notice the different working distance), using the a mix between in-lens and in-chamber SE detectors. Bronze tool image used 10kV beam and 3nd aperture, using the BSE detector which allowed a better identification of the diamonds from the bond. Back to top

Tools After 6 Cycles

	Fig 13. Bronze tool Low magnification revels that there is are approximately 5 to10 diamonds in an area of 400um^2. Fig 14. Bronze tool Diamonds seem to be very much exposed at this stage.
		Fig 15. Bronze tool Large diamonds in size ~4um have more tendency to pullout from the bond material.
Bronze tool after 6 grinding cycles of BK7 crown glass. Coated with gold for 30 seconds, in 50 millitorr vacuumed sputter machine. Tool images were taking with 10kV beam and 3nd aperture, using the BSE detector which allowed a better identification of the diamonds from the bond. The stage was in a small angle which allowed a better appreciation of diamond size and shape. The resin tool was left out from this stage onwards due to technical difficulties. Back to top

Tools After 8 Cycles

	Fig 16. Bronze tool Low magnification image show the average number of diamond on a specific area within the work-zone of the tool remain almost unchanged.   Fig 17. Bronze tool High magnification revel the phenomena of diamond blunting, and a more uniform bond surface.   Fig 18. Bronze tool Blunt diamonds in the bronze bond material. Fig 19. Bronze tool Blunt diamonds, visible tracks in the bond material.
Bronze tool after 8 grinding cycles of BK7 crown glass. Coated with gold for 30 seconds, in 50 millitorr vacuumed sputter machine. Tool images 16-18 were taking with 10kV beam and 3nd aperture, using the BSE detector which allowed a better identification of the diamonds from the bond. Fig 19 used a mix between in-lens and in-chamber SE detectors The stage was in a small angle which allowed a better appreciation of diamond size and shape.   Back to top

Pseudo Colors

	Fig 19. Bronze tool after 3 cycles Fig 20. Bronze tool after 8 cycles
		Fig 21. Bronze tool after 8 cycles   These images were taking using a BSE detector and colored for a more "alive" version than the normal black and white (More...). Back to top

Summary of Observation

Observation of the tool face after a variety of grinding and dressing operations provided characteristically different surfaces for each case. Dressing operation were conducted with a 2-4 mm bronze and resin tools, and 1500 grit dressing stick. Differences can be clearly observed from the dressed tools images, showing low (or none) diamond protrusion, with the diamonds appearing almost flush with the matrix surface, similar to SEM observation results discussed by Gough et al.

Grinding tests images show that the matrix material appears smoother with more tracks evident on the surface as grinding cycles increased. Diamonds are sticking out more from the surface with a noticeable sharper form after 6 cycles compare to 3 cycles test. 8 cycles test created a tool surface with moderate protrusion heights but significant diamond blunting, producing flat tops on the diamonds. The matrix also displays a smoother texture likely due to smaller particles impacting the surface [5].

Fig 22. Work zone tracks visible after 8 working cycles.
In-lense and In-chamber detectors mix signals.

Conclusions

As experimental data show, a blunting of the diamond grains occurs during grinding with the formation of very pronounced wear areas on some of them. The blunting of the diamond grains also leads, in particular, to a lowering of the grinding efficiency if it is not accompanied by the simultaneous removal of the blunted grains and their replacement by new ones. As a result of binder wearing the tool loses it effectiveness. Consequently, the dependence of grinding efficiency on the amount of wear of the diamond grains must be taken into consideration both in the construction as well as in the development of techniques for using the tool.

Reference

[1] A. Broese van Groenou, "Some Results on the Wear of a Bronze-Bonded Grinding Wheel", Science of Ceramic Machining and Surface Finishing II (1979).

[2] Miller, Michele and Handigund, Prasamm, "Abrasive Wear and Forces in Grinding of Silicon Carbide", Michigan Technological University (2000).

[3] Miler, Michele and Kakumanu, Thrinath, "Wear and Self-Dressing of Resin Bound Grinding Wheels", Michigan Technological University (1999).

[4] D. Gough, “The Use of Tool Probing for Process Optimization and Wear Analysis in Deterministic Microgrinding”, Master Thesis, University of Rochester (2003).

[5] J. S. Taylor, LLNL

[6] P. Funkenbusch, COM Summer School (2001).

[7] H. Karow, “Fabrication Methods for Precision Optics”, Wiley Series in Applied Optics, p. 309 (1993).

[8] Moore Nanotechnology Systems, LLC, 426A Winchester st., PO Box 605, Keen, NH 03431-0605 USA.

Acknowledgments

I would like to acknowledge the following people contributing there valuable time and experience to the completeness of this study - Sha Tong, Edward Fess, Enbal Shafrir, Zami Naim and Brian McIntyre.

Please enter any comments, criticisms, questions, etc.
below.

Your name:

Email address: